zgrep -<NUM> doesn't work if <NUM> is two digits or greater

[Thomas Bushnell, BSG]

Create a gzipped file:

   $ for i in $(seq 1 100); do echo $i; done > file.txt
   $ gzip file.txt

Then:

   $ zgrep -9 17 file.txt.gz
   8
   9
   ...
   25
   26

works. But:

   $ zgrep -10 17 file.txt.gz
   gzip: 17.gz: No such file or directory

fails.

Note that "grep -9 17 file.txt" and "grep -10 17 file.txt" work just fine.

[Marcel Stimberg (marcelstimberg)]

Thank you for your bug report. I can reproduce your issue on Ubuntu 12.04 (gzip: 1.4-1ubuntu2). This is an upstream bug, therefore it would be great if you could forward the bug to the developers. As far as I can see, they are not using a bug tracker but a mailing list: https://lists.gnu.org/mailman/listinfo/bug-gzip

Note that a workaround exists: Both
    zgrep -C 10 17 file.txt.gz
and
    zgrep --context=10 17 file.txt.gz
give the expected result.

[Jim Meyering (meyering)]

Here's the likely upstream patch:

  http://thread.gmane.org/gmane.comp.gnu.gzip.bugs/699/focus=705

[Marcel Stimberg (marcelstimberg)]

The patch has been submitted upstream now:
http://git.savannah.gnu.org/cgit/gzip.git/commit/?id=dc6196e2a041c022f8382b7e610b095718acccfa

@Thomas Bushnell: Do you want to prepare a patch/merge request based on this commit (and the next one, which tests the behaviour)? Otherwise I can do it.

[Thomas Bushnell, BSG (tbushnell)]

Given the easy workaround and the lack of urgency, I'd be happy for this to just get rolled out in the ordinary course of events.