Use camel-url-scanner instead of regex.

2002-12-10  Jeffrey Stedfast  <fejj@ximian.com>

	* camel-mime-filter-tohtml.c (html_convert): Use camel-url-scanner
	instead of regex.

svn path=/trunk/; revision=19087

3 files changed, 43 insertions, 90 deletions

diff --git a/camel/ChangeLog b/camel/ChangeLog
index 592dd9c74b..eaecfaccf7 100644
--- a/camel/ChangeLog
+++ b/camel/ChangeLog
@@ -1,3 +1,8 @@
+2002-12-10  Jeffrey Stedfast  <fejj@ximian.com>
+
+	* camel-mime-filter-tohtml.c (html_convert): Use camel-url-scanner
+	instead of regex.
+
 2002-12-09  Jeffrey Stedfast  <fejj@ximian.com>
 
 	* camel-url-scanner.c (camel_url_addrspec_end): Fixed to not be
diff --git a/camel/camel-mime-filter-tohtml.c b/camel/camel-mime-filter-tohtml.c
index db84b7f9fe..ab6cca1fd9 100644
--- a/camel/camel-mime-filter-tohtml.c
+++ b/camel/camel-mime-filter-tohtml.c
@@ -27,31 +27,32 @@
 
 #include <stdio.h>
 #include <string.h>
-#include <ctype.h>
-#include <regex.h>
 
+#include "camel-url-scanner.h"
 #include "camel-mime-filter-tohtml.h"
 
 #define d(x)
 
+#define CONVERT_WEB_URLS  CAMEL_MIME_FILTER_TOHTML_CONVERT_URLS
+#define CONVERT_ADDRSPEC  CAMEL_MIME_FILTER_TOHTML_CONVERT_ADDRESSES
 
-struct _UrlRegexPattern {
+static struct {
 	unsigned int mask;
-	char *pattern;
-	char *prefix;
-	regex_t *preg;
-	regmatch_t matches;
+	urlpattern_t pattern;
+} patterns[] = {
+	{ CONVERT_WEB_URLS, { "file://",   "",        camel_url_file_start,     camel_url_file_end     } },
+	{ CONVERT_WEB_URLS, { "ftp://",    "",        camel_url_web_start,      camel_url_web_end      } },
+	{ CONVERT_WEB_URLS, { "http://",   "",        camel_url_web_start,      camel_url_web_end      } },
+	{ CONVERT_WEB_URLS, { "https://",  "",        camel_url_web_start,      camel_url_web_end      } },
+	{ CONVERT_WEB_URLS, { "news://",   "",        camel_url_web_start,      camel_url_web_end      } },
+	{ CONVERT_WEB_URLS, { "nntp://",   "",        camel_url_web_start,      camel_url_web_end      } },
+	{ CONVERT_WEB_URLS, { "telnet://", "",        camel_url_web_start,      camel_url_web_end      } },
+	{ CONVERT_WEB_URLS, { "www.",      "http://", camel_url_web_start,      camel_url_web_end      } },
+	{ CONVERT_WEB_URLS, { "ftp.",      "ftp://",  camel_url_web_start,      camel_url_web_end      } },
+	{ CONVERT_ADDRSPEC, { "@",         "mailto:", camel_url_addrspec_start, camel_url_addrspec_end } },
 };
 
-static struct _UrlRegexPattern patterns[] = {
-	{ CAMEL_MIME_FILTER_TOHTML_CONVERT_URLS, "(news|nntp|telnet|file|ftp|http|https)://([-a-z0-9]+(:[-a-z0-9]+)?@)?[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-a-z0-9_$.+!*(),;:@%&=?/~#]*[^]'.}>\\) ,?!;:\"]?)?", "", NULL, { 0, 0 } },
-	{ CAMEL_MIME_FILTER_TOHTML_CONVERT_URLS, "www\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]'.}>\\) ,?!;:\"]?)?", "http://", NULL, { 0, 0 } },
-	{ CAMEL_MIME_FILTER_TOHTML_CONVERT_URLS, "ftp\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]'.}>\\) ,?!;:\"]?)?", "ftp://", NULL, { 0, 0 } },
-	{ CAMEL_MIME_FILTER_TOHTML_CONVERT_ADDRESSES, "([-_a-z0-9.\\+]+@[-_a-z0-9.]+)", "mailto:", NULL, { 0, 0 } }
-};
-
-#define NUM_URL_REGEX_PATTERNS (sizeof (patterns) / sizeof (patterns[0]))
-
+#define NUM_URL_PATTERNS (sizeof (patterns) / sizeof (patterns[0]))
 
 static void camel_mime_filter_tohtml_class_init (CamelMimeFilterToHTMLClass *klass);
 static void camel_mime_filter_tohtml_init       (CamelMimeFilterToHTML *filter);
@@ -83,35 +84,14 @@ static void
 camel_mime_filter_tohtml_finalize (CamelObject *obj)
 {
 	CamelMimeFilterToHTML *filter = (CamelMimeFilterToHTML *) obj;
-	int i;
 	
-	for (i = 0; i < NUM_URL_REGEX_PATTERNS; i++) {
-		if (filter->patterns[i].preg) {
-			regfree (filter->patterns[i].preg);
-			g_free (filter->patterns[i].preg);
-		}
-	}
-	
-	g_free (filter->patterns);
+	camel_url_scanner_free (filter->scanner);
 }
 
 static void
 camel_mime_filter_tohtml_init (CamelMimeFilterToHTML *filter)
 {
-	int i;
-	
-	/* FIXME: use a global set of patterns instead? */
-	filter->patterns = g_malloc (sizeof (patterns));
-	memcpy (filter->patterns, patterns, sizeof (patterns));
-	
-	for (i = 0; i < NUM_URL_REGEX_PATTERNS; i++) {
-		filter->patterns[i].preg = g_malloc (sizeof (regex_t));
-		if (regcomp (filter->patterns[i].preg, patterns[i].pattern, REG_EXTENDED) == -1) {
-			/* error building the regex_t so we can't use this pattern */
-			filter->patterns[i].preg = NULL;
-			filter->patterns[i].mask = 0;
-		}
-	}
+	filter->scanner = camel_url_scanner_new ();
 	
 	filter->flags = 0;
 	filter->colour = 0;
@@ -279,64 +259,28 @@ html_convert (CamelMimeFilter *filter, char *in, size_t inlen, size_t prespace,
 		
 #define CONVERT_URLS (CAMEL_MIME_FILTER_TOHTML_CONVERT_URLS | CAMEL_MIME_FILTER_TOHTML_CONVERT_ADDRESSES)
 		if (html->flags & CONVERT_URLS) {
-			struct _UrlRegexPattern *fmatch, *pat;
-			size_t matchlen, len;
-			regoff_t offset;
-			char *linebuf;
-			char save;
-			int i;
+						size_t matchlen, buflen, len;
+			urlmatch_t match;
 			
 			len = inptr - start;
-			linebuf = g_malloc (len + 1);
-			memcpy (linebuf, start, len);
-			linebuf[len] = '\0';
-			
-			start = linebuf;
-			save = '\0';
 			
 			do {
-				/* search for all of our patterns */
-				offset = 0;
-				fmatch = NULL;
-				for (i = 0; i < NUM_URL_REGEX_PATTERNS; i++) {
-					pat = html->patterns + i;
-					if ((html->flags & pat->mask) &&
-					    !regexec (pat->preg, start, 1, &pat->matches, 0)) {
-						if (pat->matches.rm_so < offset) {
-							*(start + offset) = save;
-							fmatch = NULL;
-						}
-						
-						if (!fmatch) {
-							fmatch = pat;
-							offset = pat->matches.rm_so;
-							
-							/* optimisation so we don't have to search the
-							   entire line buffer for the next pattern */
-							save = *(start + offset);
-							*(start + offset) = '\0';
-						}
-					}
-				}
-				
-				if (fmatch) {
-					/* restore our char */
-					*(start + offset) = save;
-					
+				if (camel_url_scanner_scan (html->scanner, start, len, &match)) {
 					/* write out anything before the first regex match */
-					outptr = writeln (filter, start, start + offset, outptr, &outend);
-					start += offset;
-					len -= offset;
+					outptr = writeln (filter, start, start + match.um_so,
+							  outptr, &outend);
 					
-#define MATCHLEN(matches) (matches.rm_eo - matches.rm_so)
-					matchlen = MATCHLEN (fmatch->matches);
+					start += match.um_so;
+					len -= match.um_so;
 					
-					i = 20 + strlen (fmatch->prefix) + matchlen + matchlen;
-					outptr = check_size (filter, outptr, &outend, i);
+					matchlen = match.um_eo - match.um_so;
+					
+					buflen = 20 + strlen (match.prefix) + matchlen + matchlen;
+					outptr = check_size (filter, outptr, &outend, buflen);
 					
 					/* write out the href tag */
 					outptr = g_stpcpy (outptr, "<a href=\"");
-					outptr = g_stpcpy (outptr, fmatch->prefix);
+					outptr = g_stpcpy (outptr, match.prefix);
 					memcpy (outptr, start, matchlen);
 					outptr += matchlen;
 					outptr = g_stpcpy (outptr, "\">");
@@ -356,8 +300,6 @@ html_convert (CamelMimeFilter *filter, char *in, size_t inlen, size_t prespace,
 					break;
 				}
 			} while (len > 0);
-			
-			g_free (linebuf);
 		} else {
 			outptr = writeln (filter, start, inptr, outptr, &outend);
 		}
@@ -448,11 +390,17 @@ CamelMimeFilter *
 camel_mime_filter_tohtml_new (guint32 flags, guint32 colour)
 {
 	CamelMimeFilterToHTML *new;
+	int i;
 	
 	new = CAMEL_MIME_FILTER_TOHTML (camel_object_new (camel_mime_filter_tohtml_get_type ()));
 	
 	new->flags = flags;
 	new->colour = colour;
 	
+	for (i = 0; i < NUM_URL_PATTERNS; i++) {
+		if (patterns[i].mask & flags)
+			camel_url_scanner_add (new->scanner, &patterns[i].pattern);
+	}
+	
 	return CAMEL_MIME_FILTER (new);
 }
diff --git a/camel/camel-mime-filter-tohtml.h b/camel/camel-mime-filter-tohtml.h
index 0dff90d651..2c1c07e5e3 100644
--- a/camel/camel-mime-filter-tohtml.h
+++ b/camel/camel-mime-filter-tohtml.h
@@ -51,7 +51,7 @@ typedef struct _CamelMimeFilterToHTML CamelMimeFilterToHTML;
 struct _CamelMimeFilterToHTML {
 	CamelMimeFilter parent;
 	
-	struct _UrlRegexPattern *patterns;
+	struct _CamelUrlScanner *scanner;
 	
 	guint32 flags;
 	guint32 colour;