summaryrefslogtreecommitdiff
path: root/plugins/SpellChecker/src/hunspell/hashmgr.cxx
diff options
context:
space:
mode:
authorTobias Weimer <wishmaster51@googlemail.com>2014-07-20 08:01:05 +0000
committerTobias Weimer <wishmaster51@googlemail.com>2014-07-20 08:01:05 +0000
commitd5dfa90eec9c108f86b9a0aa3f746a5b164c6649 (patch)
treee6eb1ae6ca89db58494b28d6c5124ea94c517c37 /plugins/SpellChecker/src/hunspell/hashmgr.cxx
parent42c1d344fd105bcc25ae35e5a82d16a978b302c4 (diff)
SpellChecker:
-Updated Hunspell to 1.3.3 -Create services in Load() git-svn-id: http://svn.miranda-ng.org/main/trunk@9876 1316c22d-e87f-b044-9b9b-93d7a3e3ba9c
Diffstat (limited to 'plugins/SpellChecker/src/hunspell/hashmgr.cxx')
-rw-r--r--plugins/SpellChecker/src/hunspell/hashmgr.cxx73
1 files changed, 41 insertions, 32 deletions
diff --git a/plugins/SpellChecker/src/hunspell/hashmgr.cxx b/plugins/SpellChecker/src/hunspell/hashmgr.cxx
index f150437ad6..7a5da77ecc 100644
--- a/plugins/SpellChecker/src/hunspell/hashmgr.cxx
+++ b/plugins/SpellChecker/src/hunspell/hashmgr.cxx
@@ -1,14 +1,23 @@
-#include "..\commons.h"
+#include "..\commons.h"
+
+#undef max
// build a hash table from a munched word list
HashMgr::HashMgr(const char * tpath, const char * apath, const char * key)
+ : tablesize(0)
+ , tableptr(NULL)
+ , userword(0)
+ , flag_mode(FLAG_CHAR)
+ , complexprefixes(0)
+ , utf8(0)
+ , forbiddenword(FORBIDDENWORD) // forbidden word signing flag
+ , numaliasf(0)
+ , aliasf(NULL)
+ , aliasflen(0)
+ , numaliasm(0)
+ , aliasm(NULL)
{
- tablesize = 0;
- tableptr = NULL;
- flag_mode = FLAG_CHAR;
- complexprefixes = 0;
- utf8 = 0;
langnum = 0;
lang = NULL;
enc = NULL;
@@ -16,11 +25,6 @@ HashMgr::HashMgr(const char * tpath, const char * apath, const char * key)
ignorechars = NULL;
ignorechars_utf16 = NULL;
ignorechars_utf16_len = 0;
- numaliasf = 0;
- aliasf = NULL;
- numaliasm = 0;
- aliasm = NULL;
- forbiddenword = FORBIDDENWORD; // forbidden word signing flag
load_config(apath, key);
int ec = load_tables(tpath, key);
if (ec) {
@@ -106,7 +110,7 @@ int HashMgr::add_word(const char * word, int wbl, int wcl, unsigned short * aff,
int al, const char * desc, bool onlyupcase)
{
bool upcasehomonym = false;
- int descl = desc ? (aliasm ? sizeof(short) : strlen(desc) + 1) : 0;
+ int descl = desc ? (aliasm ? sizeof(char *) : strlen(desc) + 1) : 0;
// variable-length hash record with word and optional fields
struct hentry* hp =
(struct hentry *) malloc (sizeof(struct hentry) + wbl + descl);
@@ -200,18 +204,21 @@ int HashMgr::add_word(const char * word, int wbl, int wcl, unsigned short * aff,
}
int HashMgr::add_hidden_capitalized_word(char * word, int wbl, int wcl,
- unsigned short * flags, int al, char * dp, int captype)
+ unsigned short * flags, int flagslen, char * dp, int captype)
{
+ if (flags == NULL)
+ flagslen = 0;
+
// add inner capitalized forms to handle the following allcap forms:
// Mixed caps: OpenOffice.org -> OPENOFFICE.ORG
// Allcaps with suffixes: CIA's -> CIA'S
if (((captype == HUHCAP) || (captype == HUHINITCAP) ||
- ((captype == ALLCAP) && (flags != NULL))) &&
- !((flags != NULL) && TESTAFF(flags, forbiddenword, al))) {
- unsigned short * flags2 = (unsigned short *) malloc (sizeof(unsigned short) * (al+1));
+ ((captype == ALLCAP) && (flagslen != 0))) &&
+ !((flagslen != 0) && TESTAFF(flags, forbiddenword, flagslen))) {
+ unsigned short * flags2 = (unsigned short *) malloc (sizeof(unsigned short) * (flagslen+1));
if (!flags2) return 1;
- if (al) memcpy(flags2, flags, al * sizeof(unsigned short));
- flags2[al] = ONLYUPCASEFLAG;
+ if (flagslen) memcpy(flags2, flags, flagslen * sizeof(unsigned short));
+ flags2[flagslen] = ONLYUPCASEFLAG;
if (utf8) {
char st[BUFSIZE];
w_char w[BUFSIZE];
@@ -219,11 +226,11 @@ int HashMgr::add_hidden_capitalized_word(char * word, int wbl, int wcl,
mkallsmall_utf(w, wlen, langnum);
mkallcap_utf(w, 1, langnum);
u16_u8(st, BUFSIZE, w, wlen);
- return add_word(st,wbl,wcl,flags2,al+1,dp, true);
+ return add_word(st,wbl,wcl,flags2,flagslen+1,dp, true);
} else {
mkallsmall(word, csconv);
mkinitcap(word, csconv);
- return add_word(word,wbl,wcl,flags2,al+1,dp, true);
+ return add_word(word,wbl,wcl,flags2,flagslen+1,dp, true);
}
}
return 0;
@@ -353,8 +360,8 @@ int HashMgr::load_tables(const char * tpath, const char * key)
if (dict == NULL) return 1;
// first read the first line of file to get hash table size */
- if (!(ts = dict->getline())) {
- HUNSPELL_WARNING(stderr, "error: empty dic file\n");
+ if ((ts = dict->getline()) == NULL) {
+ HUNSPELL_WARNING(stderr, "error: empty dic file %s\n", tpath);
delete dict;
return 2;
}
@@ -367,30 +374,32 @@ int HashMgr::load_tables(const char * tpath, const char * key)
}
tablesize = atoi(ts);
- if (tablesize == 0) {
+
+ int nExtra = 5 + USERWORD;
+
+ if (tablesize <= 0 || (tablesize >= (std::numeric_limits<int>::max() - 1 - nExtra) / int(sizeof(struct hentry *)))) {
HUNSPELL_WARNING(stderr, "error: line 1: missing or bad word count in the dic file\n");
delete dict;
return 4;
}
- tablesize = tablesize + 5 + USERWORD;
- if ((tablesize %2) == 0) tablesize++;
+ tablesize += nExtra;
+ if ((tablesize % 2) == 0) tablesize++;
// allocate the hash table
- tableptr = (struct hentry **) malloc(tablesize * sizeof(struct hentry *));
+ tableptr = (struct hentry **) calloc(tablesize, sizeof(struct hentry *));
if (! tableptr) {
delete dict;
return 3;
}
- for (int i=0; i<tablesize; i++) tableptr[i] = NULL;
// loop through all words on much list and add to hash
// table and create word and affix strings
- while ((ts = dict->getline())) {
+ while ((ts = dict->getline()) != NULL) {
mychomp(ts);
// split each line into word and morphological description
dp = ts;
- while ((dp = strchr(dp, ':'))) {
+ while ((dp = strchr(dp, ':')) != NULL) {
if ((dp > ts + 3) && (*(dp - 3) == ' ' || *(dp - 3) == '\t')) {
for (dp -= 4; dp >= ts && (*dp == ' ' || *dp == '\t'); dp--);
if (dp < ts) { // missing word
@@ -606,7 +615,7 @@ int HashMgr::load_config(const char * affpath, const char * key)
// read in each line ignoring any that do not
// start with a known line type indicator
- while ((line = afflst->getline())) {
+ while ((line = afflst->getline()) != NULL) {
mychomp(line);
/* remove byte order mark */
@@ -746,7 +755,7 @@ int HashMgr::parse_aliasf(char * line, FileMgr * af)
/* now parse the numaliasf lines to read in the remainder of the table */
char * nl;
for (int j=0; j < numaliasf; j++) {
- if (!(nl = af->getline())) return 1;
+ if ((nl = af->getline()) == NULL) return 1;
mychomp(nl);
tp = nl;
i = 0;
@@ -853,7 +862,7 @@ int HashMgr::parse_aliasm(char * line, FileMgr * af)
/* now parse the numaliasm lines to read in the remainder of the table */
char * nl = line;
for (int j=0; j < numaliasm; j++) {
- if (!(nl = af->getline())) return 1;
+ if ((nl = af->getline()) == NULL) return 1;
mychomp(nl);
tp = nl;
i = 0;