rpms/hunspell
5
0
Fork 0
Code Issues Pull Requests Releases Activity

add 8bit dictionary word chars to the unicode list

Browse Source
This commit is contained in:
Caolan McNamara 2007-05-22 11:42:41 +00:00
parent ebefbaa051
commit 0a1b5e5742
3 changed files with 106 additions and 234 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

203
hunspell-1.1.5-missingheaders.patch
View File

@ -1,203 +0,0 @@
--- /dev/null 2007-03-16 11:16:34.774869017 +0000
+++ hunspell-1.1.5/src/tools/munch.h 2007-03-20 09:19:49.000000000 +0000
@@ -0,0 +1,121 @@
+/* munch header file */
+
+#define MAX_LN_LEN 200
+#define MAX_WD_LEN 200
+#define MAX_PREFIXES 256
+#define MAX_SUFFIXES 256
+#define MAX_ROOTS 20
+#define MAX_WORDS 5000
+
+#define ROTATE_LEN 5
+
+#define ROTATE(v,q) \
+ (v) = ((v) << (q)) | (((v) >> (32 - q)) & ((1 << (q))-1));
+
+#define SET_SIZE 256
+
+#define XPRODUCT (1 << 0)
+
+/* the affix table entry */
+
+struct affent
+{
+ char * appnd;
+ char * strip;
+ short appndl;
+ short stripl;
+ char achar;
+ char xpflg;
+ short numconds;
+ char conds[SET_SIZE];
+};
+
+
+struct affixptr
+{
+ struct affent * aep;
+ int num;
+};
+
+/* the prefix and suffix table */
+int numpfx; /* Number of prefixes in table */
+int numsfx; /* Number of suffixes in table */
+
+/* the prefix table */
+struct affixptr ptable[MAX_PREFIXES];
+
+/* the suffix table */
+struct affixptr stable[MAX_SUFFIXES];
+
+
+/* data structure to store results of lookups */
+struct matches
+{
+ struct hentry * hashent; /* hash table entry */
+ struct affent * prefix; /* Prefix used, or NULL */
+ struct affent * suffix; /* Suffix used, or NULL */
+};
+
+int numroots; /* number of root words found */
+struct matches roots[MAX_ROOTS]; /* list of root words found */
+
+/* hashing stuff */
+
+struct hentry
+{
+ char * word;
+ char * affstr;
+ struct hentry * next;
+ int keep;
+};
+
+
+int tablesize;
+struct hentry * tableptr;
+
+/* unmunch stuff */
+
+int numwords; /* number of words found */
+struct dwords
+{
+ char * word;
+ int pallow;
+};
+
+struct dwords wlist[MAX_WORDS]; /* list words found */
+
+
+/* the routines */
+
+int parse_aff_file(FILE* afflst);
+
+void encodeit(struct affent * ptr, char * cs);
+
+int load_tables(FILE * wrdlst);
+
+int hash(const char *);
+
+int add_word(char *);
+
+struct hentry * lookup(const char *);
+
+void aff_chk (const char * word, int len);
+
+void pfx_chk (const char * word, int len, struct affent* ep, int num);
+
+void suf_chk (const char * word, int len, struct affent * ep, int num,
+ struct affent * pfxent, int cpflag);
+
+void add_affix_char(struct hentry * hent, char ac);
+
+int expand_rootword(const char *, int, const char*, int);
+
+void pfx_add (const char * word, int len, struct affent* ep, int num);
+
+void suf_add (const char * word, int len, struct affent * ep, int num);
+
+char * mystrsep(char ** stringp, const char delim);
+
+char * mystrdup(const char * s);
+
+void mychomp(char * s);
--- /dev/null 2007-03-16 11:16:34.774869017 +0000
+++ hunspell-1.1.5/src/tools/unmunch.h 2007-03-20 09:19:50.000000000 +0000
@@ -0,0 +1,76 @@
+/* unmunch header file */
+
+#define MAX_LN_LEN 200
+#define MAX_WD_LEN 200
+#define MAX_PREFIXES 256
+#define MAX_SUFFIXES 256
+#define MAX_WORDS 5000
+
+#define ROTATE_LEN 5
+
+#define ROTATE(v,q) \
+ (v) = ((v) << (q)) | (((v) >> (32 - q)) & ((1 << (q))-1));
+
+#define SET_SIZE 256
+
+#define XPRODUCT (1 << 0)
+
+/* the affix table entry */
+
+struct affent
+{
+ char * appnd;
+ char * strip;
+ short appndl;
+ short stripl;
+ char achar;
+ char xpflg;
+ short numconds;
+ char conds[SET_SIZE];
+};
+
+
+struct affixptr
+{
+ struct affent * aep;
+ int num;
+};
+
+/* the prefix and suffix table */
+int numpfx; /* Number of prefixes in table */
+int numsfx; /* Number of suffixes in table */
+
+/* the prefix table */
+struct affixptr ptable[MAX_PREFIXES];
+
+/* the suffix table */
+struct affixptr stable[MAX_SUFFIXES];
+
+
+int numwords; /* number of words found */
+struct dwords
+{
+ char * word;
+ int pallow;
+};
+
+struct dwords wlist[MAX_WORDS]; /* list words found */
+
+
+/* the routines */
+
+int parse_aff_file(FILE* afflst);
+
+void encodeit(struct affent * ptr, char * cs);
+
+int expand_rootword(const char *, int, const char*, int);
+
+void pfx_add (const char * word, int len, struct affent* ep, int num);
+
+void suf_add (const char * word, int len, struct affent * ep, int num);
+
+char * mystrsep(char ** stringp, const char delim);
+
+char * mystrdup(const char * s);
+
+void mychomp(char * s);

131
hunspell-1.1.5.encoding.patch
View File

@ -1,6 +1,101 @@
diff -ru hunspell-1.1.5.orig/src/hunspell/affixmgr.cxx hunspell-1.1.5/src/hunspell/affixmgr.cxx
--- hunspell-1.1.5.orig/src/hunspell/affixmgr.cxx 2007-03-12 11:38:23.000000000 +0000
+++ hunspell-1.1.5/src/hunspell/affixmgr.cxx 2007-05-22 12:35:27.000000000 +0100
@@ -19,6 +19,8 @@
#include "csutil.hxx"
+#include <iconv.h>
+
#ifndef MOZILLA_CLIENT
#ifndef W32
using namespace std;
@@ -230,6 +232,61 @@
checknum=0;
}
+char *changeenc(iconv_t cd, char *token)
+{
+ if (!cd || !token) return token;
+
+ iconv (cd, NULL, NULL, NULL, NULL);
+
+ char *inptr = token;
+ size_t insize = strlen(token);
+
+ size_t avail = (insize + 1) * 3;
+ char *wrptr = (char*)malloc(avail);
+ char *output = wrptr;
+
+ memset(wrptr, 0, avail);
+ size_t nconv = iconv (cd, &inptr, &insize, &wrptr, &avail);
+ free(token);
+
+ if (nconv == (size_t) -1)
+ {
+ free(output);
+ output = NULL;
+ }
+
+ return output;
+}
+
+void AffixMgr::fill_unicode_equivalents()
+{
+ if ((!ignorechars_utf16_len || !wordchars_utf16_len) && (ignorechars || wordchars))
+ {
+ iconv_t to_utf16 = iconv_open("UCS2", encoding);
+ if (to_utf16 == 0 || to_utf16 == (iconv_t) -1)
+ return;
+
+ unsigned short *p;
+
+ if (!ignorechars_utf16_len && ignorechars && strlen(ignorechars))
+ {
+ char *tmp = mystrdup(ignorechars);
+ ignorechars_utf16 = (unsigned short*)changeenc(to_utf16, tmp);
+ for (p = ignorechars_utf16; *p; ++p);
+ ignorechars_utf16_len = p - ignorechars_utf16;
+ flag_qsort(ignorechars_utf16, 0, ignorechars_utf16_len);
+ }
+
+ if (!wordchars_utf16_len && wordchars && strlen(wordchars))
+ {
+ char *tmp = mystrdup(wordchars);
+ wordchars_utf16 = (unsigned short*)changeenc(to_utf16, tmp);
+ for (p = wordchars_utf16; *p; ++p);
+ wordchars_utf16_len = p - wordchars_utf16;
+ flag_qsort(wordchars_utf16, 0, wordchars_utf16_len);
+ }
+ }
+}
// read in aff file and build up prefix and suffix entry objects
int AffixMgr::parse_file(const char * affpath)
@@ -659,6 +716,9 @@
breaktable[0] = mystrdup("-");
numbreak = 1;
}
+
+ fill_unicode_equivalents();
+
return 0;
}
diff -ru hunspell-1.1.5.orig/src/hunspell/affixmgr.hxx hunspell-1.1.5/src/hunspell/affixmgr.hxx
--- hunspell-1.1.5.orig/src/hunspell/affixmgr.hxx 2007-03-12 11:42:52.000000000 +0000
+++ hunspell-1.1.5/src/hunspell/affixmgr.hxx 2007-05-22 12:00:14.000000000 +0100
@@ -203,6 +203,7 @@
int process_pfx_tree_to_list();
int process_sfx_tree_to_list();
int redundant_condition(char, char * strip, int stripl, const char * cond, char *);
+ void fill_unicode_equivalents();
};
#endif
diff -ru hunspell-1.1.5.orig/src/hunspell/csutil.cxx hunspell-1.1.5/src/hunspell/csutil.cxx
--- hunspell-1.1.5.orig/src/hunspell/csutil.cxx 2007-03-12 12:01:56.000000000 +0000
+++ hunspell-1.1.5/src/hunspell/csutil.cxx 2007-05-21 13:31:46.000000000 +0100
+++ hunspell-1.1.5/src/hunspell/csutil.cxx 2007-05-22 12:21:30.000000000 +0100
@@ -5090,6 +5090,10 @@
#ifndef OPENOFFICEORG
#ifndef MOZILLA_CLIENT
@ -14,7 +109,7 @@ diff -ru hunspell-1.1.5.orig/src/hunspell/csutil.cxx hunspell-1.1.5/src/hunspell
int j;
diff -ru hunspell-1.1.5.orig/src/parsers/textparser.cxx hunspell-1.1.5/src/parsers/textparser.cxx
--- hunspell-1.1.5.orig/src/parsers/textparser.cxx 2007-03-05 12:59:53.000000000 +0000
+++ hunspell-1.1.5/src/parsers/textparser.cxx 2007-05-21 13:31:46.000000000 +0100
+++ hunspell-1.1.5/src/parsers/textparser.cxx 2007-05-22 12:02:24.000000000 +0100
@@ -5,6 +5,7 @@
#include "../hunspell/csutil.hxx"
@ -23,7 +118,7 @@ diff -ru hunspell-1.1.5.orig/src/parsers/textparser.cxx hunspell-1.1.5/src/parse
#ifndef W32
using namespace std;
@@ -46,22 +47,76 @@
@@ -46,22 +47,52 @@
#define LATIN1_LEN (sizeof(LATIN1) / sizeof(char *))
@ -33,31 +128,7 @@ diff -ru hunspell-1.1.5.orig/src/parsers/textparser.cxx hunspell-1.1.5/src/parse
}
-TextParser::TextParser(const char * wordchars)
+static char *changeenc(iconv_t cd, char *token)
+{
+ if (!cd || !token) return token;
+
+ iconv (cd, NULL, NULL, NULL, NULL);
+
+ char *inptr = token;
+ size_t insize = strlen(token);
+
+ size_t avail = (insize + 1) * 3;
+ char *wrptr = (char*)malloc(avail);
+ char *output = wrptr;
+
+ memset(wrptr, 0, avail);
+ size_t nconv = iconv (cd, &inptr, &insize, &wrptr, &avail);
+ free(token);
+
+ if (nconv == (size_t) -1)
+ {
+ free(output);
+ output = NULL;
+ }
+
+ return output;
+}
+extern char *changeenc(iconv_t cd, char *token);
+
+char *TextParser::todictenc(char *token)
+{
@ -105,7 +176,7 @@ diff -ru hunspell-1.1.5.orig/src/parsers/textparser.cxx hunspell-1.1.5/src/parse
int TextParser::is_wordchar(char * w)
diff -ru hunspell-1.1.5.orig/src/parsers/textparser.hxx hunspell-1.1.5/src/parsers/textparser.hxx
--- hunspell-1.1.5.orig/src/parsers/textparser.hxx 2007-01-19 01:01:07.000000000 +0000
+++ hunspell-1.1.5/src/parsers/textparser.hxx 2007-05-21 13:31:46.000000000 +0100
+++ hunspell-1.1.5/src/parsers/textparser.hxx 2007-05-22 11:29:52.000000000 +0100
@@ -20,12 +20,15 @@
*
*/
@ -152,7 +223,7 @@ diff -ru hunspell-1.1.5.orig/src/parsers/textparser.hxx hunspell-1.1.5/src/parse
#endif
diff -ru hunspell-1.1.5.orig/src/tools/hunspell.cxx hunspell-1.1.5/src/tools/hunspell.cxx
--- hunspell-1.1.5.orig/src/tools/hunspell.cxx 2007-05-21 13:31:55.000000000 +0100
+++ hunspell-1.1.5/src/tools/hunspell.cxx 2007-05-21 13:31:46.000000000 +0100
+++ hunspell-1.1.5/src/tools/hunspell.cxx 2007-05-22 11:15:33.000000000 +0100
@@ -89,6 +89,8 @@
#endif
#endif

6
hunspell.spec
View File

@ -1,7 +1,7 @@
Name: hunspell
Summary: Hunspell is a spell checker and morphological analyzer library
Version: 1.1.5.3
Release: 2%{?dist}
Release: 3%{?dist}
Source: http://downloads.sourceforge.net/%{name}/hunspell-1.1.5-3.tar.gz
Group: System Environment/Libraries
URL: http://hunspell.sourceforge.net/
@ -82,6 +82,10 @@ rm -rf $RPM_BUILD_ROOT
%{_libdir}/pkgconfig/hunspell.pc
%changelog
* Tue May 22 2007 Caolan McNamara <caolanm@redhat.com> - 1.1.5.3-3
- Resolves: rhbz#240696 extend encoding patch to promote and add
dictionary 8bit WORDCHARS to the ucs-2 word char list
* Mon May 21 2007 Caolan McNamara <caolanm@redhat.com> - 1.1.5.3-2
- Resolves: rhbz#240696 add hunspell-1.1.5.encoding.patch