2008-04-11 01:21:37 -05:00
--- misc/libtextcat-2.2/configure Thu May 22 13:39:55 2003
+++ misc/build/libtextcat-2.2/configure Mon Mar 31 11:29:14 2008
@@ -5391,7 +5391,8 @@
allow_undefined_flag=
no_undefined_flag=
need_lib_prefix=unknown
-need_version=unknown
+#need_version=unknown
+need_version=no
# when you set need_version to no, make sure it does not cause -set_version
# flags to be left without arguments
archive_cmds=
@@ -5785,7 +5786,7 @@
# cross-compilation, but unfortunately the echo tests do not
# yet detect zsh echo's removal of \ escapes. Also zsh mangles
# `"' quotes if we put them in here... so don't!
- archive_cmds='$CC -r -keep_private_externs -nostdlib -o ${lib}-master.o $libobjs && $CC $(test .$module = .yes && echo -bundle || echo -dynamiclib) $allow_undefined_flag -o $lib ${lib}-master.o $deplibs$linker_flags $(test .$module != .yes && echo -install_name $rpath/$soname $verstring)'
+ archive_cmds='$CC -r -keep_private_externs -nostdlib -o ${lib}-master.o $libobjs && $CC $(test .$module = .yes && echo -bundle || echo -dynamiclib) $allow_undefined_flag -o $lib ${lib}-master.o $deplibs$compiler_flags $(test .$module != .yes && echo -install_name $rpath/$soname $verstring)'
# We need to add '_' to the symbols in $export_symbols first
#archive_expsym_cmds="$archive_cmds"' && strip -s $export_symbols'
hardcode_direct=yes
@@ -6280,7 +6281,7 @@
;;
freebsd*)
- objformat=`test -x /usr/bin/objformat && /usr/bin/objformat || echo aout`
+ objformat=`test -x /usr/bin/objformat && /usr/bin/objformat || echo elf`
version_type=freebsd-$objformat
case $version_type in
freebsd-elf*)
--- misc/libtextcat-2.2/src/Makefile.in Thu May 22 13:39:52 2003
+++ misc/build/libtextcat-2.2/src/Makefile.in Mon Mar 31 11:29:14 2008
@@ -124,20 +124,20 @@
target_vendor = @target_vendor@
AUTOMAKE_OPTIONS = 1.4 foreign
-WARNS = -W -Wall -Wshadow -Wpointer-arith
-IFLAGS =
-FLAGS = -g -O3 -funroll-loops -D_THREAD_SAFE -D_GNU_SOURCE
+#WARNS = -W -Wall -Wshadow -Wpointer-arith
+IFLAGS =
+#FLAGS = -g -O3 -funroll-loops -D_THREAD_SAFE -D_GNU_SOURCE
VERBOSE = -DVERBOSE
AM_CFLAGS = $(IFLAGS) $(VERBOSE) $(WARNS) $(FLAGS)
AM_LDFLAGS = -g
noinst_HEADERS = \
- common.h constants.h fingerprint.h textcat.h wg_mempool.h
+ common.h constants.h fingerprint.h textcat.h wg_mempool.h utf8misc.h
lib_LTLIBRARIES = libtextcat.la
libtextcat_la_SOURCES = \
- common.c fingerprint.c textcat.c wg_mempool.c
+ common.c fingerprint.c textcat.c wg_mempool.c utf8misc.c
bin_PROGRAMS = createfp
@@ -156,7 +156,7 @@
libtextcat_la_LDFLAGS =
libtextcat_la_LIBADD =
am_libtextcat_la_OBJECTS = common.lo fingerprint.lo textcat.lo \
- wg_mempool.lo
+ wg_mempool.lo utf8misc.lo
libtextcat_la_OBJECTS = $(am_libtextcat_la_OBJECTS)
bin_PROGRAMS = createfp$(EXEEXT)
noinst_PROGRAMS = testtextcat$(EXEEXT)
@@ -177,7 +177,8 @@
@AMDEP_TRUE@DEP_FILES = ./$(DEPDIR)/common.Plo ./$(DEPDIR)/createfp.Po \
@AMDEP_TRUE@ ./$(DEPDIR)/fingerprint.Plo \
@AMDEP_TRUE@ ./$(DEPDIR)/testtextcat.Po ./$(DEPDIR)/textcat.Plo \
-@AMDEP_TRUE@ ./$(DEPDIR)/wg_mempool.Plo
+@AMDEP_TRUE@ ./$(DEPDIR)/wg_mempool.Plo \
+@AMDEP_TRUE@ ./$(DEPDIR)/utf8misc.Plo
COMPILE = $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) \
$(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS)
LTCOMPILE = $(LIBTOOL) --mode=compile $(CC) $(DEFS) $(DEFAULT_INCLUDES) \
@@ -213,7 +214,7 @@
@rm -f stamp-h1
cd $(top_builddir) && $(SHELL) ./config.status src/config.h
-$(srcdir)/config.h.in: $(top_srcdir)/configure.ac $(ACLOCAL_M4)
+$(srcdir)/config.h.in: $(top_srcdir)/configure.ac $(ACLOCAL_M4)
cd $(top_srcdir) && $(AUTOHEADER)
touch $(srcdir)/config.h.in
@@ -247,8 +248,8 @@
echo "rm -f \"$${dir}/so_locations\""; \
rm -f "$${dir}/so_locations"; \
done
-libtextcat.la: $(libtextcat_la_OBJECTS) $(libtextcat_la_DEPENDENCIES)
- $(LINK) -rpath $(libdir) $(libtextcat_la_LDFLAGS) $(libtextcat_la_OBJECTS) $(libtextcat_la_LIBADD) $(LIBS)
+libtextcat.la: $(libtextcat_la_OBJECTS) $(libtextcat_la_DEPENDENCIES)
+ $(LINK) -avoid-version -rpath $(libdir) $(libtextcat_la_LDFLAGS) $(libtextcat_la_OBJECTS) $(libtextcat_la_LIBADD) $(LIBS)
binPROGRAMS_INSTALL = $(INSTALL_PROGRAM)
install-binPROGRAMS: $(bin_PROGRAMS)
@$(NORMAL_INSTALL)
@@ -285,10 +286,10 @@
echo " rm -f $$p $$f"; \
rm -f $$p $$f ; \
done
-createfp$(EXEEXT): $(createfp_OBJECTS) $(createfp_DEPENDENCIES)
+createfp$(EXEEXT): $(createfp_OBJECTS) $(createfp_DEPENDENCIES)
@rm -f createfp$(EXEEXT)
$(LINK) $(createfp_LDFLAGS) $(createfp_OBJECTS) $(createfp_LDADD) $(LIBS)
-testtextcat$(EXEEXT): $(testtextcat_OBJECTS) $(testtextcat_DEPENDENCIES)
+testtextcat$(EXEEXT): $(testtextcat_OBJECTS) $(testtextcat_DEPENDENCIES)
@rm -f testtextcat$(EXEEXT)
$(LINK) $(testtextcat_LDFLAGS) $(testtextcat_OBJECTS) $(testtextcat_LDADD) $(LIBS)
@@ -304,6 +305,7 @@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/testtextcat.Po@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/textcat.Plo@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/wg_mempool.Plo@am__quote@
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/utf8misc.Plo@am__quote@
distclean-depend:
-rm -rf ./$(DEPDIR)
--- misc/libtextcat-2.2/src/common.c Thu May 22 13:32:43 2003
+++ misc/build/libtextcat-2.2/src/common.c Mon Mar 31 11:29:14 2008
@@ -3,23 +3,23 @@
*
* Copyright (c) 2003, WiseGuys Internet B.V.
* 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.
- *
+ *
* - Neither the name of the WiseGuys Internet B.V. nor the names of
* its contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
- *
+ *
* 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
@@ -114,11 +114,11 @@
wgmem_error( "Error while strduping %u bytes.\n", strlen(s) );
}
- return( result );
+ return( result );
}
-extern void* wg_realloc( void *ptr, size_t size )
-{
+extern void* wg_realloc( void *ptr, size_t size )
+{
void *result;
if (!size) {
@@ -131,7 +131,7 @@
wgmem_error( "Error while reallocing %u bytes.\n", size );
}
- return( result );
+ return( result );
}
extern void wg_free( void *mem )
@@ -148,12 +148,12 @@
if ( fgets(line, size, fp) == NULL ) {
return NULL;
}
-
+
/** kill term null **/
if ( (p = strpbrk( line, "\n\r" )) ) {
*p = '\0';
- }
-
+ }
+
return line;
}
@@ -164,39 +164,39 @@
*
* ARGUMENTS:
* - result:
- *
+ *
* After the split, this array contains pointers to the start of each
* detected segment. Must be preallocated and at least as large as
* maxsegments. The pointers point into the dest buffer.
- *
- * - dest:
- *
+ *
+ * - dest:
+ *
* String into which result points as an index. Must be preallocated, and
* at least as big as src. You can use src as dest, but in that case src
* is overwritten!
- *
- * - src:
- *
+ *
+ * - src:
+ *
* The string to split. Sequences of whitespace are treated as separators, unless
* escaped. There are two ways to escape: by using single quotes (anything
* between single quotes is treated as one segment), or by using a backslash
* to escape the next character. The backslash escape works inside quotation
* as well.
- *
+ *
* Example:
- *
+ *
* "It\'s very\ easy 'to use WiseGuys\' wg_split()' function" is split into:
- *
+ *
* "It's"
* "very easy"
* "to use WiseGuys' wg_split()"
* "function"
- *
- * - maxsegments:
- *
+ *
+ * - maxsegments:
+ *
* The maximum number of segments. If the splitter runs out of segments,
* the remainder of the string is stored in the last segment.
- *
+ *
* RETURN VALUE:
* The number of segments found.
*/
@@ -218,12 +218,12 @@
switch (state) {
case 0:
/*** Skip spaces ***/
- while ( isspace((int) *p) ) {
+ while ( isspace((unsigned char) *p) ) {
p++;
}
state = 1;
- case 1:
+ case 1:
/*** Start segment ***/
result[cnt] = w;
cnt++;
@@ -232,12 +232,12 @@
case 2:
/*** Unquoted segment ***/
while (*p) {
- if ( isspace((int) *p) ) {
+ if ( isspace((unsigned char) *p) ) {
*w++ = '\0';
p++;
state = 0;
break;
- }
+ }
else if ( *p == '\'' ) {
/*** Start quotation ***/
p++;
@@ -292,17 +292,17 @@
}
+#ifdef HAVE_GETTIMEOFDAY /* TL : no struct timeval under Win32 */
extern void wg_timerstart(wgtimer_t *t)
{
-#ifdef HAVE_GETTIMEOFDAY
gettimeofday( &(t->start), NULL );
-#endif
}
+#endif /* TL : no struct timeval under Win32 */
+#ifdef HAVE_GETTIMEOFDAY /* TL : no struct timeval under Win32 */
extern uint4 wg_timerstop(wgtimer_t *t)
{
-#ifdef HAVE_GETTIMEOFDAY
uint4 result;
gettimeofday( &(t->stop), NULL );
result = (t->stop.tv_sec - t->start.tv_sec) * 1000000 +
@@ -312,25 +312,23 @@
t->start.tv_usec = t->stop.tv_usec;
return result;
-#else
- return 0;
-#endif
}
+#endif /* TL : no struct timeval under Win32 */
/**
* wg_strgmov -- a guarded strcpy() variation
- *
+ *
* copies src to dest (including terminating zero), and returns
* pointer to position of terminating zero in dest. The function is
* guaranteed not to write past destlimit. If the copy couldn't be
- * finished, the function returns NULL after restoring the first
- * character in dest for your convenience (since this is usually a zero).
+ * finished, the function returns NULL after restoring the first
+ * character in dest for your convenience (since this is usually a zero).
*/
char *wg_strgmov( char *dest, const char *src, const char *destlimit )
{
char tmp, *w;
-
+
if ( !dest || dest >= destlimit ) {
return NULL;
}
@@ -355,7 +353,7 @@
}
/*
- * wg_trim() -- remove whitespace surrounding a string.
+ * wg_trim() -- remove whitespace surrounding a string.
*
* Example: " bla bla bla " becomes "bla bla bla" after trimming.
*
@@ -373,12 +371,12 @@
char *lastnonspace = &dest[-1];
const char *p = src;
char *w = dest;
-
- while ( isspace((int)*p) ) {
+
+ while ( isspace((unsigned char)*p) ) {
p++;
}
while (*p) {
- if ( !isspace((int)*p) ) {
+ if ( !isspace((unsigned char)*p) ) {
lastnonspace = w;
}
*w++ = *p++;
--- misc/libtextcat-2.2/src/common.h Thu May 22 15:02:29 2003
+++ misc/build/libtextcat-2.2/src/common.h Mon Mar 31 11:29:14 2008
@@ -1,28 +1,28 @@
#ifndef _COMMON_H_
#define _COMMON_H_
/**
- * common.h -- a mixed bag of helper functions
+ * common.h -- a mixed bag of helper functions
*
* Copyright (C) 2003 WiseGuys Internet B.V.
*
* THE BSD LICENSE
- *
+ *
* 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.
- *
+ *
* - Neither the name of the WiseGuys Internet B.V. nor the names of
* its contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
- *
+ *
* 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
@@ -86,10 +86,12 @@
typedef char boole;
#endif
+#ifdef HAVE_GETTIMEOFDAY /* TL : no struct timeval under Win32 */
typedef struct wgtimer_s {
struct timeval start;
struct timeval stop;
} wgtimer_t;
+#endif /* TL : no struct timeval under Win32 */
extern void *wg_malloc( size_t size );
@@ -101,13 +103,15 @@
extern char *wg_getline( char *line, int size, FILE *fp );
+#ifdef HAVE_GETTIMEOFDAY /* TL : no struct timeval under Win32 */
extern void wg_timerstart(wgtimer_t *t);
extern uint4 wg_timerstop(wgtimer_t *t);
+#endif /* TL : no struct timeval under Win32 */
extern unsigned int wg_split( char **result, char *dest, char *src, int maxsegments );
extern char *wg_strgmov( char *dest, const char *src, const char *destlimit );
extern char *wg_trim( char *dest, const char *src );
-
+
#endif
--- misc/libtextcat-2.2/src/constants.h Thu May 22 13:32:43 2003
+++ misc/build/libtextcat-2.2/src/constants.h Mon Mar 31 11:29:14 2008
@@ -39,6 +39,8 @@
*/
#include <limits.h>
+#define _UTF8_
+
#define DESCRIPTION "out of place"
/* Reported matches are those fingerprints with a score less than best
@@ -59,14 +61,21 @@
/* Maximum number of n-grams in a fingerprint */
#define MAXNGRAMS 400
-/* Maximum size of an n-gram? */
-#define MAXNGRAMSIZE 5
+/* Maximum number of character of an n-gram? */
+#define MAXNGRAMSYMBOL 5
+
+/* Maximum size of the string representing an n-gram (must be greater than number of symbol) */
+#ifdef _UTF8_
+#define MAXNGRAMSIZE 20
+#else
+#define MAXNGRAMSIZE MAXNGRAMSYMBOL
+#endif
/* Which characters are not acceptable in n-grams? */
-#define INVALID(c) (isspace((int)c) || isdigit((int)c))
+#define INVALID(c) (isspace((unsigned char)c) || isdigit((unsigned char)c))
/* Minimum size (in characters) for accepting a document */
-#define MINDOCSIZE 25
+#define MINDOCSIZE 6
/* Maximum penalty for missing an n-gram in fingerprint */
#define MAXOUTOFPLACE 400
@@ -75,5 +84,8 @@
#define TABLEPOW 13
#define MAXSCORE INT_MAX
+
+/* where the fingerprints files are stored */
+#define DEFAULT_FINGERPRINTS_PATH ""
#endif
--- misc/libtextcat-2.2/src/fingerprint.c Thu May 22 13:32:43 2003
+++ misc/build/libtextcat-2.2/src/fingerprint.c Mon Mar 31 11:29:14 2008
@@ -6,23 +6,23 @@
* All rights reserved.
*
* THE BSD LICENSE
- *
+ *
* 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.
- *
+ *
* - Neither the name of the WiseGuys Internet B.V. nor the names of
* its contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
- *
+ *
* 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
@@ -51,7 +51,7 @@
* The reason why we go through the trouble of doing a partial
* (heap)sort is that a full quicksort behaves horribly on the data:
* most n-grams have a very low count, resulting in a data set in
- * nearly-sorted order. This causes quicksort to behave very badly.
+ * nearly-sorted order. This causes quicksort to behave very badly.
* Heapsort, on the other hand, behaves handsomely: worst case is
* Mlog(N) for M n-grams filtered through a N-sized heap.
*
@@ -63,6 +63,10 @@
* - put table/heap datastructure in a separate file.
*/
+#ifndef _UTF8_
+#define _UTF8_
+#endif
+
#include "config.h"
#include <stdio.h>
#ifdef HAVE_STDLIB_H
@@ -80,10 +84,12 @@
#include "wg_mempool.h"
#include "constants.h"
+#include "utf8misc.h"
#define TABLESIZE (1<<TABLEPOW)
#define TABLEMASK ((TABLESIZE)-1)
+
typedef struct {
sint2 rank;
@@ -96,7 +102,7 @@
const char *name;
ngram_t *fprint;
uint4 size;
-
+
} fp_t;
typedef struct entry_s {
@@ -105,13 +111,13 @@
struct entry_s *next;
} entry_t;
-typedef struct table_s {
+typedef struct table_s {
void *pool;
entry_t **table;
entry_t *heap;
struct table_s *next;
-
+
uint4 heapsize;
uint4 size;
} table_t;
@@ -122,7 +128,7 @@
* fast and furious little hash function
*
* (Note that we could use some kind of rolling checksum, and update it
- * during n-gram construction)
+ * during n-gram construction)
*/
static uint4 simplehash( const char *p, int len )
{
@@ -134,29 +140,14 @@
}
-/* checks if n-gram lex is a prefix of key and of length len */
-inline int issame( char *lex, char *key, int len )
-{
- int i;
- for (i=0; i<len; i++) {
- if ( key[i] != lex[i] ) {
- return 0;
- }
- }
- if ( lex[i] != 0 ) {
- return 0;
- }
- return 1;
-}
-
/* increases frequency of ngram(p,len) */
-static inline int increasefreq( table_t *t, char *p, int len )
-{
- uint4 hash = simplehash( p, len ) & TABLEMASK;
+static int increasefreq( table_t *t, char *p, int len )
+{
+ uint4 hash = simplehash( p, len ) & TABLEMASK;
entry_t *entry = t->table[ hash ];
-
- while ( entry ) {
+
+ while ( entry ) {
if ( issame( entry->str, p, len ) ) {
/*** Found it! ***/
entry->cnt++;
@@ -168,7 +159,7 @@
}
/*** Not found, so create ***/
- entry = wgmempool_alloc( t->pool, sizeof(entry_t) );
+ entry = (entry_t*)(wgmempool_alloc( t->pool, sizeof(entry_t) ));
strcpy( entry->str, p );
entry->cnt = 1;
@@ -181,12 +172,12 @@
#if 0
/* looks up ngram(p,len) */
-static entry_t *findfreq( table_t *t, char *p, int len )
-{
- uint4 hash = simplehash( p, len ) & TABLEMASK;
+static entry_t *findfreq( table_t *t, char *p, int len )
+{
+ uint4 hash = simplehash( p, len ) & TABLEMASK;
entry_t *entry = t->table[ hash ];
-
- while ( entry ) {
+
+ while ( entry ) {
if ( issame( entry->str, p, len ) ) {
return entry;
}
@@ -219,7 +210,7 @@
#define GREATER(x,y) ((x).cnt > (y).cnt)
#define LESS(x,y) ((x).cnt < (y).cnt)
-inline static void siftup( table_t *t, unsigned int child )
+static void siftup( table_t *t, unsigned int child )
{
entry_t *heap = t->heap;
unsigned int parent = (child-1) >> 1;
@@ -241,7 +232,7 @@
}
-inline static void siftdown( table_t *t, unsigned int heapsize, uint4 parent )
+static void siftdown( table_t *t, unsigned int heapsize, uint4 parent )
{
entry_t *heap = t->heap;
unsigned int child = parent*2 + 1;
@@ -273,7 +264,7 @@
if (t->size < t->heapsize) {
memcpy( &(heap[t->size]), item, sizeof(entry_t));
siftup( t, t->size );
- t->size++;
+ t->size++;
return 0;
}
@@ -316,18 +307,18 @@
/*** Fill result heap ***/
for (i=0; i<TABLESIZE; i++) {
- entry_t *p = t->table[i];
+ entry_t *p = t->table[i];
while (p) {
heapinsert(t, p);
p = p->next;
}
- }
+ }
return 1;
}
static table_t *inittable(uint4 maxngrams)
-{
+{
table_t *result = (table_t *)wg_zalloc( sizeof(table_t) );
result->table = (entry_t **)wg_zalloc( sizeof(entry_t*) * TABLESIZE );
result->pool = wgmempool_Init( 10000, 10 );
@@ -347,14 +338,14 @@
wgmempool_Done(t->pool);
wg_free(t->table);
wg_free(t->heap);
- wg_free(t);
+ wg_free(t);
}
extern void *fp_Init(const char *name)
{
fp_t *h = (fp_t *)wg_zalloc( sizeof(fp_t) );
-
+
if ( name ) {
h->name = wg_strdup(name);
}
@@ -458,21 +449,27 @@
return dest;
}
-
+/**
+* this function extract all n-gram from past buffer and put them into the table "t"
+* [modified] by Jocelyn Merand to accept utf-8 multi-character symbols to be used in OpenOffice
+*/
static void createngramtable( table_t *t, const char *buf )
{
char n[MAXNGRAMSIZE+1];
const char *p = buf;
int i;
+ int pointer = 0;
/*** Get all n-grams where 1<=n<=MAXNGRAMSIZE. Allow underscores only at borders. ***/
- for (;;p++) {
+ while(1) {
- const char *q = p;
+ const char *q = &p[pointer]; /*[modified] previously p++ above (for(;;p++)) now, it's pointer wich is increased so we have to get the new pointer on the buffer*/
char *m = n;
/*** First char may be an underscore ***/
- *m++ = *q++;
+ int decay = charcopy(q, m); /*[modified] previously *q++ = *m++*/
+ q = &(p[pointer+decay]); /*[modified] the old copying method do not manage multi-character symbols*/
+ m += decay; /*[modified]*/
*m = '\0';
increasefreq( t, n, 1 );
@@ -482,19 +479,22 @@
}
/*** Let the compiler unroll this ***/
- for ( i=2; i<=MAXNGRAMSIZE; i++) {
+ for ( i=2; i<=MAXNGRAMSYMBOL; i++) {
- *m++ = *q;
+ decay = charcopy(q, m); /*[modified] like above*/
+ m += decay;
*m = '\0';
increasefreq( t, n, i );
if ( *q == '_' ) break;
- q++;
+ q += decay;
if ( *q == '\0' ) {
return;
}
}
+
+ pointer = nextcharstart(p,pointer); /*[modified] p[pointer] must point on the next start of symbol, but whith utf next start is not surely next char*/
}
return;
}
@@ -514,7 +514,7 @@
{
ngram_t *x = (ngram_t *)a;
ngram_t *y = (ngram_t *)b;
-
+
return mystrcmp( x->str, y->str );
}
@@ -522,12 +522,12 @@
{
ngram_t *x = (ngram_t *)a;
ngram_t *y = (ngram_t *)b;
-
+
return x->rank - y->rank;
}
/**
- * Create a fingerprint:
+ * Create a fingerprint:
* - record the frequency of each unique n-gram in a hash table
* - take the most frequent n-grams
* - sort them alphabetically, recording their relative rank
@@ -544,20 +544,21 @@
}
/*** Throw out all invalid chars ***/
- tmp = prepbuffer( buffer, bufsize );
+ tmp = prepbuffer( buffer, bufsize );
+ /*printf("Cleaned buffer : %s\n",tmp);*/
if ( tmp == NULL ) {
return 0;
}
-
h = (fp_t*)handle;
t = inittable(maxngrams);
+ /*printf("Table initialized\n");*/
/*** Create a hash table containing n-gram counts ***/
createngramtable(t, tmp);
-
+ /*printf("Table created\n");*/
/*** Take the top N n-grams and add them to the profile ***/
- table2heap(t);
- maxngrams = WGMIN( maxngrams, t->size );
+ table2heap(t);
+ maxngrams = WGMIN( maxngrams, t->size );
h->fprint = (ngram_t *)wg_malloc( sizeof(ngram_t) * maxngrams );
h->size = maxngrams;
@@ -568,7 +569,7 @@
entry_t tmp2;
heapextract(t, &tmp2);
-
+
/*** the string and its rank is all we need ***/
strcpy( h->fprint[i].str, tmp2.str );
h->fprint[i].rank = i;
@@ -578,7 +579,7 @@
wg_free(tmp);
/*** Sort n-grams alphabetically, for easy comparison ***/
- qsort( h->fprint, h->size, sizeof(ngram_t), ngramcmp_str );
+ qsort( h->fprint, h->size, sizeof(ngram_t), ngramcmp_str );
return 1;
}
@@ -608,7 +609,7 @@
#endif
return 0;
}
-
+
h->fprint = (ngram_t *)wg_malloc(maxngrams * sizeof(ngram_t));
while (cnt < maxngrams && wg_getline(line,1024,fp)) {
@@ -635,7 +636,7 @@
h->size = cnt;
/*** Sort n-grams, for easy comparison later on ***/
- qsort( h->fprint, h->size, sizeof(ngram_t), ngramcmp_str );
+ qsort( h->fprint, h->size, sizeof(ngram_t), ngramcmp_str );
fclose(fp);
@@ -648,14 +649,15 @@
{
uint4 i;
fp_t *h = (fp_t *)handle;
- ngram_t *tmp = wg_malloc( sizeof(ngram_t) * h->size );
-
+ ngram_t *tmp = (ngram_t*)wg_malloc( sizeof(ngram_t) * h->size );
+
/*** Make a temporary and sort it on rank ***/
memcpy( tmp, h->fprint, h->size * sizeof(ngram_t) );
- qsort( tmp, h->size, sizeof(ngram_t), ngramcmp_rank );
+ qsort( tmp, h->size, sizeof(ngram_t), ngramcmp_rank );
for (i=0; i<h->size; i++) {
- fprintf( fp, "%s\n", tmp[i].str );
+ /*fprintf( fp, "%s\t%i\n", tmp[i].str, tmp[i].rank );*/
+ fprintf( fp, "%s\n", tmp[i].str);
}
wg_free( tmp );
}
@@ -669,7 +671,7 @@
uint4 i = 0;
uint4 j = 0;
sint4 sum = 0;
-
+
/*** Compare the profiles in mergesort fashion ***/
while ( i < c->size && j < u->size ) {
@@ -705,7 +707,7 @@
}
return sum;
-
+
}
--- misc/libtextcat-2.2/src/fingerprint.h Mon May 19 14:16:31 2003
+++ misc/build/libtextcat-2.2/src/fingerprint.h Mon Mar 31 11:29:14 2008
@@ -41,7 +41,13 @@
extern int fp_Read( void *handle, const char *fname, int maxngrams );
extern sint4 fp_Compare( void *cat, void *unknown, int cutoff );
extern void fp_Show( void *handle );
+#ifdef __cplusplus
+extern "C" {
+#endif
extern const char *fp_Name( void *handle );
+#ifdef __cplusplus
+}
+#endif
extern void fp_Print( void *handle, FILE *fp );
#endif
--- misc/libtextcat-2.2/src/libtextcat.map Mon Mar 31 11:30:06 2008
+++ misc/build/libtextcat-2.2/src/libtextcat.map Mon Mar 31 11:29:14 2008
@@ -1 +1,40 @@
-dummy
+{
+ global:
+ charcopy
+ issame
+ nextcharstart
+ utfstrlen
+ wgmempool_Done
+ wgmempool_Init
+ wgmempool_Reset
+ wgmempool_alloc
+ wgmempool_getline
+ wgmempool_strdup
+ special_textcat_Init
+ textcat_Classify
+ textcat_Done
+ textcat_Init
+ textcat_Version
+ fp_Compare
+ fp_Create
+ fp_Debug
+ fp_Done
+ fp_Init
+ fp_Name
+ fp_Print
+ fp_Read
+ heapextract
+ wg_calloc
+ wg_free
+ wg_getline
+ wg_malloc
+ wg_split
+ wg_strdup
+ wg_strgmov
+ wg_trim
+ wg_zalloc
+ wgmem_error
+
+ local:
+ *;
+}
--- misc/libtextcat-2.2/src/makefile.mk Mon Mar 31 11:30:06 2008
+++ misc/build/libtextcat-2.2/src/makefile.mk Mon Mar 31 11:29:42 2008
@@ -1 +1,90 @@
-dummy
+#*************************************************************************
+#
+# DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+#
+# Copyright 2008 by Sun Microsystems, Inc.
+#
+# OpenOffice.org - a multi-platform office productivity suite
+#
+# $RCSfile: libtextcat-2.2.patch,v $
+#
+# $Revision: 1.8 $
+#
+# This file is part of OpenOffice.org.
+#
+# OpenOffice.org is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Lesser General Public License version 3
+# only, as published by the Free Software Foundation.
+#
+# OpenOffice.org is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Lesser General Public License version 3 for more details
+# (a copy is included in the LICENSE file that accompanied this code).
+#
+# You should have received a copy of the GNU Lesser General Public License
+# version 3 along with OpenOffice.org. If not, see
+# <http://www.openoffice.org/license.html>
+# for a copy of the LGPLv3 License.
+#
+#*************************************************************************
+
+PRJ = ..$/..$/..$/..$/..
+
+PRJNAME = libtextcat
+TARGET = libtextcat
+CFLAGSCALL=gsd
+
+USE_DEFFILE=TRUE
+EXTERNAL_WARNINGS_NOT_ERRORS := TRUE
+
+.INCLUDE : settings.mk
+
+# --- Files --------------------------------------------------------
+
+# !! not to be compiled because those belong to a stand alone programs: !!
+# $(SLO)$/createfp.obj\
+# $(SLO)$/testtextcat.obj
+
+SLOFILES= \
+ $(SLO)$/common.obj\
+ $(SLO)$/fingerprint.obj\
+ $(SLO)$/textcat.obj\
+ $(SLO)$/wg_mempool.obj\
+ $(SLO)$/utf8misc.obj
+
+#SHL1TARGET= $(TARGET)$(UPD)$(DLLPOSTFIX)
+SHL1TARGET= $(TARGET)
+
+SHL1STDLIBS=
+
+# build DLL
+SHL1LIBS= $(SLB)$/$(TARGET).lib
+SHL1IMPLIB= i$(TARGET)
+SHL1DEPN= $(SHL1LIBS)
+SHL1DEF= $(MISC)$/$(SHL1TARGET).def
+
+# build DEF file
+DEF1NAME= $(SHL1TARGET)
+DEF1DEPN=$(MISC)$/$(SHL1TARGET).flt
+
+SHL1VERSIONMAP= libtextcat.map
+
+# --- Targets ------------------------------------------------------
+
+.INCLUDE : target.mk
+
+# copy hand supplied configuration file for Win32 builds to the file
+# which is included in the source code
+$(SLOFILES) : config.h
+config.h :
+ $(GNUCOPY) $(OUT)$/misc$/build$/libtextcat-2.2$/src$/win32_config.h $(OUT)$/misc$/build$/libtextcat-2.2$/src$/config.h
+
+
+$(MISC)$/$(SHL1TARGET).flt: makefile.mk
+ @echo ------------------------------
+ @echo Making: $@
+ @echo Imp>$@
+ @echo __CT>>$@
+ @echo _real>>$@
+ @echo unnamed>>$@
--- misc/libtextcat-2.2/src/textcat.c Thu May 22 13:32:43 2003
+++ misc/build/libtextcat-2.2/src/textcat.c Mon Mar 31 11:29:14 2008
@@ -4,23 +4,23 @@
* Copyright (C) 2003 WiseGuys Internet B.V.
*
* THE BSD LICENSE
- *
+ *
* 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.
- *
+ *
* - Neither the name of the WiseGuys Internet B.V. nor the names of
* its contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
- *
+ *
* 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
@@ -74,6 +74,7 @@
typedef struct {
void **fprint;
+ char *fprint_disable;
uint4 size;
uint4 maxsize;
@@ -112,11 +113,21 @@
fp_Done( h->fprint[i] );
}
wg_free( h->fprint );
+ wg_free( h->fprint_disable );
wg_free( h );
}
-extern void *textcat_Init( const char *conffile )
+/** Replaces older function */
+extern void *textcat_Init( const char *conffile ){
+ return special_textcat_Init( conffile, DEFAULT_FINGERPRINTS_PATH );
+}
+
+/**
+ * Originaly this function had only one parameter (conffile) it has been modified since OOo use
+ * Basicaly prefix is the directory path where fingerprints are stored
+ */
+extern void *special_textcat_Init( const char *conffile, const char *prefix )
{
textcat_t *h;
char line[1024];
@@ -134,11 +145,13 @@
h->size = 0;
h->maxsize = 16;
h->fprint = (void **)wg_malloc( sizeof(void*) * h->maxsize );
+ h->fprint_disable = (char *)wg_malloc( sizeof(char*) * h->maxsize ); /*added to store the state of languages*/
while ( wg_getline( line, 1024, fp ) ) {
char *p;
char *segment[4];
- int res;
+ char finger_print_file_name[512];
+ int res;
/*** Skip comments ***/
#ifdef HAVE_STRCHR
@@ -156,17 +169,23 @@
/*** Ensure enough space ***/
if ( h->size == h->maxsize ) {
h->maxsize *= 2;
- h->fprint = (void *)wg_realloc( h->fprint, sizeof(void*) * h->maxsize );
+ h->fprint = (void **)wg_realloc( h->fprint, sizeof(void*) * h->maxsize );
+ h->fprint_disable = (char *)wg_realloc( h->fprint_disable, sizeof(char*) * h->maxsize );
}
/*** Load data ***/
if ((h->fprint[ h->size ] = fp_Init( segment[1] ))==NULL) {
goto ERROR;
}
- if ( fp_Read( h->fprint[h->size], segment[0], 400 ) == 0 ) {
+ finger_print_file_name[0] = '\0';
+ strcat(finger_print_file_name, prefix);
+ strcat(finger_print_file_name, segment[0]);
+
+ if ( fp_Read( h->fprint[h->size], finger_print_file_name, 400 ) == 0 ) {
textcat_Done(h);
goto ERROR;
- }
+ }
+ h->fprint_disable[h->size] = 0xF0; /*0xF0 is the code for enabled languages, 0x0F is for disabled*/
h->size++;
}
@@ -203,11 +222,18 @@
result = _TEXTCAT_RESULT_SHORT;
goto READY;
}
-
+
/*** Calculate the score for each category. ***/
for (i=0; i<h->size; i++) {
- int score = fp_Compare( h->fprint[i], unknown, threshold );
- candidates[i].score = score;
+ int score;
+ if(h->fprint_disable[i] & 0x0F){ /*if this language is disabled*/
+ score = MAXSCORE;
+ }
+ else{
+ score = fp_Compare( h->fprint[i], unknown, threshold );
+ /*printf("Score for %s : %i\n", fp_Name(h->fprint[i]), score);*/
+ }
+ candidates[i].score = score;
candidates[i].name = fp_Name( h->fprint[i] );
if ( score < minscore ) {
minscore = score;
@@ -218,7 +244,6 @@
/*** Find the best performers ***/
for (i=0; i<h->size; i++) {
if ( candidates[i].score < threshold ) {
-
if ( ++cnt == MAXCANDIDATES+1 ) {
break;
}
@@ -235,7 +260,7 @@
else {
char *p = result;
char *plimit = result+MAXOUTPUTSIZE;
-
+
qsort( candidates, cnt, sizeof(candidate_t), cmpcandidates );
*p = '\0';
@@ -247,7 +272,7 @@
}
READY:
fp_Done(unknown);
-#ifdef SHOULD_FREE
+#ifdef SHOULD_FREE
free(candidates);
#undef SHOULD_FREE
#endif
--- misc/libtextcat-2.2/src/textcat.h Mon May 19 14:16:31 2003
+++ misc/build/libtextcat-2.2/src/textcat.h Mon Mar 31 11:29:14 2008
@@ -40,6 +40,9 @@
#define _TEXTCAT_RESULT_UNKOWN "UNKNOWN"
#define _TEXTCAT_RESULT_SHORT "SHORT"
+#ifdef __cplusplus
+extern "C" {
+#endif
/**
* textcat_Init() - Initialize the text classifier. The textfile
@@ -51,10 +54,19 @@
* Returns: handle on success, NULL on error. (At the moment, the
* only way errors can occur, is when the library cannot read the
* conffile, or one of the fingerprint files listed in it.)
+ *
+ * Replace older function (and has exacly the same behaviour)
+ * see below
*/
extern void *textcat_Init( const char *conffile );
/**
+ * Originaly this function had only one parameter (conffile) it has been modified since OOo must be able to load alternativ DB
+ * Basicaly prefix is the directory path where fingerprints are stored
+ */
+extern void *special_textcat_Init( const char *conffile, const char *prefix );
+
+/**
* textcat_Done() - Free up resources for handle
*/
extern void textcat_Done( void *handle );
@@ -77,4 +89,8 @@
* textcat_Version() - Returns a string describing the version of this classifier.
*/
extern char *textcat_Version();
+
+#ifdef __cplusplus
+}
+#endif
#endif
--- misc/libtextcat-2.2/src/utf8misc.c Mon Mar 31 11:30:06 2008
+++ misc/build/libtextcat-2.2/src/utf8misc.c Mon Mar 31 11:29:14 2008
@@ -1 +1,132 @@
-dummy
+/***************************************************************************
+ * Copyright (C) 2006 by Jocelyn Merand *
+ * joc.mer@gmail.com *
+ * *
+ * THE BSD LICENSE
+ *
+ * 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.
+ *
+ * - Neither the name of the WiseGuys Internet B.V. nor the names of
+ * its contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * 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
+ * OWNER 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.
+ ***************************************************************************/
+
+#ifndef _UTF8_MISC_H_
+#include "utf8misc.h"
+#endif
+
+
+int nextcharstart(const char *str, int position){
+ int pointer = position;
+
+ if(str[pointer] & ESCAPE_MASK){ /*if the first bit of the current char is 1*/
+
+ /*then str[pointer] is an escape character*/
+
+ char escape_char = ((str[pointer] & WEIGHT_MASK) << 1); /*and we use it to count (by bit translation) following characters (only the weightest part)*/
+
+ while(escape_char & ESCAPE_MASK && str[pointer]){/*every step, we move the byte of 1 bit left, when first bit is 0, it's finished*/
+ escape_char = escape_char <<1;
+ ++pointer;
+ }
+ }
+ if(str[pointer]){ /*finaly, if we are not on the \0 character, we jump to the next character*/
+ ++pointer;
+ }
+ return pointer;
+}
+
+
+int charcopy(const char *str, char *dest){
+
+ int pointer = 0;
+ if(str[pointer] & ESCAPE_MASK){ /*if the first bit of the current char is 1*/
+
+ /*then str[pointer] is an escape character*/
+
+ char escape_char = ((str[pointer] & WEIGHT_MASK) << 1); /*and we use it to count following characters (only the weightest part)*/
+
+ while(escape_char & ESCAPE_MASK && str[pointer]){ /*every step, we move the byte of 1 bit left, when first bit is 0, it's finished*/
+ dest[pointer] = str[pointer];
+ escape_char = escape_char <<1;
+ ++pointer;
+ }
+ }
+ if(str[pointer]){
+ dest[pointer] = str[pointer];
+ ++pointer;
+ }
+
+ return pointer;
+}
+
+
+int issame( char *lex, char *key, int len )
+{
+ /*printf("[%s] prefix of [%s] with length %i", lex, key, len);*/
+ int char_counter = 0;
+ int pointer = 0;
+ while(char_counter < len) {
+
+ if(key[pointer] & ESCAPE_MASK){ /*if the first bit of the current char is 1*/
+
+ /*then key[pointer] is an escap character*/
+
+ char escape_char = ((key[pointer] & WEIGHT_MASK) << 1); /*and we use it to count (only the weightest part)*/
+
+ while(escape_char & ESCAPE_MASK && key[pointer] == lex[pointer] ){
+ escape_char = escape_char <<1;
+ ++pointer;
+ }
+ }
+ ++char_counter; /*and we are on a new utf8 character*/
+ if ( key[pointer] != lex[pointer] ) {
+ return 0;
+ /*printf(" NO\n", lex, key, len);*/
+ }
+ ++pointer;
+ }
+ if ( lex[pointer] != '\0' ) {
+ return 0;
+ /*printf(" NO\n");*/
+ }
+
+ /*printf(" YES\n");*/
+
+ return 1;
+}
+
+
+extern int utfstrlen(const char* str){
+ int char_counter = 0;
+ int pointer = 0;
+ while(str[pointer]) {
+ pointer = nextcharstart(str, pointer);
+
+ ++char_counter; /*and we are on a new utf8 character*/
+ }
+ return char_counter;
+}
+
--- misc/libtextcat-2.2/src/utf8misc.h Mon Mar 31 11:30:06 2008
+++ misc/build/libtextcat-2.2/src/utf8misc.h Mon Mar 31 11:29:14 2008
@@ -1 +1,88 @@
-dummy
+/***************************************************************************
+ * Copyright (C) 2006 by Jocelyn Merand *
+ * joc.mer@gmail.com *
+ * *
+ * THE BSD LICENSE
+ *
+ * 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.
+ *
+ * - Neither the name of the WiseGuys Internet B.V. nor the names of
+ * its contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * 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
+ * OWNER 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.
+ ***************************************************************************/
+
+#ifndef _UTF8_MISC_H_
+#define _UTF8_MISC_H_
+
+/**
+ * These variables are used in character processing functions
+ * These have been added to manage utf-8 symbols, particularly escape chars
+ */
+#ifdef _UTF8_
+#define ESCAPE_MASK 0x80
+#define WEIGHT_MASK 0xF0
+#else
+#define ESCAPE_MASK 0xFF
+#define WEIGHT_MASK 0x00
+#endif
+
+
+/*
+ * Is used to jump to the next start of char
+ * of course it's only usefull when encoding is utf-8
+ * This function have been added by Jocelyn Merand to use libtextcat in OOo
+ */
+int nextcharstart(const char *str, int position);
+
+
+/*Copy the char in str to dest
+ * of course it's only usefull when encoding is utf8 and the symbol is encoded with more than 1 char
+ * return the number of char jumped
+ * This function have been added by Jocelyn Merand to use libtextcat in OOo
+ */
+int charcopy(const char *str, char *dest);
+
+
+/* checks if n-gram lex is a prefix of key and of length len
+* if _UTF8_ is defined, it uses escap characters and len is not realy the length of lex
+* in this case, len is the number of utf-8 char strlen("€") == 3 but len == 1
+*/
+int issame( char *lex, char *key, int len );
+
+
+/* Counts the number of characters
+* if _UTF8_ is defined, it uses escap characters and the result is not realy the length of str
+* in this case, the result is the number of utf-8 char strlen("€") == 3 but utfstrlen("€") == 1
+*/
+#ifdef __cplusplus
+extern "C" {
+#endif
+extern int utfstrlen(const char* str);
+#ifdef __cplusplus
+}
+#endif
+
+#endif
+
--- misc/libtextcat-2.2/src/win32_config.h Mon Mar 31 11:30:06 2008
+++ misc/build/libtextcat-2.2/src/win32_config.h Mon Mar 31 11:29:14 2008
@@ -1 +1,136 @@
-dummy
+/* src/config.h. Generated by configure. */
+/* src/config.h.in. Generated from configure.ac by autoheader. */
+
+/* Define to one of `_getb67', `GETB67', `getb67' for Cray-2 and Cray-YMP
+ systems. This function is required for `alloca.c' support on those systems.
+ */
+/* #undef CRAY_STACKSEG_END */
+
+/* Define to 1 if using `alloca.c'. */
+/* #undef C_ALLOCA */
+
+/* Define to 1 if you have `alloca', as a function or macro. */
+/* #undef HAVE_ALLOCA */
+
+/* Define to 1 if you have <alloca.h> and it should be used (not on Ultrix).
+ */
+/* #undef HAVE_ALLOCA_H */
+
+/* Define to 1 if you have the <dlfcn.h> header file. */
+#define HAVE_DLFCN_H 1
+
+/* Define to 1 if you don't have `vprintf' but do have `_doprnt.' */
+/* #undef HAVE_DOPRNT */
+
+/* Define to 1 if you have the `gettimeofday' function. */
+/* #undef HAVE_GETTIMEOFDAY */
+
+/* Define to 1 if you have the <inttypes.h> header file. */
+/* #undef HAVE_INTTYPES_H */
+
+/* Define to 1 if you have the <limits.h> header file. */
+#define HAVE_LIMITS_H 1
+
+/* Define to 1 if your system has a GNU libc compatible `malloc' function, and
+ to 0 otherwise. */
+#define HAVE_MALLOC 1
+
+/* Define to 1 if you have the <memory.h> header file. */
+#define HAVE_MEMORY_H 1
+
+/* Define to 1 if you have the `memset' function. */
+#define HAVE_MEMSET 1
+
+/* Define to 1 if your system has a GNU libc compatible `realloc' function,
+ and to 0 otherwise. */
+#define HAVE_REALLOC 1
+
+/* Define to 1 if you have the <stdint.h> header file. */
+/* #undef HAVE_STDINT_H */
+
+/* Define to 1 if you have the <stdlib.h> header file. */
+#define HAVE_STDLIB_H 1
+
+/* Define to 1 if you have the `strchr' function. */
+#define HAVE_STRCHR 1
+
+/* Define to 1 if you have the `strdup' function. */
+#define HAVE_STRDUP 1
+
+/* Define to 1 if you have the <strings.h> header file. */
+/* #undef HAVE_STRINGS_H */
+
+/* Define to 1 if you have the <string.h> header file. */
+#define HAVE_STRING_H 1
+
+/* Define to 1 if you have the `strpbrk' function. */
+#define HAVE_STRPBRK 1
+
+/* Define to 1 if you have the <sys/stat.h> header file. */
+#define HAVE_SYS_STAT_H 1
+
+/* Define to 1 if you have the <sys/time.h> header file. */
+/* #undef HAVE_SYS_TIME_H */
+
+/* Define to 1 if you have the <sys/types.h> header file. */
+#define HAVE_SYS_TYPES_H 1
+
+/* Define to 1 if you have the <unistd.h> header file. */
+#define HAVE_UNISTD_H 1
+
+/* Define to 1 if you have the `vprintf' function. */
+#define HAVE_VPRINTF 1
+
+/* Name of package */
+#define PACKAGE "libtextcat"
+
+/* Define to the address where bug reports for this package should be sent. */
+#define PACKAGE_BUGREPORT ""
+
+/* Define to the full name of this package. */
+#define PACKAGE_NAME "libtextcat"
+
+/* Define to the full name and version of this package. */
+#define PACKAGE_STRING "libtextcat 2.2"
+
+/* Define to the one symbol short name of this package. */
+#define PACKAGE_TARNAME "libtextcat"
+
+/* Define to the version of this package. */
+#define PACKAGE_VERSION "2.2"
+
+/* If using the C implementation of alloca, define if you know the
+ direction of stack growth for your system; otherwise it will be
+ automatically deduced at run-time.
+ STACK_DIRECTION > 0 => grows toward higher addresses
+ STACK_DIRECTION < 0 => grows toward lower addresses
+ STACK_DIRECTION = 0 => direction of growth unknown */
+/* #undef STACK_DIRECTION */
+
+/* Define to 1 if you have the ANSI C header files. */
+#define STDC_HEADERS 1
+
+/* Define to 1 if you can safely include both <sys/time.h> and <time.h>. */
+#define TIME_WITH_SYS_TIME 1
+
+/* Define to 1 if your <sys/time.h> declares `struct tm'. */
+/* #undef TM_IN_SYS_TIME */
+
+/* Version number of package */
+#define VERSION "2.2"
+
+/* Define to empty if `const' does not conform to ANSI C. */
+/* #undef const */
+
+/* Define as `__inline' if that's what the C compiler calls it, or to nothing
+ if it is not supported. */
+/* #undef inline */
+
+/* Define to rpl_malloc if the replacement function should be used. */
+/* #undef malloc */
+
+/* Define to rpl_realloc if the replacement function should be used. */
+/* #undef realloc */
+
+/* Define to `unsigned' if <sys/types.h> does not define. */
+/* #undef size_t */
