Fix tag delete

Fix relative stats URL

.drone.yml

@@ -0,0 +1,56 @@
+kind: pipeline
+type: docker
+name: amd64
+
+platform:
+  os: linux
+  arch: amd64
+
+steps:
+  - name: build
+    image: simon987/ubuntu_ci
+    commands:
+      - ./ci/build.sh
+  - name: scp files
+    image: appleboy/drone-scp
+    settings:
+      host:
+        from_secret: SSH_HOST
+      port:
+        from_secret: SSH_PORT
+      user:
+        from_secret: SSH_USER
+      key:
+        from_secret: SSH_KEY
+      target: /files/sist2/${DRONE_REPO_OWNER}_${DRONE_REPO_NAME}/${DRONE_BRANCH}_${DRONE_BUILD_NUMBER}_${DRONE_COMMIT}/
+      source:
+        - ./sist2-x64-linux
+        - ./sist2-x64-linux-debug.tar.gz
+
+---
+kind: pipeline
+type: docker
+name: arm64
+
+platform:
+  arch: arm64
+
+steps:
+  - name: build
+    image: simon987/ubuntu_ci_arm
+    commands:
+      - ./ci/build_arm64.sh
+  - name: scp files
+    image: appleboy/drone-scp
+    settings:
+      host:
+        from_secret: SSH_HOST
+      port:
+        from_secret: SSH_PORT
+      user:
+        from_secret: SSH_USER
+      key:
+        from_secret: SSH_KEY
+      target: /files/sist2/${DRONE_REPO_OWNER}_${DRONE_REPO_NAME}/${DRONE_BRANCH}_${DRONE_BUILD_NUMBER}_${DRONE_COMMIT}/
+      source:
+        - ./sist2-arm64-linux

.github/ISSUE_TEMPLATE/issue-template.md

@@ -0,0 +1,18 @@
+---
+name: Issue template
+about: General
+title: ''
+labels: ''
+assignees: ''
+
+---
+
+sist2 version:
+
+Platform (Linux or Docker, x86-64 or arm64):
+
+Elasticsearch version:
+
+Command with arguments:  `ex: "scan ~/Documents -o ./i2 --threads 3 -q 1.0`
+
+If the issue is related to the `scan` module, please attach the files necessary to reproduce the error or email them to me[at]simon987.net.

.gitignore

@@ -1,6 +1,5 @@
 .idea
 thumbs
-test
 *.cbp
 CMakeCache.txt
 CMakeFiles
@@ -16,3 +15,4 @@ bundle.js
 *.a
 vgcore.*
 build/
+third-party/

.gitmodules

@@ -1,45 +1,6 @@
-[submodule "argparse"]
-	path = argparse
+[submodule "third-party/libscan"]
+	path = third-party/libscan
+	url = https://github.com/simon987/libscan
+[submodule "third-party/argparse"]
+	path = third-party/argparse
 	url = https://github.com/cofyc/argparse
-[submodule "cJSON"]
-	path = cJSON
-	url = https://github.com/DaveGamble/cJSON
-[submodule "lmdb"]
-	path = lmdb
-	url = https://github.com/LMDB/lmdb
-[submodule "utf8.h"]
-	path = utf8.h
-	url = https://github.com/sheredom/utf8.h
-[submodule "lib/bzip2-1.0.6"]
-	path = lib/bzip2-1.0.6
-	url = https://github.com/enthought/bzip2-1.0.6
-[submodule "lib/libmagic"]
-	path = lib/libmagic
-	url = https://github.com/threatstack/libmagic
-[submodule "lib/harfbuzz"]
-	path = lib/harfbuzz
-	url = https://github.com/harfbuzz/harfbuzz
-[submodule "lib/openjpeg"]
-	path = lib/openjpeg
-	url = https://github.com/uclouvain/openjpeg
-[submodule "lib/ffmpeg"]
-	path = lib/ffmpeg
-	url = https://git.ffmpeg.org/ffmpeg.git
-[submodule "lib/onion"]
-	path = lib/onion
-	url = https://github.com/davidmoreno/onion
-[submodule "lib/mupdf"]
-	path = lib/mupdf
-	url = git://git.ghostscript.com/mupdf.git
-[submodule "lib/tesseract"]
-	path = lib/tesseract
-	url = https://github.com/tesseract-ocr/tesseract
-[submodule "lib/leptonica"]
-	path = lib/leptonica
-	url = https://github.com/danbloomberg/leptonica
-[submodule "lib/libtiff"]
-	path = lib/libtiff
-	url = https://gitlab.com/libtiff/libtiff
-[submodule "lib/libpng"]
-	path = lib/libpng
-	url = https://github.com/glennrp/libpng

.teamcity/settings.kts

@@ -1,69 +0,0 @@
-import jetbrains.buildServer.configs.kotlin.v2019_2.*
-import jetbrains.buildServer.configs.kotlin.v2019_2.buildSteps.ExecBuildStep
-import jetbrains.buildServer.configs.kotlin.v2019_2.buildSteps.exec
-import jetbrains.buildServer.configs.kotlin.v2019_2.triggers.vcs
-import jetbrains.buildServer.configs.kotlin.v2019_2.vcs.GitVcsRoot
-
-/*
-The settings script is an entry point for defining a TeamCity
-project hierarchy. The script should contain a single call to the
-project() function with a Project instance or an init function as
-an argument.
-
-VcsRoots, BuildTypes, Templates, and subprojects can be
-registered inside the project using the vcsRoot(), buildType(),
-template(), and subProject() methods respectively.
-
-To debug settings scripts in command-line, run the
-
-    mvnDebug org.jetbrains.teamcity:teamcity-configs-maven-plugin:generate
-
-command and attach your debugger to the port 8000.
-
-To debug in IntelliJ Idea, open the 'Maven Projects' tool window (View
--> Tool Windows -> Maven Projects), find the generate task node
-(Plugins -> teamcity-configs -> teamcity-configs:generate), the
-'Debug' option is available in the context menu for the task.
-*/
-
-version = "2019.2"
-
-project {
-
-    vcsRoot(HttpsGithubComSimon987sist2refsHeadsMaster)
-
-    buildType(Build)
-}
-
-object Build : BuildType({
-    name = "Build"
-
-    artifactRules = """
-        sist2
-        sist2_scan
-    """.trimIndent()
-
-    vcs {
-        root(HttpsGithubComSimon987sist2refsHeadsMaster)
-    }
-
-    steps {
-        exec {
-            name = "Build"
-            path = "./ci/build.sh"
-            dockerImage = "simon987/general_ci"
-            dockerImagePlatform = ExecBuildStep.ImagePlatform.Linux
-            dockerPull = true
-        }
-    }
-
-    triggers {
-        vcs {
-        }
-    }
-})
-
-object HttpsGithubComSimon987sist2refsHeadsMaster : GitVcsRoot({
-    name = "https://github.com/simon987/sist2#refs/heads/master"
-    url = "https://github.com/simon987/sist2"
-})

CMakeLists.txt

@@ -2,85 +2,55 @@ cmake_minimum_required(VERSION 3.7)
 set(CMAKE_C_STANDARD 11)
 
 project(sist2 C)
-list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_LIST_DIR}/CMakeModules")
 
 option(SIST_DEBUG "Build a debug executable" on)
 
-add_executable(
-        sist2
+set(BUILD_TESTS off)
+add_subdirectory(third-party/libscan)
+set(ARGPARSE_SHARED off)
+add_subdirectory(third-party/argparse)
+
+add_executable(sist2
+
+        # argparse
+        third-party/argparse/argparse.h third-party/argparse/argparse.c
+
         src/main.c
         src/sist.h
         src/io/walk.h src/io/walk.c
-        src/parsing/media.h src/parsing/media.c
-        src/parsing/pdf.h src/parsing/pdf.c
         src/io/store.h src/io/store.c
         src/tpool.h src/tpool.c
         src/parsing/parse.h src/parsing/parse.c
         src/io/serialize.h src/io/serialize.c
         src/parsing/mime.h src/parsing/mime.c src/parsing/mime_generated.c
-        src/parsing/text.h src/parsing/text.c
         src/index/web.c src/index/web.h
         src/web/serve.c src/web/serve.h
-        src/web/auth_basic.h src/web/auth_basic.c
         src/index/elastic.c src/index/elastic.h
         src/util.c src/util.h
-        src/ctx.h src/types.h src/parsing/font.c src/parsing/font.h
-        src/parsing/arc.c src/parsing/arc.h
-        src/parsing/doc.c src/parsing/doc.h
+        src/ctx.h src/types.h
         src/log.c src/log.h
-        src/parsing/cbr.h src/parsing/cbr.c
-
-        # argparse
-        argparse/argparse.h argparse/argparse.c
-
-        # cJSON
-        cJSON/cJSON.h cJSON/cJSON.c
-
-        # LMDB
-        lmdb/libraries/liblmdb/lmdb.h lmdb/libraries/liblmdb/mdb.c
-        lmdb/libraries/liblmdb/midl.h lmdb/libraries/liblmdb/midl.c
         src/cli.c src/cli.h
+        src/stats.c src/stats.h src/ctx.c
+        src/parsing/sidecar.c src/parsing/sidecar.h)
 
-        # utf8.h
-        utf8.h/utf8.h
-)
+target_link_directories(sist2 PRIVATE BEFORE ${_VCPKG_INSTALLED_DIR}/${VCPKG_TARGET_TRIPLET}/lib/)
+set(CMAKE_FIND_LIBRARY_SUFFIXES .a .lib)
 
-find_package(PkgConfig REQUIRED)
-set(ENV{PKG_CONFIG_PATH} "$ENV{PKG_CONFIG_PATH}:/usr/local/lib/pkgconfig/")
+find_package(lmdb CONFIG REQUIRED)
+find_package(cJSON CONFIG REQUIRED)
+find_package(unofficial-glib CONFIG REQUIRED)
+find_package(unofficial-mongoose CONFIG REQUIRED)
+find_package(CURL CONFIG REQUIRED)
 
-find_package(Freetype REQUIRED)
+#find_package(OpenSSL REQUIRED)
 
-pkg_check_modules(GLIB REQUIRED glib-2.0)
-pkg_check_modules(GOBJECT REQUIRED gobject-2.0)
-pkg_check_modules(UUID REQUIRED uuid)
-
-add_definitions(${UUID_CFLAGS_OTHER})
-add_definitions(${GLIB_CFLAGS_OTHER})
-add_definitions(${GOBJECT_CFLAGS_OTHER})
-add_definitions(${FREETYPE_CFLAGS_OTHER})
-
-list(REMOVE_ITEM GLIB_LIBRARIES pcre)
-list(REMOVE_ITEM GOBJECT_LIBRARIES pcre)
-list(REMOVE_ITEM UUID_LIBRARIES pcre)
 
 target_include_directories(
         sist2 PUBLIC
-        ${GOBJECT_INCLUDE_DIRS}
-        ${GLIB_INCLUDE_DIRS}
-        ${PROJECT_SOURCE_DIR}/lib/ffmpeg/
-        ${FREETYPE_INCLUDE_DIRS}
-        ${UUID_INCLUDE_DIRS}
-        ${PROJECT_SOURCE_DIR}/
-        ${PROJECT_SOURCE_DIR}/lmdb/libraries/liblmdb/
-        ${PROJECT_SOURCE_DIR}/lib/onion/src/
-        ${PROJECT_SOURCE_DIR}/lib/mupdf/include/
-        ${PROJECT_SOURCE_DIR}/include/
-        /usr/include/libxml2/
-        ${PROJECT_SOURCE_DIR}/lib/tesseract/include/
-)
-target_link_directories(
-        sist2 PUBLIC
-        ${UUID_LIBRARY_DIRS}
+        ${CMAKE_SOURCE_DIR}/third-party/onion/src/
+        ${CMAKE_SOURCE_DIR}/third-party/utf8.h/
+        ${CMAKE_SOURCE_DIR}/third-party/libscan/
+        ${CMAKE_SOURCE_DIR}/
 )
 
 target_compile_options(
@@ -97,6 +67,8 @@ if (SIST_DEBUG)
             -fstack-protector
             -fno-omit-frame-pointer
             -fsanitize=address
+            -fno-inline
+#            -O2
     )
     target_link_options(
             sist2
@@ -118,53 +90,29 @@ else ()
     )
 endif ()
 
-TARGET_LINK_LIBRARIES(
+add_dependencies(
+        sist2
+        scan
+        argparse
+)
+
+target_link_libraries(
         sist2
 
-        ${GLIB_LIBRARIES}
-        ${GOBJECT_LIBRARIES}
-        ${UUID_LIBRARIES}
-
-        # ffmpeg
-        ${PROJECT_SOURCE_DIR}/lib/libavcodec.a
-        ${PROJECT_SOURCE_DIR}/lib/libavformat.a
-        ${PROJECT_SOURCE_DIR}/lib/libavutil.a
-        ${PROJECT_SOURCE_DIR}/lib/libswscale.a
-        ${PROJECT_SOURCE_DIR}/lib/libswresample.a
-
-        # mupdf
-        ${PROJECT_SOURCE_DIR}/lib/libmupdf.a
-        ${PROJECT_SOURCE_DIR}/lib/libmupdf-third.a
-
-        # onion
-        ${PROJECT_SOURCE_DIR}/lib/libonion_static.a
+        z
+        lmdb
+        cjson
+        argparse
+        unofficial::glib::glib
+        unofficial::mongoose::mongoose
+        CURL::libcurl
 
         pthread
-
-        m
-        bz2
-#        ${PROJECT_SOURCE_DIR}/lib/libmagic.a
         magic
-        ${PROJECT_SOURCE_DIR}/lib/libharfbuzz.a
-        ${PROJECT_SOURCE_DIR}/lib/libopenjp2.a
-        freetype
-        archive
 
-        xml2
+        c
 
-        ${PROJECT_SOURCE_DIR}/lib/libtesseract.a
-        ${PROJECT_SOURCE_DIR}/lib/liblept.a
-        ${PROJECT_SOURCE_DIR}/lib/libtiff.a
-        ${PROJECT_SOURCE_DIR}/lib/libpng16.a
-        stdc++
-
-        # curl
-        ${PROJECT_SOURCE_DIR}/lib/libcurl.a
-        ${PROJECT_SOURCE_DIR}/lib/libcrypto.a
-        ${PROJECT_SOURCE_DIR}/lib/libssl.a
-        dl
-
-        pcre
+        scan
 )
 
 add_custom_target(

CMakeModules/FindFFmpeg.cmake

@@ -1,80 +0,0 @@
-# - Try to find ffmpeg libraries (libavcodec, libavformat and libavutil)
-# Once done this will define
-#
-# FFMPEG_FOUND - system has ffmpeg or libav
-# FFMPEG_INCLUDE_DIR - the ffmpeg include directory
-# FFMPEG_LIBRARIES - Link these to use ffmpeg
-# FFMPEG_LIBAVCODEC
-# FFMPEG_LIBAVFORMAT
-# FFMPEG_LIBAVUTIL
-#
-# Copyright (c) 2008 Andreas Schneider <mail@cynapses.org>
-# Modified for other libraries by Lasse Kärkkäinen <tronic>
-# Modified for Hedgewars by Stepik777
-#
-# Redistribution and use is allowed according to the terms of the New
-# BSD license.
-#
-
-if (FFMPEG_LIBRARIES AND FFMPEG_INCLUDE_DIR)
-# in cache already
-set(FFMPEG_FOUND TRUE)
-else (FFMPEG_LIBRARIES AND FFMPEG_INCLUDE_DIR)
-# use pkg-config to get the directories and then use these values
-# in the FIND_PATH() and FIND_LIBRARY() calls
-find_package(PkgConfig)
-if (PKG_CONFIG_FOUND)
-pkg_check_modules(_FFMPEG_AVCODEC libavcodec)
-pkg_check_modules(_FFMPEG_AVFORMAT libavformat)
-pkg_check_modules(_FFMPEG_AVUTIL libavutil)
-endif (PKG_CONFIG_FOUND)
-
-find_path(FFMPEG_AVCODEC_INCLUDE_DIR
-NAMES libavcodec/avcodec.h
-PATHS ${_FFMPEG_AVCODEC_INCLUDE_DIRS} /usr/include /usr/local/include /opt/local/include /sw/include
-PATH_SUFFIXES ffmpeg libav
-)
-
-find_library(FFMPEG_LIBAVCODEC
-NAMES avcodec
-PATHS ${_FFMPEG_AVCODEC_LIBRARY_DIRS} /usr/lib /usr/local/lib /opt/local/lib /sw/lib
-)
-
-find_library(FFMPEG_LIBAVFORMAT
-NAMES avformat
-PATHS ${_FFMPEG_AVFORMAT_LIBRARY_DIRS} /usr/lib /usr/local/lib /opt/local/lib /sw/lib
-)
-
-find_library(FFMPEG_LIBAVUTIL
-NAMES avutil
-PATHS ${_FFMPEG_AVUTIL_LIBRARY_DIRS} /usr/lib /usr/local/lib /opt/local/lib /sw/lib
-)
-
-if (FFMPEG_LIBAVCODEC AND FFMPEG_LIBAVFORMAT)
-set(FFMPEG_FOUND TRUE)
-endif()
-
-if (FFMPEG_FOUND)
-set(FFMPEG_INCLUDE_DIR ${FFMPEG_AVCODEC_INCLUDE_DIR})
-
-set(FFMPEG_LIBRARIES
-${FFMPEG_LIBAVCODEC}
-${FFMPEG_LIBAVFORMAT}
-${FFMPEG_LIBAVUTIL}
-)
-
-endif (FFMPEG_FOUND)
-
-if (FFMPEG_FOUND)
-if (NOT FFMPEG_FIND_QUIETLY)
-message(STATUS "Found FFMPEG or Libav: ${FFMPEG_LIBRARIES}, ${FFMPEG_INCLUDE_DIR}")
-endif (NOT FFMPEG_FIND_QUIETLY)
-else (FFMPEG_FOUND)
-if (FFMPEG_FIND_REQUIRED)
-message(FATAL_ERROR "Could not find libavcodec or libavformat or libavutil")
-endif (FFMPEG_FIND_REQUIRED)
-endif (FFMPEG_FOUND)
-
-endif (FFMPEG_LIBRARIES AND FFMPEG_INCLUDE_DIR)
-
-

CMakeModules/FindLibMagic.cmake

@@ -1,100 +0,0 @@
-
-#-------------------------------------------------------------------------------
-# Copyright (c) 2013-2013, Lars Baehren <lbaehren@gmail.com>
-# All rights reserved.
-#
-# Redistribution and use in source and binary forms, with or without modification,
-# are permitted provided that the following conditions are met:
-#
-#  * Redistributions of source code must retain the above copyright notice, this
-#    list of conditions and the following disclaimer.
-#  * Redistributions in binary form must reproduce the above copyright notice,
-#    this list of conditions and the following disclaimer in the documentation
-#    and/or other materials provided with the distribution.
-#
-# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
-# AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
-# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
-# DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
-# FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
-# DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
-# SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
-# CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
-# OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
-# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-#-------------------------------------------------------------------------------
-
-# - Check for the presence of LIBMAGIC
-#
-# The following variables are set when LIBMAGIC is found:
-#  LIBMAGIC_FOUND      = Set to true, if all components of LIBMAGIC have been
-#                        found.
-#  LIBMAGIC_INCLUDES   = Include path for the header files of LIBMAGIC
-#  LIBMAGIC_LIBRARIES  = Link these to use LIBMAGIC
-#  LIBMAGIC_LFLAGS     = Linker flags (optional)
-
-if (NOT LIBMAGIC_FOUND)
-
-  if (NOT LIBMAGIC_ROOT_DIR)
-    set (LIBMAGIC_ROOT_DIR ${CMAKE_INSTALL_PREFIX})
-  endif (NOT LIBMAGIC_ROOT_DIR)
-
-  ##____________________________________________________________________________
-  ## Check for the header files
-
-  find_path (LIBMAGIC_FILE_H
-    NAMES file/file.h
-    HINTS ${LIBMAGIC_ROOT_DIR} ${CMAKE_INSTALL_PREFIX}
-    PATH_SUFFIXES include
-    )
-  if (LIBMAGIC_FILE_H)
-    list (APPEND LIBMAGIC_INCLUDES ${LIBMAGIC_FILE_H})
-  endif (LIBMAGIC_FILE_H)
-
-  find_path (LIBMAGIC_MAGIC_H
-    NAMES magic.h
-    HINTS ${LIBMAGIC_ROOT_DIR} ${CMAKE_INSTALL_PREFIX}
-    PATH_SUFFIXES include include/linux
-    )
-  if (LIBMAGIC_MAGIC_H)
-    list (APPEND LIBMAGIC_INCLUDES ${LIBMAGIC_MAGIC_H})
-  endif (LIBMAGIC_MAGIC_H)
-
-  list (REMOVE_DUPLICATES LIBMAGIC_INCLUDES)
-
-  ##____________________________________________________________________________
-  ## Check for the library
-
-  find_library (LIBMAGIC_LIBRARIES magic
-    HINTS ${LIBMAGIC_ROOT_DIR} ${CMAKE_INSTALL_PREFIX}
-    PATH_SUFFIXES lib
-    )
-
-  ##____________________________________________________________________________
-  ## Actions taken when all components have been found
-
-  #find_package_handle_standard_args (LIBMAGIC DEFAULT_MSG LIBMAGIC_LIBRARIES LIBMAGIC_INCLUDES)
-
-  if (LIBMAGIC_FOUND)
-    if (NOT LIBMAGIC_FIND_QUIETLY)
-      message (STATUS "Found components for LIBMAGIC")
-      message (STATUS "LIBMAGIC_ROOT_DIR  = ${LIBMAGIC_ROOT_DIR}")
-      message (STATUS "LIBMAGIC_INCLUDES  = ${LIBMAGIC_INCLUDES}")
-      message (STATUS "LIBMAGIC_LIBRARIES = ${LIBMAGIC_LIBRARIES}")
-    endif (NOT LIBMAGIC_FIND_QUIETLY)
-  else (LIBMAGIC_FOUND)
-    if (LIBMAGIC_FIND_REQUIRED)
-      message (FATAL_ERROR "Could not find LIBMAGIC!")
-    endif (LIBMAGIC_FIND_REQUIRED)
-  endif (LIBMAGIC_FOUND)
-
-  ##____________________________________________________________________________
-  ## Mark advanced variables
-
-  mark_as_advanced (
-    LIBMAGIC_ROOT_DIR
-    LIBMAGIC_INCLUDES
-    LIBMAGIC_LIBRARIES
-    )
-
-endif (NOT LIBMAGIC_FOUND)

CMakeModules/FindOpenSSL.cmake

@@ -1,478 +0,0 @@
-# Distributed under the OSI-approved BSD 3-Clause License.  See accompanying
-# file Copyright.txt or https://cmake.org/licensing for details.
-
-macro(_OpenSSL_test_and_find_dependencies ssl_library crypto_library)
-  if((CMAKE_SYSTEM_NAME STREQUAL "Linux") AND
-     (("${ssl_library}" MATCHES "\\${CMAKE_STATIC_LIBRARY_SUFFIX}$") OR
-      ("${crypto_library}" MATCHES "\\${CMAKE_STATIC_LIBRARY_SUFFIX}$")))
-    set(_OpenSSL_has_dependencies TRUE)
-    find_package(Threads)
-  else()
-    set(_OpenSSL_has_dependencies FALSE)
-  endif()
-endmacro()
-
-function(_OpenSSL_add_dependencies libraries_var library)
-  if(CMAKE_THREAD_LIBS_INIT)
-    list(APPEND ${libraries_var} ${CMAKE_THREAD_LIBS_INIT})
-  endif()
-  list(APPEND ${libraries_var} ${CMAKE_DL_LIBS})
-  set(${libraries_var} ${${libraries_var}} PARENT_SCOPE)
-endfunction()
-
-function(_OpenSSL_target_add_dependencies target)
-  if(_OpenSSL_has_dependencies)
-    set_property( TARGET ${target} APPEND PROPERTY INTERFACE_LINK_LIBRARIES Threads::Threads )
-    set_property( TARGET ${target} APPEND PROPERTY INTERFACE_LINK_LIBRARIES ${CMAKE_DL_LIBS} )
-  endif()
-endfunction()
-
-if (UNIX)
-  find_package(PkgConfig QUIET)
-  pkg_check_modules(_OPENSSL QUIET openssl)
-endif ()
-
-# Support preference of static libs by adjusting CMAKE_FIND_LIBRARY_SUFFIXES
-if(OPENSSL_USE_STATIC_LIBS)
-  set(_openssl_ORIG_CMAKE_FIND_LIBRARY_SUFFIXES ${CMAKE_FIND_LIBRARY_SUFFIXES})
-  if(WIN32)
-    set(CMAKE_FIND_LIBRARY_SUFFIXES .lib .a ${CMAKE_FIND_LIBRARY_SUFFIXES})
-  else()
-    set(CMAKE_FIND_LIBRARY_SUFFIXES .a )
-  endif()
-endif()
-
-if (WIN32)
-  # http://www.slproweb.com/products/Win32OpenSSL.html
-  set(_OPENSSL_ROOT_HINTS
-    ${OPENSSL_ROOT_DIR}
-    "[HKEY_LOCAL_MACHINE\\SOFTWARE\\Microsoft\\Windows\\CurrentVersion\\Uninstall\\OpenSSL (32-bit)_is1;Inno Setup: App Path]"
-    "[HKEY_LOCAL_MACHINE\\SOFTWARE\\Microsoft\\Windows\\CurrentVersion\\Uninstall\\OpenSSL (64-bit)_is1;Inno Setup: App Path]"
-    ENV OPENSSL_ROOT_DIR
-    )
-  file(TO_CMAKE_PATH "$ENV{PROGRAMFILES}" _programfiles)
-  set(_OPENSSL_ROOT_PATHS
-    "${_programfiles}/OpenSSL"
-    "${_programfiles}/OpenSSL-Win32"
-    "${_programfiles}/OpenSSL-Win64"
-    "C:/OpenSSL/"
-    "C:/OpenSSL-Win32/"
-    "C:/OpenSSL-Win64/"
-    )
-  unset(_programfiles)
-else ()
-  set(_OPENSSL_ROOT_HINTS
-    ${OPENSSL_ROOT_DIR}
-    ENV OPENSSL_ROOT_DIR
-    )
-endif ()
-
-set(_OPENSSL_ROOT_HINTS_AND_PATHS
-    HINTS ${_OPENSSL_ROOT_HINTS}
-    PATHS ${_OPENSSL_ROOT_PATHS}
-    )
-
-find_path(OPENSSL_INCLUDE_DIR
-  NAMES
-    openssl/ssl.h
-  ${_OPENSSL_ROOT_HINTS_AND_PATHS}
-  HINTS
-    ${_OPENSSL_INCLUDEDIR}
-  PATH_SUFFIXES
-    include
-)
-
-if(WIN32 AND NOT CYGWIN)
-  if(MSVC)
-    # /MD and /MDd are the standard values - if someone wants to use
-    # others, the libnames have to change here too
-    # use also ssl and ssleay32 in debug as fallback for openssl < 0.9.8b
-    # enable OPENSSL_MSVC_STATIC_RT to get the libs build /MT (Multithreaded no-DLL)
-    # In Visual C++ naming convention each of these four kinds of Windows libraries has it's standard suffix:
-    #   * MD for dynamic-release
-    #   * MDd for dynamic-debug
-    #   * MT for static-release
-    #   * MTd for static-debug
-
-    # Implementation details:
-    # We are using the libraries located in the VC subdir instead of the parent directory even though :
-    # libeay32MD.lib is identical to ../libeay32.lib, and
-    # ssleay32MD.lib is identical to ../ssleay32.lib
-    # enable OPENSSL_USE_STATIC_LIBS to use the static libs located in lib/VC/static
-
-    if (OPENSSL_MSVC_STATIC_RT)
-      set(_OPENSSL_MSVC_RT_MODE "MT")
-    else ()
-      set(_OPENSSL_MSVC_RT_MODE "MD")
-    endif ()
-
-    # Since OpenSSL 1.1, lib names are like libcrypto32MTd.lib and libssl32MTd.lib
-    if( "${CMAKE_SIZEOF_VOID_P}" STREQUAL "8" )
-        set(_OPENSSL_MSVC_ARCH_SUFFIX "64")
-    else()
-        set(_OPENSSL_MSVC_ARCH_SUFFIX "32")
-    endif()
-
-    if(OPENSSL_USE_STATIC_LIBS)
-      set(_OPENSSL_PATH_SUFFIXES
-        "lib/VC/static"
-        "VC/static"
-        "lib"
-        )
-    else()
-      set(_OPENSSL_PATH_SUFFIXES
-        "lib/VC"
-        "VC"
-        "lib"
-        )
-    endif ()
-
-    find_library(LIB_EAY_DEBUG
-      NAMES
-        libcrypto${_OPENSSL_MSVC_ARCH_SUFFIX}${_OPENSSL_MSVC_RT_MODE}d
-        libcrypto${_OPENSSL_MSVC_RT_MODE}d
-        libcryptod
-        libeay32${_OPENSSL_MSVC_RT_MODE}d
-        libeay32d
-        cryptod
-      NAMES_PER_DIR
-      ${_OPENSSL_ROOT_HINTS_AND_PATHS}
-      PATH_SUFFIXES
-        ${_OPENSSL_PATH_SUFFIXES}
-    )
-
-    find_library(LIB_EAY_RELEASE
-      NAMES
-        libcrypto${_OPENSSL_MSVC_ARCH_SUFFIX}${_OPENSSL_MSVC_RT_MODE}
-        libcrypto${_OPENSSL_MSVC_RT_MODE}
-        libcrypto
-        libeay32${_OPENSSL_MSVC_RT_MODE}
-        libeay32
-        crypto
-      NAMES_PER_DIR
-      ${_OPENSSL_ROOT_HINTS_AND_PATHS}
-      PATH_SUFFIXES
-        ${_OPENSSL_PATH_SUFFIXES}
-    )
-
-    find_library(SSL_EAY_DEBUG
-      NAMES
-        libssl${_OPENSSL_MSVC_ARCH_SUFFIX}${_OPENSSL_MSVC_RT_MODE}d
-        libssl${_OPENSSL_MSVC_RT_MODE}d
-        libssld
-        ssleay32${_OPENSSL_MSVC_RT_MODE}d
-        ssleay32d
-        ssld
-      NAMES_PER_DIR
-      ${_OPENSSL_ROOT_HINTS_AND_PATHS}
-      PATH_SUFFIXES
-        ${_OPENSSL_PATH_SUFFIXES}
-    )
-
-    find_library(SSL_EAY_RELEASE
-      NAMES
-        libssl${_OPENSSL_MSVC_ARCH_SUFFIX}${_OPENSSL_MSVC_RT_MODE}
-        libssl${_OPENSSL_MSVC_RT_MODE}
-        libssl
-        ssleay32${_OPENSSL_MSVC_RT_MODE}
-        ssleay32
-        ssl
-      NAMES_PER_DIR
-      ${_OPENSSL_ROOT_HINTS_AND_PATHS}
-      PATH_SUFFIXES
-        ${_OPENSSL_PATH_SUFFIXES}
-    )
-
-    set(LIB_EAY_LIBRARY_DEBUG "${LIB_EAY_DEBUG}")
-    set(LIB_EAY_LIBRARY_RELEASE "${LIB_EAY_RELEASE}")
-    set(SSL_EAY_LIBRARY_DEBUG "${SSL_EAY_DEBUG}")
-    set(SSL_EAY_LIBRARY_RELEASE "${SSL_EAY_RELEASE}")
-
-    include(${CMAKE_CURRENT_LIST_DIR}/SelectLibraryConfigurations.cmake)
-    select_library_configurations(LIB_EAY)
-    select_library_configurations(SSL_EAY)
-
-    mark_as_advanced(LIB_EAY_LIBRARY_DEBUG LIB_EAY_LIBRARY_RELEASE
-                     SSL_EAY_LIBRARY_DEBUG SSL_EAY_LIBRARY_RELEASE)
-    set(OPENSSL_SSL_LIBRARY ${SSL_EAY_LIBRARY} )
-    set(OPENSSL_CRYPTO_LIBRARY ${LIB_EAY_LIBRARY} )
-  elseif(MINGW)
-    # same player, for MinGW
-    set(LIB_EAY_NAMES crypto libeay32)
-    set(SSL_EAY_NAMES ssl ssleay32)
-    find_library(LIB_EAY
-      NAMES
-        ${LIB_EAY_NAMES}
-      NAMES_PER_DIR
-      ${_OPENSSL_ROOT_HINTS_AND_PATHS}
-      PATH_SUFFIXES
-        "lib/MinGW"
-        "lib"
-    )
-
-    find_library(SSL_EAY
-      NAMES
-        ${SSL_EAY_NAMES}
-      NAMES_PER_DIR
-      ${_OPENSSL_ROOT_HINTS_AND_PATHS}
-      PATH_SUFFIXES
-        "lib/MinGW"
-        "lib"
-    )
-
-    mark_as_advanced(SSL_EAY LIB_EAY)
-    set(OPENSSL_SSL_LIBRARY ${SSL_EAY} )
-    set(OPENSSL_CRYPTO_LIBRARY ${LIB_EAY} )
-    unset(LIB_EAY_NAMES)
-    unset(SSL_EAY_NAMES)
-  else()
-    # Not sure what to pick for -say- intel, let's use the toplevel ones and hope someone report issues:
-    find_library(LIB_EAY
-      NAMES
-        libcrypto
-        libeay32
-      NAMES_PER_DIR
-      ${_OPENSSL_ROOT_HINTS_AND_PATHS}
-      HINTS
-        ${_OPENSSL_LIBDIR}
-      PATH_SUFFIXES
-        lib
-    )
-
-    find_library(SSL_EAY
-      NAMES
-        libssl
-        ssleay32
-      NAMES_PER_DIR
-      ${_OPENSSL_ROOT_HINTS_AND_PATHS}
-      HINTS
-        ${_OPENSSL_LIBDIR}
-      PATH_SUFFIXES
-        lib
-    )
-
-    mark_as_advanced(SSL_EAY LIB_EAY)
-    set(OPENSSL_SSL_LIBRARY ${SSL_EAY} )
-    set(OPENSSL_CRYPTO_LIBRARY ${LIB_EAY} )
-  endif()
-else()
-
-  find_library(OPENSSL_SSL_LIBRARY
-    NAMES
-      ssl
-      ssleay32
-      ssleay32MD
-    NAMES_PER_DIR
-    ${_OPENSSL_ROOT_HINTS_AND_PATHS}
-    HINTS
-      ${_OPENSSL_LIBDIR}
-    PATH_SUFFIXES
-      lib
-  )
-
-  find_library(OPENSSL_CRYPTO_LIBRARY
-    NAMES
-      crypto
-    NAMES_PER_DIR
-    ${_OPENSSL_ROOT_HINTS_AND_PATHS}
-    HINTS
-      ${_OPENSSL_LIBDIR}
-    PATH_SUFFIXES
-      lib
-  )
-
-  mark_as_advanced(OPENSSL_CRYPTO_LIBRARY OPENSSL_SSL_LIBRARY)
-
-endif()
-
-# compat defines
-set(OPENSSL_SSL_LIBRARIES ${OPENSSL_SSL_LIBRARY})
-set(OPENSSL_CRYPTO_LIBRARIES ${OPENSSL_CRYPTO_LIBRARY})
-_OpenSSL_test_and_find_dependencies("${OPENSSL_SSL_LIBRARY}" "${OPENSSL_CRYPTO_LIBRARY}")
-if(_OpenSSL_has_dependencies)
-  _OpenSSL_add_dependencies( OPENSSL_SSL_LIBRARIES "${OPENSSL_SSL_LIBRARY}" )
-  _OpenSSL_add_dependencies( OPENSSL_CRYPTO_LIBRARIES "${OPENSSL_CRYPTO_LIBRARY}" )
-endif()
-
-function(from_hex HEX DEC)
-  string(TOUPPER "${HEX}" HEX)
-  set(_res 0)
-  string(LENGTH "${HEX}" _strlen)
-
-  while (_strlen GREATER 0)
-    math(EXPR _res "${_res} * 16")
-    string(SUBSTRING "${HEX}" 0 1 NIBBLE)
-    string(SUBSTRING "${HEX}" 1 -1 HEX)
-    if (NIBBLE STREQUAL "A")
-      math(EXPR _res "${_res} + 10")
-    elseif (NIBBLE STREQUAL "B")
-      math(EXPR _res "${_res} + 11")
-    elseif (NIBBLE STREQUAL "C")
-      math(EXPR _res "${_res} + 12")
-    elseif (NIBBLE STREQUAL "D")
-      math(EXPR _res "${_res} + 13")
-    elseif (NIBBLE STREQUAL "E")
-      math(EXPR _res "${_res} + 14")
-    elseif (NIBBLE STREQUAL "F")
-      math(EXPR _res "${_res} + 15")
-    else()
-      math(EXPR _res "${_res} + ${NIBBLE}")
-    endif()
-
-    string(LENGTH "${HEX}" _strlen)
-  endwhile()
-
-  set(${DEC} ${_res} PARENT_SCOPE)
-endfunction()
-
-if(OPENSSL_INCLUDE_DIR AND EXISTS "${OPENSSL_INCLUDE_DIR}/openssl/opensslv.h")
-  file(STRINGS "${OPENSSL_INCLUDE_DIR}/openssl/opensslv.h" openssl_version_str
-       REGEX "^#[\t ]*define[\t ]+OPENSSL_VERSION_NUMBER[\t ]+0x([0-9a-fA-F])+.*")
-
-  if(openssl_version_str)
-    # The version number is encoded as 0xMNNFFPPS: major minor fix patch status
-    # The status gives if this is a developer or prerelease and is ignored here.
-    # Major, minor, and fix directly translate into the version numbers shown in
-    # the string. The patch field translates to the single character suffix that
-    # indicates the bug fix state, which 00 -> nothing, 01 -> a, 02 -> b and so
-    # on.
-
-    string(REGEX REPLACE "^.*OPENSSL_VERSION_NUMBER[\t ]+0x([0-9a-fA-F])([0-9a-fA-F][0-9a-fA-F])([0-9a-fA-F][0-9a-fA-F])([0-9a-fA-F][0-9a-fA-F])([0-9a-fA-F]).*$"
-           "\\1;\\2;\\3;\\4;\\5" OPENSSL_VERSION_LIST "${openssl_version_str}")
-    list(GET OPENSSL_VERSION_LIST 0 OPENSSL_VERSION_MAJOR)
-    list(GET OPENSSL_VERSION_LIST 1 OPENSSL_VERSION_MINOR)
-    from_hex("${OPENSSL_VERSION_MINOR}" OPENSSL_VERSION_MINOR)
-    list(GET OPENSSL_VERSION_LIST 2 OPENSSL_VERSION_FIX)
-    from_hex("${OPENSSL_VERSION_FIX}" OPENSSL_VERSION_FIX)
-    list(GET OPENSSL_VERSION_LIST 3 OPENSSL_VERSION_PATCH)
-
-    if (NOT OPENSSL_VERSION_PATCH STREQUAL "00")
-      from_hex("${OPENSSL_VERSION_PATCH}" _tmp)
-      # 96 is the ASCII code of 'a' minus 1
-      math(EXPR OPENSSL_VERSION_PATCH_ASCII "${_tmp} + 96")
-      unset(_tmp)
-      # Once anyone knows how OpenSSL would call the patch versions beyond 'z'
-      # this should be updated to handle that, too. This has not happened yet
-      # so it is simply ignored here for now.
-      string(ASCII "${OPENSSL_VERSION_PATCH_ASCII}" OPENSSL_VERSION_PATCH_STRING)
-    endif ()
-
-    set(OPENSSL_VERSION "${OPENSSL_VERSION_MAJOR}.${OPENSSL_VERSION_MINOR}.${OPENSSL_VERSION_FIX}${OPENSSL_VERSION_PATCH_STRING}")
-  endif ()
-endif ()
-
-set(OPENSSL_LIBRARIES ${OPENSSL_SSL_LIBRARIES} ${OPENSSL_CRYPTO_LIBRARIES} )
-list(REMOVE_DUPLICATES OPENSSL_LIBRARIES)
-
-foreach(_comp IN LISTS OpenSSL_FIND_COMPONENTS)
-  if(_comp STREQUAL "Crypto")
-    if(EXISTS "${OPENSSL_INCLUDE_DIR}" AND
-        (EXISTS "${OPENSSL_CRYPTO_LIBRARY}" OR
-        EXISTS "${LIB_EAY_LIBRARY_DEBUG}" OR
-        EXISTS "${LIB_EAY_LIBRARY_RELEASE}")
-    )
-      set(OpenSSL_${_comp}_FOUND TRUE)
-    else()
-      set(OpenSSL_${_comp}_FOUND FALSE)
-    endif()
-  elseif(_comp STREQUAL "SSL")
-    if(EXISTS "${OPENSSL_INCLUDE_DIR}" AND
-        (EXISTS "${OPENSSL_SSL_LIBRARY}" OR
-        EXISTS "${SSL_EAY_LIBRARY_DEBUG}" OR
-        EXISTS "${SSL_EAY_LIBRARY_RELEASE}")
-    )
-      set(OpenSSL_${_comp}_FOUND TRUE)
-    else()
-      set(OpenSSL_${_comp}_FOUND FALSE)
-    endif()
-  else()
-    message(WARNING "${_comp} is not a valid OpenSSL component")
-    set(OpenSSL_${_comp}_FOUND FALSE)
-  endif()
-endforeach()
-unset(_comp)
-
-include(${CMAKE_CURRENT_LIST_DIR}/FindPackageHandleStandardArgs.cmake)
-find_package_handle_standard_args(OpenSSL
-  REQUIRED_VARS
-    OPENSSL_CRYPTO_LIBRARY
-    OPENSSL_INCLUDE_DIR
-  VERSION_VAR
-    OPENSSL_VERSION
-  HANDLE_COMPONENTS
-  FAIL_MESSAGE
-    "Could NOT find OpenSSL, try to set the path to OpenSSL root folder in the system variable OPENSSL_ROOT_DIR"
-)
-
-mark_as_advanced(OPENSSL_INCLUDE_DIR OPENSSL_LIBRARIES)
-
-if(OPENSSL_FOUND)
-  if(NOT TARGET OpenSSL::Crypto AND
-      (EXISTS "${OPENSSL_CRYPTO_LIBRARY}" OR
-        EXISTS "${LIB_EAY_LIBRARY_DEBUG}" OR
-        EXISTS "${LIB_EAY_LIBRARY_RELEASE}")
-      )
-    add_library(OpenSSL::Crypto UNKNOWN IMPORTED)
-    set_target_properties(OpenSSL::Crypto PROPERTIES
-      INTERFACE_INCLUDE_DIRECTORIES "${OPENSSL_INCLUDE_DIR}")
-    if(EXISTS "${OPENSSL_CRYPTO_LIBRARY}")
-      set_target_properties(OpenSSL::Crypto PROPERTIES
-        IMPORTED_LINK_INTERFACE_LANGUAGES "C"
-        IMPORTED_LOCATION "${OPENSSL_CRYPTO_LIBRARY}")
-    endif()
-    if(EXISTS "${LIB_EAY_LIBRARY_RELEASE}")
-      set_property(TARGET OpenSSL::Crypto APPEND PROPERTY
-        IMPORTED_CONFIGURATIONS RELEASE)
-      set_target_properties(OpenSSL::Crypto PROPERTIES
-        IMPORTED_LINK_INTERFACE_LANGUAGES_RELEASE "C"
-        IMPORTED_LOCATION_RELEASE "${LIB_EAY_LIBRARY_RELEASE}")
-    endif()
-    if(EXISTS "${LIB_EAY_LIBRARY_DEBUG}")
-      set_property(TARGET OpenSSL::Crypto APPEND PROPERTY
-        IMPORTED_CONFIGURATIONS DEBUG)
-      set_target_properties(OpenSSL::Crypto PROPERTIES
-        IMPORTED_LINK_INTERFACE_LANGUAGES_DEBUG "C"
-        IMPORTED_LOCATION_DEBUG "${LIB_EAY_LIBRARY_DEBUG}")
-    endif()
-    _OpenSSL_target_add_dependencies(OpenSSL::Crypto)
-  endif()
-
-  if(NOT TARGET OpenSSL::SSL AND
-      (EXISTS "${OPENSSL_SSL_LIBRARY}" OR
-        EXISTS "${SSL_EAY_LIBRARY_DEBUG}" OR
-        EXISTS "${SSL_EAY_LIBRARY_RELEASE}")
-      )
-    add_library(OpenSSL::SSL UNKNOWN IMPORTED)
-    set_target_properties(OpenSSL::SSL PROPERTIES
-      INTERFACE_INCLUDE_DIRECTORIES "${OPENSSL_INCLUDE_DIR}")
-    if(EXISTS "${OPENSSL_SSL_LIBRARY}")
-      set_target_properties(OpenSSL::SSL PROPERTIES
-        IMPORTED_LINK_INTERFACE_LANGUAGES "C"
-        IMPORTED_LOCATION "${OPENSSL_SSL_LIBRARY}")
-    endif()
-    if(EXISTS "${SSL_EAY_LIBRARY_RELEASE}")
-      set_property(TARGET OpenSSL::SSL APPEND PROPERTY
-        IMPORTED_CONFIGURATIONS RELEASE)
-      set_target_properties(OpenSSL::SSL PROPERTIES
-        IMPORTED_LINK_INTERFACE_LANGUAGES_RELEASE "C"
-        IMPORTED_LOCATION_RELEASE "${SSL_EAY_LIBRARY_RELEASE}")
-    endif()
-    if(EXISTS "${SSL_EAY_LIBRARY_DEBUG}")
-      set_property(TARGET OpenSSL::SSL APPEND PROPERTY
-        IMPORTED_CONFIGURATIONS DEBUG)
-      set_target_properties(OpenSSL::SSL PROPERTIES
-        IMPORTED_LINK_INTERFACE_LANGUAGES_DEBUG "C"
-        IMPORTED_LOCATION_DEBUG "${SSL_EAY_LIBRARY_DEBUG}")
-    endif()
-    if(TARGET OpenSSL::Crypto)
-      set_target_properties(OpenSSL::SSL PROPERTIES
-        INTERFACE_LINK_LIBRARIES OpenSSL::Crypto)
-    endif()
-    _OpenSSL_target_add_dependencies(OpenSSL::SSL)
-  endif()
-endif()
-
-# Restore the original find library ordering
-if(OPENSSL_USE_STATIC_LIBS)
-  set(CMAKE_FIND_LIBRARY_SUFFIXES ${_openssl_ORIG_CMAKE_FIND_LIBRARY_SUFFIXES})
-endif()

CMakeModules/FindPackageHandleStandardArgs.cmake

@@ -1,268 +0,0 @@
-# Distributed under the OSI-approved BSD 3-Clause License.  See accompanying
-# file Copyright.txt or https://cmake.org/licensing for details.
-
-include(${CMAKE_CURRENT_LIST_DIR}/FindPackageMessage.cmake)
-
-# internal helper macro
-macro(_FPHSA_FAILURE_MESSAGE _msg)
-  set (__msg "${_msg}")
-  if (FPHSA_REASON_FAILURE_MESSAGE)
-    string(APPEND __msg "\n    Reason given by package: ${FPHSA_REASON_FAILURE_MESSAGE}\n")
-  endif()
-  if (${_NAME}_FIND_REQUIRED)
-    message(FATAL_ERROR "${__msg}")
-  else ()
-    if (NOT ${_NAME}_FIND_QUIETLY)
-      message(STATUS "${__msg}")
-    endif ()
-  endif ()
-endmacro()
-
-
-# internal helper macro to generate the failure message when used in CONFIG_MODE:
-macro(_FPHSA_HANDLE_FAILURE_CONFIG_MODE)
-  # <PackageName>_CONFIG is set, but FOUND is false, this means that some other of the REQUIRED_VARS was not found:
-  if(${_NAME}_CONFIG)
-    _FPHSA_FAILURE_MESSAGE("${FPHSA_FAIL_MESSAGE}: missing:${MISSING_VARS} (found ${${_NAME}_CONFIG} ${VERSION_MSG})")
-  else()
-    # If _CONSIDERED_CONFIGS is set, the config-file has been found, but no suitable version.
-    # List them all in the error message:
-    if(${_NAME}_CONSIDERED_CONFIGS)
-      set(configsText "")
-      list(LENGTH ${_NAME}_CONSIDERED_CONFIGS configsCount)
-      math(EXPR configsCount "${configsCount} - 1")
-      foreach(currentConfigIndex RANGE ${configsCount})
-        list(GET ${_NAME}_CONSIDERED_CONFIGS ${currentConfigIndex} filename)
-        list(GET ${_NAME}_CONSIDERED_VERSIONS ${currentConfigIndex} version)
-        string(APPEND configsText "\n    ${filename} (version ${version})")
-      endforeach()
-      if (${_NAME}_NOT_FOUND_MESSAGE)
-        if (FPHSA_REASON_FAILURE_MESSAGE)
-          string(PREPEND FPHSA_REASON_FAILURE_MESSAGE "${${_NAME}_NOT_FOUND_MESSAGE}\n    ")
-        else()
-          set(FPHSA_REASON_FAILURE_MESSAGE "${${_NAME}_NOT_FOUND_MESSAGE}")
-        endif()
-      else()
-        string(APPEND configsText "\n")
-      endif()
-      _FPHSA_FAILURE_MESSAGE("${FPHSA_FAIL_MESSAGE} ${VERSION_MSG}, checked the following files:${configsText}")
-
-    else()
-      # Simple case: No Config-file was found at all:
-      _FPHSA_FAILURE_MESSAGE("${FPHSA_FAIL_MESSAGE}: found neither ${_NAME}Config.cmake nor ${_NAME_LOWER}-config.cmake ${VERSION_MSG}")
-    endif()
-  endif()
-endmacro()
-
-
-function(FIND_PACKAGE_HANDLE_STANDARD_ARGS _NAME _FIRST_ARG)
-
-# Set up the arguments for `cmake_parse_arguments`.
-  set(options  CONFIG_MODE  HANDLE_COMPONENTS)
-  set(oneValueArgs  FAIL_MESSAGE  REASON_FAILURE_MESSAGE VERSION_VAR  FOUND_VAR)
-  set(multiValueArgs REQUIRED_VARS)
-
-# Check whether we are in 'simple' or 'extended' mode:
-  set(_KEYWORDS_FOR_EXTENDED_MODE  ${options} ${oneValueArgs} ${multiValueArgs} )
-  list(FIND _KEYWORDS_FOR_EXTENDED_MODE "${_FIRST_ARG}" INDEX)
-
-  if(${INDEX} EQUAL -1)
-    set(FPHSA_FAIL_MESSAGE ${_FIRST_ARG})
-    set(FPHSA_REQUIRED_VARS ${ARGN})
-    set(FPHSA_VERSION_VAR)
-  else()
-    cmake_parse_arguments(FPHSA "${options}" "${oneValueArgs}" "${multiValueArgs}"  ${_FIRST_ARG} ${ARGN})
-
-    if(FPHSA_UNPARSED_ARGUMENTS)
-      message(FATAL_ERROR "Unknown keywords given to FIND_PACKAGE_HANDLE_STANDARD_ARGS(): \"${FPHSA_UNPARSED_ARGUMENTS}\"")
-    endif()
-
-    if(NOT FPHSA_FAIL_MESSAGE)
-      set(FPHSA_FAIL_MESSAGE  "DEFAULT_MSG")
-    endif()
-
-    # In config-mode, we rely on the variable <PackageName>_CONFIG, which is set by find_package()
-    # when it successfully found the config-file, including version checking:
-    if(FPHSA_CONFIG_MODE)
-      list(INSERT FPHSA_REQUIRED_VARS 0 ${_NAME}_CONFIG)
-      list(REMOVE_DUPLICATES FPHSA_REQUIRED_VARS)
-      set(FPHSA_VERSION_VAR ${_NAME}_VERSION)
-    endif()
-
-    if(NOT FPHSA_REQUIRED_VARS)
-      message(FATAL_ERROR "No REQUIRED_VARS specified for FIND_PACKAGE_HANDLE_STANDARD_ARGS()")
-    endif()
-  endif()
-
-# now that we collected all arguments, process them
-
-  if("x${FPHSA_FAIL_MESSAGE}" STREQUAL "xDEFAULT_MSG")
-    set(FPHSA_FAIL_MESSAGE "Could NOT find ${_NAME}")
-  endif()
-
-  list(GET FPHSA_REQUIRED_VARS 0 _FIRST_REQUIRED_VAR)
-
-  string(TOUPPER ${_NAME} _NAME_UPPER)
-  string(TOLOWER ${_NAME} _NAME_LOWER)
-
-  if(FPHSA_FOUND_VAR)
-    if(FPHSA_FOUND_VAR MATCHES "^${_NAME}_FOUND$"  OR  FPHSA_FOUND_VAR MATCHES "^${_NAME_UPPER}_FOUND$")
-      set(_FOUND_VAR ${FPHSA_FOUND_VAR})
-    else()
-      message(FATAL_ERROR "The argument for FOUND_VAR is \"${FPHSA_FOUND_VAR}\", but only \"${_NAME}_FOUND\" and \"${_NAME_UPPER}_FOUND\" are valid names.")
-    endif()
-  else()
-    set(_FOUND_VAR ${_NAME_UPPER}_FOUND)
-  endif()
-
-  # collect all variables which were not found, so they can be printed, so the
-  # user knows better what went wrong (#6375)
-  set(MISSING_VARS "")
-  set(DETAILS "")
-  # check if all passed variables are valid
-  set(FPHSA_FOUND_${_NAME} TRUE)
-  foreach(_CURRENT_VAR ${FPHSA_REQUIRED_VARS})
-    if(NOT ${_CURRENT_VAR})
-      set(FPHSA_FOUND_${_NAME} FALSE)
-      string(APPEND MISSING_VARS " ${_CURRENT_VAR}")
-    else()
-      string(APPEND DETAILS "[${${_CURRENT_VAR}}]")
-    endif()
-  endforeach()
-  if(FPHSA_FOUND_${_NAME})
-    set(${_NAME}_FOUND TRUE)
-    set(${_NAME_UPPER}_FOUND TRUE)
-  else()
-    set(${_NAME}_FOUND FALSE)
-    set(${_NAME_UPPER}_FOUND FALSE)
-  endif()
-
-  # component handling
-  unset(FOUND_COMPONENTS_MSG)
-  unset(MISSING_COMPONENTS_MSG)
-
-  if(FPHSA_HANDLE_COMPONENTS)
-    foreach(comp ${${_NAME}_FIND_COMPONENTS})
-      if(${_NAME}_${comp}_FOUND)
-
-        if(NOT DEFINED FOUND_COMPONENTS_MSG)
-          set(FOUND_COMPONENTS_MSG "found components:")
-        endif()
-        string(APPEND FOUND_COMPONENTS_MSG " ${comp}")
-
-      else()
-
-        if(NOT DEFINED MISSING_COMPONENTS_MSG)
-          set(MISSING_COMPONENTS_MSG "missing components:")
-        endif()
-        string(APPEND MISSING_COMPONENTS_MSG " ${comp}")
-
-        if(${_NAME}_FIND_REQUIRED_${comp})
-          set(${_NAME}_FOUND FALSE)
-          string(APPEND MISSING_VARS " ${comp}")
-        endif()
-
-      endif()
-    endforeach()
-    set(COMPONENT_MSG "${FOUND_COMPONENTS_MSG} ${MISSING_COMPONENTS_MSG}")
-    string(APPEND DETAILS "[c${COMPONENT_MSG}]")
-  endif()
-
-  # version handling:
-  set(VERSION_MSG "")
-  set(VERSION_OK TRUE)
-
-  # check with DEFINED here as the requested or found version may be "0"
-  if (DEFINED ${_NAME}_FIND_VERSION)
-    if(DEFINED ${FPHSA_VERSION_VAR})
-      set(_FOUND_VERSION ${${FPHSA_VERSION_VAR}})
-
-      if(${_NAME}_FIND_VERSION_EXACT)       # exact version required
-        # count the dots in the version string
-        string(REGEX REPLACE "[^.]" "" _VERSION_DOTS "${_FOUND_VERSION}")
-        # add one dot because there is one dot more than there are components
-        string(LENGTH "${_VERSION_DOTS}." _VERSION_DOTS)
-        if (_VERSION_DOTS GREATER ${_NAME}_FIND_VERSION_COUNT)
-          # Because of the C++ implementation of find_package() ${_NAME}_FIND_VERSION_COUNT
-          # is at most 4 here. Therefore a simple lookup table is used.
-          if (${_NAME}_FIND_VERSION_COUNT EQUAL 1)
-            set(_VERSION_REGEX "[^.]*")
-          elseif (${_NAME}_FIND_VERSION_COUNT EQUAL 2)
-            set(_VERSION_REGEX "[^.]*\\.[^.]*")
-          elseif (${_NAME}_FIND_VERSION_COUNT EQUAL 3)
-            set(_VERSION_REGEX "[^.]*\\.[^.]*\\.[^.]*")
-          else ()
-            set(_VERSION_REGEX "[^.]*\\.[^.]*\\.[^.]*\\.[^.]*")
-          endif ()
-          string(REGEX REPLACE "^(${_VERSION_REGEX})\\..*" "\\1" _VERSION_HEAD "${_FOUND_VERSION}")
-          unset(_VERSION_REGEX)
-          if (NOT ${_NAME}_FIND_VERSION VERSION_EQUAL _VERSION_HEAD)
-            set(VERSION_MSG "Found unsuitable version \"${_FOUND_VERSION}\", but required is exact version \"${${_NAME}_FIND_VERSION}\"")
-            set(VERSION_OK FALSE)
-          else ()
-            set(VERSION_MSG "(found suitable exact version \"${_FOUND_VERSION}\")")
-          endif ()
-          unset(_VERSION_HEAD)
-        else ()
-          if (NOT ${_NAME}_FIND_VERSION VERSION_EQUAL _FOUND_VERSION)
-            set(VERSION_MSG "Found unsuitable version \"${_FOUND_VERSION}\", but required is exact version \"${${_NAME}_FIND_VERSION}\"")
-            set(VERSION_OK FALSE)
-          else ()
-            set(VERSION_MSG "(found suitable exact version \"${_FOUND_VERSION}\")")
-          endif ()
-        endif ()
-        unset(_VERSION_DOTS)
-
-      else()     # minimum version specified:
-        if (${_NAME}_FIND_VERSION VERSION_GREATER _FOUND_VERSION)
-          set(VERSION_MSG "Found unsuitable version \"${_FOUND_VERSION}\", but required is at least \"${${_NAME}_FIND_VERSION}\"")
-          set(VERSION_OK FALSE)
-        else ()
-          set(VERSION_MSG "(found suitable version \"${_FOUND_VERSION}\", minimum required is \"${${_NAME}_FIND_VERSION}\")")
-        endif ()
-      endif()
-
-    else()
-
-      # if the package was not found, but a version was given, add that to the output:
-      if(${_NAME}_FIND_VERSION_EXACT)
-         set(VERSION_MSG "(Required is exact version \"${${_NAME}_FIND_VERSION}\")")
-      else()
-         set(VERSION_MSG "(Required is at least version \"${${_NAME}_FIND_VERSION}\")")
-      endif()
-
-    endif()
-  else ()
-    # Check with DEFINED as the found version may be 0.
-    if(DEFINED ${FPHSA_VERSION_VAR})
-      set(VERSION_MSG "(found version \"${${FPHSA_VERSION_VAR}}\")")
-    endif()
-  endif ()
-
-  if(VERSION_OK)
-    string(APPEND DETAILS "[v${${FPHSA_VERSION_VAR}}(${${_NAME}_FIND_VERSION})]")
-  else()
-    set(${_NAME}_FOUND FALSE)
-  endif()
-
-
-  # print the result:
-  if (${_NAME}_FOUND)
-    FIND_PACKAGE_MESSAGE(${_NAME} "Found ${_NAME}: ${${_FIRST_REQUIRED_VAR}} ${VERSION_MSG} ${COMPONENT_MSG}" "${DETAILS}")
-  else ()
-
-    if(FPHSA_CONFIG_MODE)
-      _FPHSA_HANDLE_FAILURE_CONFIG_MODE()
-    else()
-      if(NOT VERSION_OK)
-        _FPHSA_FAILURE_MESSAGE("${FPHSA_FAIL_MESSAGE}: ${VERSION_MSG} (found ${${_FIRST_REQUIRED_VAR}})")
-      else()
-        _FPHSA_FAILURE_MESSAGE("${FPHSA_FAIL_MESSAGE} (missing:${MISSING_VARS}) ${VERSION_MSG}")
-      endif()
-    endif()
-
-  endif ()
-
-  set(${_NAME}_FOUND ${${_NAME}_FOUND} PARENT_SCOPE)
-  set(${_NAME_UPPER}_FOUND ${${_NAME}_FOUND} PARENT_SCOPE)
-endfunction()

CMakeModules/FindPackageMessage.cmake

@@ -1,48 +0,0 @@
-# Distributed under the OSI-approved BSD 3-Clause License.  See accompanying
-# file Copyright.txt or https://cmake.org/licensing for details.
-
-#[=======================================================================[.rst:
-FindPackageMessage
-------------------
-
-.. code-block:: cmake
-
-  find_package_message(<name> "message for user" "find result details")
-
-This function is intended to be used in FindXXX.cmake modules files.
-It will print a message once for each unique find result.  This is
-useful for telling the user where a package was found.  The first
-argument specifies the name (XXX) of the package.  The second argument
-specifies the message to display.  The third argument lists details
-about the find result so that if they change the message will be
-displayed again.  The macro also obeys the QUIET argument to the
-find_package command.
-
-Example:
-
-.. code-block:: cmake
-
-  if(X11_FOUND)
-    find_package_message(X11 "Found X11: ${X11_X11_LIB}"
-      "[${X11_X11_LIB}][${X11_INCLUDE_DIR}]")
-  else()
-   ...
-  endif()
-#]=======================================================================]
-
-function(find_package_message pkg msg details)
-  # Avoid printing a message repeatedly for the same find result.
-  if(NOT ${pkg}_FIND_QUIETLY)
-    string(REPLACE "\n" "" details "${details}")
-    set(DETAILS_VAR FIND_PACKAGE_MESSAGE_DETAILS_${pkg})
-    if(NOT "${details}" STREQUAL "${${DETAILS_VAR}}")
-      # The message has not yet been printed.
-      message(STATUS "${msg}")
-
-      # Save the find details in the cache to avoid printing the same
-      # message again.
-      set("${DETAILS_VAR}" "${details}"
-        CACHE INTERNAL "Details about finding ${pkg}")
-    endif()
-  endif()
-endfunction()

Docker/build.sh

@@ -1,15 +1,14 @@
-rm ./sist2
-cp ../sist2 .
+rm ./sist2 sist2_debug
+cp ../sist2.gz .
+gzip -d sist2.gz
 strip sist2
 
 version=$(./sist2 --version)
 
 echo "Version ${version}"
-docker build . -t simon987/sist2:${version} -t simon987/sist2:latest \
-  -t docker.pkg.github.com/simon987/sist2/sist2:latest -t docker.pkg.github.com/simon987/sist2/sist2:${version}
+docker build . -t simon987/sist2:${version} -t simon987/sist2:latest
+
 docker push simon987/sist2:${version}
 docker push simon987/sist2:latest
-docker push docker.pkg.github.com/simon987/sist2/sist2:latest
-docker push docker.pkg.github.com/simon987/sist2/sist2:${version}
 
-docker run --rm -it simon987/sist2 -v
+docker run --rm simon987/sist2 -v

DockerArm64/Dockerfile

@@ -0,0 +1,22 @@
+FROM ubuntu:19.10
+MAINTAINER simon987 <me@simon987.net>
+
+RUN apt update
+RUN apt install -y libglib2.0-0 libcurl4 libmagic1 libharfbuzz-bin libopenjp2-7 libarchive13 liblzma5 libzstd1 liblz4-1 \
+ curl libtiff5 libpng16-16 libpcre3
+
+RUN mkdir -p /usr/share/tessdata && \
+    cd /usr/share/tessdata/ && \
+    curl -o /usr/share/tessdata/hin.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/hin.traineddata &&\
+    curl -o /usr/share/tessdata/jpn.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/jpn.traineddata &&\
+    curl -o /usr/share/tessdata/eng.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/eng.traineddata &&\
+    curl -o /usr/share/tessdata/fra.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/fra.traineddata &&\
+    curl -o /usr/share/tessdata/rus.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/rus.traineddata &&\
+    curl -o /usr/share/tessdata/spa.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/spa.traineddata && ls -lh
+
+ADD sist2_arm64 /root/sist2
+
+ENV LANG C.UTF-8
+ENV LC_ALL C.UTF-8
+
+ENTRYPOINT ["/root/sist2"]

DockerArm64/build.sh

@@ -0,0 +1,13 @@
+rm ./sist2_arm64
+cp ../sist2_arm64.gz .
+gzip -d sist2_arm64.gz
+
+version=$(./sist2_arm64 --version)
+
+echo "Version ${version}"
+docker build . -t simon987/sist2-arm64:"${version}" -t simon987/sist2-arm64:latest
+
+docker push simon987/sist2-arm64:"${version}"
+docker push simon987/sist2-arm64:latest
+
+docker run --rm simon987/sist2-arm64 -v

README.md

@@ -1,6 +1,8 @@
 ![GitHub](https://img.shields.io/github/license/simon987/sist2.svg)
 [![CodeFactor](https://www.codefactor.io/repository/github/simon987/sist2/badge?s=05daa325188aac4eae32c786f3d9cf4e0593f822)](https://www.codefactor.io/repository/github/simon987/sist2)
-[![Development snapshots](https://ci.simon987.net/app/rest/builds/buildType(Sist2_Build)/statusIcon)](https://files.simon987.net/artifacts/Sist2/Build/)
+[![Development snapshots](https://ci.simon987.net/api/badges/simon987/sist2/status.svg)](https://files.simon987.net/sist2/simon987_sist2/)
+
+**Demo**: [sist2.simon987.net](https://sist2.simon987.net/?i=Demo%20files)
 
 # sist2
 
@@ -8,25 +10,29 @@ sist2 (Simple incremental search tool)
 
 *Warning: sist2 is in early development*
 
-![sist2.png](sist2.png)
+![sist2.png](docs/sist2.png)
 
 ## Features
 
 * Fast, low memory usage, multi-threaded
 * Mobile-friendly Web interface
 * Portable (all its features are packaged in a single executable)
-* Extracts text from common file types \*
+* Extracts text and metadata from common file types \*
 * Generates thumbnails \*
 * Incremental scanning
-* Automatic tagging from file attributes via [user scripts](scripting/README.md)
+* Manual tagging from the UI and automatic tagging based on file attributes via [user scripts](docs/scripting.md)
 * Recursive scan inside archive files \*\*
 * OCR support with tesseract \*\*\*
+* Stats page & disk utilisation visualization
 
 
 \* See [format support](#format-support)    
 \*\* See [Archive files](#archive-files)    
 \*\*\* See [OCR](#ocr)    
 
+![stats](docs/stats.png)
+
+
 ## Getting Started
 
 1. Have an Elasticsearch (>= 6.X.X) instance running
@@ -46,10 +52,10 @@ sist2 (Simple incremental search tool)
         ```
 1. Download sist2 executable
     1. Download the [latest sist2 release](https://github.com/simon987/sist2/releases) *
-    1. *(or)* Download a [development snapshot](https://files.simon987.net/artifacts/Sist2/Build/) *(Not recommended!)*
+    1. *(or)* Download a [development snapshot](https://files.simon987.net/sist2/simon987_sist2/) *(Not recommended!)*
     1. *(or)* `docker pull simon987/sist2:latest`
 
-1. See [Usage guide](USAGE.md)
+1. See [Usage guide](docs/USAGE.md)
    
 
 \* *Windows users*: **sist2** runs under [WSL](https://en.wikipedia.org/wiki/Windows_Subsystem_for_Linux)    
@@ -57,7 +63,7 @@ sist2 (Simple incremental search tool)
 
 ## Example usage
 
-See [Usage guide](USAGE.md) for more details
+See [Usage guide](docs/USAGE.md) for more details
 
 1. Scan a directory: `sist2 scan ~/Documents -o ./docs_idx`
 1. Push index to Elasticsearch: `sist2 index ./docs_idx`
@@ -68,14 +74,19 @@ See [Usage guide](USAGE.md) for more details
 
 File type | Library | Content | Thumbnail | Metadata
 :---|:---|:---|:---|:---
-pdf,xps,cbz,cbr,fb2,epub | MuPDF | text+ocr | yes, `png` | title |
-`audio/*` | ffmpeg | - | yes, `jpeg` | ID3 tags |
-`video/*` | ffmpeg | - | yes, `jpeg` | title, comment, artist |
-`image/*` | ffmpeg | - | yes, `jpeg` | [Common EXIF tags](https://github.com/simon987/sist2/blob/efdde2734eca9b14a54f84568863b7ffd59bdba3/src/parsing/media.c#L190) |
+pdf,xps,fb2,epub | MuPDF | text+ocr | yes | author, title |
+cbz,cbr | *(none)* | - | yes | - |
+`audio/*` | ffmpeg | - | yes | ID3 tags |
+`video/*` | ffmpeg | - | yes | title, comment, artist |
+`image/*` | ffmpeg | - | yes | [Common EXIF tags](https://github.com/simon987/sist2/blob/efdde2734eca9b14a54f84568863b7ffd59bdba3/src/parsing/media.c#L190) |
+raw, rw2, dng, cr2, crw, dcr, k25, kdc, mrw, pef, xf3, arw, sr2, srf, erf  | LibRaw | - | yes | Common EXIF tags |
 ttf,ttc,cff,woff,fnt,otf | Freetype2 | - | yes, `bmp` | Name & style |
 `text/plain` | *(none)* | yes | no | - |
+html, xml | *(none)* | yes | no | - |
 tar, zip, rar, 7z, ar ...  | Libarchive | yes\* | - | no |
-docx, xlsx, pptx | libOPC | yes | no | no |
+docx, xlsx, pptx | *(none)* | yes | if embedded | creator, modified_by, title |
+doc (MS Word 97-2003) | antiword | yes | yes | author, title |
+mobi, azw, azw3 | libmobi | yes | no | author, title |
 
 \* *See [Archive files](#archive-files)*
  
@@ -85,18 +96,16 @@ they were directly in the file system. Recursive (archives inside archives)
 scan is also supported.
 
 **Limitations**:
-* Parsing media files with formats that require
-*seek* (e.g. `.gif`, `.mp4` w/ fragmented metadata etc.) is not supported.
+* Support for parsing media files with formats that require *seek* (e.g. `.gif`, `.mp4` w/ fragmented metadata etc.) 
+  is limitted (see `--mem-buffer` option)
 * Archive files are scanned sequentially, by a single thread. On systems where
 **sist2** is not I/O bound, scans might be faster when larger archives are split
  into smaller parts.
-
-To check if a media file can be parsed without *seek*, execute `cat file.mp4 | ffprobe -`
  
  
 ### OCR
 
-You can enable OCR support for pdf,xps,cbz,cbr,fb2,epub file types with the
+You can enable OCR support for pdf,xps,fb2,epub file types with the
 `--ocr <lang>` option. Download the language data files with your
 package manager (`apt install tesseract-ocr-eng`) or directly [from Github](https://github.com/tesseract-ocr/tesseract/wiki/Data-Files).
 
@@ -113,23 +122,17 @@ sist2 scan --ocr eng ~/Books/Textbooks/
 ## Build from source
 
 You can compile **sist2** by yourself if you don't want to use the pre-compiled
-binaries.
+binaries (GCC 7+ required).
 
 1. Install compile-time dependencies
 
-    *(Debian)*
-    ```bash
-    apt install git cmake pkg-config libglib2.0-dev \
-        libssl-dev uuid-dev python3 libmagic-dev libfreetype6-dev \
-        libcurl4-openssl-dev libbz2-dev yasm libharfbuzz-dev ragel \
-        libarchive-dev libtiff5 libpng16-16 libpango1.0-dev \
-        libxml2-dev libopenjp2-7-dev libleptonica-dev
+   ```bash
+   vcpkg install lmdb cjson glib libarchive[core,bzip2,libxml2,lz4,lzma,lzo] pthread tesseract libxml2 ffmpeg zstd gtest mongoose libmagic libraw curl[core,ssl] jbig2dec brotli libmupdf
    ```
 
 2. Build
     ```bash
-    git clone --recurse-submodules https://github.com/simon987/sist2
-    ./scripts/get_static_libs.sh
-    cmake .
+    git clone --recursive https://github.com/simon987/sist2/
+    cmake -DSIST_DEBUG=off -DCMAKE_TOOLCHAIN_FILE=<VCPKG_ROOT>/scripts/buildsystems/vcpkg.cmake .
     make
     ```

USAGE.md

@@ -1,275 +0,0 @@
-# Usage
-
-*More examples (specifically with docker/compose) are in progress*
-
-* [scan](#scan)
-    * [options](#scan-options)
-    * [examples](#scan-examples)
-    * [index format](#index-format)
-* [index](#index)
-    * [options](#index-options)
-    * [examples](#index-examples)
-* [web](#web)
-    * [options](#web-options)
-    * [examples](#web-examples)
-    * [rewrite_url](#rewrite_url)
-    * [link to specific indices](#link-to-specific-indices)
-
-```
-Usage: sist2 scan [OPTION]... PATH
-   or: sist2 index [OPTION]... INDEX
-   or: sist2 web [OPTION]... INDEX...
-Lightning-fast file system indexer and search tool.
-
-    -h, --help            show this help message and exit
-    -v, --version         Show version and exit
-    --verbose             Turn on logging
-    --very-verbose        Turn on debug messages
-
-Scan options
-    -t, --threads=<int>   Number of threads. DEFAULT=1
-    -q, --quality=<flt>   Thumbnail quality, on a scale of 1.0 to 31.0, 1.0 being the best. DEFAULT=5
-    --size=<int>          Thumbnail size, in pixels. Use negative value to disable. DEFAULT=500
-    --content-size=<int>  Number of bytes to be extracted from text documents. Use negative value to disable. DEFAULT=32768
-    --incremental=<str>   Reuse an existing index and only scan modified files.
-    -o, --output=<str>    Output directory. DEFAULT=index.sist2/
-    --rewrite-url=<str>   Serve files from this url instead of from disk.
-    --name=<str>          Index display name. DEFAULT: (name of the directory)
-    --depth=<int>         Scan up to DEPTH subdirectories deep. Use 0 to only scan files in PATH. DEFAULT: -1
-    --archive=<str>       Archive file mode (skip|list|shallow|recurse). skip: Don't parse, list: only get file names as text, shallow: Don't parse archives inside archives. DEFAULT: recurse
-    --ocr=<str>           Tesseract language (use tesseract --list-langs to see which are installed on your machine)
-    -e, --exclude=<str>   Files that match this regex will not be scanned
-    --fast                Only index file names & mime type
-
-Index options
-    --es-url=<str>        Elasticsearch url with port. DEFAULT=http://localhost:9200
-    -p, --print           Just print JSON documents to stdout.
-    --script-file=<str>   Path to user script.
-    --batch-size=<int>    Index batch size. DEFAULT: 100
-    -f, --force-reset     Reset Elasticsearch mappings and settings. (You must use this option the first time you use the index command)
-
-Web options
-    --es-url=<str>        Elasticsearch url. DEFAULT=http://localhost:9200
-    --bind=<str>          Listen on this address. DEFAULT=localhost
-    --port=<str>          Listen on this port. DEFAULT=4090
-    --auth=<str>          Basic auth in user:password format
-Made by simon987 <me@simon987.net>. Released under GPL-3.0
-
-```
-
-## Scan
-
-### Scan options
-
-* `-t, --threads` 
-      Number of threads for file parsing. **Do not set a number higher than `$(nproc)`!**.
-* `-q, --quality` 
-    Thumbnail quality, on a scale of 1.0 to 31.0, 1.0 being the best. *Does not affect PDF thumbnails quality*
-* `--size` 
-    Thumbnail size in pixels.
-* `--content-size` 
-    Number of bytes of text to be extracted from the content of files (plain text and PDFs).
-    Repeated whitespace and special characters do not count toward this limit.
-* `--incremental`
-    Specify an existing index. Information about files in this index that were not modified (based on *mtime* attribute)
-    will be copied to the new index and will not be parsed again.
-* `-o, --output` Output directory. 
-* `--rewrite-url` Set the `rewrite_url` option for the web module (See [rewrite_url](#rewrite_url)) 
-* `--name` Set the `name` option for the web module
-* `--depth` Maximum scan dept. Set to 0 only scan files directly in the root directory, set to -1 for infinite depth
-* `--archive` Archive file mode.
-    * skip: Don't parse
-    * list: Only get file names as text
-    * shallow: Don't parse archives inside archives.
-    * recurse: Scan archives recursively (default)
-* `--ocr` See [OCR](README.md#OCR)
-* `-e, --exclude` Regex pattern to exclude files. A file is excluded if the pattern matches any 
-    part of the full absolute path.
-    
-    Examples: 
-    * `-e ".*\.ttf"`: Ignore ttf files
-    * `-e ".*\.(ttf|rar)"`: Ignore ttf and rar files
-    * `-e "^/mnt/backups/"`: Ignore all files in the `/mnt/backups/` directory
-    * `-e "^/mnt/Data[12]/"`: Ignore all files in the `/mnt/Data1/` and `/mnt/Data2/` directory
-    * `-e "(^/usr/)|(^/var/)|(^/media/DRIVE-A/tmp/)|(^/media/DRIVE-B/Trash/)"` Exclude the
-     `/usr`, `/var`, `/media/DRIVE-A/tmp`, `/media/DRIVE-B/Trash` directories
-* `--fast` Only index file names and mime type
-
-### Scan examples
-
-Simple scan
-```bash
-sist2 scan ~/Documents
-
-sist2 scan \
-    --threads 4 --content-size 16000000 --quality 1.0 --archive shallow \
-    --name "My Documents" --rewrite-url "http://nas.domain.local/My Documents/" \
-    ~/Documents -o ./documents.idx/
-```
-
-Incremental scan
-```
-sist2 scan --incremental ./orig_idx/ -o ./updated_idx/ ~/Documents
-```
-
-### Index format
-
-A typical `binary` type index structure looks like this:
-```
-documents.idx/
-├── descriptor.json
-├── _index_139965416830720
-├── _index_139965425223424
-├── _index_139965433616128
-├── _index_139965442008832
-└── thumbs
-    ├── data.mdb
-    └── lock.mdb
-```
-
-The `_index_*` files contain the raw binary index data and are not meant to be
-read by other applications. The format is generally compatible across different 
-sist2 versions.
-
-The `thumbs/` folder is a [LMDB](https://en.wikipedia.org/wiki/Lightning_Memory-Mapped_Database)
-database containing the thumbnails.
-
-The `descriptor.json` file contains general information about the index. The 
-following fields are safe to modify manually: `root`, `name`, [rewrite_url](#rewrite_url) and `timestamp`.
-
-
-*Advanced usage*
-
-Instead of using the `scan` module, you can also import an index generated
-by a third party application. The 'external' index must have the following format:
-
-```
-my_index/
-├── descriptor.json
-├── _index_0
-└── thumbs
-    ├── data.mdb
-    └── lock.mdb
-```
-
-*descriptor.json*:
-```json
-{
-    "uuid": "<valid UUID4>",
-    "version": "_external_v1",
-    "root": "(optional)",
-    "name": "<name>",
-    "rewrite_url": "(optional)",
-    "type": "json",
-    "timestamp": 1578971024
-}
-```
-
-*_index_0*: NDJSON format (One json object per line)
-
-```json
-{
-  "_id": "unique uuid for the file",
-  "index": "index uuid4 (same one as descriptor.json!)",
-  "mime": "application/x-cbz",
-  "size": 14341204,
-  "mtime": 1578882996,
-  "extension": "cbz",
-  "name": "my_book",
-  "path": "path/to/books",
-  "content": "text contents of the book",
-  "title": "Title of the book",
-  "tag": ["genre.fiction", "author.someguy", "etc..."],
-  "_keyword": [
-    {"k": "ISBN", "v": "ABCD34789231"}
-  ],
-  "_text": [
-    {"k": "other", "v": "This will be indexed as text"}
-  ]
-}
-```
-
-You can find the full list of supported fields [here](src/io/serialize.c#L90)
-
-The `_keyword.*` items will be indexed and searchable as **keyword** fields (only full matches allowed).
-The `_text.*` items will be indexed and searchable as **text** fields (fuzzy searching allowed)
-
-
-*thumbs/*:
-
-LMDB key-value store. Keys are **binary** 128-bit UUID4s (`_id` field)
-and values are raw image bytes.
-
-Importing an external `binary` type index is technically possible but
-it is currently unsupported and has no guaranties of back/forward compatibility.
-
-
-## Index
-### Index options
- * `--es-url` 
- Elasticsearch url and port. If you are using docker, make sure that both containers are on the
- same network.
- * `-p, --print` 
-    Print index in JSON format to stdout.
- * `--script-file` 
-    Path to user script. See [Scripting](scripting/README.md).
- * `--batch-size=<int>` 
-    Index batch size. Indexing is generally faster with larger batches, but payloads that
-    are too large will fail and additional overhead for retrying with smaller sizes may slow
-    down the process.
- * `-f, --force-reset` 
-    Reset Elasticsearch mappings and settings.
-    **(You must use this option the first time you use the index command)**.
-    
-### Index examples
-
-**Push to elasticsearch**
-```bash
-sist2 index --force-reset --batch-size 1000 --es-url http://localhost:9200 ./my_index/
-sist2 index ./my_index/
-```
-
-**Save index in JSON format**
-```bash
-sist2 index --print ./my_index/ > my_index.ndjson
-```
-
-**Inspect contents of an index**
-```bash
-sist2 index --print ./my_index/ | jq | less
-```
-
-## Web
-
-### Web options
- * `--es-url=<str>` Elasticsearch url.
- * `--bind=<str>` Listen on this address.
- * `--port=<str>` Listen on this port.
- * `--auth=<str>` Basic auth in user:password format
- 
-### Web examples
-
-**Single index**
-```bash
-sist2 web --auth admin:hunter2 --bind 0.0.0.0 --port 8888 my_index
-```
-
-**Multiple indices**
-```bash
-# Indices will be displayed in this order in the web interface
-sist2 web index1 index2 index3 index4
-```
-
-### rewrite_url
-
-When the `rewrite_url` field is not empty, the web module ignores the `root`
-field and will return a HTTP redirect to `<rewrite_url><path>/<name><extension>`
-instead of serving the file from disk. 
-Both the `root` and `rewrite_url` fields are safe to manually modify from the 
-`descriptor.json` file.
-
-### Link to specific indices
-
-To link to specific indices, you can add a list of comma-separated index name to 
-the URL: `?i=<name>,<name>`. By default, indices with `"(nsfw)"` in their name are
-not displayed.

ci/build.sh

@@ -1,7 +1,20 @@
 #!/usr/bin/env bash
 
-./scripts/get_static_libs.sh
+VCPKG_ROOT="/vcpkg"
 
-cmake .
-make
+rm *.gz
+
+git submodule update --init --recursive
+
+rm -rf CMakeFiles CMakeCache.txt
+cmake -DSIST_DEBUG=off -DCMAKE_TOOLCHAIN_FILE="${VCPKG_ROOT}/scripts/buildsystems/vcpkg.cmake" .
+make -j 33
 strip sist2
+mv sist2 sist2-x64-linux
+
+rm -rf CMakeFiles CMakeCache.txt
+cmake -DSIST_DEBUG=on -DCMAKE_TOOLCHAIN_FILE="${VCPKG_ROOT}/scripts/buildsystems/vcpkg.cmake" .
+make -j 33
+cp /usr/lib/x86_64-linux-gnu/libasan.so.2.0.0 libasan.so.2
+mv sist2_debug sist2-x64-linux-debug
+tar -czf sist2-x64-linux-debug.tar.gz sist2-x64-linux-debug libasan.so.2

ci/build_arm64.sh

@@ -0,0 +1,13 @@
+#!/usr/bin/env bash
+
+VCPKG_ROOT="/vcpkg"
+
+rm *.gz
+
+git submodule update --init --recursive
+
+rm -rf CMakeFiles CMakeCache.txt
+cmake -DSIST_DEBUG=off -DCMAKE_TOOLCHAIN_FILE="${VCPKG_ROOT}/scripts/buildsystems/vcpkg.cmake" .
+make -j 4
+strip sist2
+mv sist2 sist2-arm64-linux

docs/USAGE.md

@@ -0,0 +1,403 @@
+# Usage
+
+*More examples (specifically with docker/compose) are in progress*
+
+* [scan](#scan)
+    * [options](#scan-options)
+    * [examples](#scan-examples)
+    * [index format](#index-format)
+* [index](#index)
+    * [options](#index-options)
+    * [examples](#index-examples)
+* [web](#web)
+    * [options](#web-options)
+    * [examples](#web-examples)
+    * [rewrite_url](#rewrite_url)
+    * [link to specific indices](#link-to-specific-indices)
+* [exec-script](#exec-script)
+* [tagging](#tagging)
+* [sidecar files](#sidecar-files)
+
+```
+Usage: sist2 scan [OPTION]... PATH
+   or: sist2 index [OPTION]... INDEX
+   or: sist2 web [OPTION]... INDEX...
+   or: sist2 exec-script [OPTION]... INDEX
+Lightning-fast file system indexer and search tool.
+
+    -h, --help                    show this help message and exit
+    -v, --version                 Show version and exit
+    --verbose                     Turn on logging
+    --very-verbose                Turn on debug messages
+
+Scan options
+    -t, --threads=<int>           Number of threads. DEFAULT=1
+    -q, --quality=<flt>           Thumbnail quality, on a scale of 1.0 to 31.0, 1.0 being the best. DEFAULT=5
+    --size=<int>                  Thumbnail size, in pixels. Use negative value to disable. DEFAULT=500
+    --content-size=<int>          Number of bytes to be extracted from text documents. Use negative value to disable. DEFAULT=32768
+    --incremental=<str>           Reuse an existing index and only scan modified files.
+    -o, --output=<str>            Output directory. DEFAULT=index.sist2/
+    --rewrite-url=<str>           Serve files from this url instead of from disk.
+    --name=<str>                  Index display name. DEFAULT: (name of the directory)
+    --depth=<int>                 Scan up to DEPTH subdirectories deep. Use 0 to only scan files in PATH. DEFAULT: -1
+    --archive=<str>               Archive file mode (skip|list|shallow|recurse). skip: Don't parse, list: only get file names as text, shallow: Don't parse archives inside archives. DEFAULT: recurse
+    --ocr=<str>                   Tesseract language (use tesseract --list-langs to see which are installed on your machine)
+    -e, --exclude=<str>           Files that match this regex will not be scanned
+    --fast                        Only index file names & mime type
+    --treemap-threshold=<str>     Relative size threshold for treemap (see USAGE.md). DEFAULT: 0.0005
+    --mem-buffer=<int>            Maximum memory buffer size per thread in MB for files inside archives (see USAGE.md). DEFAULT: 2000
+
+Index options
+    -t, --threads=<int>           Number of threads. DEFAULT=1
+    --es-url=<str>                Elasticsearch url with port. DEFAULT=http://localhost:9200
+    --es-index=<str>              Elasticsearch index name. DEFAULT=sist2
+    -p, --print                   Just print JSON documents to stdout.
+    --script-file=<str>           Path to user script.
+    --mappings-file=<str>         Path to Elasticsearch mappings.
+    --settings-file=<str>         Path to Elasticsearch settings.
+    --async-script                Execute user script asynchronously.
+    --batch-size=<int>            Index batch size. DEFAULT: 100
+    -f, --force-reset             Reset Elasticsearch mappings and settings. (You must use this option the first time you use the index command)
+
+Web options
+    --es-url=<str>                Elasticsearch url. DEFAULT=http://localhost:9200
+    --es-index=<str>              Elasticsearch index name. DEFAULT=sist2
+    --bind=<str>                  Listen on this address. DEFAULT=localhost:4090
+    --auth=<str>                  Basic auth in user:password format
+    --tag-auth=<str>              Basic auth in user:password format for tagging
+
+Exec-script options
+    --es-url=<str>                Elasticsearch url. DEFAULT=http://localhost:9200
+    --es-index=<str>              Elasticsearch index name. DEFAULT=sist2
+    --script-file=<str>           Path to user script.
+    --async-script                Execute user script asynchronously.
+Made by simon987 <me@simon987.net>. Released under GPL-3.0
+```
+
+## Scan
+
+### Scan options
+
+* `-t, --threads` 
+      Number of threads for file parsing. **Do not set a number higher than `$(nproc)` or `$(Get-WmiObject Win32_ComputerSystem).NumberOfLogicalProcessors` in Windows!**
+* `-q, --quality` 
+    Thumbnail quality, on a scale of 1.0 to 31.0, 1.0 being the best. *Does not affect PDF thumbnails quality*
+* `--size` 
+    Thumbnail size in pixels.
+* `--content-size` 
+    Number of bytes of text to be extracted from the content of files (plain text and PDFs).
+    Repeated whitespace and special characters do not count toward this limit.
+* `--incremental`
+    Specify an existing index. Information about files in this index that were not modified (based on *mtime* attribute)
+    will be copied to the new index and will not be parsed again.
+* `-o, --output` Output directory. 
+* `--rewrite-url` Set the `rewrite_url` option for the web module (See [rewrite_url](#rewrite_url)) 
+* `--name` Set the `name` option for the web module
+* `--depth` Maximum scan dept. Set to 0 only scan files directly in the root directory, set to -1 for infinite depth
+* `--archive` Archive file mode.
+    * skip: Don't parse
+    * list: Only get file names as text
+    * shallow: Don't parse archives inside archives.
+    * recurse: Scan archives recursively (default)
+* `--ocr` See [OCR](../README.md#OCR)
+* `-e, --exclude` Regex pattern to exclude files. A file is excluded if the pattern matches any 
+    part of the full absolute path.
+    
+    Examples: 
+    * `-e ".*\.ttf"`: Ignore ttf files
+    * `-e ".*\.(ttf|rar)"`: Ignore ttf and rar files
+    * `-e "^/mnt/backups/"`: Ignore all files in the `/mnt/backups/` directory
+    * `-e "^/mnt/Data[12]/"`: Ignore all files in the `/mnt/Data1/` and `/mnt/Data2/` directory
+    * `-e "(^/usr/)|(^/var/)|(^/media/DRIVE-A/tmp/)|(^/media/DRIVE-B/Trash/)"` Exclude the
+     `/usr`, `/var`, `/media/DRIVE-A/tmp`, `/media/DRIVE-B/Trash` directories
+* `--fast` Only index file names and mime type
+* `--treemap-threshold` Directories smaller than (`treemap-threshold` * `<total size of the index>`)
+    will not be considered for the disk utilisation visualization; their size will be added to
+    the parent directory. If the parent directory is still smaller than the threshold, it will also be "merged upwards"
+    and so on.
+    
+    In effect, smaller `treemap-threshold` values will yield a more detailed 
+    (but also a more cluttered and harder to read) visualization. 
+    
+* `--mem-buffer` Maximum memory buffer size in MB (per thread) for files inside archives. Media files 
+    larger than this number will be read sequentially and no *seek* operations will be supported.
+
+    To check if a media file can be parsed without *seek*, execute `cat file.mp4 | ffprobe -`
+
+### Scan examples
+
+Simple scan
+```bash
+sist2 scan ~/Documents
+
+sist2 scan \
+    --threads 4 --content-size 16000000 --quality 1.0 --archive shallow \
+    --name "My Documents" --rewrite-url "http://nas.domain.local/My Documents/" \
+    ~/Documents -o ./documents.idx/
+```
+
+Incremental scan
+```
+sist2 scan --incremental ./orig_idx/ -o ./updated_idx/ ~/Documents
+```
+
+### Index format
+
+A typical `binary` type index structure looks like this:
+```
+documents.idx/
+├── descriptor.json
+├── _index_139965416830720
+├── _index_139965425223424
+├── _index_139965433616128
+├── _index_139965442008832
+├── _index_139965442008832
+├── treemap.csv
+├── agg_mime.csv
+├── agg_date.csv
+├── add_size.csv
+├── thumbs/
+|   ├── data.mdb
+|   └── lock.mdb
+├── tags/
+|   ├── data.mdb
+|   └── lock.mdb
+└── meta/
+    ├── data.mdb
+    └── lock.mdb
+```
+
+The `_index_*` files contain the raw binary index data and are not meant to be
+read by other applications. The format is generally compatible across different 
+sist2 versions.
+
+The `thumbs/` folder is a [LMDB](https://en.wikipedia.org/wiki/Lightning_Memory-Mapped_Database)
+database containing the thumbnails.
+
+The `descriptor.json` file contains general information about the index. The 
+following fields are safe to modify manually: `root`, `name`, [rewrite_url](#rewrite_url) and `timestamp`.
+
+The `.csv` are pre-computed aggregations necessary for the stats page.
+
+
+*Advanced usage*
+
+Instead of using the `scan` module, you can also import an index generated
+by a third party application. The 'external' index must have the following format:
+
+```
+my_index/
+├── descriptor.json
+├── _index_0
+└── thumbs/
+|   ├── data.mdb
+|   └── lock.mdb
+└── meta/
+    └── <empty>
+```
+
+*descriptor.json*:
+```json
+{
+    "uuid": "<valid UUID4>",
+    "version": "_external_v1",
+    "root": "(optional)",
+    "name": "<name>",
+    "rewrite_url": "(optional)",
+    "type": "json",
+    "timestamp": 1578971024
+}
+```
+
+*_index_0*: NDJSON format (One json object per line)
+
+```json
+{
+  "_id": "unique uuid for the file",
+  "index": "index uuid4 (same one as descriptor.json!)",
+  "mime": "application/x-cbz",
+  "size": 14341204,
+  "mtime": 1578882996,
+  "extension": "cbz",
+  "name": "my_book",
+  "path": "path/to/books",
+  "content": "text contents of the book",
+  "title": "Title of the book",
+  "tag": ["genre.fiction", "author.someguy", "etc..."],
+  "_keyword": [
+    {"k": "ISBN", "v": "ABCD34789231"}
+  ],
+  "_text": [
+    {"k": "other", "v": "This will be indexed as text"}
+  ]
+}
+```
+
+You can find the full list of supported fields [here](../src/io/serialize.c#L90)
+
+The `_keyword.*` items will be indexed and searchable as **keyword** fields (only full matches allowed).
+The `_text.*` items will be indexed and searchable as **text** fields (fuzzy searching allowed)
+
+
+*thumbs/*:
+
+LMDB key-value store. Keys are **binary** 16-byte md5 hash* (`_id` field)
+and values are raw image bytes.
+
+*\* Hash is calculated from the full path of the file, including the extension, relative to the index root*
+
+Importing an external `binary` type index is technically possible but
+it is currently unsupported and has no guaranties of back/forward compatibility.
+
+
+## Index
+### Index options
+ * `--es-url` 
+ Elasticsearch url and port. If you are using docker, make sure that both containers are on the
+ same network.
+ * `--es-index` 
+    Elasticsearch index name. DEFAULT=sist2
+ * `-p, --print` 
+    Print index in JSON format to stdout.
+ * `--script-file` 
+    Path to user script. See [Scripting](scripting.md).
+ * `--mappings-file`
+    Path to custom Elasticsearch mappings. If none is specified, [the bundled mappings](https://github.com/simon987/sist2/tree/master/schema) will be used.
+ * `--settings-file`
+    Path to custom Elasticsearch settings. *(See above)*
+ * `--async-script` 
+    Use `wait_for_completion=false` elasticsearch option while executing user script.
+     (See [Elasticsearch documentation](https://www.elastic.co/guide/en/elasticsearch/reference/current/tasks.html))
+ * `--batch-size=<int>` 
+    Index batch size. Indexing is generally faster with larger batches, but payloads that
+    are too large will fail and additional overhead for retrying with smaller sizes may slow
+    down the process.
+ * `-f, --force-reset` 
+    Reset Elasticsearch mappings and settings.
+    
+### Index examples
+
+**Push to elasticsearch**
+```bash
+sist2 index --force-reset --batch-size 1000 --es-url http://localhost:9200 ./my_index/
+sist2 index ./my_index/
+```
+
+**Save index in JSON format**
+```bash
+sist2 index --print ./my_index/ > my_index.ndjson
+```
+
+**Inspect contents of an index**
+```bash
+sist2 index --print ./my_index/ | jq | less
+```
+
+## Web
+
+### Web options
+ * `--es-url=<str>` Elasticsearch url.
+ * `--es-index` 
+    Elasticsearch index name. DEFAULT=sist2
+ * `--bind=<str>` Listen on this address.
+ * `--auth=<str>` Basic auth in user:password format
+ * `--tag-auth=<str>` Basic auth in user:password format. Works the same way as the 
+    `--auth` argument, but authentication is only applied the `/tag/` endpoint.
+ 
+### Web examples
+
+**Single index**
+```bash
+sist2 web --auth admin:hunter2 --bind 0.0.0.0:8888 my_index
+```
+
+**Multiple indices**
+```bash
+# Indices will be displayed in this order in the web interface
+sist2 web index1 index2 index3 index4
+```
+
+### rewrite_url
+
+When the `rewrite_url` field is not empty, the web module ignores the `root`
+field and will return a HTTP redirect to `<rewrite_url><path>/<name><extension>`
+instead of serving the file from disk. 
+Both the `root` and `rewrite_url` fields are safe to manually modify from the 
+`descriptor.json` file.
+
+### Link to specific indices
+
+To link to specific indices, you can add a list of comma-separated index name to 
+the URL: `?i=<name>,<name>`. By default, indices with `"(nsfw)"` in their name are
+not displayed.
+
+## exec-script
+
+The `exec-script` command is used to execute a user script for an index that has already been imported to Elasticsearch with the `index` command. Note that the documents will not be reset to their default state before each execution as the `index` command does: if you make undesired changes to the documents by accident, you will need to run `index` again to revert to the original state.
+
+
+# Tagging
+
+### Manual tagging
+
+You can modify tags of individual documents directly from the 
+ `web` interface. Note that you can setup authentication for this feature
+ with the `--tag-auth` option (See [web options](#web-options))
+
+![manual_tag](manual_tag.png)
+
+Tags that are manually added are saved both in the 
+ index folder (in `/tags/`) and in Elasticsearch*. When re-`index`ing, 
+ they are read from the index and automatically applied.
+ 
+You can safely copy the `/tags/` database to another index.
+
+See [Automatic tagging](#automatic-tagging) for information about tag 
+ hierarchies and tag colors.
+
+\* *It can take a few seconds to take effect in new search queries, and the page needs 
+    to be reloaded for the tags tab to update*
+
+
+### Automatic tagging
+
+See [scripting](scripting.md) documentation.
+
+# Sidecar files
+
+When scanning, sist2 will read metadata from `.s2meta` JSON files and overwrite the 
+original document's metadata. Sidecar metadata files will also work inside archives.
+Sidecar files themselves are not saved in the index.
+
+This feature is useful to leverage third-party applications such as speech-to-text or
+OCR to add additional metadata to a file.
+
+**Example**
+
+```
+~/Documents/
+├── Video.mp4
+└── Video.mp4.s2meta
+```
+
+The sidecar file must have exactly the same file path and the `.s2meta` suffix.
+
+`Video.mp4.s2meta`:
+```json
+{
+  "content": "This sidecar file will overwrite some metadata fields of Video.mp4",
+  "author": "Some author",
+  "duration": 12345,
+  "bitrate": 67890,
+  "some_arbitrary_field": [1,2,3]
+}
+```
+
+```
+sist2 scan ~/Documents -o ./docs.idx
+sist2 index ./docs.idx
+```
+
+*NOTE*: It is technically possible to overwrite the `tag` value using sidecar files, however,
+it is not currently possible to restore both manual tags and sidecar tags without user scripts
+while reindexing.

docs/scripting.md

@@ -39,7 +39,7 @@ it adds the `genre.<genre>` tag.
 ArrayList tags = ctx._source.tag = new ArrayList();
 
 if (ctx._source?.genre != null) {
-    tags.add("genre." + ctx._source.genre.toLowerCase())
+    tags.add("genre." + ctx._source.genre.toLowerCase());
 }
 ```
 
@@ -67,7 +67,7 @@ ArrayList tags = ctx._source.tag = new ArrayList();
 
 Matcher m = /[\(\.+](20[0-9]{2})[\)\.+]/.matcher(ctx._source.name);
 if (m.find()) {
-    tags.add("year." + m.group(1))
+    tags.add("year." + m.group(1));
 }
 ```
 
@@ -111,16 +111,6 @@ if (ctx._source.path != "") {
 }
 ```
 
-Set the name of the last folder (`/path/to/<studio>/file.mp4`) to `studio.<studio>` tag
-```Java
-ArrayList tags = ctx._source.tag = new ArrayList();
-
-if (ctx._source.path != "") {
-    String[] names = ctx._source.path.splitOnToken('/');
-    tags.add("studio." + names[names.length-1]);
-}
-```
-
 Parse `EXIF:F Number` tag
 ```Java
 if (ctx._source?.exif_fnumber != null) {

schema/mappings.json

@@ -10,12 +10,34 @@
     "path": {
       "type": "text",
       "analyzer": "path_analyzer",
+      "copy_to": "suggest-path",
       "fielddata": true,
-      "index_prefixes": {}
+      "fields": {
+        "nGram": {
+          "type": "text",
+          "analyzer": "my_nGram"
+        },
+        "text": {
+          "type": "text",
+          "analyzer": "content_analyzer"
+        }
+      }
+    },
+    "suggest-path": {
+      "type": "completion",
+      "analyzer": "case_insensitive_kw_analyzer"
     },
     "mime": {
       "type": "keyword"
     },
+    "parent": {
+      "type": "keyword",
+      "index": false
+    },
+    "thumbnail": {
+      "type": "keyword",
+      "index": false
+    },
     "videoc": {
       "type": "keyword",
       "index": false
@@ -36,6 +58,10 @@
       "type": "integer",
       "index": false
     },
+    "pages": {
+      "type": "integer",
+      "index": false
+    },
     "mtime": {
       "type": "integer"
     },
@@ -108,7 +134,12 @@
       }
     },
     "tag": {
-      "type": "keyword"
+      "type": "keyword",
+      "copy_to": "suggest-tag"
+    },
+    "suggest-tag": {
+      "type": "completion",
+      "analyzer": "case_insensitive_kw_analyzer"
     },
     "exif_make": {
       "type": "text"
@@ -133,6 +164,12 @@
     },
     "exif_user_comment": {
       "type": "text"
+    },
+    "author": {
+      "type": "text"
+    },
+    "modified_by": {
+      "type": "text"
     }
   }
 }

schema/settings.json

@@ -1,7 +1,8 @@
 {
   "index": {
     "refresh_interval": "30s",
-    "codec": "best_compression"
+    "codec": "best_compression",
+    "number_of_replicas": 0
   },
   "analysis": {
     "tokenizer": {

scripts/before_build.sh

@@ -2,15 +2,15 @@
 
 rm -rf index.sist2/
 
-rm web/js/bundle.js 2> /dev/null
-cat `ls web/js/*.min.js` > web/js/bundle.js
-cat web/js/{util,dom,search}.js >> web/js/bundle.js
+rm src/static/js/bundle.js 2> /dev/null
+cat `ls src/static/js/*.min.js` > src/static/js/bundle.js
+cat src/static/js/{util,dom}.js >> src/static/js/bundle.js
 
-rm web/css/bundle*.css 2> /dev/null
-cat web/css/*.min.css > web/css/bundle.css
-cat web/css/light.css >> web/css/bundle.css
-cat web/css/*.min.css > web/css/bundle_dark.css
-cat web/css/dark.css >> web/css/bundle_dark.css
+rm src/static/css/bundle*.css 2> /dev/null
+cat src/static/css/*.min.css > src/static/css/bundle.css
+cat src/static/css/light.css >> src/static/css/bundle.css
+cat src/static/css/*.min.css > src/static/css/bundle_dark.css
+cat src/static/css/dark.css >> src/static/css/bundle_dark.css
 
 python3 scripts/mime.py > src/parsing/mime_generated.c
 python3 scripts/serve_static.py > src/web/static_generated.c

scripts/get_static_libs.sh

@@ -1,128 +0,0 @@
-#!/usr/bin/env bash
-
-THREADS=$(nproc)
-
-cd lib
-
-cd mupdf
-CFLAGS=-fPIC make USE_SYSTEM_HARFBUZZ=yes USE_SYSTEM_OPENJPEG=yes HAVE_X11=no HAVE_GLUT=no -j $THREADS
-cd ..
-
-mv mupdf/build/release/libmupdf.a .
-mv mupdf/build/release/libmupdf-third.a .
-
-# openjp2
-cd openjpeg
-cmake . -DCMAKE_BUILD_TYPE=Release -DCMAKE_C_FLAGS="-O3 -DNDEBUG -fPIC"
-make -j $THREADS
-cd ..
-mv openjpeg/bin/libopenjp2.a .
-
-# harfbuzz
-cd harfbuzz
-./autogen.sh
-CFLAGS=-fPIC ./configure --disable-shared --enable-static
-make -j $THREADS
-cd ..
-mv harfbuzz/src/.libs/libharfbuzz.a .
-
-# ffmpeg
-cd ffmpeg
-./configure --disable-shared --enable-static --disable-ffmpeg --disable-ffplay \
- --disable-ffprobe --disable-doc\
- --disable-manpages --disable-postproc --disable-avfilter \
- --disable-alsa --disable-lzma --disable-xlib --disable-debug\
- --disable-vdpau --disable-vaapi --disable-sdl2 --disable-network\
- --extra-cflags=-fPIC
-make -j $THREADS
-cd ..
-
-mv ffmpeg/libavcodec/libavcodec.a .
-mv ffmpeg/libavformat/libavformat.a .
-mv ffmpeg/libavutil/libavutil.a .
-mv ffmpeg/libswresample/libswresample.a .
-mv ffmpeg/libswscale/libswscale.a .
-
-# onion
-cd onion
-mkdir build 2> /dev/null
-cd build
-cmake -DONION_USE_SSL=false -DONION_USE_PAM=false -DONION_USE_PNG=false -DONION_USE_JPEG=false \
--DONION_USE_JPEG=false -DONION_USE_XML2=false -DONION_USE_SYSTEMD=false -DONION_USE_SQLITE3=false \
--DONION_USE_REDIS=false -DONION_USE_GC=false -DONION_USE_TESTS=false -DONION_EXAMPLES=false \
--DONION_USE_BINDINGS_CPP=false ..
-make -j $THREADS
-cd ../..
-
-mv onion/build/src/onion/libonion_static.a .
-
-#bzip2
-cd bzip2-1.0.6
-make -j $THREADS
-cd ..
-mv bzip2-1.0.6/libbz2.a .
-
-# magic
-cd libmagic
-./autogen.sh
-./configure --enable-static --disable-shared
-make -j $THREADS
-cd ..
-mv libmagic/src/.libs/libmagic.a .
-
-# tesseract
-cd tesseract
-mkdir build
-cd build
-cmake -DSTATIC=on -DBUILD_TRAINING_TOOLS=off -DBUILD_TESTS=off -DCMAKE_BUILD_TYPE=Release \
-  -DCMAKE_CXX_FLAGS="-fPIC" -DAUTO_OPTIMIZE=off ..
-make -j $THREADS
-cd ../..
-mv tesseract/build/libtesseract.a .
-
-# leptonica
-cd leptonica
-./autogen.sh
-CFLAGS="-fPIC" ./configure --without-zlib --without-jpeg --without-giflib \
-  --without-giflib --without-libwebp --without-libwebpmux --without-libopenjpeg \
-  --enable-static --disable-shared
-make -j $THREADS
-cd ..
-mv leptonica/src/.libs/liblept.a .
-
-# tiff
-cd libtiff
-./autogen.sh
-CFLAGS="-fPIC" CXXFLAGS="-fPIC" CXX_FLAGS="-fPIC" ./configure --enable-static --disable-shared --disable-lzw --disable-jpeg --disable-webp \
-  --disable-lzma --disable-zstd --disable-jbig
-make -j $THREADS
-cd ..
-mv libtiff/libtiff/.libs/libtiff.a .
-
-# png
-cd libpng
-CFLAGS="-fPIC" ./configure --enable-static --disable-shared
-make -j $THREADS
-cd ..
-mv libpng/.libs/libpng16.a .
-
-# openssl...
-git clone --depth 1 -b OpenSSL_1_1_0-stable https://github.com/openssl/openssl
-cd openssl
-./config --prefix=$(pwd)/../ssl
-make depend
-make -j $THREADS
-make install
-cd ..
-mv ./openssl/libcrypto.a ./openssl/libssl.a .
-
-# curl
-wget -nc https://curl.haxx.se/download/curl-7.68.0.tar.gz
-tar -xzf curl-7.68.0.tar.gz
-cd curl-7.68.0
-./configure --disable-ldap --disable-ldaps --without-librtmp --disable-rtsp --disable-crypto-auth \
-  --disable-smtp --without-libidn2 --without-nghttp2 --without-brotli --enable-static --disable-shared \
-  --without-libpsl --with-ssl=$(pwd)/../ssl
-make -j $THREADS
-cd ..
-mv curl-7.68.0/lib/.libs/libcurl.a .

scripts/index_static.py

@@ -14,4 +14,5 @@ def clean(filepath):
 for file in files:
     with open(file, "r") as f:
         data = json.dumps(json.load(f), separators=(",", ":")).encode()
+    data += b'\0'
     print("char %s[%d] = {%s};" % (clean(file), len(data), ",".join(str(int(b)) for b in data)))

scripts/mime.csv

@@ -2,14 +2,18 @@ application/arj, arj
 application/base64, mme
 application/binhex, hqx
 application/book, boo|book
+application/CDFV2-corrupt,
 application/CDFV2, sdv
 application/clariscad, ccad
 application/commonground, dp
+application/csv,
+application/dicom, dcm
 application/drafting, drw
+application/epub+zip, epub
 application/freeloader, frl
 application/futuresplash, spl
 application/groupwise, vew
-application/gzip, gz
+application/gzip, gz|tgz
 application/hta, hta
 application/i-deas, unv
 application/iges, iges|igs
@@ -17,7 +21,6 @@ application/inf, inf
 application/java-archive, jar
 application/java, class
 application/javascript,
-application/x-archive, a
 application/json, json
 application/marc, mrc
 application/mbedlet, mbd
@@ -27,7 +30,9 @@ application/msword, doc|dot|w6w|wiz|word
 application/netmc, mcp
 application/octet-stream, bin|dump|gpg
 application/oda, oda
+application/ogg, ogv
 application/pdf, pdf
+application/pgp-keys,
 application/pgp-signature, pgp
 application/pkcs7-signature, p7s
 application/pkix-cert, cer|crt
@@ -43,6 +48,10 @@ application/vda, vda
 application/vnd.fdf, fdf
 application/vnd.font-fontforge-sfd, sfd
 application/vnd.hp-hpgl, hgl|hpg|hpgl
+application/vnd.iccprofile, icm
+application/vnd.iccprofile, icm
+application/vnd.lotus-1-2-3,
+application/vnd.ms-cab-compressed, cab
 application/vnd.ms-excel, xlb|xlc|xll|xlm|xls|xlw
 application/vnd.ms-fontobject, eot
 application/vnd.ms-opentype, otf
@@ -54,45 +63,75 @@ application/vnd.ms-project, mpp
 application/vnd.oasis.opendocument.base, odb
 application/vnd.oasis.opendocument.formula, odf
 application/vnd.oasis.opendocument.graphics, odg
+application/vnd.oasis.opendocument.presentation, odp
+application/vnd.oasis.opendocument.spreadsheet, ods
 application/vnd.oasis.opendocument.text, odt
+application/vnd.openxmlformats-officedocument.presentationml.presentation, pptx
+application/vnd.openxmlformats-officedocument.spreadsheetml.sheet, xlsx
+application/vnd.openxmlformats-officedocument.wordprocessingml.document, docx
+application/vnd.symbian.install,
+application/vnd.tcpdump.pcap, pcap
 application/vnd.wap.wmlc, wmlc
 application/vnd.wap.wmlscriptc, wmlsc
 application/vnd.xara, web
 application/vocaltec-media-desc, vmd
 application/vocaltec-media-file, vmf
+application/warc, warc
+application/winhelp, hlp
 application/wordperfect6.0, w60
 application/wordperfect6.1, w61
 application/wordperfect, wp|wp5|wp6|wpd
 application/x-123, wk1
+application/x-7z-compressed, 7z
 application/x-aim, aim
+application/x-apple-diskimage,
+application/x-arc,
+application/x-archive, a
+application/x-atari-7800-rom, a78
 application/x-authorware-bin, aab
 application/x-authorware-map, aam
 application/x-authorware-seg, aas
+application/x-avira-qua,
 application/x-bcpio, bcpio
 application/x-bittorrent, torrent
 application/x-bsh, bsh
 application/x-bytecode.python, pyc
 application/x-bzip2, boz|bz2
 application/x-bzip, bz
+application/x-cbr, cbr
+application/x-cbz, cbz
 application/x-cdlink, vcd
 application/x-chat, cha|chat
+application/x-chrome-extension,
 application/x-cocoa, cco
 application/x-conference, nsc
+application/x-coredump,
 application/x-cpio, cpio
 application/x-dbf, dbf
 application/x-dbt,
+application/x-debian-package, deb
 application/x-deepv, deepv
-application/x-director, dcr|dir|dxr
+application/x-director, dir|dxr
+application/x-dmp, dmp
+application/x-dosdriver,
 application/x-dosexec, dll
 application/x-dvi, dvi
 application/x-elc, elc
+application/x-empty,
 application/x-envoy, env|evy
 application/x-esrehber, es
 application/x-excel, xla|xld|xlk|xlt|xlv
 application/x-executable, exe
+application/x-font-gdos,
+application/x-font-pf2, pf2
+application/x-font-pfm, pfm
 application/x-font-sfn,
 application/x-font-ttf, ttf|ttc
+application/x-fptapplication/x-dbt,
 application/x-freelance, pre
+application/x-gamecube-rom,
+application/x-gdbm,
+application/x-gettext-translation,
 application/x-git,
 application/x-gsp, gsp
 application/x-gss, gss
@@ -102,46 +141,68 @@ application/x-hdf, hdf
 application/x-helpfile, help
 application/x-httpd-imap, imap
 application/x-ima, ima
+application/x-innosetup,
 application/x-internett-signup, ins
 application/x-inventor, iv
 application/x-ip2, ip
 application/x-java-applet,
 application/x-java-commerce, jcm
 application/x-java-image,
+application/x-java-jmod, jmod
 application/x-java-keystore,
+application/x-kdelnk,
 application/x-koan, skd|skm|skp|skt
 application/x-latex, latex|ltx
 application/x-livescreen, ivy
 application/x-lotus, wq1
+application/x-lz4+json, jsonlz4
+application/x-lz4, lz4
+application/x-lz4, lz4
+application/x-lzh-compressed,
 application/x-lzh, lzh
+application/x-lzip, lz
+application/x-lzma, lzma
+application/x-lzop, lzo
 application/x-lzx, lzx
 application/x-mach-binary, jnilib|dylib
 application/x-mach-executable,
 application/x-magic-cap-package-1.0, mc$
 application/x-mathcad, mcd
+application/x-maxis-dbpf,
 application/x-meme, mm
 application/x-midi, midi
 application/x-mif, mif
 application/x-mix-transfer, nix
 application/xml, opf
+application/x-mobipocket-ebook, mobi
+application/vnd.amazon.mobi8-ebook, azw|azw3
+application/x-msaccess, accdb
+application/x-ms-compress-szdd, fon
 application/x-ms-pdb, pdb
+application/x-ms-reader, lit
+application/x-n64-rom, z64
 application/x-navi-animation, ani
 application/x-navidoc, nvd
 application/x-navimap, map
 application/x-navistyle, stl
+application/x-nes-rom, nes
 application/x-netcdf, cdf|nc
 application/x-newton-compatible-pkg, pkg
+application/x-nintendo-ds-rom,
 application/x-object, o
 application/x-omcdatamaker, omcd
 application/x-omc, omc
 application/x-omcregerator, omcr
 application/x-pagemaker, pm4|pm5
 application/x-pcl, pcl
+application/x-pgp-keyring,
 application/x-pixclscript, plx
 application/x-pkcs7-certreqresp, p7r
 application/x-pkcs7-signature, p7a
 application/x-project, mpc|mpt|mpv|mpx
 application/x-qpro, wb1
+application/x-rar, rar
+application/x-rpm, rpm
 application/x-sdp, sdp
 application/x-sea, sea
 application/x-seelogo, sl
@@ -149,12 +210,17 @@ application/x-setupscript,
 application/x-sharedlib, so
 application/x-shar, shar
 application/x-shockwave-flash, swf
+application/x-snappy-framed,
 application/x-sprite, spr|sprite
 application/x-sqlite3,
+application/x-stargallery-thm,
+application/x-stuffit, sit
 application/x-sv4cpio, sv4cpio
 application/x-sv4crc, sv4crc
 application/x-tar, tar
 application/x-tbook, sbk|tbk
+application/x-terminfo,
+application/x-terminfo2,
 application/x-texinfo, texi|texinfo
 application/x-tex-tfm, tfm
 application/x-ustar, ustar
@@ -163,16 +229,22 @@ application/x-vnd.audioexplosion.mzz, mzz
 application/x-vnd.ls-xpix, xpix
 application/x-vrml, vrml
 application/x-wais-source, src|wsrc
+application/x-wine-extension-ini,
 application/x-wintalk, wtk
 application/x-world, svr
 application/x-wri, wri
 application/x-x509-ca-cert, der
 application/x-xz, xz
+application/x-zip,
+application/x-zstd, zst
 application/zip, zip
+application/zlib, z
+!audio/basic, au
 audio/it, it
 audio/make, funk|my|pfunk
 audio/midi, kar
 audio/mid, rmi
+audio/mp4, m4b
 audio/mpeg, m2a|mpa
 audio/ogg, ogg
 audio/s3m, s3m
@@ -180,7 +252,10 @@ audio/tsp-audio, tsi
 audio/tsplayer, tsp
 audio/vnd.qcelp, qcp
 audio/voxware, vox
+audio/x-aiff, aiff|aif
+audio/x-flac, flac
 audio/x-gsm, gsd|gsm
+audio/x-hx-aac-adts,
 audio/x-jam, jam
 audio/x-liveaudio, lam
 audio/x-m4a, m4a
@@ -194,17 +269,24 @@ audio/x-nspaudio, lma
 audio/x-pn-realaudio, ram|rm|rmm|rmp
 audio/x-psid, sid
 audio/x-realaudio, ra
+audio/x-s3m,
 audio/x-twinvq-plugin, vqe|vql
 audio/x-twinvq, vqf
 audio/x-voc, voc
 audio/x-wav, wav
+!audio/x-xbox360-executable, xex
+!audio/x-xbox-executable, xbe
 font/otf,
 font/sfnt,
+font/woff2, woff2
+font/woff, woff
+image/bmp,
 image/cmu-raster, rast
 image/fif, fif
 image/florian, flo|turbot
 image/g3fax, g3
 image/gif, gif
+image/heic, heic
 image/ief, ief|iefs
 image/jpeg, jfif|jfif-tbnl|jpe|jpeg|jpg
 image/jutvision, jut
@@ -213,6 +295,9 @@ image/pict, pic|pict
 image/png, png|x-png
 !image/svg, svg
 !image/svg+xml,
+image/tiff,
+!image/vnd.adobe.photoshop, psd
+!image/vnd.djvu, djvu
 image/vnd.fpx, fpx
 image/vnd.microsoft.icon,
 image/vnd.rn-realflash, rf
@@ -220,9 +305,15 @@ image/vnd.rn-realpix, rp
 image/vnd.wap.wbmp, wbmp
 image/vnd.xiff, xif
 image/webp, webp
+image/wmf,
+image/x-3ds, 3ds
+image/x-award-bioslogo,
 image/x-cmu-raster, ras
+image/x-cur, tga
 image/x-dwg, dwg|dxf|svf
 image/x-eps,
+image/x-exr, exr
+image/x-gem,
 image/x-icns,
 !image/x-icon, ico
 image/x-jg, art
@@ -236,32 +327,31 @@ image/x-portable-graymap, pgm
 image/x-portable-pixmap, ppm
 image/x-quicktime, qif|qti|qtif
 image/x-rgb, rgb
+image/x-tga,
 image/x-tiff, tif|tiff
-image/tiff,
+image/x-win-bitmap,
 !image/x-xcf, xcf
 !image/x-xpixmap, xpm
+image/x-xwindowdump, xwd
+message/news,
 message/rfc822, mht|mhtml|mime
 model/vnd.dwf, dwf
+model/vnd.gdl, gdl
+model/vnd.gs.gdl, gdsl
 model/vrml, wrz
 model/x-pov, pov
 text/asp, asp
 text/css, css
-text/x-sass, sass
-text/x-scss, scss
 text/html, acgi|htm|html|htmls|htx|shtml
 text/javascript, js
 text/mcf, mcf
 text/pascal, pas
-text/plain, com|cmd|conf|def|g|idc|list|lst|mar|sdml|text|txt|md|groovy|license|properties|desktop|ini|rst|cmake|ipynb|readme|less|lo|go|yml|d|cs|hpp|srt|nfo|sfv|m3u|csv|eml
+text/PGP,
+text/plain, com|cmd|conf|def|g|idc|list|lst|mar|sdml|text|txt|md|groovy|license|properties|desktop|ini|rst|cmake|ipynb|readme|less|lo|go|yml|d|cs|hpp|srt|nfo|sfv|m3u|csv|eml|make|log|markdown|yaml
+application/vnd.coffeescript, coffee
 text/richtext, rt|rtf|rtx
 text/rtf,
 text/scriplet, wsc
-text/x-awk, awk
-!video/x-jng, jng
-video/x-mng, mng
-image/x-cur, tga
-image/x-xwindowdump, xwd
-!image/vnd.adobe.photoshop, psd
 text/tab-separated-values, tsv
 text/troff, man|me|ms|roff|t|tr
 text/uri-list, uji|unis|uri|uris
@@ -273,6 +363,7 @@ text/webviewhtml, htt
 text/x-Algol68,
 text/x-asm, asm|s
 text/x-audiosoft-intra, aip
+text/x-awk, awk
 text/x-bcpl,
 text/x-c, c|cc|h
 text/x-c++, cpp|cxx|c++
@@ -287,23 +378,31 @@ text/x-makefile, am|mak
 text/xml, xml|pom|iml|plist
 text/x-m, m
 text/x-msdos-batch, bat
+text/x-ms-regedit, reg
+text/x-objective-c,
 text/x-pascal, p
 text/x-perl, pl
 text/x-php, php
+text/x-po, po
 text/x-python, py
 text/x-ruby, rb
+text/x-sass, sass
+text/x-scss, scss
 text/x-server-parsed-html, ssi
 text/x-setext, etx
 text/x-sgml, sgm|sgml
 text/x-shellscript, sh
 text/x-speech, talk
+text/x-tcl,
 text/x-tex, tex
 text/x-uil, uil
 text/x-uuencode, uue
 text/x-vcalendar, vcs
+text/x-vcard, vcf
 video/animaflex, afl
 video/avi, avi
 video/avs-video, avs
+video/MP2T,
 video/mp4, mp4
 video/mpeg, m1v|m2v|mpe|mpeg|mpg
 video/quicktime, moov|mov|qt
@@ -318,101 +417,36 @@ video/x-atomic3d-feature, fmf
 video/x-dl, dl
 video/x-dv, dif|dv
 video/x-fli, fli
+video/x-flv, flv
 video/x-isvideo, isu
+!video/x-jng, jng
+video/x-m4v, m4v
+video/x-matroska, mkv
+video/x-mng, mng
 video/x-motion-jpeg, mjpg
 video/x-ms-asf, asf|asx|wmv
+video/x-msvideo, divx
 video/x-qtc, qtc
 video/x-sgi-movie, movie|mv
-application/x-7z-compressed, 7z
-application/vnd.openxmlformats-officedocument.wordprocessingml.document, docx
-text/x-po, po
-application/x-rpm, rpm
-application/x-debian-package, deb
-application/vnd.iccprofile, icm
-application/dicom, dcm
-image/x-exr, exr
-application/vnd.iccprofile, icm
-video/x-matroska, mkv
-application/x-empty,
-model/vnd.gdl, gdl
-model/vnd.gs.gdl, gdsl
-font/woff, woff
-font/woff2, woff2
-application/epub+zip, epub
-application/x-mobipocket-ebook, mobi
-audio/x-flac, flac
-application/x-rar, rar
-video/x-msvideo, divx
-video/x-flv, flv
-application/x-kdelnk,
-text/x-tcl,
-application/ogg, ogv
-application/vnd.openxmlformats-officedocument.spreadsheetml.sheet, xlsx
-application/vnd.ms-cab-compressed, cab
-audio/mp4, m4b
-!image/vnd.djvu, djvu
-application/x-ms-reader, lit
-application/CDFV2-corrupt,
-text/x-vcard, vcf
-application/x-innosetup,
-application/winhelp, hlp
-image/x-tga,
-application/x-wine-extension-ini,
-application/x-cbz, cbz
-application/x-cbr, cbr
-application/x-ms-compress-szdd, fon
-application/x-atari-7800-rom, a78
-application/x-nes-rom, nes
-application/x-font-pfm, pfm
-application/x-gettext-translation,
-image/wmf,
-application/pgp-keys,
-image/x-3ds, 3ds
-application/x-lz4, lz4
-application/vnd.openxmlformats-officedocument.presentationml.presentation, pptx
-application/vnd.oasis.opendocument.presentation, odp
-application/x-msaccess, accdb
-application/vnd.oasis.opendocument.spreadsheet, ods
-audio/x-aiff, aiff|aif
-text/x-ms-regedit, reg
-application/x-gamecube-rom,
-application/x-nintendo-ds-rom,
-text/x-objective-c,
-application/x-font-gdos,
-application/x-apple-diskimage,
-application/x-zstd, zst
-video/x-m4v, m4v
-message/news,
-application/vnd.symbian.install,
-application/x-lzh-compressed,
-application/x-dosdriver,
-application/vnd.tcpdump.pcap, pcap
 x-epoc/x-sisx-app,
-application/x-avira-qua,
-video/MP2T,
-application/x-snappy-framed,
-application/x-lz4+json, jsonlz4
-application/x-dmp, dmp
-application/zlib, z
-application/x-pgp-keyring,
-application/x-gdbm,
-application/x-font-pf2, pf2
-application/x-zip,
-application/x-coredump,
-application/x-java-jmod, jmod
-application/x-terminfo,
-application/x-terminfo2,
-application/x-arc,
-application/vnd.lotus-1-2-3,
-image/x-win-bitmap,
-application/x-maxis-dbpf,
-text/PGP,
-audio/x-hx-aac-adts,
-application/x-chrome-extension,
-image/heic, heic
-image/x-gem,
-application/x-lzma, lzma
-application/warc, warc
-application/x-lz4, lz4
-application/x-lzip, lz
-application/x-lzop, lzo
+application/x-zstd-dictionary,
+application/vnd.ms-outlook, msg
+image/x-olympus-orf, orf
+image/x-nikon-nef, nef
+image/x-fuji-raf, raf
+image/x-panasonic-raw, rw2|raw
+image/x-adobe-dng, dng
+image/x-canon-cr2, cr2
+image/x-canon-crw, crw
+image/x-dcraw,
+image/x-kodak-dcr, dcr
+image/x-kodak-k25, k25
+image/x-kodak-kdc, kdc
+image/x-minolta-mrw, mrw
+image/x-pentax-pef, pef
+image/x-sigma-x3f, xf3
+image/x-sony-arw, arw
+image/x-sony-sr2, sr2
+image/x-sony-srf, srf
+image/x-epson-erf, erf
+sist2/sidecar, s2meta

scripts/mime.py

@@ -3,6 +3,7 @@ noparse = set()
 ext_in_hash = set()
 
 major_mime = {
+    "sist2": 0,
     "model": 1,
     "example": 2,
     "message": 3,
@@ -18,7 +19,6 @@ major_mime = {
 
 pdf = (
     "application/pdf",
-    "application/x-cbz",
     "application/epub+zip",
     "application/vnd.ms-xpsdocument",
 )
@@ -62,6 +62,40 @@ doc = (
     "application/vnd.openxmlformats-officedocument.presentationml.presentation"
 )
 
+mobi = (
+    "application/x-mobipocket-ebook",
+    "application/vnd.amazon.mobi8-ebook"
+)
+
+markup = (
+    "text/xml",
+    "text/html",
+    "text/x-sgml"
+)
+
+raw = (
+    "image/x-olympus-orf",
+    "image/x-nikon-nef",
+    "image/x-fuji-raf",
+    "image/x-panasonic-raw",
+    "image/x-adobe-dng",
+    "image/x-canon-cr2",
+    "image/x-canon-crw",
+    "image/x-dcraw",
+    "image/x-kodak-dcr",
+    "image/x-kodak-k25",
+    "image/x-kodak-kdc",
+    "image/x-minolta-mrw",
+    "image/x-pentax-pef",
+    "image/x-sigma-x3f",
+    "image/x-sony-arw",
+    "image/x-sony-sr2",
+    "image/x-sony-srf",
+    "image/x-minolta-mrw",
+    "image/x-pentax-pef",
+    "image/x-epson-erf",
+)
+
 cnt = 1
 
 
@@ -82,8 +116,18 @@ def mime_id(mime):
         mime_id += " | 0x08000000"
     elif mime in doc:
         mime_id += " | 0x04000000"
+    elif mime in mobi:
+        mime_id += " | 0x02000000"
+    elif mime in markup:
+        mime_id += " | 0x01000000"
+    elif mime in raw:
+        mime_id += " | 0x00800000"
     elif mime == "application/x-empty":
+        cnt -= 1
         return "1"
+    elif mime == "sist2/sidecar":
+        cnt -= 1
+        return "2"
     return mime_id
 
 
@@ -91,7 +135,7 @@ def clean(t):
     return t.replace("/", "_").replace(".", "_").replace("+", "_").replace("-", "_")
 
 
-with open("mime.csv") as f:
+with open("scripts/mime.csv") as f:
     for l in f:
         mime, ext_list = l.split(",")
         if l.startswith("!"):
@@ -103,7 +147,7 @@ with open("mime.csv") as f:
     print("// **Generated by mime.py**")
     print("#ifndef MIME_GENERATED_C")
     print("#define MIME_GENERATED_C")
-    print("#include <glib-2.0/glib.h>\n")
+    print("#include <glib.h>\n")
     print("#include <stdlib.h>\n")
     # Enum
     print("enum mime {")

scripts/serve_static.py

@@ -1,10 +1,12 @@
 files = [
-    "web/css/bundle.css",
-    "web/css/bundle_dark.css",
-    "web/js/bundle.js",
-    "web/img/sprite-skin-flat.png",
-    "web/img/sprite-skin-flat-dark.png",
-    "web/search.html",
+    "src/static/css/bundle.css",
+    "src/static/css/bundle_dark.css",
+    "src/static/js/bundle.js",
+    "src/static/js/search.js",
+    "src/static/img/sprite-skin-flat.png",
+    "src/static/img/sprite-skin-flat-dark.png",
+    "src/static/search.html",
+    "src/static/stats.html",
 ]
 
 

src/cli.c

@@ -1,5 +1,6 @@
 #include "cli.h"
 #include "ctx.h"
+#include <tesseract/capi.h>
 
 #define DEFAULT_OUTPUT "index.sist2/"
 #define DEFAULT_CONTENT_SIZE 32768
@@ -8,12 +9,15 @@
 #define DEFAULT_REWRITE_URL ""
 
 #define DEFAULT_ES_URL "http://localhost:9200"
+#define DEFAULT_ES_INDEX "sist2"
 #define DEFAULT_BATCH_SIZE 100
 
-#define DEFAULT_BIND_ADDR "localhost"
-#define DEFAULT_PORT "4090"
+#define DEFAULT_LISTEN_ADDRESS "localhost:4090"
+#define DEFAULT_TREEMAP_THRESHOLD 0.0005
 
-const char* TESS_DATAPATHS[] = {
+#define DEFAULT_MAX_MEM_BUFFER 2000
+
+const char *TESS_DATAPATHS[] = {
         "/usr/share/tessdata/",
         "/usr/share/tesseract-ocr/tessdata/",
         "./",
@@ -29,10 +33,18 @@ scan_args_t *scan_args_create() {
     return args;
 }
 
+exec_args_t *exec_args_create() {
+    exec_args_t *args = calloc(sizeof(exec_args_t), 1);
+    return args;
+}
+
 void scan_args_destroy(scan_args_t *args) {
     if (args->name != NULL) {
         free(args->name);
     }
+    if (args->incremental != NULL) {
+        free(args->incremental);
+    }
     if (args->path != NULL) {
         free(args->path);
     }
@@ -44,6 +56,12 @@ void scan_args_destroy(scan_args_t *args) {
 
 void index_args_destroy(index_args_t *args) {
     //todo
+    if (args->es_mappings_path) {
+        free(args->es_mappings);
+    }
+    if (args->es_settings_path) {
+        free(args->es_settings);
+    }
     free(args);
 }
 
@@ -52,6 +70,10 @@ void web_args_destroy(web_args_t *args) {
     free(args);
 }
 
+void exec_args_destroy(exec_args_t *args) {
+    free(args);
+}
+
 int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
     if (argc < 2) {
         fprintf(stderr, "Required positional argument: PATH.\n");
@@ -67,7 +89,7 @@ int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
     }
 
     if (args->incremental != NULL) {
-        abs_path = abspath(args->incremental);
+        args->incremental = abspath(args->incremental);
         if (abs_path == NULL) {
             sist_log("main.c", SIST_WARNING, "Could not open original index! Disabled incremental scan feature.");
             args->incremental = NULL;
@@ -112,7 +134,7 @@ int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
         return 1;
     }
 
-    if (args->depth < 0) {
+    if (args->depth <= 0) {
         args->depth = G_MAXINT32;
     } else {
         args->depth += 1;
@@ -120,6 +142,10 @@ int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
 
     if (args->name == NULL) {
         args->name = g_path_get_basename(args->output);
+    } else {
+        char* tmp = malloc(strlen(args->name) + 1);
+        strcpy(tmp, args->name);
+        args->name = tmp;
     }
 
     if (args->rewrite_url == NULL) {
@@ -144,7 +170,7 @@ int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
 
         char filename[128];
         sprintf(filename, "%s.traineddata", args->tesseract_lang);
-        const char * path = find_file_in_paths(TESS_DATAPATHS, filename);
+        const char *path = find_file_in_paths(TESS_DATAPATHS, filename);
         if (path == NULL) {
             LOG_FATAL("cli.c", "Could not find tesseract language file!");
         }
@@ -180,6 +206,16 @@ int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
         ScanCtx.exclude = NULL;
     }
 
+    if (args->treemap_threshold_str == 0) {
+        args->treemap_threshold = DEFAULT_TREEMAP_THRESHOLD;
+    } else {
+        args->treemap_threshold = atof(args->treemap_threshold_str);
+    }
+
+    if (args->max_memory_buffer == 0) {
+        args->max_memory_buffer = DEFAULT_MAX_MEM_BUFFER;
+    }
+
     LOG_DEBUGF("cli.c", "arg quality=%f", args->quality)
     LOG_DEBUGF("cli.c", "arg size=%d", args->size)
     LOG_DEBUGF("cli.c", "arg content_size=%d", args->content_size)
@@ -195,6 +231,36 @@ int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
     LOG_DEBUGF("cli.c", "arg tesseract_path=%s", args->tesseract_path)
     LOG_DEBUGF("cli.c", "arg exclude=%s", args->exclude_regex)
     LOG_DEBUGF("cli.c", "arg fast=%d", args->fast)
+    LOG_DEBUGF("cli.c", "arg treemap_threshold=%f", args->treemap_threshold)
+    LOG_DEBUGF("cli.c", "arg max_memory_buffer=%d", args->max_memory_buffer)
+
+    return 0;
+}
+
+int load_external_file(const char *file_path, char **dst) {
+    struct stat info;
+    int res = stat(file_path, &info);
+
+    if (res == -1) {
+        LOG_ERRORF("cli.c", "Error opening file '%s': %s\n", file_path, strerror(errno))
+        return 1;
+    }
+
+    int fd = open(file_path, O_RDONLY);
+    if (fd == -1) {
+        LOG_ERRORF("cli.c", "Error opening file '%s': %s\n", file_path, strerror(errno))
+        return 1;
+    }
+
+    *dst = malloc(info.st_size + 1);
+    res = read(fd, *dst, info.st_size);
+    if (res < 0) {
+        LOG_ERRORF("cli.c", "Error reading file '%s': %s\n", file_path, strerror(errno))
+        return 1;
+    }
+
+    *(*dst + info.st_size) = '\0';
+    close(fd);
 
     return 0;
 }
@@ -208,6 +274,13 @@ int index_args_validate(index_args_t *args, int argc, const char **argv) {
         return 1;
     }
 
+    if (args->threads == 0) {
+        args->threads = 1;
+    } else if (args->threads < 0) {
+        fprintf(stderr, "Invalid threads: %d\n", args->threads);
+        return 1;
+    }
+
     char *index_path = abspath(argv[1]);
     if (index_path == NULL) {
         fprintf(stderr, "File not found: %s\n", argv[1]);
@@ -221,30 +294,26 @@ int index_args_validate(index_args_t *args, int argc, const char **argv) {
         args->es_url = DEFAULT_ES_URL;
     }
 
+    if (args->es_index == NULL) {
+        args->es_index = DEFAULT_ES_INDEX;
+    }
+
     if (args->script_path != NULL) {
-        struct stat info;
-        int res = stat(args->script_path, &info);
-
-        if (res == -1) {
-            fprintf(stderr, "Error opening script file '%s': %s\n", args->script_path, strerror(errno));
+        if (load_external_file(args->script_path, &args->script) != 0) {
             return 1;
         }
+    }
 
-        int fd = open(args->script_path, O_RDONLY);
-        if (fd == -1) {
-            fprintf(stderr, "Error opening script file '%s': %s\n", args->script_path, strerror(errno));
+    if (args->es_settings_path != NULL) {
+        if (load_external_file(args->es_settings_path, &args->es_settings) != 0) {
             return 1;
         }
+    }
 
-        args->script = malloc(info.st_size + 1);
-        res = read(fd, args->script, info.st_size);
-        if (res < 0) {
-            fprintf(stderr, "Error reading script file '%s': %s\n", args->script_path, strerror(errno));
+    if (args->es_mappings_path != NULL) {
+        if (load_external_file(args->es_mappings_path, &args->es_mappings) != 0) {
             return 1;
         }
-
-        *(args->script + info.st_size) = '\0';
-        close(fd);
     }
 
     if (args->batch_size == 0) {
@@ -252,10 +321,16 @@ int index_args_validate(index_args_t *args, int argc, const char **argv) {
     }
 
     LOG_DEBUGF("cli.c", "arg es_url=%s", args->es_url)
+    LOG_DEBUGF("cli.c", "arg es_index=%s", args->es_index)
     LOG_DEBUGF("cli.c", "arg index_path=%s", args->index_path)
     LOG_DEBUGF("cli.c", "arg script_path=%s", args->script_path)
+    LOG_DEBUGF("cli.c", "arg async_script=%s", args->async_script)
     LOG_DEBUGF("cli.c", "arg script=%s", args->script)
     LOG_DEBUGF("cli.c", "arg print=%d", args->print)
+    LOG_DEBUGF("cli.c", "arg es_mappings_path=%s", args->es_mappings_path)
+    LOG_DEBUGF("cli.c", "arg es_mappings=%s", args->es_mappings)
+    LOG_DEBUGF("cli.c", "arg es_settings_path=%s", args->es_settings_path)
+    LOG_DEBUGF("cli.c", "arg es_settings=%s", args->es_settings)
     LOG_DEBUGF("cli.c", "arg batch_size=%d", args->batch_size)
     LOG_DEBUGF("cli.c", "arg force_reset=%d", args->force_reset)
 
@@ -275,18 +350,57 @@ int web_args_validate(web_args_t *args, int argc, const char **argv) {
         args->es_url = DEFAULT_ES_URL;
     }
 
-    if (args->bind == NULL) {
-        args->bind = DEFAULT_BIND_ADDR;
+    if (args->listen_address == NULL) {
+        args->listen_address = DEFAULT_LISTEN_ADDRESS;
     }
 
-    if (args->port == NULL) {
-        args->port = DEFAULT_PORT;
+    if (args->es_index == NULL) {
+        args->es_index = DEFAULT_ES_INDEX;
     }
 
     if (args->credentials != NULL) {
-        args->b64credentials = onion_base64_encode(args->credentials, (int) strlen(args->credentials));
-        //Remove trailing newline
-        *(args->b64credentials + strlen(args->b64credentials) - 1) = '\0';
+        char *ptr = strstr(args->credentials, ":");
+        if (ptr == NULL) {
+            fprintf(stderr, "Invalid --auth format, see usage\n");
+            return 1;
+        }
+
+        strncpy(args->auth_user, args->credentials, (ptr - args->credentials));
+        strcpy(args->auth_pass, ptr + 1);
+
+        if (strlen(args->auth_user) == 0) {
+            fprintf(stderr, "--auth username must be at least one character long");
+            return 1;
+        }
+
+        args->auth_enabled = TRUE;
+    } else {
+        args->auth_enabled = FALSE;
+    }
+
+    if (args->tag_credentials != NULL && args->credentials != NULL) {
+        fprintf(stderr, "--auth and --tag-auth are mutually exclusive");
+        return 1;
+    }
+
+    if (args->tag_credentials != NULL) {
+        char *ptr = strstr(args->tag_credentials, ":");
+        if (ptr == NULL) {
+            fprintf(stderr, "Invalid --tag-auth format, see usage\n");
+            return 1;
+        }
+
+        strncpy(args->auth_user, args->tag_credentials, (ptr - args->tag_credentials));
+        strcpy(args->auth_pass, ptr + 1);
+
+        if (strlen(args->auth_user) == 0) {
+            fprintf(stderr, "--tag-auth username must be at least one character long");
+            return 1;
+        }
+
+        args->tag_auth_enabled = TRUE;
+    } else {
+        args->tag_auth_enabled = FALSE;
     }
 
     args->index_count = argc - 1;
@@ -301,10 +415,12 @@ int web_args_validate(web_args_t *args, int argc, const char **argv) {
     }
 
     LOG_DEBUGF("cli.c", "arg es_url=%s", args->es_url)
-    LOG_DEBUGF("cli.c", "arg bind=%s", args->bind)
-    LOG_DEBUGF("cli.c", "arg port=%s", args->port)
+    LOG_DEBUGF("cli.c", "arg es_index=%s", args->es_index)
+    LOG_DEBUGF("cli.c", "arg listen=%s", args->listen_address)
     LOG_DEBUGF("cli.c", "arg credentials=%s", args->credentials)
-    LOG_DEBUGF("cli.c", "arg b64credentials=%s", args->b64credentials)
+    LOG_DEBUGF("cli.c", "arg tag_credentials=%s", args->tag_credentials)
+    LOG_DEBUGF("cli.c", "arg auth_user=%s", args->auth_user)
+    LOG_DEBUGF("cli.c", "arg auth_pass=%s", args->auth_pass)
     LOG_DEBUGF("cli.c", "arg index_count=%d", args->index_count)
     for (int i = 0; i < args->index_count; i++) {
         LOG_DEBUGF("cli.c", "arg indices[%d]=%s", i, args->indices[i])
@@ -323,3 +439,39 @@ web_args_t *web_args_create() {
     return args;
 }
 
+int exec_args_validate(exec_args_t *args, int argc, const char **argv) {
+
+    if (argc < 2) {
+        fprintf(stderr, "Required positional argument: PATH.\n");
+        return 1;
+    }
+
+    char *index_path = abspath(argv[1]);
+    if (index_path == NULL) {
+        fprintf(stderr, "File not found: %s\n", argv[1]);
+        return 1;
+    } else {
+        args->index_path = argv[1];
+        free(index_path);
+    }
+
+    if (args->es_url == NULL) {
+        args->es_url = DEFAULT_ES_URL;
+    }
+
+    if (args->es_index == NULL) {
+        args->es_index = DEFAULT_ES_INDEX;
+    }
+
+    if (args->script_path == NULL) {
+        LOG_FATAL("cli.c", "--script-file argument is required");
+    }
+
+    if (load_external_file(args->script_path, &args->script) != 0) {
+        return 1;
+    }
+
+    LOG_DEBUGF("cli.c", "arg script_path=%s", args->script_path)
+    LOG_DEBUGF("cli.c", "arg script=%s", args->script)
+    return 0;
+}

src/cli.h

@@ -3,6 +3,8 @@
 
 #include "sist.h"
 
+#include "libscan/arc/arc.h"
+
 typedef struct scan_args {
     float quality;
     int size;
@@ -20,6 +22,9 @@ typedef struct scan_args {
     const char *tesseract_path;
     char *exclude_regex;
     int fast;
+    const char* treemap_threshold_str;
+    double treemap_threshold;
+    int max_memory_buffer;
 } scan_args_t;
 
 scan_args_t *scan_args_create();
@@ -30,24 +35,44 @@ int scan_args_validate(scan_args_t *args, int argc, const char **argv);
 
 typedef struct index_args {
     char *es_url;
+    char *es_index;
     const char *index_path;
     const char *script_path;
     char *script;
+    const char *es_settings_path;
+    char *es_settings;
+    const char *es_mappings_path;
+    char *es_mappings;
     int print;
     int batch_size;
+    int async_script;
     int force_reset;
+    int threads;
 } index_args_t;
 
 typedef struct web_args {
     char *es_url;
-    char *bind;
-    char *port;
+    char *es_index;
+    char *listen_address;
     char *credentials;
-    char *b64credentials;
+    char *tag_credentials;
+    char auth_user[256];
+    char auth_pass[256];
+    int auth_enabled;
+    int tag_auth_enabled;
     int index_count;
     const char **indices;
 } web_args_t;
 
+typedef struct exec_args {
+    char *es_url;
+    char *es_index;
+    const char *index_path;
+    const char *script_path;
+    int async_script;
+    char *script;
+} exec_args_t;
+
 index_args_t *index_args_create();
 
 void index_args_destroy(index_args_t *args);
@@ -60,4 +85,10 @@ int index_args_validate(index_args_t *args, int argc, const char **argv);
 
 int web_args_validate(web_args_t *args, int argc, const char **argv);
 
+exec_args_t *exec_args_create();
+
+void exec_args_destroy(exec_args_t *args);
+
+int exec_args_validate(exec_args_t *args, int argc, const char **argv);
+
 #endif

src/ctx.c

@@ -0,0 +1,6 @@
+#include "ctx.h"
+
+ScanCtx_t ScanCtx;
+WebCtx_t WebCtx;
+IndexCtx_t IndexCtx;
+LogCtx_t LogCtx;

src/ctx.h

@@ -2,8 +2,24 @@
 #define SIST2_CTX_H
 
 #include "sist.h"
+#include "tpool.h"
+#include "libscan/scan.h"
+#include "libscan/arc/arc.h"
+#include "libscan/comic/comic.h"
+#include "libscan/ebook/ebook.h"
+#include "libscan/font/font.h"
+#include "libscan/media/media.h"
+#include "libscan/ooxml/ooxml.h"
+#include "libscan/text/text.h"
+#include "libscan/mobi/scan_mobi.h"
+#include "libscan/raw/raw.h"
+#include "libscan/msdoc/msdoc.h"
+#include "src/io/store.h"
 
-struct {
+#include <glib.h>
+#include <pcre.h>
+
+typedef struct {
     struct index_t index;
 
     GHashTable *mime_table;
@@ -11,14 +27,8 @@ struct {
 
     tpool_t *pool;
 
-    int tn_size;
     int threads;
-    int content_size;
-    float tn_qscale;
     int depth;
-    archive_mode_t archive_mode;
-    int verbose;
-    int very_verbose;
 
     size_t stat_tn_size;
     size_t stat_index_size;
@@ -26,31 +36,54 @@ struct {
     GHashTable *original_table;
     GHashTable *copy_table;
 
-    pthread_mutex_t mupdf_mu;
-    char * tesseract_lang;
-    const char * tesseract_path;
     pcre *exclude;
     pcre_extra *exclude_extra;
     int fast;
-} ScanCtx;
 
-struct {
+    scan_arc_ctx_t arc_ctx;
+    scan_comic_ctx_t comic_ctx;
+    scan_ebook_ctx_t ebook_ctx;
+    scan_font_ctx_t font_ctx;
+    scan_media_ctx_t media_ctx;
+    scan_ooxml_ctx_t ooxml_ctx;
+    scan_text_ctx_t text_ctx;
+    scan_mobi_ctx_t mobi_ctx;
+    scan_raw_ctx_t raw_ctx;
+    scan_msdoc_ctx_t msdoc_ctx;
+} ScanCtx_t;
+
+typedef struct {
     int verbose;
     int very_verbose;
     int no_color;
-} LogCtx;
+} LogCtx_t;
 
-struct {
+typedef struct {
     char *es_url;
+    char *es_index;
     int batch_size;
-} IndexCtx;
+    tpool_t *pool;
+    store_t *tag_store;
+    GHashTable *tags;
+    store_t *meta_store;
+    GHashTable *meta;
+} IndexCtx_t;
 
-struct {
+typedef struct {
     char *es_url;
+    char *es_index;
     int index_count;
-    char *b64credentials;
-    struct index_t indices[16];
-} WebCtx;
+    char *auth_user;
+    char *auth_pass;
+    int auth_enabled;
+    int tag_auth_enabled;
+    struct index_t indices[64];
+} WebCtx_t;
+
+extern ScanCtx_t ScanCtx;
+extern WebCtx_t WebCtx;
+extern IndexCtx_t IndexCtx;
+extern LogCtx_t LogCtx;
 
 
 #endif

src/index/elastic.c

@@ -1,27 +1,41 @@
 #include "elastic.h"
 #include "src/ctx.h"
 
+#include "web.h"
+
 #include "static_generated.c"
 
 
 typedef struct es_indexer {
     int queued;
     char *es_url;
+    char *es_index;
     es_bulk_line_t *line_head;
     es_bulk_line_t *line_tail;
 } es_indexer_t;
 
 
-static es_indexer_t *Indexer;
+static __thread es_indexer_t *Indexer;
 
 void delete_queue(int max);
 
-void print_json(cJSON *document, const char uuid_str[UUID_STR_LEN]) {
+void elastic_flush();
+
+void elastic_cleanup() {
+    elastic_flush();
+    if (Indexer != NULL) {
+        free(Indexer->es_index);
+        free(Indexer->es_url);
+        free(Indexer);
+    }
+}
+
+void print_json(cJSON *document, const char id_str[MD5_STR_LENGTH]) {
 
     cJSON *line = cJSON_CreateObject();
 
-    cJSON_AddStringToObject(line, "_id", uuid_str);
-    cJSON_AddStringToObject(line, "_index", "sist2");
+    cJSON_AddStringToObject(line, "_id", id_str);
+    cJSON_AddStringToObject(line, "_index", IndexCtx.es_index);
     cJSON_AddStringToObject(line, "_type", "_doc");
     cJSON_AddItemReferenceToObject(line, "_source", document);
 
@@ -33,23 +47,31 @@ void print_json(cJSON *document, const char uuid_str[UUID_STR_LEN]) {
     cJSON_Delete(line);
 }
 
-void index_json(cJSON *document, const char uuid_str[UUID_STR_LEN]) {
+void index_json_func(void *arg) {
+    es_bulk_line_t *line = arg;
+    elastic_index_line(line);
+}
 
+void index_json(cJSON *document, const char index_id_str[MD5_STR_LENGTH]) {
     char *json = cJSON_PrintUnformatted(document);
 
     size_t json_len = strlen(json);
     es_bulk_line_t *bulk_line = malloc(sizeof(es_bulk_line_t) + json_len + 2);
     memcpy(bulk_line->line, json, json_len);
-    memcpy(bulk_line->uuid_str, uuid_str, UUID_STR_LEN);
+    memcpy(bulk_line->path_md5_str, index_id_str, MD5_STR_LENGTH);
     *(bulk_line->line + json_len) = '\n';
     *(bulk_line->line + json_len + 1) = '\0';
     bulk_line->next = NULL;
 
     cJSON_free(json);
-    elastic_index_line(bulk_line);
+    tpool_add_work(IndexCtx.pool, index_json_func, bulk_line);
 }
 
-void execute_update_script(const char *script, const char index_id[UUID_STR_LEN]) {
+void execute_update_script(const char *script, int async, const char index_id[MD5_STR_LENGTH]) {
+
+    if (Indexer == NULL) {
+        Indexer = create_indexer(IndexCtx.es_url, IndexCtx.es_index);
+    }
 
     cJSON *body = cJSON_CreateObject();
     cJSON *script_obj = cJSON_AddObjectToObject(body, "script");
@@ -63,9 +85,16 @@ void execute_update_script(const char *script, const char index_id[UUID_STR_LEN]
     char *str = cJSON_Print(body);
 
     char bulk_url[4096];
-    snprintf(bulk_url, 4096, "%s/sist2/_update_by_query?pretty", Indexer->es_url);
-    response_t *r = web_post(bulk_url, str, "Content-Type: application/json");
-    LOG_INFOF("elastic.c", "Executed user script <%d>", r->status_code);
+    if (async) {
+        snprintf(bulk_url, sizeof(bulk_url), "%s/%s/_update_by_query?wait_for_completion=false", Indexer->es_url,
+                 Indexer->es_index);
+    } else {
+        snprintf(bulk_url, sizeof(bulk_url), "%s/%s/_update_by_query", Indexer->es_url, Indexer->es_index);
+    }
+    response_t *r = web_post(bulk_url, str);
+    if (!async) {
+        LOG_INFOF("elastic.c", "Executed user script <%d>", r->status_code);
+    }
     cJSON *resp = cJSON_Parse(r->body);
 
     cJSON_free(str);
@@ -80,6 +109,11 @@ void execute_update_script(const char *script, const char index_id[UUID_STR_LEN]
         cJSON_free(error_str);
     }
 
+    if (async) {
+        cJSON *task = cJSON_GetObjectItem(resp, "task");
+        LOG_INFOF("elastic.c", "User script queued: %s/_tasks/%s", Indexer->es_url, task->valuestring);
+    }
+
     cJSON_Delete(resp);
 }
 
@@ -89,16 +123,25 @@ void *create_bulk_buffer(int max, int *count, size_t *buf_len) {
 
     size_t buf_size = 0;
     size_t buf_cur = 0;
-    char *buf = malloc(1);
+    char *buf = malloc(8192);
+    size_t buf_capacity = 8192;
 
     while (line != NULL && *count < max) {
-        char action_str[512];
-        snprintf(action_str, 512,
-                 "{\"index\":{\"_id\":\"%s\", \"_type\":\"_doc\", \"_index\":\"sist2\"}}\n", line->uuid_str);
-        size_t action_str_len = strlen(action_str);
+        char action_str[256];
+        snprintf(
+                action_str, sizeof(action_str),
+                "{\"index\":{\"_id\":\"%s\",\"_type\":\"_doc\",\"_index\":\"%s\"}}\n",
+                line->path_md5_str, Indexer->es_index
+        );
 
+        size_t action_str_len = strlen(action_str);
         size_t line_len = strlen(line->line);
-        buf = realloc(buf, buf_size + line_len + action_str_len);
+
+        while (buf_size + line_len + action_str_len > buf_capacity) {
+            buf_capacity *= 2;
+            buf = realloc(buf, buf_capacity);
+        }
+
         buf_size += line_len + action_str_len;
 
         memcpy(buf + buf_cur, action_str, action_str_len);
@@ -109,15 +152,23 @@ void *create_bulk_buffer(int max, int *count, size_t *buf_len) {
         line = line->next;
         (*count)++;
     }
-    buf = realloc(buf, buf_size + 1);
+
+    if (buf_size + 1 > buf_capacity) {
+        buf = realloc(buf, buf_capacity + 1);
+    }
+
     *(buf + buf_cur) = '\0';
 
     *buf_len = buf_cur;
     return buf;
 }
 
-void *print_errors(response_t *r) {
-    cJSON *ret_json = cJSON_Parse(r->body);
+void print_errors(response_t *r) {
+    char *tmp = malloc(r->size + 1);
+    memcpy(tmp, r->body, r->size);
+    *(tmp + r->size) = '\0';
+
+    cJSON *ret_json = cJSON_Parse(tmp);
     if (cJSON_GetObjectItem(ret_json, "errors")->valueint != 0) {
         cJSON *err;
         cJSON_ArrayForEach(err, cJSON_GetObjectItem(ret_json, "items")) {
@@ -129,16 +180,38 @@ void *print_errors(response_t *r) {
         }
     }
     cJSON_Delete(ret_json);
+    free(tmp);
+}
+
+void print_error(response_t *r) {
+    char *tmp = malloc(r->size + 1);
+    memcpy(tmp, r->body, r->size);
+    *(tmp + r->size) = '\0';
+
+    cJSON *ret_json = cJSON_Parse(tmp);
+    if (cJSON_GetObjectItem(ret_json, "error") != NULL) {
+        char *str = cJSON_Print(cJSON_GetObjectItem(ret_json, "error"));
+        LOG_ERRORF("elastic.c", "%s\n", str);
+        cJSON_free(str);
+    }
+    cJSON_Delete(ret_json);
+    free(tmp);
 }
 
 void _elastic_flush(int max) {
+
+    if (max == 0) {
+        LOG_WARNING("elastic.c", "calling _elastic_flush with 0 in queue")
+        return;
+    }
+
     size_t buf_len;
     int count;
     void *buf = create_bulk_buffer(max, &count, &buf_len);
 
     char bulk_url[4096];
-    snprintf(bulk_url, 4096, "%s/sist2/_bulk?pipeline=tie", Indexer->es_url);
-    response_t *r = web_post(bulk_url, buf, "Content-Type: application/x-ndjson");
+    snprintf(bulk_url, sizeof(bulk_url), "%s/%s/_bulk?pipeline=tie", Indexer->es_url, Indexer->es_index);
+    response_t *r = web_post(bulk_url, buf);
 
     if (r->status_code == 0) {
         LOG_FATALF("elastic.c", "Could not connect to %s, make sure that elasticsearch is running!\n", IndexCtx.es_url)
@@ -147,7 +220,7 @@ void _elastic_flush(int max) {
     if (r->status_code == 413) {
 
         if (max <= 1) {
-            LOG_ERRORF("elastic.c", "Single document too large, giving up: {%s}", Indexer->line_head->uuid_str)
+            LOG_ERRORF("elastic.c", "Single document too large, giving up: {%s}", Indexer->line_head->path_md5_str)
             free_response(r);
             free(buf);
             delete_queue(1);
@@ -164,6 +237,15 @@ void _elastic_flush(int max) {
         _elastic_flush(max / 2);
         return;
 
+    } else if (r->status_code == 429) {
+
+        free_response(r);
+        free(buf);
+        LOG_WARNING("elastic.c", "Got 429 status, will retry after delay")
+        usleep(1000000 * 20);
+        _elastic_flush(max);
+        return;
+
     } else if (r->status_code != 200) {
         print_errors(r);
         delete_queue(Indexer->queued);
@@ -189,9 +271,8 @@ void delete_queue(int max) {
         Indexer->line_head = tmp->next;
         if (Indexer->line_head == NULL) {
             Indexer->line_tail = NULL;
-        } else {
-            free(tmp);
         }
+        free(tmp);
         Indexer->queued -= 1;
     }
 }
@@ -199,7 +280,7 @@ void delete_queue(int max) {
 void elastic_flush() {
 
     if (Indexer == NULL) {
-        Indexer = create_indexer(IndexCtx.es_url);
+        Indexer = create_indexer(IndexCtx.es_url, IndexCtx.es_index);
     }
 
     _elastic_flush(Indexer->queued);
@@ -208,7 +289,7 @@ void elastic_flush() {
 void elastic_index_line(es_bulk_line_t *line) {
 
     if (Indexer == NULL) {
-        Indexer = create_indexer(IndexCtx.es_url);
+        Indexer = create_indexer(IndexCtx.es_url, IndexCtx.es_index);
     }
 
     if (Indexer->line_head == NULL) {
@@ -226,14 +307,18 @@ void elastic_index_line(es_bulk_line_t *line) {
     }
 }
 
-es_indexer_t *create_indexer(const char *url) {
+es_indexer_t *create_indexer(const char *url, const char *index) {
 
     char *es_url = malloc(strlen(url) + 1);
     strcpy(es_url, url);
 
+    char *es_index = malloc(strlen(index) + 1);
+    strcpy(es_index, index);
+
     es_indexer_t *indexer = malloc(sizeof(es_indexer_t));
 
     indexer->es_url = es_url;
+    indexer->es_index = es_index;
     indexer->queued = 0;
     indexer->line_head = NULL;
     indexer->line_tail = NULL;
@@ -241,41 +326,42 @@ es_indexer_t *create_indexer(const char *url) {
     return indexer;
 }
 
-void destroy_indexer(char *script, char index_id[UUID_STR_LEN]) {
+void finish_indexer(char *script, int async_script, char *index_id) {
 
     char url[4096];
 
-    snprintf(url, sizeof(url), "%s/sist2/_refresh", IndexCtx.es_url);
-    response_t *r = web_post(url, "", NULL);
+    snprintf(url, sizeof(url), "%s/%s/_refresh", IndexCtx.es_url, IndexCtx.es_index);
+    response_t *r = web_post(url, "");
     LOG_INFOF("elastic.c", "Refresh index <%d>", r->status_code);
     free_response(r);
 
     if (script != NULL) {
-        execute_update_script(script, index_id);
+        execute_update_script(script, async_script, index_id);
+        free(script);
+
+        snprintf(url, sizeof(url), "%s/%s/_refresh", IndexCtx.es_url, IndexCtx.es_index);
+        r = web_post(url, "");
+        LOG_INFOF("elastic.c", "Refresh index <%d>", r->status_code);
+        free_response(r);
     }
 
-    snprintf(url, sizeof(url), "%s/sist2/_refresh", IndexCtx.es_url);
-    r = web_post(url, "", NULL);
-    LOG_INFOF("elastic.c", "Refresh index <%d>", r->status_code);
-    free_response(r);
-
-    snprintf(url, sizeof(url), "%s/sist2/_forcemerge", IndexCtx.es_url);
-    r = web_post(url, "", NULL);
+    snprintf(url, sizeof(url), "%s/%s/_forcemerge", IndexCtx.es_url, IndexCtx.es_index);
+    r = web_post(url, "");
     LOG_INFOF("elastic.c", "Merge index <%d>", r->status_code);
     free_response(r);
 
-    if (Indexer != NULL) {
-        free(Indexer->es_url);
-        free(Indexer);
-    }
+    snprintf(url, sizeof(url), "%s/%s/_settings", IndexCtx.es_url, IndexCtx.es_index);
+    r = web_put(url, "{\"index\":{\"refresh_interval\":\"1s\"}}");
+    LOG_INFOF("elastic.c", "Set refresh interval <%d>", r->status_code);
+    free_response(r);
 }
 
-void elastic_init(int force_reset) {
+void elastic_init(int force_reset, const char* user_mappings, const char* user_settings) {
 
     // Check if index exists
     char url[4096];
-    snprintf(url, 4096, "%s/sist2", IndexCtx.es_url);
-    response_t *r = web_get(url);
+    snprintf(url, sizeof(url), "%s/%s", IndexCtx.es_url, IndexCtx.es_index);
+    response_t *r = web_get(url, 30);
     int index_exists = r->status_code == 200;
     free_response(r);
 
@@ -284,46 +370,56 @@ void elastic_init(int force_reset) {
         LOG_INFOF("elastic.c", "Delete index <%d>", r->status_code);
         free_response(r);
 
-        snprintf(url, 4096, "%s/sist2", IndexCtx.es_url);
-        r = web_put(url, "", NULL);
+        snprintf(url, sizeof(url), "%s/%s", IndexCtx.es_url, IndexCtx.es_index);
+        r = web_put(url, "");
+
+        if (r->status_code != 200) {
+            print_error(r);
+            LOG_FATAL("elastic.c", "Could not create index")
+        }
+
         LOG_INFOF("elastic.c", "Create index <%d>", r->status_code);
         free_response(r);
 
-        snprintf(url, 4096, "%s/sist2/_close", IndexCtx.es_url);
-        r = web_post(url, "", NULL);
+        snprintf(url, sizeof(url), "%s/%s/_close", IndexCtx.es_url, IndexCtx.es_index);
+        r = web_post(url, "");
         LOG_INFOF("elastic.c", "Close index <%d>", r->status_code);
         free_response(r);
 
-        snprintf(url, 4096, "%s/_ingest/pipeline/tie", IndexCtx.es_url);
-        r = web_put(url, pipeline_json, "Content-Type: application/json");
+        snprintf(url, sizeof(url), "%s/_ingest/pipeline/tie", IndexCtx.es_url);
+        r = web_put(url, pipeline_json);
         LOG_INFOF("elastic.c", "Create pipeline <%d>", r->status_code);
         free_response(r);
 
-        snprintf(url, 4096, "%s/sist2/_settings", IndexCtx.es_url);
-        r = web_put(url, settings_json, "Content-Type: application/json");
-        LOG_INFOF("elastic.c", "Update settings <%d>", r->status_code);
+        snprintf(url, sizeof(url), "%s/%s/_settings", IndexCtx.es_url, IndexCtx.es_index);
+        r = web_put(url, user_settings ? user_settings : settings_json);
+        LOG_INFOF("elastic.c", "Update user_settings <%d>", r->status_code);
         free_response(r);
 
-        snprintf(url, 4096, "%s/sist2/_mappings/_doc?include_type_name=true", IndexCtx.es_url);
-        r = web_put(url, mappings_json, "Content-Type: application/json");
-        LOG_INFOF("elastic.c", "Update mappings <%d>", r->status_code);
+        snprintf(url, sizeof(url), "%s/%s/_mappings/_doc?include_type_name=true", IndexCtx.es_url, IndexCtx.es_index);
+        r = web_put(url, user_mappings ? user_mappings : mappings_json);
+        LOG_INFOF("elastic.c", "Update user_mappings <%d>", r->status_code);
         free_response(r);
 
-        snprintf(url, 4096, "%s/sist2/_open", IndexCtx.es_url);
-        r = web_post(url, "", NULL);
+        snprintf(url, sizeof(url), "%s/%s/_open", IndexCtx.es_url, IndexCtx.es_index);
+        r = web_post(url, "");
         LOG_INFOF("elastic.c", "Open index <%d>", r->status_code);
         free_response(r);
     }
 }
 
-cJSON *elastic_get_document(const char *uuid_str) {
+cJSON *elastic_get_document(const char *id_str) {
     char url[4096];
-    snprintf(url, 4096, "%s/sist2/_doc/%s", WebCtx.es_url, uuid_str);
+    snprintf(url, sizeof(url), "%s/%s/_doc/%s", WebCtx.es_url, WebCtx.es_index, id_str);
 
-    response_t *r = web_get(url);
+    response_t *r = web_get(url, 3);
     cJSON *json = NULL;
     if (r->status_code == 200) {
-        json = cJSON_Parse(r->body);
+        char *tmp = malloc(r->size + 1);
+        memcpy(tmp, r->body, r->size);
+        *(tmp + r->size) = '\0';
+        json = cJSON_Parse(tmp);
+        free(tmp);
     }
     free_response(r);
     return json;
@@ -331,21 +427,25 @@ cJSON *elastic_get_document(const char *uuid_str) {
 
 char *elastic_get_status() {
     char url[4096];
-    snprintf(url, 4096,
-             "%s/_cluster/state/metadata/sist2?filter_path=metadata.indices.*.state", WebCtx.es_url);
+    snprintf(url, sizeof(url),
+             "%s/_cluster/state/metadata/%s?filter_path=metadata.indices.*.state", WebCtx.es_url, WebCtx.es_index);
 
-    response_t *r = web_get(url);
+    response_t *r = web_get(url, 30);
     cJSON *json = NULL;
     char *status = malloc(128 * sizeof(char));
     status[0] = '\0';
 
     if (r->status_code == 200) {
-        json = cJSON_Parse(r->body);
+        char *tmp = malloc(r->size + 1);
+        memcpy(tmp, r->body, r->size);
+        *(tmp + r->size) = '\0';
+        json = cJSON_Parse(tmp);
+        free(tmp);
         const cJSON *metadata = cJSON_GetObjectItem(json, "metadata");
         if (metadata != NULL) {
             const cJSON *indices = cJSON_GetObjectItem(metadata, "indices");
-            const cJSON *sist2 = cJSON_GetObjectItem(indices, "sist2");
-            const cJSON *state = cJSON_GetObjectItem(sist2, "state");
+            const cJSON *index = cJSON_GetObjectItem(indices, WebCtx.es_index);
+            const cJSON *state = cJSON_GetObjectItem(index, "state");
             strcpy(status, state->valuestring);
         }
     }

src/index/elastic.h

@@ -5,7 +5,7 @@
 
 typedef struct es_bulk_line {
     struct es_bulk_line *next;
-    char uuid_str[UUID_STR_LEN];
+    char path_md5_str[MD5_STR_LENGTH];
     char line[0];
 } es_bulk_line_t;
 
@@ -16,20 +16,21 @@ typedef struct es_indexer es_indexer_t;
 
 void elastic_index_line(es_bulk_line_t *line);
 
-void elastic_flush();
+void print_json(cJSON *document, const char index_id_str[MD5_STR_LENGTH]);
 
-void print_json(cJSON *document, const char uuid_str[UUID_STR_LEN]);
+void index_json(cJSON *document, const char index_id_str[MD5_STR_LENGTH]);
 
-void index_json(cJSON *document, const char uuid_str[UUID_STR_LEN]);
+es_indexer_t *create_indexer(const char *url, const char *index);
 
-es_indexer_t *create_indexer(const char* es_url);
+void elastic_cleanup();
+void finish_indexer(char *script, int async_script, char *index_id);
 
-void destroy_indexer(char *script, char index_id[UUID_STR_LEN]);
+void elastic_init(int force_reset, const char* user_mappings, const char* user_settings);
 
-void elastic_init(int force_reset);
-
-cJSON *elastic_get_document(const char *uuid_str);
+cJSON *elastic_get_document(const char *id_str);
 
 char *elastic_get_status();
 
+void execute_update_script(const char *script, int async, const char index_id[MD5_STR_LENGTH]);
+
 #endif

src/index/web.c

@@ -1,4 +1,11 @@
 #include "web.h"
+#include "src/sist.h"
+#include "src/ctx.h"
+
+#include <mongoose.h>
+#include <pthread.h>
+#include <curl/curl.h>
+
 
 size_t write_cb(char *ptr, size_t size, size_t nmemb, void *user_data) {
 
@@ -9,11 +16,91 @@ size_t write_cb(char *ptr, size_t size, size_t nmemb, void *user_data) {
 }
 
 void free_response(response_t *resp) {
-    free(resp->body);
+    if (resp->body != NULL) {
+        free(resp->body);
+    }
     free(resp);
 }
 
-response_t *web_get(const char *url) {
+void web_post_async_poll(subreq_ctx_t* req) {
+    fd_set fdread;
+    fd_set fdwrite;
+    fd_set fdexcep;
+    int maxfd = -1;
+
+    FD_ZERO(&fdread);
+    FD_ZERO(&fdwrite);
+    FD_ZERO(&fdexcep);
+
+    CURLMcode mc = curl_multi_fdset(req->multi, &fdread, &fdwrite, &fdexcep, &maxfd);
+
+    if(mc != CURLM_OK) {
+        req->done = TRUE;
+        return;
+    }
+
+    if (maxfd == -1) {
+        // no fds ready yet
+        return;
+    }
+
+    struct timeval timeout = {1, 0};
+    int rc = select(maxfd + 1, &fdread, &fdwrite, &fdexcep, &timeout);
+
+    switch(rc) {
+        case -1:
+            req->done = TRUE;
+            break;
+        case 0:
+            break;
+        default:
+            curl_multi_perform(req->multi, &req->running_handles);
+            break;
+    }
+
+    if (req->running_handles == 0) {
+        req->done = TRUE;
+        req->response->body = req->response_buf.buf;
+        req->response->size = req->response_buf.cur;
+        curl_easy_getinfo(req->handle, CURLINFO_RESPONSE_CODE, &req->response->status_code);
+
+        curl_multi_cleanup(req->multi);
+        curl_easy_cleanup(req->handle);
+        curl_slist_free_all(req->headers);
+        return;
+    }
+}
+
+subreq_ctx_t *web_post_async(const char *url, char *data) {
+    subreq_ctx_t *req = calloc(1, sizeof(subreq_ctx_t));
+    req->response = calloc(1, sizeof(response_t));
+    req->data = data;
+    req->response_buf = dyn_buffer_create();
+
+    req->handle = curl_easy_init();
+    CURL *curl = req->handle;
+    curl_easy_setopt(curl, CURLOPT_URL, url);
+    curl_easy_setopt(curl, CURLOPT_WRITEDATA, (void *) (&req->response_buf));
+    curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, write_cb);
+    curl_easy_setopt(curl, CURLOPT_POST, 1);
+    curl_easy_setopt(curl, CURLOPT_USERAGENT, "sist2");
+
+    struct curl_slist *headers = NULL;
+    headers = curl_slist_append(headers, "Content-Type: application/json");
+    curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
+
+    curl_easy_setopt(curl, CURLOPT_POSTFIELDS, data);
+
+    req->multi = curl_multi_init();
+    curl_multi_add_handle(req->multi, curl);
+    curl_multi_perform(req->multi, &req->running_handles);
+
+    LOG_DEBUGF("web.c", "async request POST %s", url)
+
+    return req;
+}
+
+response_t *web_get(const char *url, int timeout) {
     response_t *resp = malloc(sizeof(response_t));
 
     CURL *curl;
@@ -24,18 +111,24 @@ response_t *web_get(const char *url) {
     curl_easy_setopt(curl, CURLOPT_WRITEDATA, (void *) (&buffer));
     curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, write_cb);
     curl_easy_setopt(curl, CURLOPT_USERAGENT, "sist2");
+    curl_easy_setopt(curl, CURLOPT_TIMEOUT, timeout);
+
+    struct curl_slist *headers = NULL;
+    headers = curl_slist_append(headers, "Content-Type: application/json");
+    curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
 
     curl_easy_perform(curl);
     curl_easy_getinfo(curl, CURLINFO_RESPONSE_CODE, &resp->status_code);
 
     curl_easy_cleanup(curl);
+    curl_slist_free_all(headers);
 
     resp->body = buffer.buf;
     resp->size = buffer.cur;
     return resp;
 }
 
-response_t *web_post(const char *url, const char *data, const char *header) {
+response_t *web_post(const char *url, const char *data) {
 
     response_t *resp = malloc(sizeof(response_t));
 
@@ -50,10 +143,8 @@ response_t *web_post(const char *url, const char *data, const char *header) {
     curl_easy_setopt(curl, CURLOPT_USERAGENT, "sist2");
 
     struct curl_slist *headers = NULL;
-    if (header != NULL) {
-        headers = curl_slist_append(headers, header);
-        curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
-    }
+    headers = curl_slist_append(headers, "Content-Type: application/json");
+    curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
 
     curl_easy_setopt(curl, CURLOPT_POSTFIELDS, data);
 
@@ -70,7 +161,7 @@ response_t *web_post(const char *url, const char *data, const char *header) {
 }
 
 
-response_t *web_put(const char *url, const char *data, const char *header) {
+response_t *web_put(const char *url, const char *data) {
 
     response_t *resp = malloc(sizeof(response_t));
 
@@ -86,11 +177,9 @@ response_t *web_put(const char *url, const char *data, const char *header) {
     curl_easy_setopt(curl, CURLOPT_DNS_USE_GLOBAL_CACHE, 0);
     curl_easy_setopt(curl, CURLOPT_IPRESOLVE, CURLOPT_DNS_LOCAL_IP4 );
 
-    if (header != NULL) {
-        struct curl_slist *headers = NULL;
-        headers = curl_slist_append(headers, header);
-        curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
-    }
+    struct curl_slist *headers = NULL;
+    headers = curl_slist_append(headers, "Content-Type: application/json");
+    curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
 
     curl_easy_setopt(curl, CURLOPT_POSTFIELDS, data);
 
@@ -98,6 +187,7 @@ response_t *web_put(const char *url, const char *data, const char *header) {
     curl_easy_getinfo(curl, CURLINFO_RESPONSE_CODE, &resp->status_code);
 
     curl_easy_cleanup(curl);
+    curl_slist_free_all(headers);
 
     resp->body = buffer.buf;
     resp->size = buffer.cur;
@@ -119,13 +209,17 @@ response_t *web_delete(const char *url) {
     curl_easy_setopt(curl, CURLOPT_USERAGENT, "sist2");
 
     curl_easy_setopt(curl, CURLOPT_POSTFIELDS, "");
+    struct curl_slist *headers = NULL;
+    headers = curl_slist_append(headers, "Content-Type: application/json");
+    curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
 
     curl_easy_perform(curl);
     curl_easy_getinfo(curl, CURLINFO_RESPONSE_CODE, &resp->status_code);
 
     curl_easy_cleanup(curl);
+    curl_slist_free_all(headers);
 
     resp->body = buffer.buf;
     resp->size = buffer.cur;
     return resp;
-}
+}

src/index/web.h

@@ -2,6 +2,8 @@
 #define SIST2_WEB_H
 
 #include "src/sist.h"
+#include <mongoose.h>
+#include <curl/curl.h>
 
 typedef struct response {
     char *body;
@@ -9,9 +11,27 @@ typedef struct response {
     int status_code;
 } response_t;
 
-response_t *web_get(const char *url);
-response_t *web_post(const char * url, const char * data, const char* header);
-response_t *web_put(const char *url, const char *data, const char *header);
+typedef struct {
+    response_t *resp;
+    int done;
+} http_ev_data_t;
+
+typedef struct {
+    char* data;
+    dyn_buffer_t response_buf;
+    struct curl_slist *headers;
+    CURL *handle;
+    CURLM *multi;
+    response_t *response;
+    int running_handles;
+    int done;
+} subreq_ctx_t;
+
+response_t *web_get(const char *url, int timeout);
+response_t *web_post(const char * url, const char * data);
+void web_post_async_poll(subreq_ctx_t* req);
+subreq_ctx_t *web_post_async(const char *url, char *data);
+response_t *web_put(const char *url, const char *data);
 response_t *web_delete(const char *url);
 
 void free_response(response_t *resp);

src/io/serialize.c

@@ -1,16 +1,18 @@
 #include "src/ctx.h"
 #include "serialize.h"
+#include "src/parsing/parse.h"
+#include "src/parsing/mime.h"
 
 static __thread int index_fd = -1;
 
 typedef struct {
-    unsigned char uuid[16];
-    unsigned long ino;
+    unsigned char path_md5[MD5_DIGEST_LENGTH];
     unsigned long size;
     unsigned int mime;
     int mtime;
     short base;
     short ext;
+    char has_parent;
 } line_t;
 
 void skip_meta(FILE *file) {
@@ -30,7 +32,7 @@ void skip_meta(FILE *file) {
 
 void write_index_descriptor(char *path, index_descriptor_t *desc) {
     cJSON *json = cJSON_CreateObject();
-    cJSON_AddStringToObject(json, "uuid", desc->uuid);
+    cJSON_AddStringToObject(json, "id", desc->id);
     cJSON_AddStringToObject(json, "version", desc->version);
     cJSON_AddStringToObject(json, "root", desc->root);
     cJSON_AddStringToObject(json, "name", desc->name);
@@ -40,10 +42,13 @@ void write_index_descriptor(char *path, index_descriptor_t *desc) {
 
     int fd = open(path, O_CREAT | O_WRONLY, S_IRUSR | S_IWUSR);
     if (fd < 0) {
-        LOG_FATALF("serialize.c", "Could not write index descriptor: %s", strerror(errno));
+        LOG_FATALF("serialize.c", "Could not open index descriptor: %s", strerror(errno));
     }
     char *str = cJSON_Print(json);
-    write(fd, str, strlen(str));
+    int ret = write(fd, str, strlen(str));
+    if (ret == -1) {
+        LOG_FATALF("serialize.c", "Could not write index descriptor: %s", strerror(errno));
+    }
     free(str);
     close(fd);
 
@@ -57,11 +62,14 @@ index_descriptor_t read_index_descriptor(char *path) {
     int fd = open(path, O_RDONLY);
 
     if (fd == -1) {
-        LOG_FATALF("serialize.c", "Invalid/corrupt index (Could not find descriptor): %s: %s\n", path ,strerror(errno))
+        LOG_FATALF("serialize.c", "Invalid/corrupt index (Could not find descriptor): %s: %s\n", path, strerror(errno))
     }
 
     char *buf = malloc(info.st_size + 1);
-    read(fd, buf, info.st_size);
+    int ret = read(fd, buf, info.st_size);
+    if (ret == -1) {
+        LOG_FATALF("serialize.c", "Could not read index descriptor: %s", strerror(errno));
+    }
     *(buf + info.st_size) = '\0';
     close(fd);
 
@@ -74,7 +82,7 @@ index_descriptor_t read_index_descriptor(char *path) {
     strcpy(descriptor.rewrite_url, cJSON_GetObjectItem(json, "rewrite_url")->valuestring);
     descriptor.root_len = (short) strlen(descriptor.root);
     strcpy(descriptor.version, cJSON_GetObjectItem(json, "version")->valuestring);
-    strcpy(descriptor.uuid, cJSON_GetObjectItem(json, "uuid")->valuestring);
+    strcpy(descriptor.id, cJSON_GetObjectItem(json, "id")->valuestring);
     if (cJSON_GetObjectItem(json, "type") == NULL) {
         strcpy(descriptor.type, INDEX_TYPE_BIN);
     } else {
@@ -136,6 +144,14 @@ char *get_meta_key_text(enum metakey meta_key) {
             return "exif_model";
         case MetaExifDateTime:
             return "exif_datetime";
+        case MetaAuthor:
+            return "author";
+        case MetaModifiedBy:
+            return "modified_by";
+        case MetaThumbnail:
+            return "thumbnail";
+        case MetaPages:
+            return "pages";
         default:
             return NULL;
     }
@@ -158,8 +174,8 @@ void write_document(document_t *doc) {
     dyn_buffer_t buf = dyn_buffer_create();
 
     // Ignore root directory in the file path
-    doc->ext = doc->ext - ScanCtx.index.desc.root_len;
-    doc->base = doc->base - ScanCtx.index.desc.root_len;
+    doc->ext = (short) (doc->ext - ScanCtx.index.desc.root_len);
+    doc->base = (short) (doc->base - ScanCtx.index.desc.root_len);
     doc->filepath += ScanCtx.index.desc.root_len;
 
     dyn_buffer_write(&buf, doc, sizeof(line_t));
@@ -170,11 +186,11 @@ void write_document(document_t *doc) {
         dyn_buffer_write_char(&buf, meta->key);
 
         if (IS_META_INT(meta->key)) {
-            dyn_buffer_write_int(&buf, meta->intval);
+            dyn_buffer_write_int(&buf, meta->int_val);
         } else if (IS_META_LONG(meta->key)) {
-            dyn_buffer_write_long(&buf, meta->longval);
+            dyn_buffer_write_long(&buf, meta->long_val);
         } else {
-            dyn_buffer_write_str(&buf, meta->strval);
+            dyn_buffer_write_str(&buf, meta->str_val);
         }
 
         meta_line_t *tmp = meta;
@@ -193,6 +209,8 @@ void write_document(document_t *doc) {
 
 void thread_cleanup() {
     close(index_fd);
+    cleanup_parse();
+    cleanup_font();
 }
 
 
@@ -201,9 +219,9 @@ void read_index_bin(const char *path, const char *index_id, index_func func) {
     dyn_buffer_t buf = dyn_buffer_create();
 
     FILE *file = fopen(path, "rb");
-    while (1) {
+    while (TRUE) {
         buf.cur = 0;
-        fread((void *) &line, 1, sizeof(line_t), file);
+        size_t _ = fread((void *) &line, 1, sizeof(line_t), file);
         if (feof(file)) {
             break;
         }
@@ -211,10 +229,10 @@ void read_index_bin(const char *path, const char *index_id, index_func func) {
         cJSON *document = cJSON_CreateObject();
         cJSON_AddStringToObject(document, "index", index_id);
 
-        char uuid_str[UUID_STR_LEN];
-        uuid_unparse(line.uuid, uuid_str);
+        char path_md5_str[MD5_STR_LENGTH];
+        buf2hex(line.path_md5, sizeof(line.path_md5), path_md5_str);
 
-        const char* mime_text = mime_get_mime_text(line.mime);
+        const char *mime_text = mime_get_mime_text(line.mime);
         if (mime_text == NULL) {
             cJSON_AddNullToObject(document, "mime");
         } else {
@@ -223,7 +241,7 @@ void read_index_bin(const char *path, const char *index_id, index_func func) {
         cJSON_AddNumberToObject(document, "size", (double) line.size);
         cJSON_AddNumberToObject(document, "mtime", line.mtime);
 
-        int c;
+        int c = 0;
         while ((c = getc(file)) != 0) {
             dyn_buffer_write_char(&buf, (char) c);
         }
@@ -235,42 +253,42 @@ void read_index_bin(const char *path, const char *index_id, index_func func) {
         } else {
             *(buf.buf + line.ext) = '\0';
         }
-        cJSON_AddStringToObject(document, "name", buf.buf + line.base);
+
+        char tmp[PATH_MAX * 3];
+
+        str_escape(tmp, buf.buf + line.base);
+        cJSON_AddStringToObject(document, "name", tmp);
+
         if (line.base > 0) {
             *(buf.buf + line.base - 1) = '\0';
-            cJSON_AddStringToObject(document, "path", buf.buf);
+
+            str_escape(tmp, buf.buf);
+            cJSON_AddStringToObject(document, "path", tmp);
         } else {
             cJSON_AddStringToObject(document, "path", "");
         }
 
         enum metakey key = getc(file);
+        size_t ret = 0;
         while (key != '\n') {
             switch (key) {
+                case MetaPages:
                 case MetaWidth:
                 case MetaHeight: {
                     int value;
-                    fread(&value, sizeof(int), 1, file);
+                    ret = fread(&value, sizeof(int), 1, file);
                     cJSON_AddNumberToObject(document, get_meta_key_text(key), value);
                     break;
                 }
                 case MetaMediaDuration:
                 case MetaMediaBitrate: {
                     long value;
-                    fread(&value, sizeof(long), 1, file);
+                    ret = fread(&value, sizeof(long), 1, file);
                     cJSON_AddNumberToObject(document, get_meta_key_text(key), (double) value);
                     break;
                 }
                 case MetaMediaAudioCodec:
-                case MetaMediaVideoCodec: {
-                    int value;
-                    fread(&value, sizeof(int), 1, file);
-                    const AVCodecDescriptor *desc = avcodec_descriptor_get(value);
-                    if (desc != NULL) {
-                        cJSON_AddStringToObject(document, get_meta_key_text(key), desc->name);
-                    }
-                    break;
-                }
-
+                case MetaMediaVideoCodec:
                 case MetaContent:
                 case MetaArtist:
                 case MetaAlbum:
@@ -287,6 +305,9 @@ void read_index_bin(const char *path, const char *index_id, index_func func) {
                 case MetaExifIsoSpeedRatings:
                 case MetaExifDateTime:
                 case MetaExifModel:
+                case MetaAuthor:
+                case MetaModifiedBy:
+                case MetaThumbnail:
                 case MetaTitle: {
                     buf.cur = 0;
                     while ((c = getc(file)) != 0) {
@@ -305,8 +326,36 @@ void read_index_bin(const char *path, const char *index_id, index_func func) {
             key = getc(file);
         }
 
-        func(document, uuid_str);
+        cJSON *meta_obj = NULL;
+        if (IndexCtx.meta != NULL) {
+            const char *meta_string = g_hash_table_lookup(IndexCtx.meta, path_md5_str);
+            if (meta_string != NULL) {
+                meta_obj = cJSON_Parse(meta_string);
+
+                cJSON *child;
+                for (child = meta_obj->child; child != NULL; child = child->next) {
+                    char meta_key[4096];
+                    strcpy(meta_key, child->string);
+                    cJSON_DeleteItemFromObject(document, meta_key);
+                    cJSON_AddItemReferenceToObject(document, meta_key, child);
+                }
+            }
+        }
+
+        if (IndexCtx.tags != NULL) {
+            const char *tags_string = g_hash_table_lookup(IndexCtx.tags, path_md5_str);
+            if (tags_string != NULL) {
+                cJSON *tags_arr = cJSON_Parse(tags_string);
+                cJSON_DeleteItemFromObject(document, "tag");
+                cJSON_AddItemToObject(document, "tag", tags_arr);
+            }
+        }
+
+        func(document, path_md5_str);
         cJSON_Delete(document);
+        if (meta_obj) {
+            cJSON_Delete(meta_obj);
+        }
     }
     dyn_buffer_destroy(&buf);
     fclose(file);
@@ -330,7 +379,7 @@ const char *json_type_array_fields[] = {
 void read_index_json(const char *path, UNUSED(const char *index_id), index_func func) {
 
     FILE *file = fopen(path, "r");
-    while (1) {
+    while (TRUE) {
         char *line = NULL;
         size_t len;
         size_t read = getline(&line, &len, file);
@@ -350,7 +399,7 @@ void read_index_json(const char *path, UNUSED(const char *index_id), index_func
         }
 
         cJSON *document = cJSON_CreateObject();
-        const char *uuid_str = cJSON_GetObjectItem(input, "_id")->valuestring;
+        const char *id_str = cJSON_GetObjectItem(input, "_id")->valuestring;
 
         for (int i = 0; i < (sizeof(json_type_copy_fields) / sizeof(json_type_copy_fields[0])); i++) {
             cJSON *value = cJSON_GetObjectItem(input, json_type_copy_fields[i]);
@@ -378,7 +427,7 @@ void read_index_json(const char *path, UNUSED(const char *index_id), index_func
             }
         }
 
-        func(document, uuid_str);
+        func(document, id_str);
         cJSON_Delete(document);
         cJSON_Delete(input);
 
@@ -386,7 +435,7 @@ void read_index_json(const char *path, UNUSED(const char *index_id), index_func
     fclose(file);
 }
 
-void read_index(const char *path, const char index_id[UUID_STR_LEN], const char *type, index_func func) {
+void read_index(const char *path, const char index_id[MD5_STR_LENGTH], const char *type, index_func func) {
 
     if (strcmp(type, INDEX_TYPE_BIN) == 0) {
         read_index_bin(path, index_id, func);
@@ -399,13 +448,15 @@ void incremental_read(GHashTable *table, const char *filepath) {
     FILE *file = fopen(filepath, "rb");
     line_t line;
 
+    LOG_DEBUGF("serialize.c", "Incremental read %s", filepath)
+
     while (1) {
-        fread((void *) &line, 1, sizeof(line_t), file);
-        if (feof(file)) {
+        size_t ret = fread((void *) &line, sizeof(line_t), 1, file);
+        if (ret != 1 || feof(file)) {
             break;
         }
 
-        incremental_put(table, line.ino, line.mtime);
+        incremental_put(table, line.path_md5, line.mtime);
 
         while ((getc(file))) {}
         skip_meta(file);
@@ -423,41 +474,55 @@ void incremental_copy(store_t *store, store_t *dst_store, const char *filepath,
     FILE *dst_file = fopen(dst_filepath, "ab");
     line_t line;
 
-    while (1) {
-        fread((void *) &line, 1, sizeof(line_t), file);
-        if (feof(file)) {
+    LOG_DEBUGF("serialize.c", "Incremental copy %s", filepath)
+
+    while (TRUE) {
+        size_t ret = fread((void *) &line, sizeof(line_t), 1, file);
+        if (ret != 1 || feof(file)) {
             break;
         }
 
-        if (incremental_get(copy_table, line.ino)) {
+        // Assume that files with parents still exist.
+        //  One way to "fix" this would be to check if the parent is marked for copy but it would consistently
+        //  delete files with grandparents, which is a side-effect worse than having orphaned files
+        if (line.has_parent || incremental_get(copy_table, line.path_md5)) {
             fwrite(&line, sizeof(line), 1, dst_file);
 
-            size_t buf_len;
-            char *buf = store_read(store, (char *) line.uuid, 16, &buf_len);
-            store_write(dst_store, (char *) line.uuid, 16, buf, buf_len);
-            free(buf);
-
+            // Copy filepath
+            char filepath_buf[PATH_MAX];
             char c;
+            char *ptr = filepath_buf;
             while ((c = (char) getc(file))) {
-                fwrite(&c, sizeof(c), 1, dst_file);
+                *ptr++ = c;
+            }
+            *ptr = '\0';
+            fwrite(filepath_buf, (ptr - filepath_buf) + 1, 1, dst_file);
+
+            // Copy tn store contents
+            size_t buf_len;
+            char path_md5[MD5_DIGEST_LENGTH];
+            MD5((unsigned char *) filepath_buf, (ptr - filepath_buf), (unsigned char *) path_md5);
+            char *buf = store_read(store, path_md5, sizeof(path_md5), &buf_len);
+            if (buf_len != 0) {
+                store_write(dst_store, path_md5, sizeof(path_md5), buf, buf_len);
+                free(buf);
             }
-            fwrite("\0", sizeof(c), 1, dst_file);
 
             enum metakey key;
             while (1) {
                 key = getc(file);
+                fwrite(&key, sizeof(char), 1, dst_file);
                 if (key == '\n') {
                     break;
                 }
-                fwrite(&key, sizeof(char), 1, dst_file);
 
                 if (IS_META_INT(key)) {
                     int val;
-                    fread(&val, sizeof(val), 1, file);
+                    ret = fread(&val, sizeof(val), 1, file);
                     fwrite(&val, sizeof(val), 1, dst_file);
                 } else if (IS_META_LONG(key)) {
                     long val;
-                    fread(&val, sizeof(val), 1, file);
+                    ret = fread(&val, sizeof(val), 1, file);
                     fwrite(&val, sizeof(val), 1, dst_file);
                 } else {
                     while ((c = (char) getc(file))) {
@@ -467,8 +532,10 @@ void incremental_copy(store_t *store, store_t *dst_store, const char *filepath,
                 }
             }
         } else {
+            while ((getc(file))) {}
             skip_meta(file);
         }
     }
     fclose(file);
+    fclose(dst_file);
 }

src/io/serialize.h

@@ -2,16 +2,19 @@
 #define SIST2_SERIALIZE_H
 
 #include "src/sist.h"
-#include <sys/syscall.h>
+#include "store.h"
 
-typedef void(*index_func)(cJSON *, const char[UUID_STR_LEN]);
+#include <sys/syscall.h>
+#include <glib.h>
+
+typedef void(*index_func)(cJSON *, const char[MD5_STR_LENGTH]);
 
 void incremental_copy(store_t *store, store_t *dst_store, const char *filepath,
                       const char *dst_filepath, GHashTable *copy_table);
 
 void write_document(document_t *doc);
 
-void read_index(const char *path, const char[UUID_STR_LEN], const char *type, index_func);
+void read_index(const char *path, const char[MD5_STR_LENGTH], const char *type, index_func);
 
 void incremental_read(GHashTable *table, const char *filepath);
 

src/io/store.c

@@ -1,9 +1,10 @@
 #include "store.h"
 #include "src/ctx.h"
 
-store_t *store_create(char *path) {
+store_t *store_create(char *path, size_t chunk_size) {
 
     store_t *store = malloc(sizeof(struct store_t));
+    store->chunk_size = chunk_size;
     pthread_rwlock_init(&store->lock, NULL);
 
     mdb_env_create(&store->env);
@@ -18,7 +19,7 @@ store_t *store_create(char *path) {
         LOG_FATALF("store.c", "Error while opening store: %s (%s)\n", mdb_strerror(open_ret), path)
     }
 
-    store->size = (size_t) 1024 * 1024 * 5;
+    store->size = (size_t) store->chunk_size;
     ScanCtx.stat_tn_size = 0;
     mdb_env_set_mapsize(store->env, store->size);
 
@@ -39,12 +40,20 @@ void store_destroy(store_t *store) {
     free(store);
 }
 
+void store_flush(store_t *store) {
+    mdb_env_sync(store->env, TRUE);
+}
+
 void store_write(store_t *store, char *key, size_t key_len, char *buf, size_t buf_len) {
 
     if (LogCtx.very_verbose) {
-        char uuid_str[UUID_STR_LEN];
-        uuid_unparse((unsigned char *) key, uuid_str);
-        LOG_DEBUGF("store.c", "Store write {%s} %lu bytes", uuid_str, buf_len)
+        if (key_len == MD5_DIGEST_LENGTH) {
+            char path_md5_str[MD5_STR_LENGTH];
+            buf2hex((unsigned char *) key, MD5_DIGEST_LENGTH, path_md5_str);
+            LOG_DEBUGF("store.c", "Store write {%s} %lu bytes", path_md5_str, buf_len)
+        } else {
+            LOG_DEBUGF("store.c", "Store write {%s} %lu bytes", key, buf_len)
+        }
     }
 
     MDB_val mdb_key;
@@ -69,7 +78,7 @@ void store_write(store_t *store, char *key, size_t key_len, char *buf, size_t bu
         // Cannot resize when there is a opened transaction.
         //  Resize take effect on the next commit.
         pthread_rwlock_wrlock(&store->lock);
-        store->size += 1024 * 1024 * 50;
+        store->size += store->chunk_size;
         mdb_env_set_mapsize(store->env, store->size);
         mdb_txn_begin(store->env, NULL, 0, &txn);
         put_ret = mdb_put(txn, store->dbi, &mdb_key, &mdb_value, 0);
@@ -110,3 +119,42 @@ char *store_read(store_t *store, char *key, size_t key_len, size_t *ret_vallen)
     return buf;
 }
 
+GHashTable *store_read_all(store_t *store) {
+
+    int count = 0;
+
+    GHashTable *table = g_hash_table_new_full(g_str_hash, g_str_equal, free, free);
+
+    MDB_txn *txn = NULL;
+    mdb_txn_begin(store->env, NULL, MDB_RDONLY, &txn);
+
+    MDB_cursor *cur = NULL;
+    mdb_cursor_open(txn, store->dbi, &cur);
+
+    MDB_val key;
+    MDB_val value;
+
+    while (mdb_cursor_get(cur, &key, &value, MDB_NEXT) == 0) {
+        char *key_str = malloc(key.mv_size);
+        memcpy(key_str, key.mv_data, key.mv_size);
+        char *val_str = malloc(value.mv_size);
+        memcpy(val_str, value.mv_data, value.mv_size);
+
+        g_hash_table_insert(table, key_str, val_str);
+        count += 1;
+    }
+
+    const char *path;
+    mdb_env_get_path(store->env, &path);
+    LOG_DEBUGF("store.c", "Read %d entries from %s", count, path);
+
+    mdb_cursor_close(cur);
+    mdb_txn_abort(txn);
+    return table;
+}
+
+
+void store_copy(store_t *store, const char *destination) {
+    mkdir(destination, S_IWUSR | S_IRUSR | S_IXUSR);
+    mdb_env_copy(store->env, destination);
+}

src/io/store.h

@@ -4,21 +4,32 @@
 #include <pthread.h>
 #include <lmdb.h>
 
+#include <glib.h>
+
+#define STORE_SIZE_TN 1024 * 1024 * 5
+#define STORE_SIZE_TAG 1024 * 16
+#define STORE_SIZE_META STORE_SIZE_TAG
+
 typedef struct store_t {
     MDB_dbi dbi;
     MDB_env *env;
     size_t size;
+    size_t chunk_size;
     pthread_rwlock_t lock;
 } store_t;
 
-#include "src/sist.h"
-
-store_t *store_create(char *path);
+store_t *store_create(char *path, size_t chunk_size);
 
 void store_destroy(store_t *store);
 
 void store_write(store_t *store, char *key, size_t key_len, char *buf, size_t buf_len);
 
+void store_flush(store_t *store);
+
 char *store_read(store_t *store, char *key, size_t key_len, size_t *ret_vallen);
 
+GHashTable *store_read_all(store_t *store);
+
+void store_copy(store_t *store, const char *destination);
+
 #endif

src/io/walk.c

@@ -1,5 +1,8 @@
 #include "walk.h"
 #include "src/ctx.h"
+#include "src/parsing/parse.h"
+
+#include <ftw.h>
 
 __always_inline
 parse_job_t *create_fs_parse_job(const char *filepath, const struct stat *info, int base) {
@@ -15,12 +18,13 @@ parse_job_t *create_fs_parse_job(const char *filepath, const struct stat *info,
         job->ext = len;
     }
 
-    job->info = *info;
+    job->vfile.info = *info;
 
-    memset(job->parent, 0, 16);
+    memset(job->parent, 0, MD5_DIGEST_LENGTH);
 
     job->vfile.filepath = job->filepath;
     job->vfile.read = fs_read;
+    job->vfile.reset = fs_reset;
     job->vfile.close = fs_close;
     job->vfile.fd = -1;
     job->vfile.is_fs_file = TRUE;

src/io/walk.h

@@ -3,8 +3,6 @@
 
 #define _XOPEN_SOURCE 500
 
-#include "src/sist.h"
-
 int walk_directory_tree(const char *);
 
 #endif

src/log.c

@@ -1,15 +1,17 @@
 #include "log.h"
 
+#include <pthread.h>
+#include <stdarg.h>
+
 const char *log_colors[] = {
-        "\033[34m", "\033[01;34m", "\033[0m",
-        "\033[01;33m", "\033[31m", "\033[01;31m"
+        "\033[34m", "\033[01;34m", "\033[01;33m", "\033[0m", "\033[31m", "\033[01;31m"
 };
 
 const char *log_levels[] = {
         "DEBUG", "INFO", "WARNING", "ERROR", "FATAL"
 };
 
-void sist_logf(char *filepath, int level, char *format, ...) {
+void vsist_logf(const char *filepath, int level, char *format, va_list ap) {
 
     static int is_tty = -1;
     if (is_tty == -1) {
@@ -31,23 +33,20 @@ void sist_logf(char *filepath, int level, char *format, ...) {
     if (is_tty) {
         log_len = snprintf(
                 log_str, sizeof(log_str),
-                "\033[%dm[%04X]%s [%s] [%s %s] ",
+                "\033[%dm[%04llX]%s [%s] [%s %s] ",
                 31 + ((unsigned int) (pid)) % 7, pid, log_colors[level],
                 datetime, log_levels[level], filepath
         );
     } else {
         log_len = snprintf(
                 log_str, sizeof(log_str),
-                "[%04X] [%s] [%s %s] ",
+                "[%04llX] [%s] [%s %s] ",
                 pid, datetime, log_levels[level], filepath
         );
     }
 
-    va_list ap;
-    va_start(ap, format);
     size_t maxsize = sizeof(log_str) - log_len;
     log_len += vsnprintf(log_str + log_len, maxsize, format, ap);
-    va_end(ap);
 
     if (is_tty) {
         log_len += sprintf(log_str + log_len, "\033[0m\n");
@@ -56,10 +55,20 @@ void sist_logf(char *filepath, int level, char *format, ...) {
         log_len += 1;
     }
 
-    write(STDERR_FILENO, log_str, log_len);
+    int ret = write(STDERR_FILENO, log_str, log_len);
+    if (ret == -1) {
+        LOG_FATALF("serialize.c", "Could not write index descriptor: %s", strerror(errno))
+    }
 }
 
-void sist_log(char *filepath, int level, char *str) {
+void sist_logf(const char *filepath, int level, char *format, ...) {
+    va_list ap;
+    va_start(ap, format);
+    vsist_logf(filepath, level, format, ap);
+    va_end(ap);
+}
+
+void sist_log(const char *filepath, int level, char *str) {
 
     static int is_tty = -1;
     if (is_tty == -1) {
@@ -81,7 +90,7 @@ void sist_log(char *filepath, int level, char *str) {
     if (is_tty) {
         log_len = snprintf(
                 log_str, sizeof(log_str),
-                "\033[%dm[%04X]%s [%s] [%s %s] %s \033[0m\n",
+                "\033[%dm[%04llX]%s [%s] [%s %s] %s \033[0m\n",
                 31 + ((unsigned int) (pid)) % 7, pid, log_colors[level],
                 datetime, log_levels[level], filepath,
                 str
@@ -89,11 +98,14 @@ void sist_log(char *filepath, int level, char *str) {
     } else {
         log_len = snprintf(
                 log_str, sizeof(log_str),
-                "[%04X] [%s] [%s %s] %s \n",
+                "[%04llX] [%s] [%s %s] %s \n",
                 pid, datetime, log_levels[level], filepath,
                 str
         );
     }
 
-    write(STDERR_FILENO, log_str, log_len);
+    int ret = write(STDERR_FILENO, log_str, log_len);
+    if (ret == -1) {
+        LOG_FATALF("serialize.c", "Could not write index descriptor: %s", strerror(errno));
+    }
 }

src/log.h

@@ -1,6 +1,7 @@
 #ifndef SIST2_LOG_H
 #define SIST2_LOG_H
 
+
 #define LOG_MAX_LENGTH 8192
 
 #define SIST_DEBUG 0
@@ -36,10 +37,11 @@
     sist_log(filepath, SIST_FATAL, str);\
     exit(-1);
 
-#include "src/sist.h"
+#include "sist.h"
 
-void sist_logf(char *filepath, int level, char *format, ...);
+void sist_logf(const char *filepath, int level, char *format, ...);
+void vsist_logf(const char *filepath, int level, char *format, va_list ap);
 
-void sist_log(char *filepath, int level, char *str);
+void sist_log(const char *filepath, int level, char *str);
 
 #endif

src/main.c

@@ -1,31 +1,43 @@
 #include "sist.h"
 #include "ctx.h"
 
+#include <third-party/argparse/argparse.h>
+#include <locale.h>
+
+#include "cli.h"
+#include "io/serialize.h"
+#include "io/store.h"
+#include "tpool.h"
+#include "io/walk.h"
+#include "index/elastic.h"
+#include "web/serve.h"
+#include "parsing/mime.h"
+#include "parsing/parse.h"
+
+#include "stats.h"
+
 #define DESCRIPTION "Lightning-fast file system indexer and search tool."
 
 #define EPILOG "Made by simon987 <me@simon987.net>. Released under GPL-3.0"
 
 
-static const char *const Version = "1.3.1";
+static const char *const Version = "2.9.0";
 static const char *const usage[] = {
         "sist2 scan [OPTION]... PATH",
         "sist2 index [OPTION]... INDEX",
         "sist2 web [OPTION]... INDEX...",
+        "sist2 exec-script [OPTION]... INDEX",
         NULL,
 };
 
-void global_init() {
-    curl_global_init(CURL_GLOBAL_NOTHING);
-    av_log_set_level(AV_LOG_QUIET);
-}
-
 void init_dir(const char *dirpath) {
     char path[PATH_MAX];
     snprintf(path, PATH_MAX, "%sdescriptor.json", dirpath);
 
-    uuid_t uuid;
-    uuid_generate(uuid);
-    uuid_unparse(uuid, ScanCtx.index.desc.uuid);
+    unsigned char index_md5[MD5_DIGEST_LENGTH];
+    MD5((unsigned char *) ScanCtx.index.desc.name, strlen(ScanCtx.index.desc.name), index_md5);
+    buf2hex(index_md5, MD5_DIGEST_LENGTH, ScanCtx.index.desc.id);
+
     time(&ScanCtx.index.desc.timestamp);
     strcpy(ScanCtx.index.desc.version, Version);
     strcpy(ScanCtx.index.desc.type, INDEX_TYPE_BIN);
@@ -37,34 +49,151 @@ void scan_print_header() {
     LOG_INFOF("main.c", "sist2 v%s", Version)
 }
 
-void sist2_scan(scan_args_t *args) {
+void _store(char *key, size_t key_len, char *buf, size_t buf_len) {
+    store_write(ScanCtx.index.store, key, key_len, buf, buf_len);
+}
+
+void _log(const char *filepath, int level, char *str) {
+    if (level == LEVEL_FATAL) {
+        sist_log(filepath, level, str);
+        exit(-1);
+    }
+
+    if (LogCtx.verbose) {
+        if (level == LEVEL_DEBUG) {
+            if (LogCtx.very_verbose) {
+                sist_log(filepath, level, str);
+            }
+        } else {
+            sist_log(filepath, level, str);
+        }
+    }
+}
+
+void _logf(const char *filepath, int level, char *format, ...) {
+
+    va_list args;
+
+    va_start(args, format);
+    if (level == LEVEL_FATAL) {
+        vsist_logf(filepath, level, format, args);
+        exit(-1);
+    }
+
+    if (LogCtx.verbose) {
+        if (level == LEVEL_DEBUG) {
+            if (LogCtx.very_verbose) {
+                vsist_logf(filepath, level, format, args);
+            }
+        } else {
+            vsist_logf(filepath, level, format, args);
+        }
+    }
+    va_end(args);
+}
+
+void initialize_scan_context(scan_args_t *args) {
+
+    // Arc
+    ScanCtx.arc_ctx.mode = args->archive_mode;
+    ScanCtx.arc_ctx.log = _log;
+    ScanCtx.arc_ctx.logf = _logf;
+    ScanCtx.arc_ctx.parse = (parse_callback_t) parse;
+
+    // Comic
+    ScanCtx.comic_ctx.log = _log;
+    ScanCtx.comic_ctx.logf = _logf;
+    ScanCtx.comic_ctx.store = _store;
+    ScanCtx.comic_ctx.tn_size = args->size;
+    ScanCtx.comic_ctx.tn_qscale = args->quality;
+    ScanCtx.comic_ctx.cbr_mime = mime_get_mime_by_string(ScanCtx.mime_table, "application/x-cbr");
+    ScanCtx.comic_ctx.cbz_mime = mime_get_mime_by_string(ScanCtx.mime_table, "application/x-cbz");
+
+    // Ebook
+    pthread_mutex_init(&ScanCtx.ebook_ctx.mupdf_mutex, NULL);
+    ScanCtx.ebook_ctx.content_size = args->content_size;
+    ScanCtx.ebook_ctx.tn_size = args->size;
+    ScanCtx.ebook_ctx.tesseract_lang = args->tesseract_lang;
+    ScanCtx.ebook_ctx.tesseract_path = args->tesseract_path;
+    ScanCtx.ebook_ctx.log = _log;
+    ScanCtx.ebook_ctx.logf = _logf;
+    ScanCtx.ebook_ctx.store = _store;
+
+    // Font
+    ScanCtx.font_ctx.enable_tn = args->size > 0;
+    ScanCtx.font_ctx.log = _log;
+    ScanCtx.font_ctx.logf = _logf;
+    ScanCtx.font_ctx.store = _store;
+
+    // Media
+    ScanCtx.media_ctx.tn_qscale = args->quality;
+    ScanCtx.media_ctx.tn_size = args->size;
+    ScanCtx.media_ctx.log = _log;
+    ScanCtx.media_ctx.logf = _logf;
+    ScanCtx.media_ctx.store = _store;
+    ScanCtx.media_ctx.max_media_buffer = (long) args->max_memory_buffer * 1024 * 1024;
+    init_media();
+
+    // OOXML
+    ScanCtx.ooxml_ctx.content_size = args->content_size;
+    ScanCtx.ooxml_ctx.log = _log;
+    ScanCtx.ooxml_ctx.logf = _logf;
+    ScanCtx.ooxml_ctx.store = _store;
+
+    // MOBI
+    ScanCtx.mobi_ctx.content_size = args->content_size;
+    ScanCtx.mobi_ctx.log = _log;
+    ScanCtx.mobi_ctx.logf = _logf;
+
+    // TEXT
+    ScanCtx.text_ctx.content_size = args->content_size;
+    ScanCtx.text_ctx.log = _log;
+    ScanCtx.text_ctx.logf = _logf;
+
+    // MSDOC
+    ScanCtx.msdoc_ctx.tn_size = args->size;
+    ScanCtx.msdoc_ctx.content_size = args->content_size;
+    ScanCtx.msdoc_ctx.log = _log;
+    ScanCtx.msdoc_ctx.logf = _logf;
+    ScanCtx.msdoc_ctx.store = _store;
+    ScanCtx.msdoc_ctx.msdoc_mime = mime_get_mime_by_string(ScanCtx.mime_table, "application/msword");
 
-    ScanCtx.tn_qscale = args->quality;
-    ScanCtx.tn_size = args->size;
-    ScanCtx.content_size = args->content_size;
     ScanCtx.threads = args->threads;
     ScanCtx.depth = args->depth;
-    ScanCtx.archive_mode = args->archive_mode;
+
     strncpy(ScanCtx.index.path, args->output, sizeof(ScanCtx.index.path));
     strncpy(ScanCtx.index.desc.name, args->name, sizeof(ScanCtx.index.desc.name));
     strncpy(ScanCtx.index.desc.root, args->path, sizeof(ScanCtx.index.desc.root));
     strncpy(ScanCtx.index.desc.rewrite_url, args->rewrite_url, sizeof(ScanCtx.index.desc.rewrite_url));
     ScanCtx.index.desc.root_len = (short) strlen(ScanCtx.index.desc.root);
-    ScanCtx.tesseract_lang = args->tesseract_lang;
-    ScanCtx.tesseract_path = args->tesseract_path;
     ScanCtx.fast = args->fast;
 
-    init_dir(ScanCtx.index.path);
+    // Raw
+    ScanCtx.raw_ctx.tn_qscale = args->quality;
+    ScanCtx.raw_ctx.tn_size = args->size;
+    ScanCtx.raw_ctx.log = _log;
+    ScanCtx.raw_ctx.logf = _logf;
+    ScanCtx.raw_ctx.store = _store;
+}
+
+
+void sist2_scan(scan_args_t *args) {
 
     ScanCtx.mime_table = mime_get_mime_table();
     ScanCtx.ext_table = mime_get_ext_table();
 
-    cbr_init();
+    initialize_scan_context(args);
+
+    init_dir(ScanCtx.index.path);
 
     char store_path[PATH_MAX];
     snprintf(store_path, PATH_MAX, "%sthumbs", ScanCtx.index.path);
     mkdir(store_path, S_IWUSR | S_IRUSR | S_IXUSR);
-    ScanCtx.index.store = store_create(store_path);
+    ScanCtx.index.store = store_create(store_path, STORE_SIZE_TN);
+
+    snprintf(store_path, PATH_MAX, "%smeta", ScanCtx.index.path);
+    mkdir(store_path, S_IWUSR | S_IRUSR | S_IXUSR);
+    ScanCtx.index.meta_store = store_create(store_path, STORE_SIZE_META);
 
     scan_print_header();
 
@@ -90,7 +219,7 @@ void sist2_scan(scan_args_t *args) {
         while ((de = readdir(dir)) != NULL) {
             if (strncmp(de->d_name, "_index_", sizeof("_index_") - 1) == 0) {
                 char file_path[PATH_MAX];
-                snprintf(file_path, PATH_MAX, "%s/%s", args->incremental, de->d_name);
+                snprintf(file_path, PATH_MAX, "%s%s", args->incremental, de->d_name);
                 incremental_read(ScanCtx.original_table, file_path);
             }
         }
@@ -99,7 +228,7 @@ void sist2_scan(scan_args_t *args) {
         LOG_INFOF("main.c", "Loaded %d items in to mtime table.", g_hash_table_size(ScanCtx.original_table))
     }
 
-    ScanCtx.pool = tpool_create(args->threads, thread_cleanup);
+    ScanCtx.pool = tpool_create(args->threads, thread_cleanup, TRUE);
     tpool_start(ScanCtx.pool);
     walk_directory_tree(ScanCtx.index.desc.root);
     tpool_wait(ScanCtx.pool);
@@ -109,7 +238,7 @@ void sist2_scan(scan_args_t *args) {
         char dst_path[PATH_MAX];
         snprintf(store_path, PATH_MAX, "%sthumbs", args->incremental);
         snprintf(dst_path, PATH_MAX, "%s_index_original", ScanCtx.index.path);
-        store_t *source = store_create(store_path);
+        store_t *source = store_create(store_path, STORE_SIZE_TN);
 
         DIR *dir = opendir(args->incremental);
         if (dir == NULL) {
@@ -120,24 +249,34 @@ void sist2_scan(scan_args_t *args) {
         while ((de = readdir(dir)) != NULL) {
             if (strncmp(de->d_name, "_index_", sizeof("_index_") - 1) == 0) {
                 char file_path[PATH_MAX];
-                snprintf(file_path, PATH_MAX, "%s/%s", args->incremental, de->d_name);
+                snprintf(file_path, PATH_MAX, "%s%s", args->incremental, de->d_name);
                 incremental_copy(source, ScanCtx.index.store, file_path, dst_path, ScanCtx.copy_table);
             }
         }
         closedir(dir);
         store_destroy(source);
+
+        snprintf(store_path, PATH_MAX, "%stags", args->incremental);
+        snprintf(dst_path, PATH_MAX, "%stags", ScanCtx.index.path);
+        mkdir(store_path, S_IWUSR | S_IRUSR | S_IXUSR);
+        store_t *source_tags = store_create(store_path, STORE_SIZE_TAG);
+        store_copy(source_tags, dst_path);
+        store_destroy(source_tags);
     }
 
+    generate_stats(&ScanCtx.index, args->treemap_threshold, ScanCtx.index.path);
+
     store_destroy(ScanCtx.index.store);
 }
 
 void sist2_index(index_args_t *args) {
 
     IndexCtx.es_url = args->es_url;
+    IndexCtx.es_index = args->es_index;
     IndexCtx.batch_size = args->batch_size;
 
     if (!args->print) {
-        elastic_init(args->force_reset);
+        elastic_init(args->force_reset, args->es_mappings, args->es_settings);
     }
 
     char descriptor_path[PATH_MAX];
@@ -157,6 +296,16 @@ void sist2_index(index_args_t *args) {
         LOG_FATALF("main.c", "Could not open index %s: %s", args->index_path, strerror(errno))
     }
 
+    char path_tmp[PATH_MAX];
+    snprintf(path_tmp, sizeof(path_tmp), "%s/tags", args->index_path);
+    mkdir(path_tmp, S_IWUSR | S_IRUSR | S_IXUSR);
+    IndexCtx.tag_store = store_create(path_tmp, STORE_SIZE_TAG);
+    IndexCtx.tags = store_read_all(IndexCtx.tag_store);
+
+    snprintf(path_tmp, sizeof(path_tmp), "%s/meta", args->index_path);
+    IndexCtx.meta_store = store_create(path_tmp, STORE_SIZE_META);
+    IndexCtx.meta = store_read_all(IndexCtx.meta_store);
+
     index_func f;
     if (args->print) {
         f = print_json;
@@ -164,27 +313,64 @@ void sist2_index(index_args_t *args) {
         f = index_json;
     }
 
+    void (*cleanup)();
+    if (args->print) {
+        cleanup = NULL;
+    } else {
+        cleanup = elastic_cleanup;
+    }
+
+    IndexCtx.pool = tpool_create(args->threads, cleanup, FALSE);
+    tpool_start(IndexCtx.pool);
+
     struct dirent *de;
     while ((de = readdir(dir)) != NULL) {
         if (strncmp(de->d_name, "_index_", sizeof("_index_") - 1) == 0) {
             char file_path[PATH_MAX];
             snprintf(file_path, PATH_MAX, "%s/%s", args->index_path, de->d_name);
-            read_index(file_path, desc.uuid, desc.type, f);
+            read_index(file_path, desc.id, desc.type, f);
         }
     }
     closedir(dir);
 
+    tpool_wait(IndexCtx.pool);
+
+    tpool_destroy(IndexCtx.pool);
+
     if (!args->print) {
-        elastic_flush();
-        destroy_indexer(args->script, desc.uuid);
+        finish_indexer(args->script, args->async_script, desc.id);
     }
+
+    store_destroy(IndexCtx.tag_store);
+    g_hash_table_remove_all(IndexCtx.tags);
+    g_hash_table_destroy(IndexCtx.tags);
+}
+
+void sist2_exec_script(exec_args_t *args) {
+
+    LogCtx.verbose = TRUE;
+
+    char descriptor_path[PATH_MAX];
+    snprintf(descriptor_path, PATH_MAX, "%s/descriptor.json", args->index_path);
+    index_descriptor_t desc = read_index_descriptor(descriptor_path);
+
+    IndexCtx.es_url = args->es_url;
+
+    LOG_DEBUGF("main.c", "descriptor version %s (%s)", desc.version, desc.type)
+
+    execute_update_script(args->script, args->async_script, desc.id);
+    free(args->script);
 }
 
 void sist2_web(web_args_t *args) {
 
     WebCtx.es_url = args->es_url;
+    WebCtx.es_index = args->es_index;
     WebCtx.index_count = args->index_count;
-    WebCtx.b64credentials = args->b64credentials;
+    WebCtx.auth_user = args->auth_user;
+    WebCtx.auth_pass = args->auth_pass;
+    WebCtx.auth_enabled = args->auth_enabled;
+    WebCtx.tag_auth_enabled = args->tag_auth_enabled;
 
     for (int i = 0; i < args->index_count; i++) {
         char *abs_path = abspath(args->indices[i]);
@@ -194,7 +380,11 @@ void sist2_web(web_args_t *args) {
         char path_tmp[PATH_MAX];
 
         snprintf(path_tmp, PATH_MAX, "%sthumbs", abs_path);
-        WebCtx.indices[i].store = store_create(path_tmp);
+        WebCtx.indices[i].store = store_create(path_tmp, STORE_SIZE_TN);
+
+        snprintf(path_tmp, PATH_MAX, "%stags", abs_path);
+        mkdir(path_tmp, S_IWUSR | S_IRUSR | S_IXUSR);
+        WebCtx.indices[i].tag_store = store_create(path_tmp, STORE_SIZE_TAG);
 
         snprintf(path_tmp, PATH_MAX, "%sdescriptor.json", abs_path);
         WebCtx.indices[i].desc = read_index_descriptor(path_tmp);
@@ -204,21 +394,25 @@ void sist2_web(web_args_t *args) {
         free(abs_path);
     }
 
-    serve(args->bind, args->port);
+    serve(args->listen_address);
 }
 
 
 int main(int argc, const char *argv[]) {
-
-    global_init();
+    setlocale(LC_ALL, "");
 
     scan_args_t *scan_args = scan_args_create();
     index_args_t *index_args = index_args_create();
     web_args_t *web_args = web_args_create();
+    exec_args_t *exec_args = exec_args_create();
 
     int arg_version = 0;
 
     char *common_es_url = NULL;
+    char *common_es_index = NULL;
+    char *common_script_path = NULL;
+    int common_async_script = 0;
+    int common_threads = 0;
 
     struct argparse_option options[] = {
             OPT_HELP(),
@@ -228,7 +422,7 @@ int main(int argc, const char *argv[]) {
             OPT_BOOLEAN(0, "very-verbose", &LogCtx.very_verbose, "Turn on debug messages"),
 
             OPT_GROUP("Scan options"),
-            OPT_INTEGER('t', "threads", &scan_args->threads, "Number of threads. DEFAULT=1"),
+            OPT_INTEGER('t', "threads", &common_threads, "Number of threads. DEFAULT=1"),
             OPT_FLOAT('q', "quality", &scan_args->quality,
                       "Thumbnail quality, on a scale of 1.0 to 31.0, 1.0 being the best. DEFAULT=5"),
             OPT_INTEGER(0, "size", &scan_args->size,
@@ -249,20 +443,37 @@ int main(int argc, const char *argv[]) {
                                                              "which are installed on your machine)"),
             OPT_STRING('e', "exclude", &scan_args->exclude_regex, "Files that match this regex will not be scanned"),
             OPT_BOOLEAN(0, "fast", &scan_args->fast, "Only index file names & mime type"),
+            OPT_STRING(0, "treemap-threshold", &scan_args->treemap_threshold_str, "Relative size threshold for treemap "
+                                                                                  "(see USAGE.md). DEFAULT: 0.0005"),
+            OPT_INTEGER(0, "mem-buffer", &scan_args->max_memory_buffer,
+                        "Maximum memory buffer size per thread in MB for files inside archives "
+                        "(see USAGE.md). DEFAULT: 2000"),
 
             OPT_GROUP("Index options"),
+            OPT_INTEGER('t', "threads", &common_threads, "Number of threads. DEFAULT=1"),
             OPT_STRING(0, "es-url", &common_es_url, "Elasticsearch url with port. DEFAULT=http://localhost:9200"),
+            OPT_STRING(0, "es-index", &common_es_index, "Elasticsearch index name. DEFAULT=sist2"),
             OPT_BOOLEAN('p', "print", &index_args->print, "Just print JSON documents to stdout."),
-            OPT_STRING(0, "script-file", &index_args->script_path, "Path to user script."),
+            OPT_STRING(0, "script-file", &common_script_path, "Path to user script."),
+            OPT_STRING(0, "mappings-file", &index_args->es_mappings_path, "Path to Elasticsearch mappings."),
+            OPT_STRING(0, "settings-file", &index_args->es_settings_path, "Path to Elasticsearch settings."),
+            OPT_BOOLEAN(0, "async-script", &common_async_script, "Execute user script asynchronously."),
             OPT_INTEGER(0, "batch-size", &index_args->batch_size, "Index batch size. DEFAULT: 100"),
             OPT_BOOLEAN('f', "force-reset", &index_args->force_reset, "Reset Elasticsearch mappings and settings. "
                                                                       "(You must use this option the first time you use the index command)"),
 
             OPT_GROUP("Web options"),
             OPT_STRING(0, "es-url", &common_es_url, "Elasticsearch url. DEFAULT=http://localhost:9200"),
-            OPT_STRING(0, "bind", &web_args->bind, "Listen on this address. DEFAULT=localhost"),
-            OPT_STRING(0, "port", &web_args->port, "Listen on this port. DEFAULT=4090"),
+            OPT_STRING(0, "es-index", &common_es_index, "Elasticsearch index name. DEFAULT=sist2"),
+            OPT_STRING(0, "bind", &web_args->listen_address, "Listen on this address. DEFAULT=localhost:4090"),
             OPT_STRING(0, "auth", &web_args->credentials, "Basic auth in user:password format"),
+            OPT_STRING(0, "tag-auth", &web_args->tag_credentials, "Basic auth in user:password format for tagging"),
+
+            OPT_GROUP("Exec-script options"),
+            OPT_STRING(0, "es-url", &common_es_url, "Elasticsearch url. DEFAULT=http://localhost:9200"),
+            OPT_STRING(0, "es-index", &common_es_index, "Elasticsearch index name. DEFAULT=sist2"),
+            OPT_STRING(0, "script-file", &common_script_path, "Path to user script."),
+            OPT_BOOLEAN(0, "async-script", &common_async_script, "Execute user script asynchronously."),
 
             OPT_END(),
     };
@@ -283,6 +494,18 @@ int main(int argc, const char *argv[]) {
 
     web_args->es_url = common_es_url;
     index_args->es_url = common_es_url;
+    exec_args->es_url = common_es_url;
+
+    web_args->es_index = common_es_index;
+    index_args->es_index = common_es_index;
+    exec_args->es_index = common_es_index;
+
+    index_args->script_path = common_script_path;
+    exec_args->script_path = common_script_path;
+    index_args->threads = common_threads;
+    scan_args->threads = common_threads;
+    exec_args->async_script = common_async_script;
+    index_args->async_script = common_async_script;
 
     if (argc == 0) {
         argparse_usage(&argparse);
@@ -311,6 +534,14 @@ int main(int argc, const char *argv[]) {
         }
         sist2_web(web_args);
 
+    } else if (strcmp(argv[0], "exec-script") == 0) {
+
+        int err = exec_args_validate(exec_args, argc, argv);
+        if (err != 0) {
+            goto end;
+        }
+        sist2_exec_script(exec_args);
+
     } else {
         fprintf(stderr, "Invalid command: '%s'\n", argv[0]);
         argparse_usage(&argparse);
@@ -322,6 +553,7 @@ int main(int argc, const char *argv[]) {
     scan_args_destroy(scan_args);
     index_args_destroy(index_args);
     web_args_destroy(web_args);
+    exec_args_destroy(exec_args);
 
     return 0;
 }

src/parsing/arc.c

@@ -1,155 +0,0 @@
-#include "arc.h"
-#include "src/ctx.h"
-
-int should_parse_filtered_file(const char *filepath, int ext) {
-    char tmp[PATH_MAX * 2];
-
-    if (ext == 0) {
-        return FALSE;
-    }
-
-    memcpy(tmp, filepath, ext - 1);
-    *(tmp + ext - 1) = '\0';
-
-    char *idx = strrchr(tmp, '.');
-
-    if (idx == NULL) {
-        return FALSE;
-    }
-
-    if (strcmp(idx, ".tar") == 0) {
-        return TRUE;
-    }
-
-    return FALSE;
-}
-
-int arc_read(struct vfile *f, void *buf, size_t size) {
-    return archive_read_data(f->arc, buf, size);
-}
-
-typedef struct arc_data {
-    vfile_t *f;
-    char buf[ARC_BUF_SIZE];
-} arc_data_f;
-
-int vfile_open_callback(struct archive *a, void *user_data) {
-    arc_data_f *data = user_data;
-
-    if (data->f->is_fs_file && data->f->fd == -1) {
-        data->f->fd = open(data->f->filepath, O_RDONLY);
-    }
-
-    return ARCHIVE_OK;
-}
-
-long vfile_read_callback(struct archive *a, void *user_data, const void **buf) {
-    arc_data_f *data = user_data;
-
-    *buf = data->buf;
-    return data->f->read(data->f, data->buf, ARC_BUF_SIZE);
-}
-
-int vfile_close_callback(struct archive *a, void *user_data) {
-    arc_data_f *data = user_data;
-
-    if (data->f->close != NULL) {
-        data->f->close(data->f);
-    }
-
-    return ARCHIVE_OK;
-}
-
-void parse_archive(vfile_t *f, document_t *doc) {
-
-    struct archive *a;
-    struct archive_entry *entry;
-
-
-    arc_data_f data;
-    data.f = f;
-
-    int ret = 0;
-    if (data.f->is_fs_file) {
-
-        a = archive_read_new();
-        archive_read_support_filter_all(a);
-        archive_read_support_format_all(a);
-
-        ret = archive_read_open_filename(a, doc->filepath, ARC_BUF_SIZE);
-    } else if (ScanCtx.archive_mode == ARC_MODE_RECURSE) {
-
-        a = archive_read_new();
-        archive_read_support_filter_all(a);
-        archive_read_support_format_all(a);
-
-        ret = archive_read_open(
-                a, &data,
-                vfile_open_callback,
-                vfile_read_callback,
-                vfile_close_callback
-        );
-    } else {
-        return;
-    }
-
-    if (ret != ARCHIVE_OK) {
-        LOG_ERRORF(doc->filepath, "(arc.c) [%d] %s", ret, archive_error_string(a))
-        archive_read_free(a);
-        return;
-    }
-
-    if (ScanCtx.archive_mode == ARC_MODE_LIST) {
-
-        dyn_buffer_t buf = dyn_buffer_create();
-
-        while (archive_read_next_header(a, &entry) == ARCHIVE_OK) {
-            if (S_ISREG(archive_entry_stat(entry)->st_mode)) {
-
-                char *path = (char *) archive_entry_pathname(entry);
-
-                dyn_buffer_append_string(&buf, path);
-                dyn_buffer_write_char(&buf, '\n');
-            }
-        }
-        dyn_buffer_write_char(&buf, '\0');
-
-        meta_line_t *meta_list = malloc(sizeof(meta_line_t) + buf.cur);
-        meta_list->key = MetaContent;
-        strcpy(meta_list->strval, buf.buf);
-        APPEND_META(doc, meta_list);
-        dyn_buffer_destroy(&buf);
-
-    } else {
-
-        parse_job_t *sub_job = malloc(sizeof(parse_job_t) + PATH_MAX * 2);
-
-        sub_job->vfile.close = NULL;
-        sub_job->vfile.read = arc_read;
-        sub_job->vfile.arc = a;
-        sub_job->vfile.filepath = sub_job->filepath;
-        sub_job->vfile.is_fs_file = FALSE;
-        memcpy(sub_job->parent, doc->uuid, sizeof(uuid_t));
-
-        while (archive_read_next_header(a, &entry) == ARCHIVE_OK) {
-            sub_job->info = *archive_entry_stat(entry);
-            if (S_ISREG(sub_job->info.st_mode)) {
-                sprintf(sub_job->filepath, "%s#/%s", f->filepath, archive_entry_pathname(entry));
-                sub_job->base = (int) (strrchr(sub_job->filepath, '/') - sub_job->filepath) + 1;
-
-                char *p = strrchr(sub_job->filepath, '.');
-                if (p != NULL) {
-                    sub_job->ext = (int) (p - sub_job->filepath + 1);
-                } else {
-                    sub_job->ext = (int) strlen(sub_job->filepath);
-                }
-
-                parse(sub_job);
-            }
-        }
-
-        free(sub_job);
-    }
-
-    archive_read_free(a);
-}

src/parsing/arc.h

@@ -1,13 +0,0 @@
-#ifndef SIST2_ARC_H
-#define SIST2_ARC_H
-
-#include "src/sist.h"
-#define ARC_BUF_SIZE 8192
-
-int should_parse_filtered_file(const char *filepath, int ext);
-
-void parse_archive(vfile_t *f, document_t *doc);
-
-int arc_read(struct vfile * f, void *buf, size_t size);
-
-#endif

src/parsing/cbr.c

@@ -1,52 +0,0 @@
-#include "cbr.h"
-#include "src/ctx.h"
-
-unsigned int cbr_mime;
-unsigned int cbz_mime;
-
-void cbr_init() {
-    cbr_mime = mime_get_mime_by_string(ScanCtx.mime_table, "application/x-cbr");
-    cbz_mime = mime_get_mime_by_string(ScanCtx.mime_table, "application/x-cbz");
-}
-
-int is_cbr(unsigned int mime) {
-    return mime == cbr_mime;
-}
-
-void parse_cbr(void *buf, size_t buf_len, document_t *doc) {
-    char *out_buf = malloc(buf_len * 2);
-    size_t out_buf_used = 0;
-
-    struct archive *rar_in = archive_read_new();
-    archive_read_support_filter_none(rar_in);
-    archive_read_support_format_rar(rar_in);
-
-    archive_read_open_memory(rar_in, buf, buf_len);
-
-    struct archive *zip_out = archive_write_new();
-    archive_write_set_format_zip(zip_out);
-    archive_write_open_memory(zip_out, out_buf, buf_len * 2, &out_buf_used);
-
-    struct archive_entry *entry;
-    while (archive_read_next_header(rar_in, &entry) == ARCHIVE_OK) {
-        archive_write_header(zip_out, entry);
-
-        char arc_buf[ARC_BUF_SIZE];
-        int len = archive_read_data(rar_in, arc_buf, ARC_BUF_SIZE);
-        while (len > 0) {
-            archive_write_data(zip_out, arc_buf, len);
-            len = archive_read_data(rar_in, arc_buf, ARC_BUF_SIZE);
-        }
-    }
-
-    archive_write_close(zip_out);
-    archive_write_free(zip_out);
-
-    archive_read_close(rar_in);
-    archive_read_free(rar_in);
-
-    doc->mime = cbz_mime;
-    parse_pdf(out_buf, out_buf_used, doc);
-    doc->mime = cbr_mime;
-    free(out_buf);
-}

src/parsing/cbr.h

@@ -1,12 +0,0 @@
-#ifndef SIST2_CBR_H
-#define SIST2_CBR_H
-
-#include "src/sist.h"
-
-void cbr_init();
-
-int is_cbr(unsigned int mime);
-
-void parse_cbr(void *buf, size_t buf_len, document_t *doc);
-
-#endif

src/parsing/doc.c

@@ -1,140 +0,0 @@
-#include "doc.h"
-#include "src/ctx.h"
-
-
-#define STR_STARTS_WITH(x, y) (strncmp(y, x, sizeof(y) - 1) == 0)
-
-__always_inline
-static int should_read_part(const char *part) {
-
-    LOG_DEBUGF("doc.c", "Got part : %s", part)
-
-    if (part == NULL) {
-        return FALSE;
-    }
-
-    if (    // Word
-            STR_STARTS_WITH(part, "word/document.xml")
-            || STR_STARTS_WITH(part, "word/footnotes.xml")
-            || STR_STARTS_WITH(part, "word/endnotes.xml")
-            || STR_STARTS_WITH(part, "word/footer")
-            || STR_STARTS_WITH(part, "word/header")
-            // PowerPoint
-            || STR_STARTS_WITH(part, "ppt/slides/slide")
-            || STR_STARTS_WITH(part, "ppt/notesSlides/slide")
-            // Excel
-            || STR_STARTS_WITH(part, "xl/worksheets/sheet")
-            || STR_STARTS_WITH(part, "xl/sharedStrings.xml")
-            || STR_STARTS_WITH(part, "xl/workbook.xml")
-            ) {
-        return TRUE;
-    }
-
-    return FALSE;
-}
-
-int extract_text(xmlDoc *xml, xmlNode *node, text_buffer_t *buf) {
-    //TODO: Check which nodes are likely to have a 't' child, and ignore nodes that aren't
-    xmlErrorPtr err = xmlGetLastError();
-    if (err != NULL) {
-        if (err->level == XML_ERR_FATAL) {
-            LOG_ERRORF("doc.c", "Got fatal XML error while parsing document: %s", err->message)
-            return -1;
-        } else {
-            LOG_ERRORF("doc.c", "Got recoverable XML error while parsing document: %s", err->message)
-        }
-    }
-
-    for (xmlNode *child = node; child; child = child->next) {
-        if (*child->name == 't' && *(child->name + 1) == '\0') {
-            xmlChar *text = xmlNodeListGetString(xml, child->xmlChildrenNode, 1);
-
-            if (text) {
-                text_buffer_append_string0(buf, (char *) text);
-                text_buffer_append_char(buf, ' ');
-                xmlFree(text);
-            }
-        }
-
-        extract_text(xml, child->children, buf);
-    }
-}
-
-int xml_io_read(void *context, char *buffer, int len) {
-    struct archive *a = context;
-    return archive_read_data(a, buffer, len);
-}
-
-int xml_io_close(UNUSED(void *context)) {
-    //noop
-    return 0;
-}
-
-__always_inline
-static int read_part(struct archive *a, text_buffer_t *buf, document_t *doc) {
-
-    xmlDoc *xml = xmlReadIO(xml_io_read, xml_io_close, a, "/", NULL, XML_PARSE_RECOVER | XML_PARSE_NOWARNING | XML_PARSE_NOERROR | XML_PARSE_NONET);
-
-    if (xml == NULL) {
-        LOG_ERROR(doc->filepath, "Could not parse XML")
-        return -1;
-    }
-
-    xmlNode *root = xmlDocGetRootElement(xml);
-    if (root == NULL) {
-        LOG_ERROR(doc->filepath, "Empty document")
-        xmlFreeDoc(xml);
-        return -1;
-    }
-
-    extract_text(xml, root, buf);
-    xmlFreeDoc(xml);
-
-    return 0;
-}
-
-void parse_doc(void *mem, size_t mem_len, document_t *doc) {
-
-    if (mem == NULL) {
-        return;
-    }
-
-    struct archive *a = archive_read_new();
-    archive_read_support_format_zip(a);
-
-    int ret = archive_read_open_memory(a, mem, mem_len);
-    if (ret != ARCHIVE_OK) {
-        LOG_ERRORF(doc->filepath, "Could not read archive: %s", archive_error_string(a))
-        archive_read_free(a);
-        return;
-    }
-
-    text_buffer_t buf = text_buffer_create(ScanCtx.content_size);
-
-    struct archive_entry *entry;
-    while (archive_read_next_header(a, &entry) == ARCHIVE_OK) {
-        if (S_ISREG(archive_entry_stat(entry)->st_mode)) {
-            const char *path = archive_entry_pathname(entry);
-
-            if (should_read_part(path)) {
-                ret = read_part(a, &buf, doc);
-                if (ret != 0) {
-                    break;
-                }
-            }
-        }
-    }
-
-    if (buf.dyn_buffer.cur > 0) {
-        text_buffer_terminate_string(&buf);
-
-        meta_line_t *meta = malloc(sizeof(meta_line_t) + buf.dyn_buffer.cur);
-        meta->key = MetaContent;
-        strcpy(meta->strval, buf.dyn_buffer.buf);
-        APPEND_META(doc, meta)
-    }
-
-    archive_read_close(a);
-    archive_read_free(a);
-    text_buffer_destroy(&buf);
-}

src/parsing/doc.h

@@ -1,8 +0,0 @@
-#ifndef SIST2_DOC_H
-#define SIST2_DOC_H
-
-#include "src/sist.h"
-
-void parse_doc(void *buf, size_t buf_len, document_t *doc);
-
-#endif

src/parsing/font.c

@@ -1,226 +0,0 @@
-#include "font.h"
-
-
-#include "src/ctx.h"
-
-__thread FT_Library ft_lib = NULL;
-
-
-typedef struct text_dimensions {
-    unsigned int width;
-    unsigned int height;
-    unsigned int baseline;
-} text_dimensions_t;
-
-typedef struct glyph {
-    int top;
-    int height;
-    int width;
-    int descent;
-    int ascent;
-    int advance_width;
-    unsigned char *pixmap;
-} glyph_t;
-
-
-__always_inline
-int kerning_offset(char c, char pc, FT_Face face) {
-    FT_Vector kerning;
-    FT_Get_Kerning(face, c, pc, FT_KERNING_DEFAULT, &kerning);
-
-    return (int) (kerning.x / 64);
-}
-
-__always_inline
-glyph_t ft_glyph_to_glyph(FT_GlyphSlot slot) {
-    glyph_t glyph;
-
-    glyph.pixmap = slot->bitmap.buffer;
-
-    glyph.width = (int) slot->bitmap.width;
-    glyph.height = (int) slot->bitmap.rows;
-    glyph.top = slot->bitmap_top;
-    glyph.advance_width = (int) slot->advance.x / 64;
-
-    glyph.descent = MAX(0, glyph.height - glyph.top);
-    glyph.ascent = MAX(0, MAX(glyph.top, glyph.height) - glyph.descent);
-
-    return glyph;
-}
-
-text_dimensions_t text_dimension(char *text, FT_Face face) {
-    text_dimensions_t dimensions;
-
-    dimensions.width = 0;
-
-    int num_chars = (int) strlen(text);
-
-    unsigned int max_ascent = 0;
-    int max_descent = 0;
-
-    char pc = 0;
-    for (int i = 0; i < num_chars; i++) {
-        char c = text[i];
-
-        FT_Load_Char(face, c, 0);
-        glyph_t glyph = ft_glyph_to_glyph(face->glyph);
-
-        max_descent = MAX(max_descent, glyph.descent);
-        max_ascent = MAX(max_ascent, MAX(glyph.height, glyph.ascent));
-
-        int kerning_x = kerning_offset(c, pc, face);
-        dimensions.width += MAX(glyph.advance_width, glyph.width) + kerning_x;
-
-        pc = c;
-    }
-
-    dimensions.height = max_ascent + max_descent;
-    dimensions.baseline = max_descent;
-
-    return dimensions;
-}
-
-void draw_glyph(glyph_t *glyph, int x, int y, struct text_dimensions text_info, unsigned char *bitmap) {
-    unsigned int src = 0;
-    unsigned int dst = y * text_info.width + x;
-    unsigned int row_offset = text_info.width - glyph->width;
-    unsigned int buf_len = text_info.width * text_info.height;
-
-    for (unsigned int sy = 0; sy < glyph->height; sy++) {
-        for (unsigned int sx = 0; sx < glyph->width; sx++) {
-            if (dst < buf_len) {
-                bitmap[dst] |= glyph->pixmap[src];
-            }
-            src++;
-            dst++;
-        }
-        dst += row_offset;
-    }
-}
-
-void bmp_format(dyn_buffer_t *buf, text_dimensions_t dimensions, const unsigned char *bitmap) {
-
-    dyn_buffer_write_short(buf, 0x4D42); // Magic
-    dyn_buffer_write_int(buf, 0); // Size placeholder
-    dyn_buffer_write_int(buf, 0x5157); //Reserved
-    dyn_buffer_write_int(buf, 14 + 40 + 256 * 4); // pixels offset
-
-    dyn_buffer_write_int(buf, 40); // DIB size
-    dyn_buffer_write_int(buf, (int) dimensions.width);
-    dyn_buffer_write_int(buf, (int) dimensions.height);
-    dyn_buffer_write_short(buf, 1); // Color planes
-    dyn_buffer_write_short(buf, 8); // bits per pixel
-    dyn_buffer_write_int(buf, 0); // compression
-    dyn_buffer_write_int(buf, 0); // Ignored
-    dyn_buffer_write_int(buf, 3800); // hres
-    dyn_buffer_write_int(buf, 3800); // vres
-    dyn_buffer_write_int(buf, 256); // Color count
-    dyn_buffer_write_int(buf, 0); // Ignored
-
-    // RGBA32 Color table (Grayscale)
-    for (int i = 255; i >= 0; i--) {
-        dyn_buffer_write_int(buf, i + (i << 8) + (i << 16));
-    }
-
-    // Pixel array: write from bottom to top, with rows padded to multiples of 4-bytes
-    for (int y = (int) dimensions.height - 1; y >= 0; y--) {
-        for (unsigned int x = 0; x < dimensions.width; x++) {
-            dyn_buffer_write_char(buf, (char) bitmap[y * dimensions.width + x]);
-        }
-        while (buf->cur % 4 != 0) {
-            dyn_buffer_write_char(buf, 0);
-        }
-    }
-
-    // Size
-    *(int *) ((char *) buf->buf + 2) = buf->cur;
-}
-
-void parse_font(const char *buf, size_t buf_len, document_t *doc) {
-    if (ft_lib == NULL) {
-        FT_Init_FreeType(&ft_lib);
-    }
-    if (buf == NULL) {
-        return;
-    }
-
-    FT_Face face;
-    FT_Error err = FT_New_Memory_Face(ft_lib, (unsigned char *) buf, buf_len, 0, &face);
-    if (err != 0) {
-        LOG_ERRORF(doc->filepath, "(font.c) FT_New_Memory_Face() returned error code [%d] %s", err, ft_error_string(err));
-        return;
-    }
-
-    char font_name[1024];
-
-    if (face->style_name == NULL || *(face->style_name) == '?') {
-        if (face->family_name == NULL) {
-            strcpy(font_name, "(null)");
-        } else {
-            strcpy(font_name, face->family_name);
-        }
-    } else {
-        snprintf(font_name, sizeof(font_name), "%s %s", face->family_name, face->style_name);
-    }
-
-    meta_line_t *meta_name = malloc(sizeof(meta_line_t) + strlen(font_name));
-    meta_name->key = MetaFontName;
-    strcpy(meta_name->strval, font_name);
-    APPEND_META(doc, meta_name)
-
-    if (ScanCtx.tn_size <= 0) {
-        return;
-    }
-
-    int pixel = 64;
-    int num_chars = (int) strlen(font_name);
-
-    err = FT_Set_Pixel_Sizes(face, 0, pixel);
-    if (err != 0) {
-        LOG_WARNINGF(doc->filepath, "(font.c) FT_Set_Pixel_Sizes() returned error code [%d] %s", err, ft_error_string(err))
-        return;
-    }
-
-    text_dimensions_t dimensions = text_dimension(font_name, face);
-    unsigned char *bitmap = calloc(dimensions.width * dimensions.height, 1);
-
-    FT_Vector pen;
-    pen.x = 0;
-
-    char pc = 0;
-    for (int i = 0; i < num_chars; i++) {
-        char c = font_name[i];
-
-        err = FT_Load_Char(face, c, FT_LOAD_NO_HINTING | FT_LOAD_RENDER);
-        if (err != 0) {
-            c = c >= 'a' && c <= 'z' ? c - 32 : c + 32;
-            err = FT_Load_Char(face, c, FT_LOAD_NO_HINTING | FT_LOAD_RENDER);
-            if (err != 0) {
-                LOG_WARNINGF(doc->filepath, "(font.c) FT_Load_Char() returned error code [%d] %s", err, ft_error_string(err));
-                continue;
-            }
-        }
-        glyph_t glyph = ft_glyph_to_glyph(face->glyph);
-
-        pen.x += kerning_offset(c, pc, face);
-        if (pen.x <= 0) {
-            pen.x = ABS(glyph.advance_width - glyph.width);
-        }
-        pen.y = dimensions.height - glyph.ascent - dimensions.baseline;
-
-        draw_glyph(&glyph, pen.x, pen.y, dimensions, bitmap);
-
-        pen.x += glyph.advance_width;
-        pc = c;
-    }
-
-    dyn_buffer_t bmp_data = dyn_buffer_create();
-    bmp_format(&bmp_data, dimensions, bitmap);
-
-    store_write(ScanCtx.index.store, (char *) doc->uuid, sizeof(doc->uuid), (char *) bmp_data.buf, bmp_data.cur);
-
-    dyn_buffer_destroy(&bmp_data);
-    free(bitmap);
-
-    FT_Done_Face(face);
-}

src/parsing/font.h

@@ -1,9 +0,0 @@
-#ifndef SIST2_FONT_H
-#define SIST2_FONT_H
-
-#include "src/sist.h"
-
-
-void parse_font(const char * buf, size_t buf_len, document_t *doc);
-
-#endif

src/parsing/media.c

@@ -1,398 +0,0 @@
-#include "src/sist.h"
-#include "src/ctx.h"
-
-#define MIN_SIZE 32
-#define AVIO_BUF_SIZE 8192
-
-__always_inline
-static AVCodecContext *alloc_jpeg_encoder(int dstW, int dstH, float qscale) {
-
-    AVCodec *jpeg_codec = avcodec_find_encoder(AV_CODEC_ID_MJPEG);
-    AVCodecContext *jpeg = avcodec_alloc_context3(jpeg_codec);
-    jpeg->width = dstW;
-    jpeg->height = dstH;
-    jpeg->time_base.den = 1000000;
-    jpeg->time_base.num = 1;
-    jpeg->i_quant_factor = qscale;
-
-    jpeg->pix_fmt = AV_PIX_FMT_YUVJ420P;
-    int ret = avcodec_open2(jpeg, jpeg_codec, NULL);
-
-    if (ret != 0) {
-        printf("Could not open jpeg encoder: %s!\n", av_err2str(ret));
-        return NULL;
-    }
-
-    return jpeg;
-}
-
-__always_inline
-AVFrame *scale_frame(const AVCodecContext *decoder, const AVFrame *frame, int size) {
-
-    int dstW;
-    int dstH;
-    if (frame->width <= size && frame->height <= size) {
-        dstW = frame->width;
-        dstH = frame->height;
-    } else {
-        double ratio = (double) frame->width / frame->height;
-        if (frame->width > frame->height) {
-            dstW = size;
-            dstH = (int) (size / ratio);
-        } else {
-            dstW = (int) (size * ratio);
-            dstH = size;
-        }
-    }
-
-    if (dstW <= MIN_SIZE || dstH <= MIN_SIZE) {
-        return NULL;
-    }
-
-    AVFrame *scaled_frame = av_frame_alloc();
-
-    struct SwsContext *ctx = sws_getContext(
-            decoder->width, decoder->height, decoder->pix_fmt,
-            dstW, dstH, AV_PIX_FMT_YUVJ420P,
-            SWS_FAST_BILINEAR, 0, 0, 0
-    );
-
-    int dst_buf_len = av_image_get_buffer_size(AV_PIX_FMT_YUV420P, dstW, dstH, 1);
-    uint8_t *dst_buf = (uint8_t *) av_malloc(dst_buf_len);
-
-    av_image_fill_arrays(scaled_frame->data, scaled_frame->linesize, dst_buf, AV_PIX_FMT_YUV420P, dstW, dstH, 1);
-
-    sws_scale(ctx,
-              (const uint8_t *const *) frame->data, frame->linesize,
-              0, decoder->height,
-              scaled_frame->data, scaled_frame->linesize
-    );
-
-    scaled_frame->width = dstW;
-    scaled_frame->height = dstH;
-    scaled_frame->format = AV_PIX_FMT_YUV420P;
-
-    sws_freeContext(ctx);
-
-    return scaled_frame;
-}
-
-__always_inline
-static AVFrame *read_frame(AVFormatContext *pFormatCtx, AVCodecContext *decoder, int stream_idx, document_t *doc) {
-    AVFrame *frame = av_frame_alloc();
-
-    AVPacket avPacket;
-    av_init_packet(&avPacket);
-
-    int receive_ret = -EAGAIN;
-    while (receive_ret == -EAGAIN) {
-        // Get video frame
-        while (1) {
-            int read_frame_ret = av_read_frame(pFormatCtx, &avPacket);
-
-            if (read_frame_ret != 0) {
-                if (read_frame_ret != AVERROR_EOF) {
-                    LOG_WARNINGF(doc->filepath,
-                                 "(media.c) avcodec_read_frame() returned error code [%d] %s",
-                                 read_frame_ret, av_err2str(read_frame_ret)
-                    )
-                }
-                av_frame_free(&frame);
-                av_packet_unref(&avPacket);
-                return NULL;
-            }
-
-            //Ignore audio/other frames
-            if (avPacket.stream_index != stream_idx) {
-                av_packet_unref(&avPacket);
-                continue;
-            }
-            break;
-        }
-
-        // Feed it to decoder
-        int decode_ret = avcodec_send_packet(decoder, &avPacket);
-        if (decode_ret != 0) {
-            LOG_WARNINGF(doc->filepath,
-                         "(media.c) avcodec_send_packet() returned error code [%d] %s",
-                         decode_ret, av_err2str(decode_ret)
-            )
-        }
-        av_packet_unref(&avPacket);
-        receive_ret = avcodec_receive_frame(decoder, frame);
-    }
-    return frame;
-}
-
-#define APPEND_TAG_META(doc, tag_, keyname) \
-    text_buffer_t tex = text_buffer_create(-1); \
-    text_buffer_append_string0(&tex, tag_->value); \
-    text_buffer_terminate_string(&tex); \
-    meta_line_t *meta_tag = malloc(sizeof(meta_line_t) + tex.dyn_buffer.cur); \
-    meta_tag->key = keyname; \
-    strcpy(meta_tag->strval, tex.dyn_buffer.buf); \
-    APPEND_META(doc, meta_tag) \
-    text_buffer_destroy(&tex);
-
-__always_inline
-static void append_audio_meta(AVFormatContext *pFormatCtx, document_t *doc) {
-
-    AVDictionaryEntry *tag = NULL;
-    while ((tag = av_dict_get(pFormatCtx->metadata, "", tag, AV_DICT_IGNORE_SUFFIX))) {
-        char key[32];
-        strncpy(key, tag->key, sizeof(key));
-
-        char *ptr = key;
-        for (; *ptr; ++ptr) *ptr = (char) tolower(*ptr);
-
-        if (strcmp(key, "artist") == 0) {
-            APPEND_TAG_META(doc, tag, MetaArtist)
-        } else if (strcmp(key, "genre") == 0) {
-            APPEND_TAG_META(doc, tag, MetaGenre)
-        } else if (strcmp(key, "title") == 0) {
-            APPEND_TAG_META(doc, tag, MetaTitle)
-        } else if (strcmp(key, "album_artist") == 0) {
-            APPEND_TAG_META(doc, tag, MetaAlbumArtist)
-        } else if (strcmp(key, "album") == 0) {
-            APPEND_TAG_META(doc, tag, MetaAlbum)
-        }
-    }
-}
-
-__always_inline
-static void append_video_meta(AVFormatContext *pFormatCtx, AVFrame *frame, document_t *doc, int include_audio_tags, int is_video) {
-
-    if (is_video) {
-        meta_line_t *meta_duration = malloc(sizeof(meta_line_t));
-        meta_duration->key = MetaMediaDuration;
-        meta_duration->longval = pFormatCtx->duration / AV_TIME_BASE;
-        APPEND_META(doc, meta_duration)
-
-        meta_line_t *meta_bitrate = malloc(sizeof(meta_line_t));
-        meta_bitrate->key = MetaMediaBitrate;
-        meta_bitrate->longval = pFormatCtx->bit_rate;
-        APPEND_META(doc, meta_bitrate)
-    }
-
-    AVDictionaryEntry *tag = NULL;
-    if (is_video) {
-        while ((tag = av_dict_get(pFormatCtx->metadata, "", tag, AV_DICT_IGNORE_SUFFIX))) {
-            if (include_audio_tags && strcmp(tag->key, "title") == 0) {
-                APPEND_TAG_META(doc, tag, MetaTitle)
-            } else if (strcmp(tag->key, "comment") == 0) {
-                APPEND_TAG_META(doc, tag, MetaContent)
-            } else if (include_audio_tags && strcmp(tag->key, "artist") == 0) {
-                APPEND_TAG_META(doc, tag, MetaArtist)
-            }
-        }
-    } else {
-        // EXIF metadata
-        while ((tag = av_dict_get(frame->metadata, "", tag, AV_DICT_IGNORE_SUFFIX))) {
-            if (include_audio_tags && strcmp(tag->key, "Artist") == 0) {
-                APPEND_TAG_META(doc, tag, MetaArtist)
-            } else if (strcmp(tag->key, "ImageDescription") == 0) {
-                APPEND_TAG_META(doc, tag, MetaContent)
-            } else if (strcmp(tag->key, "Make") == 0) {
-                APPEND_TAG_META(doc, tag, MetaExifMake)
-            } else if (strcmp(tag->key, "Model") == 0) {
-                APPEND_TAG_META(doc, tag, MetaExifModel)
-            } else if (strcmp(tag->key, "Software") == 0) {
-                APPEND_TAG_META(doc, tag, MetaExifSoftware)
-            } else if (strcmp(tag->key, "FNumber") == 0) {
-                APPEND_TAG_META(doc, tag, MetaExifFNumber)
-            } else if (strcmp(tag->key, "FocalLength") == 0) {
-                APPEND_TAG_META(doc, tag, MetaExifFocalLength)
-            } else if (strcmp(tag->key, "UserComment") == 0) {
-                APPEND_TAG_META(doc, tag, MetaExifUserComment)
-            } else if (strcmp(tag->key, "ISOSpeedRatings") == 0) {
-                APPEND_TAG_META(doc, tag, MetaExifIsoSpeedRatings)
-            } else if (strcmp(tag->key, "ExposureTime") == 0) {
-                APPEND_TAG_META(doc, tag, MetaExifExposureTime)
-            } else if (strcmp(tag->key, "DateTime") == 0) {
-                APPEND_TAG_META(doc, tag, MetaExifDateTime)
-            }
-        }
-    }
-}
-
-void parse_media(AVFormatContext *pFormatCtx, document_t *doc) {
-
-    int video_stream = -1;
-    int audio_stream = -1;
-
-    avformat_find_stream_info(pFormatCtx, NULL);
-
-    for (int i = (int) pFormatCtx->nb_streams - 1; i >= 0; i--) {
-        AVStream *stream = pFormatCtx->streams[i];
-
-        if (stream->codecpar->codec_type == AVMEDIA_TYPE_AUDIO) {
-            if (audio_stream == -1) {
-                meta_line_t *meta_audio = malloc(sizeof(meta_line_t));
-                meta_audio->key = MetaMediaAudioCodec;
-                meta_audio->intval = stream->codecpar->codec_id;
-                APPEND_META(doc, meta_audio)
-
-                append_audio_meta(pFormatCtx, doc);
-                audio_stream = i;
-            }
-        } else if (stream->codecpar->codec_type == AVMEDIA_TYPE_VIDEO) {
-
-            if (video_stream == -1) {
-                meta_line_t *meta_vid = malloc(sizeof(meta_line_t));
-                meta_vid->key = MetaMediaVideoCodec;
-                meta_vid->intval = stream->codecpar->codec_id;
-                APPEND_META(doc, meta_vid)
-
-                meta_line_t *meta_w = malloc(sizeof(meta_line_t));
-                meta_w->key = MetaWidth;
-                meta_w->intval = stream->codecpar->width;
-                APPEND_META(doc, meta_w)
-
-                meta_line_t *meta_h = malloc(sizeof(meta_line_t));
-                meta_h->key = MetaHeight;
-                meta_h->intval = stream->codecpar->height;
-                APPEND_META(doc, meta_h)
-
-                video_stream = i;
-            }
-        }
-    }
-
-    if (video_stream != -1 && ScanCtx.tn_size > 0) {
-        AVStream *stream = pFormatCtx->streams[video_stream];
-
-        if (stream->codecpar->width <= MIN_SIZE || stream->codecpar->height <= MIN_SIZE) {
-            avformat_close_input(&pFormatCtx);
-            avformat_free_context(pFormatCtx);
-            return;
-        }
-
-        // Decoder
-        AVCodec *video_codec = avcodec_find_decoder(stream->codecpar->codec_id);
-        AVCodecContext *decoder = avcodec_alloc_context3(video_codec);
-        avcodec_parameters_to_context(decoder, stream->codecpar);
-        avcodec_open2(decoder, video_codec, NULL);
-
-        //Seek
-        if (stream->nb_frames > 1 && stream->codecpar->codec_id != AV_CODEC_ID_GIF) {
-            int seek_ret = 0;
-            for (int i = 20; i >= 0; i--) {
-                seek_ret = av_seek_frame(pFormatCtx, video_stream,
-                                         stream->duration * 0.10, 0);
-                if (seek_ret == 0) {
-                    break;
-                }
-            }
-        }
-
-        AVFrame *frame = read_frame(pFormatCtx, decoder, video_stream, doc);
-        if (frame == NULL) {
-            avcodec_free_context(&decoder);
-            avformat_close_input(&pFormatCtx);
-            avformat_free_context(pFormatCtx);
-            return;
-        }
-
-        append_video_meta(pFormatCtx, frame, doc, audio_stream == -1, stream->nb_frames > 1);
-
-        // Scale frame
-        AVFrame *scaled_frame = scale_frame(decoder, frame, ScanCtx.tn_size);
-
-        if (scaled_frame == NULL) {
-            av_frame_free(&frame);
-            avcodec_free_context(&decoder);
-            avformat_close_input(&pFormatCtx);
-            avformat_free_context(pFormatCtx);
-            return;
-        }
-
-        // Encode frame to jpeg
-        AVCodecContext *jpeg_encoder = alloc_jpeg_encoder(scaled_frame->width, scaled_frame->height, ScanCtx.tn_qscale);
-        avcodec_send_frame(jpeg_encoder, scaled_frame);
-
-        AVPacket jpeg_packet;
-        av_init_packet(&jpeg_packet);
-        avcodec_receive_packet(jpeg_encoder, &jpeg_packet);
-
-        // Save thumbnail
-        store_write(ScanCtx.index.store, (char *) doc->uuid, sizeof(doc->uuid), (char *) jpeg_packet.data,
-                    jpeg_packet.size);
-
-        av_packet_unref(&jpeg_packet);
-        av_frame_free(&frame);
-        av_free(*scaled_frame->data);
-        av_frame_free(&scaled_frame);
-        avcodec_free_context(&jpeg_encoder);
-        avcodec_free_context(&decoder);
-    }
-
-    avformat_close_input(&pFormatCtx);
-    avformat_free_context(pFormatCtx);
-}
-
-void parse_media_filename(const char *filepath, document_t *doc) {
-
-    AVFormatContext *pFormatCtx = avformat_alloc_context();
-    if (pFormatCtx == NULL) {
-        LOG_ERROR(doc->filepath, "(media.c) Could not allocate context with avformat_alloc_context()")
-        return;
-    }
-    int res = avformat_open_input(&pFormatCtx, filepath, NULL, NULL);
-    if (res < 0) {
-        LOG_ERRORF(doc->filepath, "(media.c) avformat_open_input() returned [%d] %s", res, av_err2str(res))
-        avformat_close_input(&pFormatCtx);
-        avformat_free_context(pFormatCtx);
-        return;
-    }
-
-    parse_media(pFormatCtx, doc);
-}
-
-
-int vfile_read(void *ptr, uint8_t *buf, int buf_size) {
-    struct vfile *f = ptr;
-
-    int ret = f->read(f, buf, buf_size);
-
-    if (ret == 0) {
-        return AVERROR_EOF;
-    }
-    return ret;
-}
-
-void parse_media_vfile(struct vfile *f, document_t *doc) {
-
-    AVFormatContext *pFormatCtx = avformat_alloc_context();
-    if (pFormatCtx == NULL) {
-        LOG_ERROR(doc->filepath, "(media.c) Could not allocate context with avformat_alloc_context()")
-        return;
-    }
-
-    unsigned char *buffer = (unsigned char *) av_malloc(AVIO_BUF_SIZE);
-    AVIOContext *io_ctx = avio_alloc_context(buffer, AVIO_BUF_SIZE, 0, f, vfile_read, NULL, NULL);
-
-    pFormatCtx->pb = io_ctx;
-    pFormatCtx->flags |= AVFMT_FLAG_CUSTOM_IO;
-
-    int res = avformat_open_input(&pFormatCtx, "", NULL, NULL);
-    if (res == -5) {
-        // Tried to parse media that requires seek
-        av_free(io_ctx->buffer);
-        avio_context_free(&io_ctx);
-        avformat_close_input(&pFormatCtx);
-        avformat_free_context(pFormatCtx);
-        return;
-    } else if (res < 0) {
-        LOG_ERRORF(doc->filepath, "(media.c) avformat_open_input() returned [%d] %s", res, av_err2str(res))
-        av_free(io_ctx->buffer);
-        avio_context_free(&io_ctx);
-        avformat_close_input(&pFormatCtx);
-        avformat_free_context(pFormatCtx);
-        return;
-    }
-
-    parse_media(pFormatCtx, doc);
-    av_free(io_ctx->buffer);
-    avio_context_free(&io_ctx);
-}
-

src/parsing/media.h

@@ -1,14 +0,0 @@
-#ifndef SIST2_MEDIA_H
-#define SIST2_MEDIA_H
-
-
-#include "src/sist.h"
-
-#define MIN_VIDEO_SIZE 1024 * 64
-#define MIN_IMAGE_SIZE 1024 * 2
-
-void parse_media_filename(const char * filepath, document_t *doc);
-
-void parse_media_vfile(struct vfile *f, document_t *doc);
-
-#endif

src/parsing/mime.h

@@ -1,11 +1,12 @@
 #ifndef SIST2_MIME_H
 #define SIST2_MIME_H
 
-#include "src/sist.h"
+#include "../sist.h"
 
-#define MAJOR_MIME(mime_id) (mime_id & 0x0FFF0000) >> 16
+#define MAJOR_MIME(mime_id) (mime_id & 0x000F0000) >> 16
 
 #define MIME_EMPTY 1
+#define MIME_SIST2_SIDECAR 2
 
 #define DONT_PARSE 0x80000000
 #define SHOULD_PARSE(mime_id) (ScanCtx.fast == 0 && (mime_id & DONT_PARSE) != DONT_PARSE && mime_id != 0)
@@ -25,6 +26,15 @@
 #define DOC_MASK 0x04000000
 #define IS_DOC(mime_id) (mime_id & DOC_MASK) == DOC_MASK
 
+#define MOBI_MASK 0x02000000
+#define IS_MOBI(mime_id) (mime_id & MOBI_MASK) == MOBI_MASK
+
+#define MARKUP_MASK 0x01000000
+#define IS_MARKUP(mime_id) (mime_id & MARKUP_MASK) == MARKUP_MASK
+
+#define RAW_MASK 0x00800000
+#define IS_RAW(mime_id) (mime_id & RAW_MASK) == RAW_MASK
+
 enum major_mime {
     MimeInvalid = 0,
     MimeModel = 1,

src/parsing/parse.c

@@ -1,7 +1,16 @@
+#include "parse.h"
+
 #include "src/sist.h"
 #include "src/ctx.h"
+#include "mime.h"
+#include "src/io/serialize.h"
+#include "src/parsing/sidecar.h"
 
-__thread magic_t Magic = NULL;
+#include <magic.h>
+
+
+#define MIN_VIDEO_SIZE 1024 * 64
+#define MIN_IMAGE_SIZE 1024 * 2
 
 int fs_read(struct vfile *f, void *buf, size_t size) {
 
@@ -24,63 +33,47 @@ void fs_close(struct vfile *f) {
     }
 }
 
-void *read_all(parse_job_t *job, const char *buf, int bytes_read) {
-
-    void *full_buf;
-
-    if (job->info.st_size <= bytes_read) {
-        full_buf = malloc(job->info.st_size);
-        memcpy(full_buf, buf, job->info.st_size);
-    } else {
-        full_buf = malloc(job->info.st_size);
-        memcpy(full_buf, buf, bytes_read);
-
-        int ret = job->vfile.read(&job->vfile, full_buf + bytes_read, job->info.st_size - bytes_read);
-        if (ret < 0) {
-            LOG_ERRORF(job->filepath, "read(): [%d] %s", errno, strerror(errno))
-            return NULL;
-        }
+void fs_reset(struct vfile *f) {
+    if (f->fd != -1) {
+        lseek(f->fd, 0, SEEK_SET);
     }
-
-    return full_buf;
 }
 
+#define IS_GIT_OBJ (strlen(doc.filepath + doc.base) == 38 && (strstr(doc.filepath, "objects") != NULL))
+
 void parse(void *arg) {
 
     parse_job_t *job = arg;
     document_t doc;
 
-    int inc_ts = incremental_get(ScanCtx.original_table, job->info.st_ino);
-    if (inc_ts != 0 && inc_ts == job->info.st_mtim.tv_sec) {
-        incremental_mark_file_for_copy(ScanCtx.copy_table, job->info.st_ino);
-        return;
-    }
-
-    if (Magic == NULL) {
-        Magic = magic_open(MAGIC_MIME_TYPE);
-        magic_load(Magic, NULL);
-    }
-
     doc.filepath = job->filepath;
     doc.ext = (short) job->ext;
     doc.base = (short) job->base;
+
+    char *rel_path = doc.filepath + ScanCtx.index.desc.root_len;
+    MD5((unsigned char *) rel_path, strlen(rel_path), doc.path_md5);
+
     doc.meta_head = NULL;
     doc.meta_tail = NULL;
     doc.mime = 0;
-    doc.size = job->info.st_size;
-    doc.ino = job->info.st_ino;
-    doc.mtime = job->info.st_mtim.tv_sec;
+    doc.size = job->vfile.info.st_size;
+    doc.mtime = job->vfile.info.st_mtim.tv_sec;
 
-    uuid_generate(doc.uuid);
-    char *buf[PARSE_BUF_SIZE];
-
-    if (LogCtx.very_verbose) {
-        char uuid_str[UUID_STR_LEN];
-        uuid_unparse(doc.uuid, uuid_str);
-        LOG_DEBUGF(job->filepath, "Starting parse job {%s}", uuid_str)
+    int inc_ts = incremental_get(ScanCtx.original_table, doc.path_md5);
+    if (inc_ts != 0 && inc_ts == job->vfile.info.st_mtim.tv_sec) {
+        incremental_mark_file_for_copy(ScanCtx.copy_table, doc.path_md5);
+        return;
     }
 
-    if (job->info.st_size == 0) {
+    char *buf[MAGIC_BUF_SIZE];
+
+    if (LogCtx.very_verbose) {
+        char path_md5_str[MD5_STR_LENGTH];
+        buf2hex(doc.path_md5, MD5_DIGEST_LENGTH, path_md5_str);
+        LOG_DEBUGF(job->filepath, "Starting parse job {%s}", path_md5_str)
+    }
+
+    if (job->vfile.info.st_size == 0) {
         doc.mime = MIME_EMPTY;
     } else if (*(job->filepath + job->ext) != '\0' && (job->ext - job->base != 1)) {
         doc.mime = mime_get_mime_by_ext(ScanCtx.ext_table, job->filepath + job->ext);
@@ -89,15 +82,34 @@ void parse(void *arg) {
     int bytes_read = 0;
 
     if (doc.mime == 0 && !ScanCtx.fast) {
+        if (IS_GIT_OBJ) {
+            goto abort;
+        }
+
         // Get mime type with libmagic
-        bytes_read = job->vfile.read(&job->vfile, buf, PARSE_BUF_SIZE);
+        if (!job->vfile.is_fs_file) {
+            LOG_WARNING(job->filepath,
+                        "Guessing mime type with libmagic inside archive files is not currently supported");
+            goto abort;
+        }
+
+        bytes_read = job->vfile.read(&job->vfile, buf, MAGIC_BUF_SIZE);
         if (bytes_read < 0) {
-            LOG_WARNINGF(job->filepath, "read() Error: %s", strerror(errno))
+
+            if (job->vfile.is_fs_file) {
+                LOG_ERRORF(job->filepath, "read(): [%d] %s", errno, strerror(errno))
+            } else {
+                LOG_ERRORF(job->filepath, "(virtual) read(): [%d] %s", bytes_read, archive_error_string(job->vfile.arc))
+            }
+
             CLOSE_FILE(job->vfile)
             return;
         }
 
-        const char *magic_mime_str = magic_buffer(Magic, buf, bytes_read);
+        magic_t magic = magic_open(MAGIC_MIME_TYPE);
+        magic_load(magic, NULL);
+
+        const char *magic_mime_str = magic_buffer(magic, buf, bytes_read);
         if (magic_mime_str != NULL) {
             doc.mime = mime_get_mime_by_string(ScanCtx.mime_table, magic_mime_str);
 
@@ -108,75 +120,74 @@ void parse(void *arg) {
             }
         }
 
-        magic_close(Magic);
-        Magic = NULL;
+        job->vfile.reset(&job->vfile);
+
+        magic_close(magic);
     }
 
     int mmime = MAJOR_MIME(doc.mime);
 
     if (!(SHOULD_PARSE(doc.mime))) {
 
+    } else if (IS_RAW(doc.mime)) {
+        parse_raw(&ScanCtx.raw_ctx, &job->vfile, &doc);
     } else if ((mmime == MimeVideo && doc.size >= MIN_VIDEO_SIZE) ||
                (mmime == MimeImage && doc.size >= MIN_IMAGE_SIZE) || mmime == MimeAudio) {
 
-//        if (job->vfile.is_fs_file) {
-//            parse_media_filename(job->filepath, &doc);
-//        } else {
-//            parse_media_vfile(&job->vfile, &doc);
-//        }
+        parse_media(&ScanCtx.media_ctx, &job->vfile, &doc);
 
     } else if (IS_PDF(doc.mime)) {
-        void *pdf_buf = read_all(job, (char *) buf, bytes_read);
-        parse_pdf(pdf_buf, doc.size, &doc);
+        parse_ebook(&ScanCtx.ebook_ctx, &job->vfile, mime_get_mime_text(doc.mime), &doc);
 
-        if (pdf_buf != buf && pdf_buf != NULL) {
-            free(pdf_buf);
+    } else if (mmime == MimeText && ScanCtx.text_ctx.content_size > 0) {
+        if (IS_MARKUP(doc.mime)) {
+            parse_markup(&ScanCtx.text_ctx, &job->vfile, &doc);
+        } else {
+            parse_text(&ScanCtx.text_ctx, &job->vfile, &doc);
         }
 
-    } else if (mmime == MimeText && ScanCtx.content_size > 0) {
-        parse_text(bytes_read, &job->vfile, (char *) buf, &doc);
-
     } else if (IS_FONT(doc.mime)) {
-        void *font_buf = read_all(job, (char *) buf, bytes_read);
-        parse_font(font_buf, doc.size, &doc);
+        parse_font(&ScanCtx.font_ctx, &job->vfile, &doc);
 
-        if (font_buf != buf && font_buf != NULL) {
-            free(font_buf);
-        }
     } else if (
-            ScanCtx.archive_mode != ARC_MODE_SKIP && (
+            ScanCtx.arc_ctx.mode != ARC_MODE_SKIP && (
                     IS_ARC(doc.mime) ||
                     (IS_ARC_FILTER(doc.mime) && should_parse_filtered_file(doc.filepath, doc.ext))
             )) {
-        parse_archive(&job->vfile, &doc);
-    } else if (ScanCtx.content_size > 0 && IS_DOC(doc.mime)) {
-        void *doc_buf = read_all(job, (char *) buf, bytes_read);
-        parse_doc(doc_buf, doc.size, &doc);
-
-        if (doc_buf != buf && doc_buf != NULL) {
-            free(doc_buf);
-        }
-    } else if (is_cbr(doc.mime)) {
-        void *cbr_buf = read_all(job, (char *) buf, bytes_read);
-        parse_cbr(cbr_buf, doc.size, &doc);
-
-        if (cbr_buf != buf && cbr_buf != NULL) {
-            free(cbr_buf);
-        }
+        parse_archive(&ScanCtx.arc_ctx, &job->vfile, &doc);
+    } else if ((ScanCtx.ooxml_ctx.content_size > 0 || ScanCtx.media_ctx.tn_size > 0) && IS_DOC(doc.mime)) {
+        parse_ooxml(&ScanCtx.ooxml_ctx, &job->vfile, &doc);
+    } else if (is_cbr(&ScanCtx.comic_ctx, doc.mime) || is_cbz(&ScanCtx.comic_ctx, doc.mime)) {
+        parse_comic(&ScanCtx.comic_ctx, &job->vfile, &doc);
+    } else if (IS_MOBI(doc.mime)) {
+        parse_mobi(&ScanCtx.mobi_ctx, &job->vfile, &doc);
+    } else if (doc.mime == MIME_SIST2_SIDECAR) {
+        parse_sidecar(&job->vfile, &doc);
+        CLOSE_FILE(job->vfile)
+        return;
+    } else if (is_msdoc(&ScanCtx.msdoc_ctx, doc.mime)) {
+        parse_msdoc(&ScanCtx.msdoc_ctx, &job->vfile, &doc);
     }
 
-    //Parent meta
-    if (!uuid_is_null(job->parent)) {
-        char tmp[UUID_STR_LEN];
-        uuid_unparse(job->parent, tmp);
+    abort:
 
-        meta_line_t *meta_parent = malloc(sizeof(meta_line_t) + UUID_STR_LEN + 1);
+    //Parent meta
+    if (!md5_digest_is_null(job->parent)) {
+        meta_line_t *meta_parent = malloc(sizeof(meta_line_t) + MD5_STR_LENGTH);
         meta_parent->key = MetaParent;
-        strcpy(meta_parent->strval, tmp);
+        buf2hex(job->parent, MD5_DIGEST_LENGTH, meta_parent->str_val);
         APPEND_META((&doc), meta_parent)
+
+        doc.has_parent = TRUE;
+    } else {
+        doc.has_parent = FALSE;
     }
 
     write_document(&doc);
 
     CLOSE_FILE(job->vfile)
 }
+
+void cleanup_parse() {
+    // noop
+}

src/parsing/parse.h

@@ -1,13 +1,16 @@
 #ifndef SIST2_PARSE_H
 #define SIST2_PARSE_H
 
-#include "src/sist.h"
+#include "../sist.h"
 
-#define PARSE_BUF_SIZE 4096
+#define MAGIC_BUF_SIZE 4096 * 6
 
 int fs_read(struct vfile *f, void *buf, size_t size);
 void fs_close(struct vfile *f);
+void fs_reset(struct vfile *f);
 
 void parse(void *arg);
 
+void cleanup_parse();
+
 #endif

src/parsing/pdf.c

@@ -1,331 +0,0 @@
-#include "pdf.h"
-#include "src/ctx.h"
-
-#define MIN_OCR_SIZE 350
-#define MIN_OCR_LEN 10
-__thread text_buffer_t thread_buffer;
-
-
-int render_cover(fz_context *ctx, document_t *doc, fz_document *fzdoc) {
-
-    int err = 0;
-    fz_page *cover = NULL;
-
-    fz_var(cover);
-    fz_var(err);
-    fz_try(ctx)
-        cover = fz_load_page(ctx, fzdoc, 0);
-    fz_catch(ctx)
-        err = 1;
-
-    if (err != 0) {
-        fz_drop_page(ctx, cover);
-        LOG_WARNINGF(doc->filepath, "fz_load_page() returned error code [%d] %s", err, ctx->error.message)
-        return FALSE;
-    }
-
-    fz_rect bounds = fz_bound_page(ctx, cover);
-
-    float scale;
-    float w = (float) bounds.x1 - bounds.x0;
-    float h = (float) bounds.y1 - bounds.y0;
-    if (w > h) {
-        scale = (float) ScanCtx.tn_size / w;
-    } else {
-        scale = (float) ScanCtx.tn_size / h;
-    }
-    fz_matrix m = fz_scale(scale, scale);
-
-    bounds = fz_transform_rect(bounds, m);
-    fz_irect bbox = fz_round_rect(bounds);
-    fz_pixmap *pixmap = fz_new_pixmap_with_bbox(ctx, ctx->colorspace->rgb, bbox, NULL, 0);
-
-    fz_clear_pixmap_with_value(ctx, pixmap, 0xFF);
-    fz_device *dev = fz_new_draw_device(ctx, m, pixmap);
-
-    fz_var(err);
-    fz_try(ctx)
-    {
-        pthread_mutex_lock(&ScanCtx.mupdf_mu);
-        fz_run_page(ctx, cover, dev, fz_identity, NULL);
-    }
-    fz_always(ctx)
-    {
-        fz_close_device(ctx, dev);
-        fz_drop_device(ctx, dev);
-        pthread_mutex_unlock(&ScanCtx.mupdf_mu);
-    }
-    fz_catch(ctx)
-        err = ctx->error.errcode;
-
-    if (err != 0) {
-        LOG_WARNINGF(doc->filepath, "fz_run_page() returned error code [%d] %s", err, ctx->error.message)
-        fz_drop_page(ctx, cover);
-        fz_drop_pixmap(ctx, pixmap);
-        return FALSE;
-    }
-
-    fz_buffer *fzbuf = NULL;
-    fz_var(fzbuf);
-    fz_var(err);
-
-    fz_try(ctx)
-        fzbuf = fz_new_buffer_from_pixmap_as_png(ctx, pixmap, fz_default_color_params);
-    fz_catch(ctx)
-        err = ctx->error.errcode;
-
-    if (err == 0) {
-        unsigned char *tn_buf;
-        size_t tn_len = fz_buffer_storage(ctx, fzbuf, &tn_buf);
-        store_write(ScanCtx.index.store, (char *) doc->uuid, sizeof(doc->uuid), (char *) tn_buf, tn_len);
-    }
-
-    fz_drop_buffer(ctx, fzbuf);
-    fz_drop_pixmap(ctx, pixmap);
-    fz_drop_page(ctx, cover);
-
-    if (err != 0) {
-        LOG_WARNINGF(doc->filepath, "fz_new_buffer_from_pixmap_as_png() returned error code [%d] %s", err,
-                     ctx->error.message)
-        return FALSE;
-    }
-
-    return TRUE;
-}
-
-void fz_err_callback(void *user, UNUSED(const char *message)) {
-    if (LogCtx.verbose) {
-        document_t *doc = (document_t *) user;
-        LOG_WARNINGF(doc->filepath, "FZ: %s", message)
-    }
-}
-
-__always_inline
-static void init_ctx(fz_context *ctx, document_t *doc) {
-    fz_disable_icc(ctx);
-    fz_register_document_handlers(ctx);
-
-    ctx->warn.print_user = doc;
-    ctx->warn.print = fz_err_callback;
-    ctx->error.print_user = doc;
-    ctx->error.print = fz_err_callback;
-}
-
-__always_inline
-static int read_stext_block(fz_stext_block *block, text_buffer_t *tex) {
-    if (block->type != FZ_STEXT_BLOCK_TEXT) {
-        return 0;
-    }
-
-    fz_stext_line *line = block->u.t.first_line;
-    while (line != NULL) {
-        fz_stext_char *c = line->first_char;
-        while (c != NULL) {
-            if (text_buffer_append_char(tex, c->c) == TEXT_BUF_FULL) {
-                return TEXT_BUF_FULL;
-            }
-            c = c->next;
-        }
-        line = line->next;
-    }
-    return 0;
-}
-
-#define IS_VALID_BPP(d) (d==1 || d==2 || d==4 || d==8 || d==16 || d==24 || d==32)
-
-void fill_image(fz_context *ctx, UNUSED(fz_device *dev),
-                fz_image *img, UNUSED(fz_matrix ctm), UNUSED(float alpha),
-                UNUSED(fz_color_params color_params)) {
-
-    int l2factor = 0;
-
-    if (img->w > MIN_OCR_SIZE && img->h > MIN_OCR_SIZE && IS_VALID_BPP(img->n)) {
-
-        fz_pixmap *pix = img->get_pixmap(ctx, img, NULL, img->w, img->h, &l2factor);
-
-        if (pix->h > MIN_OCR_SIZE && img->h > MIN_OCR_SIZE && img->xres != 0) {
-            TessBaseAPI *api = TessBaseAPICreate();
-            TessBaseAPIInit3(api, ScanCtx.tesseract_path, ScanCtx.tesseract_lang);
-
-            TessBaseAPISetImage(api, pix->samples, pix->w, pix->h, pix->n, pix->stride);
-            TessBaseAPISetSourceResolution(api, pix->xres);
-
-            char *text = TessBaseAPIGetUTF8Text(api);
-            size_t len = strlen(text);
-            if (len >= MIN_OCR_LEN) {
-                text_buffer_append_string(&thread_buffer, text, len - 1);
-                LOG_DEBUGF(
-                        "pdf.c",
-                        "(OCR) %dx%d got %dB from tesseract (%s), buffer:%dB",
-                        pix->w, pix->h, len, ScanCtx.tesseract_lang, thread_buffer.dyn_buffer.cur
-                )
-            }
-
-            TessBaseAPIEnd(api);
-            TessBaseAPIDelete(api);
-        }
-        fz_drop_pixmap(ctx, pix);
-    }
-}
-
-void parse_pdf(void *buf, size_t buf_len, document_t *doc) {
-
-    if (buf == NULL) {
-        return;
-    }
-
-    static int mu_is_initialized = 0;
-    if (!mu_is_initialized) {
-        pthread_mutex_init(&ScanCtx.mupdf_mu, NULL);
-        mu_is_initialized = 1;
-    }
-    fz_context *ctx = fz_new_context(NULL, NULL, FZ_STORE_UNLIMITED);
-
-    init_ctx(ctx, doc);
-
-    int err = 0;
-
-    fz_document *fzdoc = NULL;
-    fz_stream *stream = NULL;
-    fz_var(fzdoc);
-    fz_var(stream);
-    fz_var(err);
-
-    fz_try(ctx)
-    {
-        stream = fz_open_memory(ctx, buf, buf_len);
-        fzdoc = fz_open_document_with_stream(ctx, mime_get_mime_text(doc->mime), stream);
-    }
-    fz_catch(ctx)
-        err = ctx->error.errcode;
-
-    if (err != 0) {
-        fz_drop_stream(ctx, stream);
-        fz_drop_document(ctx, fzdoc);
-        fz_drop_context(ctx);
-        return;
-    }
-
-    char title[4096] = {'\0',};
-    fz_try(ctx)
-        fz_lookup_metadata(ctx, fzdoc, FZ_META_INFO_TITLE, title, sizeof(title));
-    fz_catch(ctx)
-        ;
-
-    if (strlen(title) > 0) {
-        meta_line_t *meta_content = malloc(sizeof(meta_line_t) + strlen(title));
-        meta_content->key = MetaTitle;
-        strcpy(meta_content->strval, title);
-        APPEND_META(doc, meta_content)
-    }
-
-    int page_count = -1;
-    fz_var(err);
-    fz_try(ctx)
-        page_count = fz_count_pages(ctx, fzdoc);
-    fz_catch(ctx)
-        err = ctx->error.errcode;
-
-    if (err) {
-        LOG_WARNINGF(doc->filepath, "fz_count_pages() returned error code [%d] %s", err, ctx->error.message)
-        fz_drop_stream(ctx, stream);
-        fz_drop_document(ctx, fzdoc);
-        fz_drop_context(ctx);
-        return;
-    }
-
-    if (ScanCtx.tn_size > 0) {
-        err = render_cover(ctx, doc, fzdoc);
-    }
-
-    if (err == TRUE) {
-        fz_drop_stream(ctx, stream);
-        fz_drop_document(ctx, fzdoc);
-        fz_drop_context(ctx);
-        return;
-    }
-
-    if (ScanCtx.content_size > 0) {
-        fz_stext_options opts = {0};
-        thread_buffer = text_buffer_create(ScanCtx.content_size);
-
-        for (int current_page = 0; current_page < page_count; current_page++) {
-            fz_page *page = NULL;
-            fz_var(err);
-            fz_try(ctx)
-                page = fz_load_page(ctx, fzdoc, current_page);
-            fz_catch(ctx)
-                err = ctx->error.errcode;
-            if (err != 0) {
-                LOG_WARNINGF(doc->filepath, "fz_load_page() returned error code [%d] %s", err, ctx->error.message)
-                text_buffer_destroy(&thread_buffer);
-                fz_drop_page(ctx, page);
-                fz_drop_stream(ctx, stream);
-                fz_drop_document(ctx, fzdoc);
-                fz_drop_context(ctx);
-                return;
-            }
-
-            fz_stext_page *stext = fz_new_stext_page(ctx, fz_bound_page(ctx, page));
-            fz_device *dev = fz_new_stext_device(ctx, stext, &opts);
-            dev->stroke_path = NULL;
-            dev->stroke_text = NULL;
-            dev->clip_text = NULL;
-            dev->clip_stroke_path = NULL;
-            dev->clip_stroke_text = NULL;
-
-            if (ScanCtx.tesseract_lang != NULL) {
-                dev->fill_image = fill_image;
-            }
-
-            fz_var(err);
-            fz_try(ctx)
-                fz_run_page(ctx, page, dev, fz_identity, NULL);
-            fz_always(ctx)
-            {
-                fz_close_device(ctx, dev);
-                fz_drop_device(ctx, dev);
-            }
-            fz_catch(ctx)
-                err = ctx->error.errcode;
-
-            if (err != 0) {
-                LOG_WARNINGF(doc->filepath, "fz_run_page() returned error code [%d] %s", err, ctx->error.message)
-                text_buffer_destroy(&thread_buffer);
-                fz_drop_page(ctx, page);
-                fz_drop_stext_page(ctx, stext);
-                fz_drop_stream(ctx, stream);
-                fz_drop_document(ctx, fzdoc);
-                fz_drop_context(ctx);
-                return;
-            }
-
-            fz_stext_block *block = stext->first_block;
-            while (block != NULL) {
-                int ret = read_stext_block(block, &thread_buffer);
-                if (ret == TEXT_BUF_FULL) {
-                    break;
-                }
-                block = block->next;
-            }
-            fz_drop_stext_page(ctx, stext);
-            fz_drop_page(ctx, page);
-
-            if (thread_buffer.dyn_buffer.cur >= thread_buffer.dyn_buffer.size) {
-                break;
-            }
-        }
-        text_buffer_terminate_string(&thread_buffer);
-
-        meta_line_t *meta_content = malloc(sizeof(meta_line_t) + thread_buffer.dyn_buffer.cur);
-        meta_content->key = MetaContent;
-        memcpy(meta_content->strval, thread_buffer.dyn_buffer.buf, thread_buffer.dyn_buffer.cur);
-        APPEND_META(doc, meta_content)
-
-        text_buffer_destroy(&thread_buffer);
-    }
-
-    fz_drop_stream(ctx, stream);
-    fz_drop_document(ctx, fzdoc);
-    fz_drop_context(ctx);
-}

src/parsing/pdf.h

@@ -1,9 +0,0 @@
-#ifndef SIST2_PDF_H
-#define SIST2_PDF_H
-
-#include "src/sist.h"
-
-
-void parse_pdf(void *buf, size_t buf_len, document_t *doc);
-
-#endif

src/parsing/sidecar.c

@@ -0,0 +1,35 @@
+#include "sidecar.h"
+
+#include "src/ctx.h"
+
+void parse_sidecar(vfile_t *vfile, document_t *doc) {
+
+    LOG_DEBUGF("sidecar.c", "Parsing sidecar file %s", vfile->filepath)
+
+    size_t size;
+    char *buf = read_all(vfile, &size);
+    if (buf == NULL) {
+        LOG_ERRORF("sidecar.c", "Read error for %s", vfile->filepath)
+        return;
+    }
+
+    buf = realloc(buf, size + 1);
+    *(buf + size) = '\0';
+
+    cJSON *json = cJSON_Parse(buf);
+    if (json == NULL) {
+        LOG_ERRORF("sidecar.c", "Could not parse JSON sidecar %s", vfile->filepath)
+        return;
+    }
+    char *json_str = cJSON_PrintUnformatted(json);
+
+    unsigned char path_md5[MD5_DIGEST_LENGTH];
+    MD5((unsigned char *) vfile->filepath + ScanCtx.index.desc.root_len, doc->ext - 1 - ScanCtx.index.desc.root_len,
+        path_md5);
+
+    store_write(ScanCtx.index.meta_store, (char *) path_md5, sizeof(path_md5), json_str, strlen(json_str) + 1);
+
+    cJSON_Delete(json);
+    free(json_str);
+    free(buf);
+}

src/parsing/sidecar.h

@@ -0,0 +1,8 @@
+#ifndef SIST2_SIDECAR_H
+#define SIST2_SIDECAR_H
+
+#include "src/sist.h"
+
+void parse_sidecar(vfile_t *vfile, document_t *doc);
+
+#endif

src/parsing/text.c

@@ -1,37 +0,0 @@
-#include "text.h"
-#include "src/ctx.h"
-
-void parse_text(int bytes_read, struct vfile *f, char *buf, document_t *doc) {
-
-    char *intermediate_buf;
-    int intermediate_buf_len;
-
-    if (bytes_read == doc->size || bytes_read >= ScanCtx.content_size) {
-        int to_copy = MIN(bytes_read, ScanCtx.content_size);
-        intermediate_buf = malloc(to_copy);
-        intermediate_buf_len = to_copy;
-        memcpy(intermediate_buf, buf, to_copy);
-
-    } else {
-        int to_read = MIN(ScanCtx.content_size, doc->size) - bytes_read;
-
-        intermediate_buf = malloc(to_read + bytes_read);
-        intermediate_buf_len = to_read + bytes_read;
-        if (bytes_read != 0) {
-            memcpy(intermediate_buf, buf, bytes_read);
-        }
-
-        f->read(f, intermediate_buf + bytes_read, to_read);
-    }
-    text_buffer_t tex = text_buffer_create(ScanCtx.content_size);
-    text_buffer_append_string(&tex, intermediate_buf, intermediate_buf_len);
-    text_buffer_terminate_string(&tex);
-
-    meta_line_t *meta = malloc(sizeof(meta_line_t) + tex.dyn_buffer.cur);
-    meta->key = MetaContent;
-    strcpy(meta->strval, tex.dyn_buffer.buf);
-    APPEND_META(doc, meta)
-
-    free(intermediate_buf);
-    text_buffer_destroy(&tex);
-}

src/parsing/text.h

@@ -1,8 +0,0 @@
-#ifndef SIST2_TEXT_H
-#define SIST2_TEXT_H
-
-#include "src/sist.h"
-
-void parse_text(int bytes_read, struct vfile *f, char *buf, document_t *doc);
-
-#endif

src/sist.h

@@ -1,75 +1,51 @@
 #ifndef SIST_H
 #define SIST_H
 
-#define UUID_STR_LEN 37
+#ifndef	FALSE
+#define	FALSE	(0)
+#define BOOL int
+#endif
+
+#ifndef	TRUE
+#define	TRUE	(!FALSE)
+#endif
+
+#undef	MAX
+#define MAX(a, b)  (((a) > (b)) ? (a) : (b))
+
+#undef	MIN
+#define MIN(a, b)  (((a) < (b)) ? (a) : (b))
+
+#ifndef PATH_MAX
+#define PATH_MAX 4096
+#endif
+
+#undef ABS
+#define ABS(a)	   (((a) < 0) ? -(a) : (a))
+
 #define UNUSED(x) __attribute__((__unused__))  x
 
-#include <glib-2.0/glib.h>
-#include <unistd.h>
+#define MD5_STR_LENGTH 33
+
+#include "util.h"
+#include "log.h"
+#include "types.h"
+
+#include "libscan/scan.h"
+
+#include <cjson/cJSON.h>
+
+#include <string.h>
 #include <stdio.h>
 #include <stdlib.h>
-#include <string.h>
+#include <stdarg.h>
+
+#include <unistd.h>
 #include <fcntl.h>
-#include <ftw.h>
-#include <uuid.h>
-#include <magic.h>
-#include "libavformat/avformat.h"
-#include "libswscale/swscale.h"
-#include "libswresample/swresample.h"
-#include "libavcodec/avcodec.h"
-#include "libavutil/imgutils.h"
-#include <ctype.h>
-#include <mupdf/fitz.h>
-#include <mupdf/pdf.h>
-#include "argparse/argparse.h"
-#include <time.h>
-#include <limits.h>
-#include <pthread.h>
+#include <dirent.h>
 #include <sys/stat.h>
-#include <wordexp.h>
-#include "ft2build.h"
-#include "freetype/freetype.h"
-#include <archive.h>
-#include <archive_entry.h>
-#include <libxml/xmlstring.h>
-#include <libxml/parser.h>
-#define BOOL int
-#include <tesseract/capi.h>
-#include <pcre.h>
-
-#include <onion/onion.h>
-#include <onion/handler.h>
-#include <onion/block.h>
-#include <onion/shortcuts.h>
-#include <onion/codecs.h>
-#include <curl/curl.h>
-
-#include "cJSON/cJSON.h"
-
-#include "types.h"
-#include "tpool.h"
-#include "utf8.h/utf8.h"
-#include "util.h"
-#include "io/store.h"
-#include "io/serialize.h"
-#include "io/walk.h"
-#include "parsing/parse.h"
-#include "parsing/mime.h"
-#include "parsing/text.h"
-#include "parsing/pdf.h"
-#include "parsing/media.h"
-#include "parsing/font.h"
-#include "parsing/arc.h"
-#include "parsing/doc.h"
-#include "parsing/cbr.h"
-#include "cli.h"
-#include "log.h"
-
-#include "src/index/elastic.h"
-#include "index/web.h"
-#include "web/serve.h"
-#include "web/auth_basic.h"
-
-;
+#include <sys/types.h>
+#include <errno.h>
+#include <ctype.h>
 
 #endif

src/static/css/autocomplete.min.css

@@ -0,0 +1,4 @@
+.autocomplete-suggestions { text-align: left; cursor: default; border: 1px solid #ccc; border-top: 0; background: #fff; box-shadow: -1px 1px 3px rgba(0,0,0,.1); position: absolute; display: none; z-index: 9999; max-height: 254px; overflow: hidden; overflow-y: auto; box-sizing: border-box; }
+.autocomplete-suggestion { position: relative; padding: 0 .6em; line-height: 23px; white-space: nowrap; overflow: hidden; text-overflow: ellipsis; font-size: 1.02em; color: #333; }
+.autocomplete-suggestion b { font-weight: normal; color: #1f8dd6; }
+.autocomplete-suggestion.selected { background: #f0f0f0; }

src/static/css/bricklayer.min.css

@@ -0,0 +1 @@
+.bricklayer{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-box-align:start;-webkit-align-items:flex-start;-ms-flex-align:start;align-items:flex-start;-webkit-box-pack:center;-webkit-justify-content:center;-ms-flex-pack:center;justify-content:center;-webkit-flex-wrap:wrap;-ms-flex-wrap:wrap;flex-wrap:wrap}.bricklayer-column-sizer{width:100%;display:none}@media screen and (min-width:640px){.bricklayer-column-sizer{width:50%}}@media screen and (min-width:980px){.bricklayer-column-sizer{width:33.333%}}@media screen and (min-width:1200px){.bricklayer-column-sizer{width:25%}}.bricklayer-column{-webkit-box-flex:1;-webkit-flex:1;-ms-flex:1;flex:1;padding-left:5px;padding-right:5px}

src/static/css/dark.css

@@ -6,9 +6,10 @@
     width: 1rem;
     margin-right: 0.2rem;
     cursor: pointer;
-    color: #757575;
     line-height: 1rem;
-    height: 1.1rem;
+    height: 1rem;
+    background-image: url(data:image/svg+xml;base64,PHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHhtbG5zOnhsaW5rPSJodHRwOi8vd3d3LnczLm9yZy8xOTk5L3hsaW5rIiB4PSIwcHgiIHk9IjBweCIKICAgICB2aWV3Qm94PSIwIDAgNDI2LjY2NyA0MjYuNjY3IiBzdHlsZT0iZW5hYmxlLWJhY2tncm91bmQ6bmV3IDAgMCA0MjYuNjY3IDQyNi42Njc7IiBmaWxsPSIjZmZmIj4KPGc+CiAgICA8Zz4KICAgICAgICA8Zz4KICAgICAgICAgICAgPHJlY3QgeD0iMTkyIiB5PSIxOTIiIHdpZHRoPSI0Mi42NjciIGhlaWdodD0iMTI4Ii8+CiAgICAgICAgICAgIDxwYXRoIGQ9Ik0yMTMuMzMzLDBDOTUuNDY3LDAsMCw5NS40NjcsMCwyMTMuMzMzczk1LjQ2NywyMTMuMzMzLDIxMy4zMzMsMjEzLjMzM1M0MjYuNjY3LDMzMS4yLDQyNi42NjcsMjEzLjMzMwogICAgICAgICAgICAgICAgUzMzMS4yLDAsMjEzLjMzMywweiBNMjEzLjMzMywzODRjLTk0LjA4LDAtMTcwLjY2Ny03Ni41ODctMTcwLjY2Ny0xNzAuNjY3UzExOS4yNTMsNDIuNjY3LDIxMy4zMzMsNDIuNjY3CiAgICAgICAgICAgICAgICBTMzg0LDExOS4yNTMsMzg0LDIxMy4zMzNTMzA3LjQxMywzODQsMjEzLjMzMywzODR6Ii8+CiAgICAgICAgICAgIDxyZWN0IHg9IjE5MiIgeT0iMTA2LjY2NyIgd2lkdGg9IjQyLjY2NyIgaGVpZ2h0PSI0Mi42NjciLz4KICAgICAgICA8L2c+CiAgICA8L2c+CjwvZz4KPC9zdmc+Cg==);
+    filter: brightness(65%);
 }
 
 .info-icon:hover {
@@ -37,6 +38,7 @@
     color: #BBB;
     text-overflow: ellipsis;
     overflow: hidden;
+    white-space: nowrap;
 }
 
 a {
@@ -119,7 +121,7 @@ body {
     background: #546b7a;
 }
 
-.navbar a:hover {
+a:hover, .btn:hover {
     color: #fff;
 }
 
@@ -128,7 +130,11 @@ body {
 }
 
 .document {
-    padding: 0.5rem;
+    padding: 0.3rem;
+}
+
+.card-text:last-child {
+    margin-top: -1px;
 }
 
 .document p {
@@ -164,6 +170,12 @@ body {
     background-color: #FAAB3C;
 }
 
+.add-tag-button {
+    cursor: pointer;
+    color: #212529;
+    background-color: #e0e0e0;
+}
+
 .card-img-overlay {
     pointer-events: none;
     padding: 0.75rem;
@@ -189,22 +201,42 @@ body {
     margin-right: 3px;
 }
 
+.badge-delete {
+    margin-right: -2px;
+    margin-left: 2px;
+    margin-top: -1px;
+    font-family: monospace;
+    font-size: 90%;
+    background: rgba(0, 0, 0, 0.2);
+    padding: 0.1em 0.4em;
+    color: white;
+    cursor: pointer;
+}
+
 .badge-user {
     color: #212529;
     background-color: #e0e0e0;
 }
 
+.card-img-top {
+    border-top-left-radius: 0;
+    border-top-right-radius: 0;
+}
+
 .fit {
     display: block;
     min-width: 64px;
     max-width: 100%;
-    max-height: 175px;
+    max-height: 400px;
     margin: 0 auto 0;
-    padding: 3px 3px 0;
     width: auto;
     height: auto;
 }
 
+.img-padding {
+    padding: 4px 4px 0 4px;
+}
+
 .fit-sm {
     display: block;
     max-width: 64px;
@@ -221,23 +253,7 @@ body {
     width: 100%;
 }
 
-@media (min-width: 1200px) {
-    .card-columns {
-        column-count: 4;
-    }
-}
-
-@media (min-width: 1500px) {
-    .container {
-        max-width: 1440px;
-    }
-
-    .card-columns {
-        column-count: 5;
-    }
-}
-
-@media (min-width: 1800px) {
+@media screen and (min-width: 1800px) {
     .container {
         max-width: 1550px;
     }
@@ -266,6 +282,7 @@ mark {
     margin: 3px;
     white-space: normal;
     color: rgb(224, 224, 224);
+    overflow: hidden;
 }
 
 .irs-single, .irs-from, .irs-to {
@@ -411,7 +428,7 @@ option {
     max-height: 250px;
 }
 
-@media (max-width: 600px) {
+@media (max-width: 650px) {
     .media-expanded .fit {
         max-height: none;
     }
@@ -432,6 +449,7 @@ option {
     .small-btn {
         display: none;
     }
+
     .large-btn {
         display: inherit;
     }
@@ -441,6 +459,7 @@ option {
     .small-btn {
         display: inherit;
     }
+
     .large-btn {
         display: none;
     }
@@ -457,3 +476,69 @@ option {
 svg {
     fill: white;
 }
+
+.play {
+    position: absolute;
+    width: 50px;
+    height: 50px;
+    left: 50%;
+    top: 50%;
+    transform: translate(-50%, -50%);
+    pointer-events: none;
+}
+
+.play svg {
+    fill: rgba(255, 255, 255, 0.7);
+}
+
+.img-wrapper:hover svg {
+    fill: rgba(255, 255, 255, 1);
+}
+
+.pointer {
+    cursor: pointer;
+}
+
+.stats-card {
+    text-align: center;
+    margin-top: 1em;
+    padding: 1em;
+
+    box-shadow: 0 .125rem .25rem rgba(0, 0, 0, .08) !important;
+    border-radius: 0;
+    border: none;
+
+    background: #212121;
+}
+
+.graph {
+    display: inline-block;
+    width: 40%;
+}
+
+.full-screen {
+    position: absolute;
+    left: 0;
+    width: 100%;
+}
+
+.stats-btn {
+    float: right;
+    margin-bottom: 10px;
+}
+
+#graphs-card svg text {
+    fill: #eee;
+}
+
+.wholerow {
+    outline: none !important;
+}
+
+.stat > .card-body {
+    padding: 0.7em 1.25em;
+}
+
+#modal-body > .img-wrapper {
+    margin-bottom: 1em;
+}

src/static/css/ion.rangeSlider.skinFlat.min.css

@@ -0,0 +1 @@
+.irs-bar,.irs-bar-edge,.irs-line-left,.irs-line-mid,.irs-line-right,.irs-slider{background:url("./img/sprite-skin-flat.png") repeat-x}.irs{height:40px}.irs-with-grid{height:60px}.irs-line{height:12px;top:25px}.irs-line-left{height:12px;background-position:0 -30px}.irs-line-mid{height:12px;background-position:0 0}.irs-line-right{height:12px;background-position:100% -30px}.irs-bar{height:12px;top:25px;background-position:0 -60px}.irs-bar-edge{top:25px;height:12px;width:9px;background-position:0 -90px}.irs-shadow{height:3px;top:34px;background:#000;opacity:0.25}.lt-ie9 .irs-shadow{filter: alpha(opacity=25)}.irs-slider{width:16px;height:18px;top:22px;background-position:0 -120px}.irs-slider.state_hover,.irs-slider:hover{background-position:0 -150px}.irs-max,.irs-min{color:#999;font-size:10px;line-height:1.333;text-shadow:none;top:0;padding:1px 3px;background:#e1e4e9;-moz-border-radius:4px;border-radius:4px}.irs-from,.irs-single,.irs-to{color:#fff;font-size:10px;line-height:1.333;text-shadow:none;padding:1px 5px;background:#2196F3;-moz-border-radius:4px;border-radius:4px}.irs-from:after,.irs-single:after,.irs-to:after{position:absolute;display:block;content:"";bottom:-6px;left:50%;width:0;height:0;margin-left:-3px;overflow:hidden;border:3px solid transparent;border-top-color:#2196F3}.irs-grid-pol{background:#e1e4e9}.irs-grid-text{color:#999}.irs-disabled{}

src/static/css/light.css

@@ -6,9 +6,10 @@
     width: 1rem;
     margin-right: 0.2rem;
     cursor: pointer;
-    color: #757575;
     line-height: 1rem;
     height: 1rem;
+    background-image: url(data:image/svg+xml;base64,PHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHhtbG5zOnhsaW5rPSJodHRwOi8vd3d3LnczLm9yZy8xOTk5L3hsaW5rIiB4PSIwcHgiIHk9IjBweCIKICAgICB2aWV3Qm94PSIwIDAgNDI2LjY2NyA0MjYuNjY3IiBzdHlsZT0iZW5hYmxlLWJhY2tncm91bmQ6bmV3IDAgMCA0MjYuNjY3IDQyNi42Njc7IiBmaWxsPSIjZmZmIj4KPGc+CiAgICA8Zz4KICAgICAgICA8Zz4KICAgICAgICAgICAgPHJlY3QgeD0iMTkyIiB5PSIxOTIiIHdpZHRoPSI0Mi42NjciIGhlaWdodD0iMTI4Ii8+CiAgICAgICAgICAgIDxwYXRoIGQ9Ik0yMTMuMzMzLDBDOTUuNDY3LDAsMCw5NS40NjcsMCwyMTMuMzMzczk1LjQ2NywyMTMuMzMzLDIxMy4zMzMsMjEzLjMzM1M0MjYuNjY3LDMzMS4yLDQyNi42NjcsMjEzLjMzMwogICAgICAgICAgICAgICAgUzMzMS4yLDAsMjEzLjMzMywweiBNMjEzLjMzMywzODRjLTk0LjA4LDAtMTcwLjY2Ny03Ni41ODctMTcwLjY2Ny0xNzAuNjY3UzExOS4yNTMsNDIuNjY3LDIxMy4zMzMsNDIuNjY3CiAgICAgICAgICAgICAgICBTMzg0LDExOS4yNTMsMzg0LDIxMy4zMzNTMzA3LjQxMywzODQsMjEzLjMzMywzODR6Ii8+CiAgICAgICAgICAgIDxyZWN0IHg9IjE5MiIgeT0iMTA2LjY2NyIgd2lkdGg9IjQyLjY2NyIgaGVpZ2h0PSI0Mi42NjciLz4KICAgICAgICA8L2c+CiAgICA8L2c+CjwvZz4KPC9zdmc+Cg==);
+    filter: brightness(45%);
 }
 
 .info-icon:hover {
@@ -36,6 +37,7 @@
     color: #444;
     text-overflow: ellipsis;
     overflow: hidden;
+    white-space: nowrap;
 }
 
 body {
@@ -48,7 +50,9 @@ body {
 
 .card {
     margin-top: 1em;
-    box-shadow: 0 .125rem .25rem rgba(0, 0, 0, .075) !important;
+    box-shadow: 0 .125rem .25rem rgba(0, 0, 0, .08) !important;
+    border-radius: 0;
+    border: none;
 }
 
 .sub-document {
@@ -66,7 +70,11 @@ body {
 }
 
 .document {
-    padding: 0.5rem;
+   padding: 0.3rem;
+}
+
+.card-text:last-child {
+    margin-top: -1px;
 }
 
 .document p {
@@ -102,11 +110,33 @@ body {
     background-color: #e0e0e0;
 }
 
+.badge {
+    margin-right: 3px;
+}
+
+.badge-delete {
+    margin-right: -2px;
+    margin-left: 2px;
+    margin-top: -1px;
+    font-family: monospace;
+    font-size: 90%;
+    background: rgba(0,0,0,0.2);
+    padding: 0.1em 0.4em;
+    color: white;
+    cursor: pointer;
+}
+
 .badge-text {
     color: #FFFFFF;
     background-color: #FAAB3C;
 }
 
+.add-tag-button {
+    cursor: pointer;
+    color: #212529;
+    background-color: #e0e0e0;
+}
+
 .card-img-overlay {
     pointer-events: none;
     padding: 0.75rem;
@@ -127,21 +157,25 @@ body {
     overflow: hidden;
 }
 
-.badge {
-    margin-right: 3px;
+.card-img-top {
+    border-top-left-radius: 0;
+    border-top-right-radius: 0;
 }
 
 .fit {
     display: block;
     min-width: 64px;
     max-width: 100%;
-    max-height: 175px;
+    max-height: 400px;
     margin: 0 auto 0;
-    padding: 3px 3px 0 3px;
     width: auto;
     height: auto;
 }
 
+.img-padding {
+    padding: 4px 4px 0 4px;
+}
+
 .fit-sm {
     display: block;
     max-width: 64px;
@@ -158,23 +192,31 @@ body {
     width: 100%;
 }
 
-@media (min-width: 1200px) {
-    .card-columns {
-        column-count: 4;
+.bricklayer {
+    /*max-width: 100%;*/
+}
+
+@media screen and (max-width: 1200px) {
+    .bricklayer-column {
+        max-width: 100%;
     }
 }
 
-@media (min-width: 1500px) {
+@media screen and (min-width: 1500px) {
     .container {
         max-width: 1440px;
     }
 
-    .card-columns {
-        column-count: 5;
+    .bricklayer-column-sizer {
+        width: 20% !important;
+    }
+
+    .bricklayer-column {
+        max-width: 20%;
     }
 }
 
-@media (min-width: 1800px) {
+@media screen and (min-width: 1800px) {
     .container {
         max-width: 1550px;
     }
@@ -197,6 +239,7 @@ mark {
     margin: 3px;
     white-space: normal;
     color: #000;
+    overflow: hidden;
 }
 
 .irs-single, .irs-from, .irs-to {
@@ -272,7 +315,7 @@ mark {
     max-height: 250px;
 }
 
-@media (max-width: 600px) {
+@media (max-width: 650px) {
     .media-expanded .fit {
         max-height: none;
     }
@@ -293,6 +336,7 @@ mark {
     .small-btn {
         display: none;
     }
+
     .large-btn {
         display: inherit;
     }
@@ -302,6 +346,7 @@ mark {
     .small-btn {
         display: inherit;
     }
+
     .large-btn {
         display: none;
     }
@@ -314,3 +359,65 @@ mark {
 #pathTree .title {
     cursor: pointer;
 }
+
+.play {
+    position: absolute;
+    width: 50px;
+    height: 50px;
+    left: 50%;
+    top: 50%;
+    transform: translate(-50%, -50%);
+    pointer-events: none;
+}
+
+.play svg {
+    fill: rgba(0, 0, 0, 0.7);
+}
+
+.img-wrapper:hover svg {
+    fill: rgba(0, 0, 0, 1);
+}
+
+.pointer {
+    cursor: pointer;
+}
+
+.stats-card {
+    text-align: center;
+    margin-top: 1em;
+    padding: 1em;
+
+    box-shadow: 0 .125rem .25rem rgba(0, 0, 0, .08) !important;
+    border-radius: 0;
+    border: none;
+
+    background: #fff;
+}
+
+.graph {
+    display: inline-block;
+    width: 40%;
+}
+
+.full-screen {
+    position: absolute;
+    left: 0;
+    width: 100%;
+}
+
+.stats-btn {
+    float: right;
+    margin-bottom: 10px;
+}
+
+.wholerow {
+    outline: none !important;
+}
+
+.stat > .card-body {
+    padding: 0.7em 1.25em;
+}
+
+#modal-body > .img-wrapper {
+    margin-bottom: 1em;
+}

src/static/css/lity.min.css

@@ -0,0 +1,3 @@
+/*! Lity - v2.4.0 - 2019-08-10
+* http://sorgalla.com/lity/
+* Copyright (c) 2015-2019 Jan Sorgalla; Licensed MIT */.lity{z-index:9990;position:fixed;top:0;right:0;bottom:0;left:0;white-space:nowrap;background:#0b0b0b;background:rgba(0,0,0,0.9);outline:none !important;opacity:0;-webkit-transition:opacity .3s ease;-o-transition:opacity .3s ease;transition:opacity .3s ease}.lity.lity-opened{opacity:1}.lity.lity-closed{opacity:0}.lity *{-webkit-box-sizing:border-box;-moz-box-sizing:border-box;box-sizing:border-box}.lity-wrap{z-index:9990;position:fixed;top:0;right:0;bottom:0;left:0;text-align:center;outline:none !important}.lity-wrap:before{content:'';display:inline-block;height:100%;vertical-align:middle;margin-right:-0.25em}.lity-loader{z-index:9991;color:#fff;position:absolute;top:50%;margin-top:-0.8em;width:100%;text-align:center;font-size:14px;font-family:Arial,Helvetica,sans-serif;opacity:0;-webkit-transition:opacity .3s ease;-o-transition:opacity .3s ease;transition:opacity .3s ease}.lity-loading .lity-loader{opacity:1}.lity-container{z-index:9992;position:relative;text-align:left;vertical-align:middle;display:inline-block;white-space:normal;max-width:100%;max-height:100%;outline:none !important}.lity-content{z-index:9993;width:100%;-webkit-transform:scale(1);-ms-transform:scale(1);-o-transform:scale(1);transform:scale(1);-webkit-transition:-webkit-transform .3s ease;transition:-webkit-transform .3s ease;-o-transition:-o-transform .3s ease;transition:transform .3s ease;transition:transform .3s ease, -webkit-transform .3s ease, -o-transform .3s ease}.lity-loading .lity-content,.lity-closed .lity-content{-webkit-transform:scale(.8);-ms-transform:scale(.8);-o-transform:scale(.8);transform:scale(.8)}.lity-content:after{content:'';position:absolute;left:0;top:0;bottom:0;display:block;right:0;width:auto;height:auto;z-index:-1;-webkit-box-shadow:0 0 8px rgba(0,0,0,0.6);box-shadow:0 0 8px rgba(0,0,0,0.6)}.lity-close{z-index:9994;width:35px;height:35px;position:fixed;right:0;top:0;-webkit-appearance:none;cursor:pointer;text-decoration:none;text-align:center;padding:0;color:#fff;font-style:normal;font-size:35px;font-family:Arial,Baskerville,monospace;line-height:35px;text-shadow:0 1px 2px rgba(0,0,0,0.6);border:0;background:none;outline:none;-webkit-box-shadow:none;box-shadow:none}.lity-close::-moz-focus-inner{border:0;padding:0}.lity-close:hover,.lity-close:focus,.lity-close:active,.lity-close:visited{text-decoration:none;text-align:center;padding:0;color:#fff;font-style:normal;font-size:35px;font-family:Arial,Baskerville,monospace;line-height:35px;text-shadow:0 1px 2px rgba(0,0,0,0.6);border:0;background:none;outline:none;-webkit-box-shadow:none;box-shadow:none}.lity-close:active{top:1px}.lity-image img{max-width:100%;display:block;line-height:0;border:0}.lity-iframe .lity-container,.lity-youtube .lity-container,.lity-vimeo .lity-container,.lity-facebookvideo .lity-container,.lity-googlemaps .lity-container{width:100%;max-width:964px}.lity-iframe-container{width:100%;height:0;padding-top:56.25%;overflow:hidden;pointer-events:all;-webkit-transform:translateZ(0);transform:translateZ(0);-webkit-overflow-scrolling:touch}.lity-iframe-container iframe{position:absolute;display:block;top:0;left:0;width:100%;height:100%;-webkit-box-shadow:0 0 8px rgba(0,0,0,0.6);box-shadow:0 0 8px rgba(0,0,0,0.6);background:#000}.lity-hide{display:none}