Package: ssed
Version: 3.62-6.1
Severity: normal
Tags: patch

Hi,

Attached is the diff for my ssed 3.62-6.2 NMU.

-- 
Homepage: http://www.sesse.net/
diff -Nru /tmp/GZ2sljtrdc/ssed-3.62/debian/changelog 
/tmp/oucE3Awpnq/ssed-3.62/debian/changelog
--- /tmp/GZ2sljtrdc/ssed-3.62/debian/changelog  2006-11-07 23:56:49.000000000 
+0100
+++ /tmp/oucE3Awpnq/ssed-3.62/debian/changelog  2006-11-07 23:56:50.000000000 
+0100
@@ -1,3 +1,24 @@
+ssed (3.62-6.2) unstable; urgency=high
+
+  * Non-maintainer upload.
+  * Actually include the change to fix the mips/mipsel FTBFS bug, which now
+    has a bug number: Closes: #397520.
+
+ -- Steinar H. Gunderson <[EMAIL PROTECTED]>  Tue,  7 Nov 2006 23:40:19 +0100
+
+ssed (3.62-6.1) unstable; urgency=high
+
+  * Non-maintainer upload.
+  * Backport fix from PCRE upstream; in pcre/regexec.c, "start_match" was
+    never properly initialized. This lead to test suite failures on
+    mips/mipsel with optimizing builds of this file on gcc4; why it didn't on
+    any other platforms is a bit of a mystery, but with the fix it compiles
+    and passes the tests properly. This fixes an unreported FTBFS on
+    mips/mipsel, and hopefully lets a working version of ssed propagate to
+    etch, fixing at least one FTBFS there.
+
+ -- Steinar H. Gunderson <[EMAIL PROTECTED]>  Tue,  7 Nov 2006 18:24:01 +0100
+
 ssed (3.62-6) unstable; urgency=low
 
   * Fix real breakage, &size_t being passed where &int expected.
diff -Nru /tmp/GZ2sljtrdc/ssed-3.62/pcre/regexec.c 
/tmp/oucE3Awpnq/ssed-3.62/pcre/regexec.c
--- /tmp/GZ2sljtrdc/ssed-3.62/pcre/regexec.c    2004-11-06 12:37:29.000000000 
+0100
+++ /tmp/oucE3Awpnq/ssed-3.62/pcre/regexec.c    2006-11-07 23:56:50.000000000 
+0100
@@ -1737,7 +1737,7 @@
   match_data match_block;
   const uschar *start_bits = NULL;
   const uschar *bmtable = NULL;
-  const uschar *start_match;
+  const uschar *start_match = (uschar *)subject + start_offset;
   const uschar *end_subject;
   const uschar *req_char_ptr = start_match - 1;
   BOOL using_temporary_offsets = FALSE;

Reply via email to