js-cfb/bits/40_parse.js
SheetJS 1a428c8f5b version bump 0.10.3: ignore bad FAT
- sector checked before reading (h/t @e12009)
  See https://github.com/SheetJS/js-xlsx/issues/569
- flow type checking
2017-02-23 21:11:45 -08:00

102 lines
2.8 KiB
JavaScript

function parse(file) {
var mver = 3; // major version
var ssz = 512; // sector size
var nmfs = 0; // number of mini FAT sectors
var ndfs = 0; // number of DIFAT sectors
var dir_start = 0; // first directory sector location
var minifat_start = 0; // first mini FAT sector location
var difat_start = 0; // first mini FAT sector location
var fat_addrs = []; // locations of FAT sectors
/* [MS-CFB] 2.2 Compound File Header */
var blob/*:any*/ = file.slice(0,512);
prep_blob(blob, 0);
/* major version */
var mv = check_get_mver(blob);
mver = mv[0];
switch(mver) {
case 3: ssz = 512; break; case 4: ssz = 4096; break;
default: throw "Major Version: Expected 3 or 4 saw " + mver;
}
/* reprocess header */
if(ssz !== 512) { blob = file.slice(0,ssz); prep_blob(blob, 28 /* blob.l */); }
/* Save header for final object */
var header = file.slice(0,ssz);
check_shifts(blob, mver);
// Number of Directory Sectors
var nds = blob.read_shift(4, 'i');
if(mver === 3 && nds !== 0) throw '# Directory Sectors: Expected 0 saw ' + nds;
// Number of FAT Sectors
//var nfs = blob.read_shift(4, 'i');
blob.l += 4;
// First Directory Sector Location
dir_start = blob.read_shift(4, 'i');
// Transaction Signature
blob.l += 4;
// Mini Stream Cutoff Size
blob.chk('00100000', 'Mini Stream Cutoff Size: ');
// First Mini FAT Sector Location
minifat_start = blob.read_shift(4, 'i');
// Number of Mini FAT Sectors
nmfs = blob.read_shift(4, 'i');
// First DIFAT sector location
difat_start = blob.read_shift(4, 'i');
// Number of DIFAT Sectors
ndfs = blob.read_shift(4, 'i');
// Grab FAT Sector Locations
for(var q, j = 0; j < 109; ++j) { /* 109 = (512 - blob.l)>>>2; */
q = blob.read_shift(4, 'i');
if(q<0) break;
fat_addrs[j] = q;
}
/** Break the file up into sectors */
var sectors = sectorify(file, ssz);
sleuth_fat(difat_start, ndfs, sectors, ssz, fat_addrs);
/** Chains */
var sector_list/*:SectorList*/ = make_sector_list(sectors, dir_start, fat_addrs, ssz);
sector_list[dir_start].name = "!Directory";
if(nmfs > 0 && minifat_start !== ENDOFCHAIN) sector_list[minifat_start].name = "!MiniFAT";
sector_list[fat_addrs[0]].name = "!FAT";
sector_list.fat_addrs = fat_addrs;
sector_list.ssz = ssz;
/* [MS-CFB] 2.6.1 Compound File Directory Entry */
var files = {}, Paths/*:any*/ = [], FileIndex = [], FullPaths = [], FullPathDir = {};
read_directory(dir_start, sector_list, sectors, Paths, nmfs, files, FileIndex);
build_full_paths(FileIndex, FullPathDir, FullPaths, Paths);
var root_name = Paths.shift();
Paths.root = root_name;
/* [MS-CFB] 2.6.4 (Unicode 3.0.1 case conversion) */
var find_path = make_find_path(FullPaths, Paths, FileIndex, files, root_name);
return {
raw: {header: header, sectors: sectors},
FileIndex: FileIndex,
FullPaths: FullPaths,
FullPathDir: FullPathDir,
find: find_path
};
} // parse