hashirama/goldendict-ng
hashirama/goldendict-ng
0
0
Fork 0
mirror of https://github.com/xiaoyifang/goldendict-ng.git synced 2024-11-24 00:14:06 +00:00
Code Issues Actions 1 Projects Releases Wiki Activity

Support 64-bit index in AARD dictionaries

Browse source
This commit is contained in:
Abs62 2012-09-17 19:47:45 +04:00
parent bc7a8b39c8
commit 3abb74ad51
1 changed files with 44 additions and 8 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

52
aard.cc
View file

@ -97,6 +97,7 @@ __attribute__((packed))
typedef BigEndian< uint16_t > uint16_be;
typedef BigEndian< uint32_t > uint32_be;
typedef BigEndian< uint64_t > uint64_be;
/// AAR file header
struct AAR_header
@ -129,6 +130,16 @@ __attribute__((packed))
#endif
;
struct IndexElement64
{
uint32_be wordOffset;
uint64_be articleOffset;
}
#ifndef _MSC_VER
__attribute__((packed))
#endif
;
enum
{
Signature = 0x58524141, // AARX on little-endian, XRAA on big-endian
@ -795,13 +806,23 @@ vector< sptr< Dictionary::Class > > makeDictionaries(
{
try
{
{
QFileInfo info( FsEncoding::decode( i->c_str() ) );
if( info.size() > ULONG_MAX )
{
DPRINTF( "File %s is too large", i->c_str() );
continue;
}
}
File::Class df( *i, "rb" );
AAR_header dictHeader;
df.read( &dictHeader, sizeof(dictHeader) );
bool has64bitIndex = !strncmp( dictHeader.indexItemFormat, ">LQ", 4 );
if( strncmp( dictHeader.signature, "aard", 4 )
|| strncmp( dictHeader.indexItemFormat, ">LL", 4 )
|| ( !has64bitIndex && strncmp( dictHeader.indexItemFormat, ">LL", 4 ) )
|| strncmp( dictHeader.keyLengthFormat, ">H", 2 )
|| strncmp( dictHeader.articleLengthFormat, ">L", 2) )
{
@ -877,17 +898,32 @@ vector< sptr< Dictionary::Class > > makeDictionaries(
uint32_t wordCount = dictHeader.wordsCount;
set< uint32_t > articleOffsets;
uint32_t pos = df.tell();
uint32_t wordsBase = pos + wordCount * sizeof( IndexElement );
uint32_t wordsBase = pos + wordCount * ( has64bitIndex ? sizeof( IndexElement64 ) : sizeof( IndexElement ) );
uint32_t articlesBase = dictHeader.articleOffset;
for( uint32_t j = 0; j < wordCount; j++ )
{
IndexElement el;
uint32_t articleOffset;
uint32_t wordOffset;
df.seek( pos );
df.read( &el, sizeof(el) );
uint32_t articleOffset = articlesBase + el.articleOffset;
uint32_t wordOffset = wordsBase + el.wordOffset;
if( has64bitIndex )
{
IndexElement64 el64;
df.seek( pos );
df.read( &el64, sizeof(el64) );
articleOffset = articlesBase + el64.articleOffset;
wordOffset = wordsBase + el64.wordOffset;
}
else
{
IndexElement el;
df.seek( pos );
df.read( &el, sizeof(el) );
articleOffset = articlesBase + el.articleOffset;
wordOffset = wordsBase + el.wordOffset;
}
df.seek( wordOffset );
@ -903,7 +939,7 @@ vector< sptr< Dictionary::Class > > makeDictionaries(
// Insert new entry
indexedWords.addWord( Utf8::decode( string( data.data(), wordSize ) ), articleOffset);
pos += sizeof(el);
pos += has64bitIndex ? sizeof( IndexElement64 ) : sizeof( IndexElement );
}
// Finish with the chunks