home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!munnari.oz.au!ariel.ucs.unimelb.EDU.AU!ariel.ucs.unimelb.EDU.AU!not-for-mail
- From: duty@ariel.ucs.unimelb.EDU.AU (Duty Programmer)
- Newsgroups: comp.os.linux
- Subject: GNU (?) compress (4.0.1) -c bug
- Date: 23 Nov 1992 16:39:26 +1100
- Organization: Information Technology Services, University of Melbourne
- Lines: 39
- Sender: rab@tauon.ph.unimelb.EDU.AU
- Distribution: world
- Message-ID: <1epqqeINN3s7@ariel.ucs.unimelb.EDU.AU>
- Reply-To: rab@tauon.ph.unimelb.EDU.AU
- NNTP-Posting-Host: ariel.ucs.unimelb.edu.au
-
- After installing some new man pages the other day i tried to look at
- one using the man-1.1 (?) package, and received the error message
- "stdin: not in compressed format". As man has always worked well for
- me, i looked into this further and discovered that the problem is that
- the version of compress i was using (recently updated from hlu's disks)
- has a trivial bug which prevents it from working (as is) with the man
- package (as configured).
-
- The 'bug' is this: (un)compress has a '-c' switch, which is supposed
- to mean "write output on standard output, don't delete original".
- When used with compress, '-c' makes compress act like zcat.
-
- I have no idea where my old version of compress was from, but it worked
- quite happily in this regard, as the man package is configured to run
- pilelines of the form:
-
- groff ... /usr/man/man1/blah.1 | compress -c > /usr/man/cat1/blah.1.Z
-
- when it produces compressed pre-roff'ed man pages. I ftp'd to prep and
- downloaded compress-4.0.1.tar only to find that it was the *same*
- compress that hlu had on his bootable rootdisk and that the bug is
- there in the source code, viz:
- > case 'c':
- > zcat_flg = 1;
- The '-c' switch is clearly intended to work when uncompressing (in fact
- zcat-ing!), but is broken when compressing. (Actually, it is totally
- redundant in this context, since compress will write to stdout when
- reading from stdin anyway.)
-
- So, although the solution is entirely trivial wrt man (just get rid of the
- '-c' option), it appears to be a bug. I'll mail this to the authors, but
- i thought i'd mention it here in case other people were bitten by this.
-
- rab
- ------------------------------------------------------------------------
- Richard Brown | E-mail: rab@tauon.ph.unimelb.EDU.AU
- School of Physics | Phone : +61 3 344 5081
- University of Melbourne | Fax : +61 3 347 4783
- Parkville Victoria AUSTRALIA 3052 | Telex : AA35185
-