/*
* symmetric patMatch that checks if text is shorter than pattern
* used for artist checking in addNewTitle
*/
static int checkSim( const char *text, const char *pat ) {
if( strlen( text ) < strlen( pat ) ) {
return patMatch( pat, text );
}
else {
return patMatch( text, pat );
}
}
/*
* checks if a title entry 'title' matches the search term 'pat'
* the test is driven by the first two characters in the
* search term. The first character gives the range (taLgd(p))
* the second character notes if the search should be
* exact or fuzzy (=*)
*/
static int matchTitle( mptitle *title, const char* pat ) {
int fuzzy=0;
char loname[1024];
char *lopat;
int res=0;
if( ( '=' == pat[1] ) || ( '*' == pat[1] ) ) {
if( '*' == pat[1] ) {
fuzzy=1;
}
switch( pat[0] ) {
case 't':
strltcpy( loname, title->title, 1024 );
break;
case 'a':
strltcpy( loname, title->artist, 1024 );
break;
case 'l':
strltcpy( loname, title->album, 1024 );
break;
case 'g':
strltcpy( loname, title->genre, 1024 );
break;
case 'd':
strltcpy( loname, title->display, 1024 );
break;
case 'p': /* @obsolete! */
strltcpy( loname, title->path, 1024 );
break;
default:
addMessage( 0, "Unknown range %c in %s!", pat[0], pat );
addMessage( 0, "Using display instead" );
strltcpy( loname, title->display, 1024 );
break;
}
}
else {
strltcpy( loname, title->display, 1024 );
}
if( fuzzy ) {
res=patMatch( loname, pat+2 );
}
else {
lopat=(char*)falloc( strlen( pat+1 ), 1 );
strltcpy( lopat, pat+2, strlen( pat+1 ) );
res=( strstr( loname, lopat ) != NULL );
free( lopat );
}
return res;
}
/*
* matches term with pattern in search
*/
static int isMatch( const char *term, const char *pat, const int fuzzy ) {
char loterm[MAXPATHLEN];
if( fuzzy ){
return patMatch( term, pat);
}
else {
strltcpy( loterm, term, MAXPATHLEN );
return ( strstr( loterm, pat ) != NULL );
}
}
int main()
{
char text[] = "ABABDABACDABABCABABABABCABAB";
char patt[] = "ABABCABAZ";
int tLen = strlen(text);
int pLen = strlen(patt);
patMatch(text, patt, tLen, pLen);
return 0;
}
struct nameOff *scanIntronFile(char *preIntronQ, char *startIntronQ,
char *endIntronQ, char *postIntronQ, boolean invert)
{
char intronFileName[600];
FILE *f;
char lineBuf[4*1024];
char *words[4*128];
int wordCount;
int lineCount = 0;
int preLenQ = strlen(preIntronQ);
int startLenQ = strlen(startIntronQ);
int endLenQ = strlen(endIntronQ);
int postLenQ = strlen(postIntronQ);
char *preIntronF, *startIntronF, *endIntronF, *postIntronF;
int preLenF, startLenF, endLenF, postLenF;
int preIx = 6, startIx = 7, endIx =8, postIx = 9;
struct nameOff *list = NULL, *el;
boolean addIt;
int i;
if (preLenQ > 25 || postLenQ > 25 || startLenQ > 40 || endLenQ > 40)
{
errAbort("Can only handle queries up to 25 bases on either side of the intron "
"and 40 bases inside the intron.");
}
sprintf(intronFileName, "%s%s", wormCdnaDir(), "introns.txt");
f = mustOpen(intronFileName, "r");
while (fgets(lineBuf, sizeof(lineBuf), f) != NULL)
{
++lineCount;
wordCount = chopByWhite(lineBuf, words, ArraySize(words));
if (wordCount == ArraySize(words))
{
warn("May have truncated end of line %d of %s",
lineCount, intronFileName);
}
if (wordCount == 0)
continue;
if (wordCount < 11)
errAbort("Unexpected short line %d of %s", lineCount, intronFileName);
preIntronF = words[preIx];
startIntronF = words[startIx];
endIntronF = words[endIx];
postIntronF = words[postIx];
preLenF = strlen(preIntronF);
startLenF = strlen(startIntronF);
endLenF = strlen(endIntronF);
postLenF = strlen(postIntronF);
addIt = FALSE;
if ( ( preLenQ == 0 || patMatch(preIntronQ, preIntronF+preLenF-preLenQ+countSpecial(preIntronQ), preLenQ))
&& (startLenQ == 0 || patMatch(startIntronQ, startIntronF, startLenQ))
&& ( endLenQ == 0 || patMatch(endIntronQ, endIntronF+endLenF-endLenQ+countSpecial(endIntronQ), endLenQ))
&& ( postLenQ == 0 || patMatch(postIntronQ, postIntronF, postLenQ)) )
{
addIt = TRUE;
}
if (invert)
addIt = !addIt;
if (addIt)
{
addIntronToHistogram(preIntronF+preLenF, startIntronF, endIntronF+endLenF, postIntronF);
AllocVar(el);
el->chrom = cloneString(words[1]);
el->name = cloneString(words[5]);
el->start = atoi(words[2]);
el->end = atoi(words[3]);
el->cdnaCount = atoi(words[0]);
memcpy(el->startI, startIntronF, 2);
memcpy(el->endI, endIntronF + endLenF - 2, 2);
assert(wordCount == el->cdnaCount + 10);
for (i=10; i<wordCount; ++i)
{
struct slName *name = newSlName(words[i]);
slAddHead(&el->cdnaNames, name);
}
slReverse(&el->cdnaNames);
assert(slCount(el->cdnaNames) == el->cdnaCount);
slAddHead(&list, el);
}
}
fclose(f);
slSort(&list, cmpCounts);
return list;
}
