add exif dateTime, allow some special characters in text meta

scripts/get_static_libs.sh

@@ -75,7 +75,7 @@ cd tesseract
 mkdir build
 cd build
 cmake -DSTATIC=on -DBUILD_TRAINING_TOOLS=off -DBUILD_TESTS=off -DCMAKE_BUILD_TYPE=Release \
-  -DCMAKE_CXX_FLAGS="-fPIC"  ..
+  -DCMAKE_CXX_FLAGS="-fPIC" -DAUTO_OPTIMIZE=off ..
 make -j $THREADS
 cd ../..
 mv tesseract/build/libtesseract.a .

src/io/serialize.c

@@ -134,6 +134,8 @@ char *get_meta_key_text(enum metakey meta_key) {
             return "exif_iso_speed_ratings";
         case MetaExifModel:
             return "exif_model";
+        case MetaExifDateTime:
+            return "exif_datetime";
         default:
             return NULL;
     }
@@ -278,6 +280,7 @@ void read_index_bin(const char *path, const char *index_id, index_func func) {
                 case MetaExifFocalLength:
                 case MetaExifUserComment:
                 case MetaExifIsoSpeedRatings:
+                case MetaExifDateTime:
                 case MetaExifModel:
                 case MetaTitle: {
                     buf.cur = 0;

src/main.c

@@ -6,7 +6,7 @@
 #define EPILOG "Made by simon987 <me@simon987.net>. Released under GPL-3.0"
 
 
-static const char *const Version = "1.2.7";
+static const char *const Version = "1.2.9";
 static const char *const usage[] = {
         "sist2 scan [OPTION]... PATH",
         "sist2 index [OPTION]... INDEX",
@@ -49,6 +49,7 @@ void sist2_scan(scan_args_t *args) {
     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;

src/parsing/media.c

@@ -209,6 +209,8 @@ append_video_meta(AVFormatContext *pFormatCtx, AVFrame *frame, document_t *doc,
                 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)
             }
         }
     }

src/types.h

@@ -39,6 +39,7 @@ enum metakey {
     MetaExifUserComment = 20 | META_STR_MASK,
     MetaExifModel = 21 | META_STR_MASK,
     MetaExifIsoSpeedRatings = 22 | META_STR_MASK,
+    MetaExifDateTime = 23 | META_STR_MASK,
     //Note to self: this will break after 31 entries
 };
 

src/util.h

@@ -7,7 +7,7 @@
 #define INITIAL_BUF_SIZE 1024 * 16
 
 #define SHOULD_IGNORE_CHAR(c) !(SHOULD_KEEP_CHAR(c))
-#define SHOULD_KEEP_CHAR(c) ((c >= '0' && c <= '9') || (c >= 'A' && c <= 'z') || (c > 127))
+#define SHOULD_KEEP_CHAR(c) ((c >= '\'' && c <= ';') || (c >= 'A' && c <= 'z') || (c > 127))
 
 
 typedef struct dyn_buffer {

web/js/dom.js

@@ -499,8 +499,7 @@ function makePreloader() {
 function makePageIndicator(searchResult) {
     let pageIndicator = document.createElement("div");
     pageIndicator.setAttribute("class", "page-indicator font-weight-light");
-    const totalHits = searchResult["hits"]["total"].hasOwnProperty("value")
-        ? searchResult["hits"]["total"]["value"] : searchResult["hits"]["total"];
+    const totalHits = searchResult["aggregations"]["total_count"]["value"];
     pageIndicator.appendChild(document.createTextNode(docCount + " / " + totalHits));
     return pageIndicator;
 }
@@ -547,8 +546,7 @@ function makeStatsCard(searchResult) {
     });
 
     let stat = document.createElement("span");
-    const totalHits = searchResult["hits"]["total"].hasOwnProperty("value")
-        ? searchResult["hits"]["total"]["value"] : searchResult["hits"]["total"];
+    const totalHits = searchResult["aggregations"]["total_count"]["value"];
     stat.appendChild(document.createTextNode(totalHits + " results in " + searchResult["took"] + "ms"));
 
     statsCardBody.appendChild(stat);

web/js/search.js

@@ -363,7 +363,7 @@ function search(after = null) {
         },
         "sort": [
             {"_score": {"order": "desc"}},
-            {"_tie": {"order":"asc"}}
+            {"_tie": {"order": "asc"}}
         ],
         highlight: {
             pre_tags: ["<mark>"],
@@ -376,11 +376,13 @@ function search(after = null) {
                 font_name: {},
             }
         },
-        aggs: {
-            total_size: {"sum": {"field": "size"}}
-        },
+        aggs:
+            {
+                total_size: {"sum": {"field": "size"}},
+                total_count: {"value_count": {"field": "size"}}
+            },
         size: SIZE,
-    }
+    };
 
     if (after) {
         q.search_after = [after["_score"], after["_id"]];

web/search.html

@@ -11,7 +11,7 @@
 
 <nav class="navbar navbar-expand-lg">
     <a class="navbar-brand" href="/">sist2</a>
-    <span class="badge badge-pill version">v1.2.7</span>
+    <span class="badge badge-pill version">v1.2.9</span>
     <span class="tagline">Lightning-fast file system indexer and search tool </span>
     <a style="margin-left: auto" id="theme" class="btn" title="Toggle theme" href="/">Theme</a>
 </nav>