I guess this is pure paranoia... but should not the return value of
mkstemp be sanity-checked, e.g. like:
int j;
sprintf(tempFilename,"%sgv_%s.%s.XX",tmpDirBuf,tmpName,tmpExt);
file_translateTildeInPath(tempFilename);
no_such_file = 1;
j = mkstemp(tempFi
package gv
notfixed 583668 1:3.6.5.90-1
fixed 583668 1:3.6.8.91-1
thanks
* Markus Steinborn [100529 16:09]:
> Just for the records: In 3.6.5.90 (upstream) the configure-script was
> broken. Commit 73bb88a65dc1c6c9dc309b60b5454d9475cfccd9 repaired the
> defect (and changed other things):
Than
Bernhard R. Link schrieb:
This bug was fixed upstream in 3.6.5.90-1, the first version
after lenny. :-(
Just for the records: In 3.6.5.90 (upstream) the configure-script was
broken. Commit 73bb88a65dc1c6c9dc309b60b5454d9475cfccd9 repaired the
defect (and changed other things):
--- a/gv/co
package gv
tags 583668 + security lenny
fixed 583668 3.6.5.90-1
thanks
This bug was fixed upstream in 3.6.5.90-1, the first version
after lenny. :-(
Attached is a simplified version (without the configure changes
as Debian has mkstemp) that should fix this in lenny.
Bernhard R. Link
---
4 matches
Mail list logo