void init_header(INDEXDATAHEADER *header)
{
header->lenwordchars=header->lenbeginchars=header->lenendchars=header->lenignorelastchar=header->lenignorefirstchar=header->lenbumpposchars=MAXCHARDEFINED;
header->wordchars = (char *)emalloc(header->lenwordchars + 1);
header->wordchars = SafeStrCopy(header->wordchars,WORDCHARS,&header->lenwordchars);
sortstring(header->wordchars); /* Sort chars and remove dups */
makelookuptable(header->wordchars,header->wordcharslookuptable);
header->beginchars = (char *)emalloc(header->lenbeginchars + 1);
header->beginchars = SafeStrCopy(header->beginchars,BEGINCHARS,&header->lenbeginchars);
sortstring(header->beginchars); /* Sort chars and remove dups */
makelookuptable(header->beginchars,header->begincharslookuptable);
header->endchars = (char *)emalloc(header->lenendchars + 1);
header->endchars = SafeStrCopy(header->endchars,ENDCHARS,&header->lenendchars);
sortstring(header->endchars); /* Sort chars and remove dups */
makelookuptable(header->endchars,header->endcharslookuptable);
header->ignorelastchar = (char *)emalloc(header->lenignorelastchar + 1);
header->ignorelastchar = SafeStrCopy(header->ignorelastchar,IGNORELASTCHAR,&header->lenignorelastchar);
sortstring(header->ignorelastchar); /* Sort chars and remove dups */
makelookuptable(header->ignorelastchar,header->ignorelastcharlookuptable);
header->ignorefirstchar = (char *)emalloc(header->lenignorefirstchar + 1);
header->ignorefirstchar = SafeStrCopy(header->ignorefirstchar,IGNOREFIRSTCHAR,&header->lenignorefirstchar);
sortstring(header->ignorefirstchar); /* Sort chars and remove dups */
makelookuptable(header->ignorefirstchar,header->ignorefirstcharlookuptable);
header->bumpposchars = (char *)emalloc(header->lenbumpposchars + 1);
header->bumpposchars[0]='\0';
header->lenindexedon=header->lensavedasheader=header->lenindexn=header->lenindexd=header->lenindexp=header->lenindexa=MAXSTRLEN;
header->indexn = (char *)emalloc(header->lenindexn + 1);header->indexn[0]='\0';
header->indexd = (char *)emalloc(header->lenindexd + 1);header->indexd[0]='\0';
header->indexp = (char *)emalloc(header->lenindexp + 1);header->indexp[0]='\0';
header->indexa = (char *)emalloc(header->lenindexa + 1);header->indexa[0]='\0';
header->savedasheader = (char *)emalloc(header->lensavedasheader + 1);header->savedasheader[0]='\0';
header->indexedon = (char *)emalloc(header->lenindexedon + 1);header->indexedon[0]='\0';
header->ignoreTotalWordCountWhenRanking = 1;
header->minwordlimit = MINWORDLIMIT;
header->maxwordlimit = MAXWORDLIMIT;
makelookuptable("",header->bumpposcharslookuptable);
BuildTranslateChars(header->translatecharslookuptable,(unsigned char *)"",(unsigned char *)"");
/* this is to ignore numbers */
header->numberchars_used_flag = 0; /* not used by default*/
/* initialize the stemmer/fuzzy structure to None */
header->fuzzy_data = set_fuzzy_mode( header->fuzzy_data, "None" );
}
size_t SafeStrLower( _In_ LPCWSTR wszSrc, _In_ size_t cchSrc, _Out_ LPWSTR wszDest, _In_ size_t cchDest)
{
if (cchSrc == (size_t)-1)
cchSrc = wcslen(wszSrc);
if (cchSrc >= cchDest) {
SetLastError(ERROR_FILENAME_EXCED_RANGE);
return 0;
}
SafeStrCopy(wszSrc, cchSrc, wszDest, cchDest);
_wcslwr_s((WCHAR*)wszDest, cchDest);
return wcslen(wszDest);
}
void read_header(SWISH *sw, INDEXDATAHEADER *header, void *DB)
{
int id,
len;
unsigned long tmp, tmp1, tmp2, tmp3, tmp4;
unsigned char *buffer;
DB_InitReadHeader(sw, DB);
DB_ReadHeaderData(sw, &id,&buffer,&len,DB);
while (id)
{
switch (id)
{
case INDEXHEADER_ID:
case INDEXVERSION_ID:
case MERGED_ID:
case DOCPROPENHEADER_ID:
break;
case WORDCHARSHEADER_ID:
header->wordchars = SafeStrCopy(header->wordchars, (char *)buffer, &header->lenwordchars);
sortstring(header->wordchars);
makelookuptable(header->wordchars, header->wordcharslookuptable);
break;
case BEGINCHARSHEADER_ID:
header->beginchars = SafeStrCopy(header->beginchars, (char *)buffer, &header->lenbeginchars);
sortstring(header->beginchars);
makelookuptable(header->beginchars, header->begincharslookuptable);
break;
case ENDCHARSHEADER_ID:
header->endchars = SafeStrCopy(header->endchars, (char *)buffer, &header->lenendchars);
sortstring(header->endchars);
makelookuptable(header->endchars, header->endcharslookuptable);
break;
case IGNOREFIRSTCHARHEADER_ID:
header->ignorefirstchar = SafeStrCopy(header->ignorefirstchar, (char *)buffer, &header->lenignorefirstchar);
sortstring(header->ignorefirstchar);
makelookuptable(header->ignorefirstchar, header->ignorefirstcharlookuptable);
break;
case IGNORELASTCHARHEADER_ID:
header->ignorelastchar = SafeStrCopy(header->ignorelastchar, (char *)buffer, &header->lenignorelastchar);
sortstring(header->ignorelastchar);
makelookuptable(header->ignorelastchar, header->ignorelastcharlookuptable);
break;
/* replaced by fuzzy_mode Aug 20, 2002
case STEMMINGHEADER_ID:
parse_int_from_buffer(tmp,buffer);
header-> = tmp;
break;
case SOUNDEXHEADER_ID:
parse_int_from_buffer(tmp,buffer);
header->applySoundexRules = tmp;
break;
*/
case FUZZYMODEHEADER_ID:
parse_int_from_buffer(tmp,buffer);
header->fuzzy_data = get_fuzzy_mode(header->fuzzy_data, tmp);
break;
case IGNORETOTALWORDCOUNTWHENRANKING_ID:
parse_int_from_buffer(tmp,buffer);
header->ignoreTotalWordCountWhenRanking = tmp;
break;
case MINWORDLIMHEADER_ID:
parse_int_from_buffer(tmp,buffer);
header->minwordlimit = tmp;
break;
case MAXWORDLIMHEADER_ID:
parse_int_from_buffer(tmp,buffer);
header->maxwordlimit = tmp;
break;
case SAVEDASHEADER_ID:
header->savedasheader = SafeStrCopy(header->savedasheader, (char *)buffer, &header->lensavedasheader);
break;
case NAMEHEADER_ID:
header->indexn = SafeStrCopy(header->indexn, (char *)buffer, &header->lenindexn);
break;
case DESCRIPTIONHEADER_ID:
header->indexd = SafeStrCopy(header->indexd, (char *)buffer, &header->lenindexd);
break;
case POINTERHEADER_ID:
header->indexp = SafeStrCopy(header->indexp, (char *)buffer, &header->lenindexp);
break;
case MAINTAINEDBYHEADER_ID:
header->indexa = SafeStrCopy(header->indexa, (char *)buffer, &header->lenindexa);
break;
case INDEXEDONHEADER_ID:
header->indexedon = SafeStrCopy(header->indexedon, (char *)buffer, &header->lenindexedon);
break;
case COUNTSHEADER_ID:
parse_int4_from_buffer(tmp1,tmp2,tmp3,tmp4,buffer);
header->totalwords = tmp1;
header->totalfiles = tmp2;
header->total_word_positions = tmp3;
header->removedfiles = tmp4;
break;
/* removed due to patents problems
case FILEINFOCOMPRESSION_ID:
ReadHeaderInt(itmp, fp);
header->applyFileInfoCompression = itmp;
break;
*/
case TRANSLATECHARTABLE_ID:
parse_integer_table_from_buffer(header->translatecharslookuptable, sizeof(header->translatecharslookuptable) / sizeof(int), (char *)buffer);
break;
case STOPWORDS_ID:
load_word_hash_from_buffer(&header->hashstoplist, (char *)buffer);
break;
case METANAMES_ID:
parse_MetaNames_from_buffer(header, (char *)buffer);
break;
case BUZZWORDS_ID:
load_word_hash_from_buffer(&header->hashbuzzwordlist, (char *)buffer);
break;
#ifndef USE_BTREE
case TOTALWORDSPERFILE_ID:
if ( !header->ignoreTotalWordCountWhenRanking )
{
header->TotalWordsPerFile = emalloc( header->totalfiles * sizeof(int) );
parse_integer_table_from_buffer(header->TotalWordsPerFile, header->totalfiles, (char *)buffer);
}
break;
#endif
default:
progerr("Severe index error in header");
break;
}
efree(buffer);
DB_ReadHeaderData(sw, &id,&buffer,&len,DB);
}
DB_EndReadHeader(sw, DB);
}
void getdefaults(SWISH * sw, char *conffile, int *hasdir, int *hasindex, int hasverbose)
{
int i,
gotdir,
gotindex;
char *line = NULL;
FILE *fp;
int linenumber = 0;
int baddirective = 0;
StringList *sl;
IndexFILE *indexf = NULL;
unsigned char *StringValue = NULL;
struct swline *tmplist;
char *w0;
gotdir = gotindex = 0;
if ((fp = fopen(conffile, F_READ_TEXT)) == NULL || !isfile(conffile))
progerrno("Couldn't open the configuration file '%s': ", conffile);
if ( sw->verbose >= 2 )
printf("Parsing config file '%s'\n", conffile );
/* Init default index file */
addindexfile(sw, INDEXFILE);
indexf = sw->indexlist;
sl = NULL;
while ( !feof( fp ) )
{
/* Free previous line */
if ( line )
efree( line );
/* Read a line */
line = read_line_from_file( &linenumber, fp );
if ( sl )
freeStringList(sl);
/* Parse line */
if (!(sl = parse_line(line)))
continue;
if (!sl->n)
continue;
w0 = sl->word[0]; /* Config Direct. = 1. word */
if (w0[0] == '#')
continue; /* comment */
if (strcasecmp(w0, "IndexDir") == 0)
{
if (sl->n > 1)
{
if (!*hasdir)
{
gotdir = 1;
grabCmdOptions(sl, 1, &sw->dirlist);
}
}
else
progerr("%s: requires at least one value", w0);
continue;
}
if (strcasecmp(w0, "IncludeConfigFile") == 0)
{
if (sl->n == 2)
{
normalize_path( sl->word[1] );
getdefaults(sw, sl->word[1], hasdir, hasindex, hasverbose);
}
else
progerr("%s: requires one value", w0);
continue;
}
if (strcasecmp(w0, "NoContents") == 0)
{
if (sl->n > 1)
{
grabCmdOptions(sl, 1, &sw->nocontentslist);
}
else
progerr("%s: requires at least one value", w0);
continue;
}
if (strcasecmp(w0, "IndexFile") == 0)
{
if (!(*hasindex))
{
if (sl->n == 2)
{
gotindex = 1;
if (indexf->line)
efree(indexf->line);
indexf->line = estrdup(sl->word[1]);
normalize_path( indexf->line );
}
else
progerr("%s: requires one value", w0);
}
continue;
}
if (strcasecmp(w0, "IndexReport") == 0)
{
if (sl->n == 2)
{
if (!hasverbose)
sw->verbose = read_integer( sl->word[1], w0, 0, 4 );
}
else
progerr("%s: requires one value", w0);
continue;
}
if (strcasecmp(w0, "ParserWarnLevel") == 0)
{
if (sl->n == 2)
sw->parser_warn_level = read_integer( sl->word[1], w0, 0, 9 );
else
progerr("%s: requires one value", w0);
continue;
}
if (strcasecmp(w0, "obeyRobotsNoIndex") == 0)
{
sw->obeyRobotsNoIndex = getYesNoOrAbort(sl, 1, 1);
continue;
}
if (strcasecmp(w0, "AbsoluteLinks") == 0)
{
sw->AbsoluteLinks = getYesNoOrAbort(sl, 1, 1);
continue;
}
if (strcasecmp(w0, "MinWordLimit") == 0)
{
if (sl->n == 2)
{
indexf->header.minwordlimit = read_integer( sl->word[1], w0, 0, INT_MAX );
}
else
progerr("%s: requires one value", w0);
continue;
}
if (strcasecmp(w0, "MaxWordLimit") == 0)
{
if (sl->n == 2)
{
indexf->header.maxwordlimit = read_integer( sl->word[1], w0, 0, INT_MAX );
}
else
progerr("%s: requires one value", w0);
continue;
}
if (strcasecmp(w0, "IndexComments") == 0)
{
sw->indexComments = getYesNoOrAbort(sl, 1, 1);
continue;
}
if (strcasecmp(w0, "IgnoreNumberChars") == 0)
{
if (sl->n == 2)
{
indexf->header.numberchars = SafeStrCopy(indexf->header.numberchars, sl->word[1], &indexf->header.lennumberchars);
sortstring(indexf->header.numberchars);
makelookuptable(indexf->header.numberchars, indexf->header.numbercharslookuptable);
indexf->header.numberchars_used_flag = 1; /* Flag that it is used */
}
else
progerr("%s: requires one value (a set of characters)", w0);
continue;
}
if (strcasecmp(w0, "WordCharacters") == 0)
{
if (sl->n == 2)
{
indexf->header.wordchars = SafeStrCopy(indexf->header.wordchars, sl->word[1], &indexf->header.lenwordchars);
sortstring(indexf->header.wordchars);
makelookuptable(indexf->header.wordchars, indexf->header.wordcharslookuptable);
}
else
progerr("%s: requires one value", w0);
continue;
}
if (strcasecmp(w0, "BeginCharacters") == 0)
{
if (sl->n == 2)
{
indexf->header.beginchars = SafeStrCopy(indexf->header.beginchars, sl->word[1], &indexf->header.lenbeginchars);
sortstring(indexf->header.beginchars);
makelookuptable(indexf->header.beginchars, indexf->header.begincharslookuptable);
}
else
progerr("%s: requires one value", w0);
continue;
}
if (strcasecmp(w0, "EndCharacters") == 0)
{
if (sl->n == 2)
{
indexf->header.endchars = SafeStrCopy(indexf->header.endchars, sl->word[1], &indexf->header.lenendchars);
sortstring(indexf->header.endchars);
makelookuptable(indexf->header.endchars, indexf->header.endcharslookuptable);
}
else
progerr("%s: requires one value", w0);
continue;
}
if (strcasecmp(w0, "IgnoreLastChar") == 0)
{
if (sl->n == 2)
{
indexf->header.ignorelastchar = SafeStrCopy(indexf->header.ignorelastchar, sl->word[1], &indexf->header.lenignorelastchar);
sortstring(indexf->header.ignorelastchar);
makelookuptable(indexf->header.ignorelastchar, indexf->header.ignorelastcharlookuptable);
} /* Do nothing */
/* else progerr("%s: requires one value",w0); */
continue;
}
if (strcasecmp(w0, "IgnoreFirstChar") == 0)
{
if (sl->n == 2)
{
indexf->header.ignorefirstchar = SafeStrCopy(indexf->header.ignorefirstchar, sl->word[1], &indexf->header.lenignorefirstchar);
sortstring(indexf->header.ignorefirstchar);
makelookuptable(indexf->header.ignorefirstchar, indexf->header.ignorefirstcharlookuptable);
} /* Do nothing */
/* else progerr("%s: requires one value",w0); */
continue;
}
if (strcasecmp(w0, "ReplaceRules") == 0)
{
if (sl->n > 2)
Build_ReplaceRules( w0, sl->word, &sw->replaceRegexps );
else
progerr("%s: requires at least two values", w0);
continue;
}
if (strcasecmp(w0, "IndexName") == 0)
{
if (sl->n > 1)
{
StringValue = StringListToString(sl, 1);
indexf->header.indexn = SafeStrCopy(indexf->header.indexn, (char *)StringValue, &indexf->header.lenindexn);
efree(StringValue);
}
else
progerr("%s: requires a value", w0);
continue;
}
if (strcasecmp(w0, "IndexDescription") == 0)
{
if (sl->n > 1)
{
StringValue = StringListToString(sl, 1);
indexf->header.indexd = SafeStrCopy(indexf->header.indexd, (char *)StringValue, &indexf->header.lenindexd);
efree(StringValue);
}
else
progerr("%s: requires a value", w0);
continue;
}
if (strcasecmp(w0, "IndexPointer") == 0)
{
if (sl->n > 1)
{
StringValue = StringListToString(sl, 1);
indexf->header.indexp = SafeStrCopy(indexf->header.indexp, (char *)StringValue, &indexf->header.lenindexp);
efree(StringValue);
}
else
progerr("%s: requires a value", w0);
continue;
}
if (strcasecmp(w0, "IndexAdmin") == 0)
{
if (sl->n > 1)
{
StringValue = StringListToString(sl, 1);
indexf->header.indexa = SafeStrCopy(indexf->header.indexa, (char *)StringValue, &indexf->header.lenindexa);
efree(StringValue);
}
else
progerr("%s: requires one value", w0);
continue;
}
if (strcasecmp(w0, "UseStemming") == 0)
{
progwarn("UseStemming is deprecated. See FuzzyIndexingMode in the docs");
if ( getYesNoOrAbort(sl, 1, 1) )
fuzzy_or_die( indexf, "Stemming_en" );
continue;
}
if (strcasecmp(w0, "UseSoundex") == 0)
{
if ( getYesNoOrAbort(sl, 1, 1) )
fuzzy_or_die( indexf, "Soundex" );
continue;
}
if (strcasecmp(w0, "FuzzyIndexingMode") == 0)
{
if (sl->n != 2)
progerr("%s: requires one value", w0);
fuzzy_or_die( indexf, sl->word[1] );
continue;
}
if (strcasecmp(w0, "IgnoreTotalWordCountWhenRanking") == 0)
{
indexf->header.ignoreTotalWordCountWhenRanking = getYesNoOrAbort(sl, 1, 1);
continue;
}
if (strcasecmp(w0, "TranslateCharacters") == 0)
{
if (sl->n >= 2)
{
if (!BuildTranslateChars(indexf->header.translatecharslookuptable, (unsigned char *)sl->word[1], (unsigned char *)sl->word[2]))
{
progerr("%s: requires two values (same length) or one translation rule", w0);
}
}
continue;
}
if (strcasecmp(w0, "ExtractPath") == 0)
{
struct metaEntry *m;
char **words;
if (sl->n < 4)
progerr("%s: requires at least three values: metaname expression type and a expression/strings", w0);
if ( !( m = getMetaNameByName( &indexf->header, sl->word[1])) )
m = addMetaEntry(&indexf->header, sl->word[1], META_INDEX, 0);
words = sl->word;
words++; /* past metaname */
add_ExtractPath( w0, sw, m, words );
continue;
}
if (strcasecmp(w0, "ExtractPathDefault") == 0)
{
struct metaEntry *m;
if (sl->n != 3)
progerr("%s: requires two values: metaname default_value", w0);
if ( !( m = getMetaNameByName( &indexf->header, sl->word[1])) )
m = addMetaEntry(&indexf->header, sl->word[1], META_INDEX, 0);
if ( m->extractpath_default )
progerr("%s already defined for meta '%s' as '%s'", w0, m->metaName, m->extractpath_default );
m->extractpath_default = estrdup( sl->word[2] );
continue;
}
if (strcasecmp(w0, "MetaNames") == 0)
{
if (sl->n <= 1)
progerr("%s: requires at least one value", w0);
for (i = 1; i < sl->n; i++)
{
if ( getMetaNameByName( &indexf->header, sl->word[i]) )
progerr("%s - name '%s' is already a MetaName", w0, sl->word[i] );
addMetaEntry(&indexf->header, sl->word[i], META_INDEX, 0);
}
continue;
}
if (strcasecmp(w0, "MetaNameAlias") == 0)
{
struct metaEntry *meta_entry;
struct metaEntry *new_meta;
if (sl->n < 3)
progerr("%s: requires at least two values", w0);
/* Make sure first entry is not an alias */
/* Lookup entry, and do not follow alias */
if ( !(meta_entry = getMetaNameByNameNoAlias( &indexf->header, sl->word[1]) ) )
progerr("%s - name '%s' not a MetaName", w0, sl->word[1] );
if ( meta_entry->alias )
progerr("%s - name '%s' must not be an alias", w0, sl->word[1] );
for (i = 2; i < sl->n; i++)
{
if ( getMetaNameByNameNoAlias( &indexf->header, sl->word[i]) )
progerr("%s - name '%s' is already a MetaName or MetaNameAlias", w0, sl->word[i] );
new_meta = addMetaEntry(&indexf->header, sl->word[i], meta_entry->metaType, 0);
new_meta->alias = meta_entry->metaID;
}
continue;
}
/* Allow setting a bias on MetaNames */
if (strcasecmp(w0, "MetaNamesRank") == 0)
{
struct metaEntry *meta_entry;
int rank = 0;
if (sl->n < 3)
progerr("%s: requires only two or more values, a rank (integer) and a list of property names", w0);
rank = read_integer( sl->word[1], w0, -RANK_BIAS_RANGE, RANK_BIAS_RANGE ); // NOTE: if this is changed db.c must match
for (i = 2; i < sl->n; i++)
{
/* already exists? */
if ( (meta_entry = getMetaNameByNameNoAlias( &indexf->header, sl->word[i])) )
{
if ( meta_entry->alias )
progerr("Can't assign a rank to metaname '%s': it is an alias", meta_entry->metaName );
if ( meta_entry->rank_bias )
progwarn("Why are you redefining the rank of metaname '%s'?", meta_entry->metaName );
}
else
meta_entry = addMetaEntry(&indexf->header, sl->word[i], META_INDEX, 0);
meta_entry->rank_bias = rank;
}
continue;
}
/* Meta name to extract out <a href> links */
if (strcasecmp(w0, "HTMLLinksMetaName") == 0)
{
if (sl->n <= 1)
progerr("%s: requires one value", w0);
if ( !( sw->links_meta = getMetaNameByName( &indexf->header, sl->word[1]) ))
sw->links_meta = addMetaEntry(&indexf->header, sl->word[1], META_INDEX, 0);
continue;
}
/* What to do with IMG ATL tags? */
if (strcasecmp(w0, "IndexAltTagMetaName") == 0)
{
if (sl->n <= 1)
progerr("%s: requires one value", w0);
if ( strcasecmp( sl->word[1], "as-text" ) == 0)
{
sw->IndexAltTag = 1;
if ( sw->IndexAltTagMeta )
{
efree( sw->IndexAltTagMeta );
sw->IndexAltTagMeta = NULL;
}
}
else
{
sw->IndexAltTag = 1;
if ( sw->IndexAltTagMeta )
{
efree( sw->IndexAltTagMeta );
sw->IndexAltTagMeta = NULL;
}
sw->IndexAltTagMeta = estrdup( sl->word[1] );
}
continue;
}
/* Meta name to extract out <img src> links */
if (strcasecmp(w0, "ImageLinksMetaName") == 0)
{
if (sl->n <= 1)
progerr("%s: requires one value", w0);
if ( !( sw->images_meta = getMetaNameByName( &indexf->header, sl->word[1]) ))
sw->images_meta = addMetaEntry(&indexf->header, sl->word[1], META_INDEX, 0);
continue;
}
if (strcasecmp(w0, "PropCompressionLevel") == 0)
{
#ifdef HAVE_ZLIB
if (sl->n == 2)
{
sw->PropCompressionLevel = read_integer( sl->word[1], w0, 0, 9 );
}
else
progerr("%s: requires one value", w0);
#else
progwarn("%s: Swish not built with zlib support -- cannot compress", w0);
#endif
continue;
}
if (strcasecmp(w0, "PropertyNames") == 0)
{
if (sl->n <= 1)
progerr("%s: requires at least one value", w0);
for (i = 1; i < sl->n; i++)
{
if ( getPropNameByName( &indexf->header, sl->word[i]) )
progerr("%s - name '%s' is already a PropertyName", w0, sl->word[i] );
addMetaEntry(&indexf->header, sl->word[i], META_PROP|META_STRING|META_IGNORE_CASE, 0);
}
continue;
}
if (strcasecmp(w0, "PropertyNamesIgnoreCase") == 0)
{
struct metaEntry *m;
if (sl->n <= 1)
progerr("%s: requires at least one value", w0);
for (i = 1; i < sl->n; i++)
{
if ( !(m = getPropNameByName( &indexf->header, sl->word[i])) )
addMetaEntry(&indexf->header, sl->word[i], META_PROP|META_STRING|META_IGNORE_CASE, 0);
else
{
if ( !is_meta_string( m ) )
progerr("%s - name '%s' is not a STRING type of Property", w0, sl->word[i] );
m->metaType |= META_IGNORE_CASE;
}
}
continue;
}
if (strcasecmp(w0, "PropertyNamesCompareCase") == 0)
{
struct metaEntry *m;
if (sl->n <= 1)
progerr("%s: requires at least one value", w0);
for (i = 1; i < sl->n; i++)
{
if ( !(m = getPropNameByName( &indexf->header, sl->word[i])) )
addMetaEntry(&indexf->header, sl->word[i], META_PROP|META_STRING, 0);
else
{
if ( !is_meta_string( m ) )
progerr("%s - name '%s' is not a STRING type of Property", w0, sl->word[i] );
m->metaType &= ~META_IGNORE_CASE;
}
}
continue;
}
/* --- this is duplicating.. */
if (strcasecmp(w0, "PropertyNamesNoStripChars") == 0)
{
struct metaEntry *m;
if (sl->n <= 1)
progerr("%s: requires at least one value", w0);
for (i = 1; i < sl->n; i++)
{
if ( !(m = getPropNameByName( &indexf->header, sl->word[i])) )
addMetaEntry(&indexf->header, sl->word[i], META_PROP|META_STRING|META_IGNORE_CASE|META_NOSTRIP, 0);
else
{
if ( !is_meta_string( m ) )
progerr("%s - name '%s' is not a STRING type of Property", w0, sl->word[i] );
m->metaType |= META_NOSTRIP;
}
}
continue;
}
if (strcasecmp(w0, "PropertyNamesStripChars") == 0)
{
struct metaEntry *m;
if (sl->n <= 1)
progerr("%s: requires at least one value", w0);
for (i = 1; i < sl->n; i++)
{
if ( !(m = getPropNameByName( &indexf->header, sl->word[i])) )
addMetaEntry(&indexf->header, sl->word[i], META_PROP|META_STRING|META_IGNORE_CASE, 0);
else
{
if ( !is_meta_string( m ) )
progerr("%s - name '%s' is not a STRING type of Property", w0, sl->word[i] );
m->metaType &= ~META_NOSTRIP;
}
}
continue;
}
if (strcasecmp(w0, "PropertyNamesNumeric") == 0)
{
if (sl->n <= 1)
progerr("%s: requires at least one value", w0);
for (i = 1; i < sl->n; i++)
{
if ( getPropNameByName( &indexf->header, sl->word[i]) )
progerr("%s - name '%s' is already a PropertyName", w0, sl->word[i] );
addMetaEntry(&indexf->header, sl->word[i], META_PROP|META_NUMBER, 0);
}
continue;
}
if (strcasecmp(w0, "PropertyNamesDate") == 0)
{
if (sl->n <= 1)
progerr("%s: requires at least one value", w0);
for (i = 1; i < sl->n; i++)
{
if ( getPropNameByName( &indexf->header, sl->word[i]) )
progerr("%s - name '%s' is already a PropertyName", w0, sl->word[i] );
addMetaEntry(&indexf->header, sl->word[i], META_PROP|META_DATE, 0);
}
continue;
}
if (strcasecmp(w0, "PropertyNameAlias") == 0)
{
struct metaEntry *meta_entry;
struct metaEntry *new_meta;
if (sl->n < 3)
progerr("%s: requires at least two values", w0);
/* Make sure first entry is not an alias */
/* Lookup entry, and do not follow alias */
if ( !(meta_entry = getPropNameByNameNoAlias( &indexf->header, sl->word[1]) ) )
progerr("%s - name '%s' not a PropertyName", w0, sl->word[1] );
if ( meta_entry->alias )
progerr("%s - name '%s' must not be an alias", w0, sl->word[1] );
for (i = 2; i < sl->n; i++)
{
if ( getPropNameByNameNoAlias( &indexf->header, sl->word[i]) )
progerr("%s - name '%s' is already a PropertyName or PropertyNameAlias", w0, sl->word[i] );
new_meta = addMetaEntry(&indexf->header, sl->word[i], meta_entry->metaType, 0);
new_meta->alias = meta_entry->metaID;
}
continue;
}
/* This allows setting a limit on a property's string length */
// One question would be if this should set the length on the alias, or the real property. */
// If on the alias then you could really fine tune:
// PropertyNames description
// PropertyNameAlias description td h1 h2 h3
// PropertyNameMaxLength 5000 description
// PropertyNameMaxLength 100 td
// PropertyNameMaxLength 10 h1 h2 h3
// then the total length would be 5000, but each one would be limited, too. I find that hard to imagine
// it would be useful. So the current design is you can only assign to a non-alias.
if (strcasecmp(w0, "PropertyNamesMaxLength") == 0)
{
struct metaEntry *meta_entry;
int max_length = 0;
if (sl->n < 3)
progerr("%s: requires only two or more values, a length and a list of property names", w0);
max_length = read_integer( sl->word[1], w0, 0, INT_MAX );
for (i = 2; i < sl->n; i++)
{
/* already exists? */
if ( (meta_entry = getPropNameByNameNoAlias( &indexf->header, sl->word[i])) )
{
if ( meta_entry->alias )
progerr("Can't assign a length to property '%s': it is an alias", meta_entry->metaName );
if ( meta_entry->max_len )
progwarn("Why are you redefining the max length of property '%s'?", meta_entry->metaName );
if ( !is_meta_string( meta_entry ) )
progerr("%s - name '%s' is not a STRING type of Property", w0, sl->word[i] );
}
else
meta_entry = addMetaEntry(&indexf->header, sl->word[i], META_PROP|META_STRING, 0);
meta_entry->max_len = max_length;
}
continue;
}
/* Set the sort length */
if (strcasecmp(w0, "PropertyNamesSortKeyLength") == 0)
{
struct metaEntry *meta_entry;
int max_length = 0;
if (sl->n < 3)
progerr("%s: requires only two or more values, a length and a list of property names", w0);
max_length = read_integer( sl->word[1], w0, 1, INT_MAX );
for (i = 2; i < sl->n; i++)
{
/* already exists? */
if ( (meta_entry = getPropNameByNameNoAlias( &indexf->header, sl->word[i])) )
{
if ( meta_entry->alias )
progerr("Can't assign a length to property '%s': it is an alias", meta_entry->metaName );
if ( meta_entry->max_len )
progwarn("Why are you redefining the max sort key length of property '%s'?", meta_entry->metaName );
if ( !is_meta_string( meta_entry ) )
progerr("%s - name '%s' is not a STRING type of Property", w0, sl->word[i] );
}
else
meta_entry = addMetaEntry(&indexf->header, sl->word[i], META_PROP|META_STRING, 0);
meta_entry->sort_len = max_length;
}
continue;
}
/* Hashed word lists */
if ( !strcasecmp(w0, "IgnoreWords") || !strcasecmp(w0, "StopWords"))
{
word_hash_config( sl, &indexf->header.hashstoplist );
continue;
}
if (strcasecmp(w0, "BuzzWords") == 0) /* 2001-04-24 moseley */
{
word_hash_config( sl, &indexf->header.hashbuzzwordlist );
continue;
}
if (strcasecmp(w0, "UseWords") == 0)
{
word_hash_config( sl, &indexf->header.hashuselist );
continue;
}
/* IndexVerbose is supported for backwards compatibility */
if (strcasecmp(w0, "IndexVerbose") == 0)
{
sw->verbose = getYesNoOrAbort(sl, 1, 1);
if (sw->verbose)
sw->verbose = 3;
continue;
}
if (strcasecmp(w0, "IndexOnly") == 0)
{
if (sl->n > 1)
{
grabCmdOptions(sl, 1, &sw->suffixlist);
}
else
progerr("%s: requires at least one value", w0);
continue;
}
if (strcasecmp(w0, "IndexContents") == 0)
{
if (sl->n > 2)
{
struct IndexContents *ic = (struct IndexContents *) emalloc(sizeof(struct IndexContents));
ic->DocType = getDocTypeOrAbort(sl, 1);
ic->patt = NULL;
for (i = 2; i < sl->n; i++)
ic->patt = addswline(ic->patt, sl->word[i]);
if (sw->indexcontents)
ic->next = sw->indexcontents;
else
ic->next = NULL;
sw->indexcontents = ic;
}
else
progerr("%s: requires at least two values", w0);
continue;
}
/* $$$ this needs fixing */
if (strcasecmp(w0, "StoreDescription") == 0)
{
if (sl->n == 3 || sl->n == 4)
{
struct StoreDescription *sd = (struct StoreDescription *) emalloc(sizeof(struct StoreDescription));
sd->DocType = getDocTypeOrAbort(sl, 1);
sd->size = 0;
sd->field = NULL;
i = 2;
if (sl->word[i][0] == '<' && sl->word[i][strlen(sl->word[i]) - 1] == '>')
{
sl->word[i][strlen(sl->word[i]) - 1] = '\0';
sd->field = estrdup(sl->word[i] + 1);
i++;
}
if (i < sl->n && isnumstring( (unsigned char *)sl->word[i] ))
{
sd->size = read_integer( sl->word[i], w0, 0, INT_MAX );
}
if (sl->n == 3 && !sd->field && !sd->size)
progerr("%s: second parameter must be <fieldname> or a number", w0);
if (sl->n == 4 && sd->field && !sd->size)
progerr("%s: third parameter must be empty or a number", w0);
if (sw->storedescription)
sd->next = sw->storedescription;
else
sd->next = NULL;
sw->storedescription = sd;
/* Make sure there's a property name */
if ( !getPropNameByName( &indexf->header, AUTOPROPERTY_SUMMARY) )
addMetaEntry(&indexf->header, AUTOPROPERTY_SUMMARY, META_PROP|META_STRING, 0);
}
else
progerr("%s: requires two or three values", w0);
continue;
}
if (strcasecmp(w0, "DefaultContents") == 0)
{
if (sl->n == 2 )
{
sw->DefaultDocType = getDocTypeOrAbort(sl, 1);
}
else
progerr("%s: requires one value -- a parser type", w0);
continue;
}
if (strcasecmp(w0, "BumpPositionCounterCharacters") == 0)
{
if (sl->n > 1)
{
indexf->header.bumpposchars = SafeStrCopy(indexf->header.bumpposchars, sl->word[1], &indexf->header.lenbumpposchars);
sortstring(indexf->header.bumpposchars);
makelookuptable(indexf->header.bumpposchars, indexf->header.bumpposcharslookuptable);
}
else
progerr("%s: requires at least one value", w0);
continue;
}
/* #### Added UndefinedMetaTags as defined by Bill Moseley */
if (strcasecmp(w0, "UndefinedMetaTags") == 0)
{
get_undefined_meta_flags( w0, sl, &sw->UndefinedMetaTags );
if ( !sw->UndefinedMetaTags )
progerr("%s: possible values are error, ignore, index or auto", w0);
continue;
}
if (strcasecmp(w0, "UndefinedXMLAttributes") == 0)
{
get_undefined_meta_flags( w0, sl, &sw->UndefinedXMLAttributes );
continue;
}
if (strcasecmp(w0, "IgnoreMetaTags") == 0)
{
if (sl->n > 1)
{
grabCmdOptions(sl, 1, &sw->ignoremetalist);
/* Go lowercase */
for (tmplist = sw->ignoremetalist; tmplist; tmplist = tmplist->next)
(void)strtolower(tmplist->line);
}
else
progerr("%s: requires at least one value", w0);
continue;
}
if (strcasecmp(w0, "XMLClassAttributes") == 0)
{
if (sl->n > 1)
{
grabCmdOptions(sl, 1, &sw->XMLClassAttributes);
/* Go lowercase */
for (tmplist = sw->XMLClassAttributes; tmplist; tmplist = tmplist->next)
(void)strtolower(tmplist->line);
}
else
progerr("%s: requires at least one value", w0);
continue;
}
if (strcasecmp(w0, "DontBumpPositionOnStartTags") == 0)
{
if (sl->n > 1)
grabCmdOptions(sl, 1, &sw->dontbumpstarttagslist);
else
progerr("%s: requires at least one value", w0);
continue;
}
if (strcasecmp(w0, "DontBumpPositionOnEndTags") == 0)
{
if (sl->n > 1)
grabCmdOptions(sl, 1, &sw->dontbumpendtagslist);
else
progerr("%s: requires at least one value", w0);
continue;
}
if (strcasecmp(w0, "TruncateDocSize") == 0)
{ /* rasc 2001-03 */
if (sl->n == 2 && isnumstring( (unsigned char *)sl->word[1] ))
sw->truncateDocSize = atol(sl->word[1]);
else
progerr("%s: requires size parameter in bytes", w0);
continue;
}
if (strcasecmp(w0, "CompressPositions") == 0)
{
sw->compressPositions = getYesNoOrAbort(sl, 1, 1);
continue;
}
else if (configModule_Entities(sw, sl));
else if (configModule_Filter(sw, sl)); /* rasc */
else if (configModule_ResultOutput(sw, sl)); /* rasc */
else if (configModule_ResultSort(sw, sl)); /* jmruiz */
else if (configModule_Index(sw, sl)); /* jmruiz */
else if (configModule_Prog(sw, sl));
else if (!parseconfline(sw, sl))
{
printf("Bad directive on line #%d of file %s: %s\n", linenumber, conffile, line);
if ( ++baddirective > 30 )
progerr("Too many errors. Can not continue.");
}
}
freeStringList(sl);
fclose(fp);
if (baddirective)
exit(1);
if (gotdir && !(*hasdir))
*hasdir = 1;
if (gotindex && !(*hasindex))
*hasindex = 1;
}
void PrintErrorText(HRESULT hr){
// First check whether it is a UPnP error
DWORD dwRet = 0;
TCHAR tszErrorString[1024];
switch(hr){
// All the upnp error codes are from upnp.h
case UPNP_E_ROOT_ELEMENT_EXPECTED:
SafeStrCopy(
tszErrorString,
"Root Element Expected",
1024
);
break;
case UPNP_E_DEVICE_ELEMENT_EXPECTED:
SafeStrCopy(
tszErrorString,
"Device Element Expected",
1024
);
break;
case UPNP_E_SERVICE_ELEMENT_EXPECTED:
SafeStrCopy(
tszErrorString,
"Service Element Expected",
1024
);
break;
case UPNP_E_SERVICE_NODE_INCOMPLETE:
SafeStrCopy(
tszErrorString,
"Service Node Incomplete",
1024
);
break;
case UPNP_E_DEVICE_NODE_INCOMPLETE:
SafeStrCopy(
tszErrorString,
"Device Node Incomplete",
1024
);
break;
case UPNP_E_ICON_ELEMENT_EXPECTED:
SafeStrCopy(
tszErrorString,
"Icon Element Expected",
1024
);
break;
case UPNP_E_ICON_NODE_INCOMPLETE:
SafeStrCopy(
tszErrorString,
"Icon Node Incomplete",
1024
);
break;
case UPNP_E_INVALID_ACTION:
SafeStrCopy(
tszErrorString,
"Invalid Action",
1024
);
break;
case UPNP_E_INVALID_ARGUMENTS:
SafeStrCopy(
tszErrorString,
"Invalid Arguments",
1024
);
break;
case UPNP_E_OUT_OF_SYNC:
SafeStrCopy(
tszErrorString,
"Out of Sync",
1024
);
break;
case UPNP_E_ACTION_REQUEST_FAILED:
SafeStrCopy(
tszErrorString,
"Action Request Failed",
1024
);
break;
case UPNP_E_TRANSPORT_ERROR:
SafeStrCopy(
tszErrorString,
"Transport Error",
1024
);
break;
case UPNP_E_VARIABLE_VALUE_UNKNOWN:
SafeStrCopy(
tszErrorString,
"Variable Value Unknown",
1024
);
break;
case UPNP_E_INVALID_VARIABLE:
SafeStrCopy(
tszErrorString,
"Invalid Variable",
1024
);
break;
case UPNP_E_DEVICE_ERROR:
SafeStrCopy(
tszErrorString,
"Device Error",
1024
);
break;
case UPNP_E_PROTOCOL_ERROR:
SafeStrCopy(
tszErrorString,
"Protocol Error",
1024
);
break;
case UPNP_E_ERROR_PROCESSING_RESPONSE:
SafeStrCopy(
tszErrorString,
"Error Processing Response",
1024
);
break;
case UPNP_E_DEVICE_TIMEOUT:
SafeStrCopy(
tszErrorString,
"Device Timeout",
1024
);
break;
case UPNP_E_INVALID_DOCUMENT:
SafeStrCopy(
tszErrorString,
"Invalid Document",
1024
);
break;
case UPNP_E_EVENT_SUBSCRIPTION_FAILED:
SafeStrCopy(
tszErrorString,
"Event Subscription Failed",
1024
);
break;
default:
if ((hr>=UPNP_E_ACTION_SPECIFIC_BASE) &&
(hr <= UPNP_E_ACTION_SPECIFIC_MAX))
{
SafeStrCopy(
tszErrorString,
"Action Specific Error",
1024
);
break;
}
LPVOID lpMsgBuf = NULL;
dwRet = FormatMessage(
FORMAT_MESSAGE_ALLOCATE_BUFFER |
FORMAT_MESSAGE_FROM_SYSTEM |
FORMAT_MESSAGE_IGNORE_INSERTS,
NULL,
GetLastError(),
MAKELANGID(LANG_NEUTRAL, SUBLANG_DEFAULT), // Default language
(LPTSTR) &lpMsgBuf,
0,
NULL
);
if ( dwRet != 0 )
{
// Display the string.
MessageBox( NULL, (LPCTSTR)lpMsgBuf, "Error", MB_OK );
// Free the buffer.
LocalFree( lpMsgBuf );
}
return;
}
MessageBox( NULL, (LPCTSTR)tszErrorString, "Error", MB_OK );
}
