sist2/scripts/mime.py

import zlib

mimes = {}
noparse = set()
ext_in_hash = set()
mime_ids = {}

major_mime = {
    "sist2": 0,
    "model": 1,
    "example": 2,
    "message": 3,
    "multipart": 4,
    "font": 5,
    "video": 6,
    "audio": 7,
    "image": 8,
    "text": 9,
    "application": 10,
    "x-epoc": 11,
}

pdf = (
    "application/pdf",
    "application/epub+zip",
    "application/vnd.ms-xpsdocument",
)

font = (
    "application/vnd.ms-opentype",
    "application/x-ms-compress-szdd"
    "application/x-font-sfn",
    "application/x-font-ttf",
    "font/otf",
    "font/sfnt",
    "font/woff",
    "font/woff2"
)

# Archive "formats"
archive = (
    "application/x-tar",
    "application/zip",
    "application/x-rar",
    "application/x-arc",
    "application/x-warc",
    "application/x-7z-compressed",
)

# Archive "filters"
arc_filter = (
    "application/gzip",
    "application/x-bzip2",
    "application/x-xz",
    "application/x-zstd",
    "application/x-lzma",
    "application/x-lz4",
    "application/x-lzip",
    "application/x-lzop",
)

doc = (
    "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
    "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
    "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


def mime_id(mime):
    if mime in mime_ids:
        return mime_ids[mime]

    global cnt
    major = mime.split("/")[0]
    mime_id = str((major_mime[major] << 16) + cnt)
    cnt += 1
    if mime in noparse:
        mime_id += " | 0x80000000"
    elif mime in pdf:
        mime_id += " | 0x40000000"
    elif mime in font:
        mime_id += " | 0x20000000"
    elif mime in archive:
        mime_id += " | 0x10000000"
    elif mime in arc_filter:
        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"
    mime_ids[mime] = mime_id
    return mime_id


def clean(t):
    return t.replace("/", "_").replace(".", "_").replace("+", "_").replace("-", "_")


def crc(s):
    return zlib.crc32(s.encode()) & 0xffffffff


with open("scripts/mime.csv") as f:
    for l in f:
        mime, ext_list = l.split(",")
        if l.startswith("!"):
            mime = mime[1:]
            noparse.add(mime)
        ext = [x.strip() for x in ext_list.split("|") if x.strip() != ""]
        mimes[mime] = ext

    seen_crc = set()
    for ext in mimes.values():
        for e in ext:
            if crc(e) in seen_crc:
                raise Exception("CRC32 collision")
            seen_crc.add(crc(e))

    seen_crc = set()
    for mime in mimes.keys():
        if crc(mime) in seen_crc:
            raise Exception("CRC32 collision")
        seen_crc.add(crc(mime))

    print("// **Generated by mime.py**")
    print("#ifndef MIME_GENERATED_C")
    print("#define MIME_GENERATED_C")
    print("#include <stdlib.h>\n")
    # Enum
    print("enum mime {")
    for mime, ext in sorted(mimes.items()):
        print(f"{clean(mime)}={mime_id(mime)},")
    print("};")

    # Enum -> string
    print("char *mime_get_mime_text(unsigned int mime_id) {"
          "switch (mime_id) {")
    for mime, ext in mimes.items():
        print("case " + clean(mime) + ": return \"" + mime + "\";")
    print("default: return NULL;}}")

    # Ext -> Enum
    print("unsigned int mime_extension_lookup(unsigned long extension_crc32) {"
          "switch (extension_crc32) {")
    for mime, ext in mimes.items():
        if len(ext) > 0:
            for e in ext:
                print(f"case {crc(e)}:", end="")
            print(f"return {clean(mime)};")
    print("default: return 0;}}")

    # string -> Enum
    print("unsigned int mime_name_lookup(unsigned long mime_crc32) {"
          "switch (mime_crc32) {")
    for mime in mimes.keys():
        print(f"case {crc(mime)}: return {clean(mime)};")

    print("default: return 0;}}")

    # mime list

    mime_list = ",".join(mime_id(x) for x in mimes.keys()) + ",0"

    print(f"unsigned int mime_ids[] = {{{mime_list}}};")
    print("unsigned int* get_mime_ids() { return mime_ids; }")

    print("#endif")