From 949608fb1efab569db8c10fed4e7d2f6793bc144 Mon Sep 17 00:00:00 2001
From: lxoliva <lxoliva@559672b5-ba27-0410-b829-e8f1faed8b1b>
Date: Fri, 22 Jan 2010 07:19:43 +0000
Subject: [PATCH] Fix matching of false positives adjusting matchlen to RLENGTH
 + 1 ??? Add more verbosity to python and awk scripts.

git-svn-id: http://www.fsfla.org/svn/fsfla/software/linux-libre/scripts@5866 559672b5-ba27-0410-b829-e8f1faed8b1b
---
 deblob-check-awk | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/deblob-check-awk b/deblob-check-awk
index 45d6d79d3d71..1f6837e007ec 100755
--- a/deblob-check-awk
+++ b/deblob-check-awk
@@ -3339,6 +3339,7 @@ for line in sys.stdin:
             what = match.lastgroup
 
             if what == 'cblob':
+		if verbose: print 'match is a blob context'
 		pend = match.end()
                 p = match.start() + 1
                 match = rxc.search (s, p)
@@ -3348,8 +3349,10 @@ for line in sys.stdin:
             assert blob_p or what == 'falsepos'
 
             if blob_p:
+		if verbose: print 'match is a blob'
                 blobs += 1
             else:
+		if verbose: print 'match is a false positive'
                 falses += 1
             
             assert match.end(what) == match.end()
@@ -3517,23 +3520,28 @@ BEGIN {
 	firstmatchstart = RSTART + p - 1;
 	for (;;) {
 	    matchstart = RSTART + p - 1;
-	    matchlen = RLENGTH;
+	    matchlen = RLENGTH + 1;
 	    if (verbose)
 		print "found match", matchstart, matchlen;
+	    if (verbose > 1)
+		print substr (s, matchstart, matchlen);
 	    if (match (substr (s, matchstart, matchlen), /^($falsepos)$/) == 1) {
+		if (verbose) print "match is a false positive";
 		blob_p = 0;
 		falses++;
 	    } else if (match (substr (s, matchstart, matchlen), /^($cblob)$/) == 1) {
+		if (verbose) print "match is a blob context";
 		pend = p + matchlen;
 		p = matchstart + 1;
 		if (verbose) print "search after cblob starting at", p;
-		match (substr (s, p, matchlen), /^($falsepos)|^($cblob)|[^\n]*($blob)/);
+		match (substr (s, p, matchlen - 1), /^($falsepos)|^($cblob)|[^\n]*($blob)/);
 		continue;
 	    } else {
-		# Match again, to skip 
+		if (verbose) print "match is a blob";
+		# Match again, to skip the leading non-blob characters.
 		match (substr (s, matchstart), /$blob/);
 		matchstart = RSTART + matchstart - 1;
-		matchlen = RLENGTH;
+		matchlen = RLENGTH + 1;
 		blob_p = 1;
 		blobs++;
 	    }
-- 
2.31.1