changeset 4369:d7ad1e42a368

util._matcher: speed up regexp matching. In 4babaa52badf, Benoit made a change that substantially slows matching when a big .hgignore file is in play, because it calls into the regexp matching engine potentially hundreds of times per file to be matched. I've partly rolled back his change, so that we only call into the matcher once per file, but preserved the ability to report a meaningful error message if there's a syntax error in the regexp.
author Bryan O'Sullivan <bos@serpentine.com>
date Tue, 24 Apr 2007 10:53:25 -0700
parents 6af107c742bf
children 4ddc6d374265
files mercurial/util.py
diffstat 1 files changed, 14 insertions(+), 16 deletions(-) [+]
line wrap: on
line diff
--- a/mercurial/util.py
+++ b/mercurial/util.py
@@ -469,22 +469,20 @@ def _matcher(canonroot, cwd, names, inc,
         """build a matching function from a set of patterns"""
         if not pats:
             return
-        matches = []
-        for k, p in pats:
-            try:
-                pat = '(?:%s)' % regex(k, p, tail)
-                matches.append(re.compile(pat).match)
-            except re.error:
-                if src: raise Abort("%s: invalid pattern (%s): %s" % (src, k, p))
-                else: raise Abort("invalid pattern (%s): %s" % (k, p))
-
-        def buildfn(text):
-            for m in matches:
-                r = m(text)
-                if r:
-                    return r
-
-        return buildfn
+        try:
+            pat = '(?:%s)' % '|'.join([regex(k, p, tail) for (k, p) in pats])
+            return re.compile(pat).match
+        except re.error:
+            for k, p in pats:
+                try:
+                    re.compile('(?:%s)' % regex(k, p, tail))
+                except re.error:
+                    if src:
+                        raise Abort("%s: invalid pattern (%s): %s" %
+                                    (src, k, p))
+                    else:
+                        raise Abort("invalid pattern (%s): %s" % (k, p))
+            raise Abort("invalid pattern")
 
     def globprefix(pat):
         '''return the non-glob prefix of a path, e.g. foo/* -> foo'''