Date: Sun, 10 Aug 2008 08:47:09 GMT From: Gabor Kovesdan <gabor@FreeBSD.org> To: Perforce Change Reviews <perforce@freebsd.org> Subject: PERFORCE change 147061 for review Message-ID: <200808100847.m7A8l9BD063387@repoman.freebsd.org>
next in thread | raw e-mail | index | archive | help
http://perforce.freebsd.org/chv.cgi?CH=147061 Change 147061 by gabor@gabor_server on 2008/08/10 08:46:33 - Cut out -w code. It is handled in util.c independently from the searching method. Affected files ... .. //depot/projects/soc2008/gabor_textproc/grep/fastgrep.c#2 edit Differences ... ==== //depot/projects/soc2008/gabor_textproc/grep/fastgrep.c#2 (text+ko) ==== @@ -52,7 +52,6 @@ fg->patternLen = strlen(pattern); fg->bol = 0; fg->eol = 0; - fg->wmatch = wflag; fg->reversedSearch = 0; /* @@ -101,7 +100,6 @@ fg->patternLen = strlen(pattern); fg->bol = 0; fg->eol = 0; - fg->wmatch = 0; fg->reversedSearch = 0; /* Remove end-of-line character ('$'). */ @@ -118,26 +116,13 @@ fg->patternLen--; } - /* Remove enclosing [[:<:]] and [[:>:]] (word match). */ - if (wflag) { - /* basic re's use \( \), extended re's ( ) */ - int extra = (grepbehave == GREP_EXTENDED) ? 1 : 2; - fg->patternLen -= 14 + 2 * extra; - fg->wmatch = 7 + extra; - } else if (fg->patternLen >= 14 && - strncmp(pattern + fg->bol, "[[:<:]]", 7) == 0 && - strncmp(pattern + fg->bol + fg->patternLen - 7, "[[:>:]]", 7) == 0) { - fg->patternLen -= 14; - fg->wmatch = 7; - } - /* * Copy pattern minus '^' and '$' characters as well as word * match character classes at the beginning and ending of the * string respectively. */ fg->pattern = grep_malloc(fg->patternLen + 1); - memcpy(fg->pattern, pattern + bol + fg->wmatch, fg->patternLen); + memcpy(fg->pattern, pattern + bol, fg->patternLen); fg->pattern[fg->patternLen] = '\0'; /* Look for ways to cheat...er...avoid the full regex engine. */ @@ -230,18 +215,6 @@ return (0); } -/* - * Word boundaries using regular expressions are defined as the point - * of transition from a non-word char to a word char, or vice versa. - * This means that grep -w +a and grep -w a+ never match anything, - * because they lack a starting or ending transition, but grep -w a+b - * does match a line containing a+b. - */ -#define isword(x) (isalnum(x) || (x) == '_') -#define wmatch(d, l, s, e) \ - ((s == 0 || !isword(d[s-1])) && (e == l || !isword(d[e])) && \ - e > s && isword(d[s]) && isword(d[e-1])) - int grep_search(fastgrep_t *fg, unsigned char *data, size_t dataLen, regmatch_t *pmatch) { @@ -270,9 +243,6 @@ fg->patternLen) == -1) { pmatch->rm_so = j; pmatch->rm_eo = j + fg->patternLen; - if (!fg->wmatch || wmatch(data, dataLen, - pmatch->rm_so, pmatch->rm_eo)) - rtrnVal = 0; } } } else if (fg->reversedSearch) { @@ -283,11 +253,6 @@ fg->patternLen) == -1) { pmatch->rm_so = j - fg->patternLen; pmatch->rm_eo = j; - if (!fg->wmatch || wmatch(data, dataLen, - pmatch->rm_so, pmatch->rm_eo)) { - rtrnVal = 0; - break; - } } /* Shift if within bounds, otherwise, we are done. */ if (j == fg->patternLen) @@ -301,12 +266,6 @@ if (grep_cmp(fg->pattern, data + j, fg->patternLen) == -1) { pmatch->rm_so = j; pmatch->rm_eo = j + fg->patternLen; - if (fg->patternLen == 0 || !fg->wmatch || - wmatch(data, dataLen, pmatch->rm_so, - pmatch->rm_eo)) { - rtrnVal = 0; - break; - } } /* Shift if within bounds, otherwise, we are done. */
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200808100847.m7A8l9BD063387>