produce broken scan image with snapscan S300 scanner

[Hiroshi Miura]

As I described in BTS of gscan2pdf, https://sourceforge.net/tracker/?func=detail&aid=3485387&group_id=174140&atid=868098
 libsane-perl: Sane.xs have a bug when length=0 is returned from backend while sane_read().
It is occured with epjitsu backend driver that is used for Fujitsu SnapScan S300 scanner.

Sane.xs:
------------------
void
sane_read (handle, max_length)
  SANE_Handle handle
  SANE_Int max_length
 INIT:
  SANE_Status status;
                SANE_Byte * data;
                SANE_Int length;
        PPCODE:
  data = malloc (max_length);
  status = sane_read (handle, data, max_length, &length);
         SV* sv = get_sv("Sane::_status", FALSE);
                sv_setiv(sv, status);
                if (status) {
                 XPUSHs(sv_2mortal(newSV(0)));
                 XPUSHs(sv_2mortal(newSViv(0)));
                }
                else {
                 XPUSHs(sv_2mortal(newSVpv(data, length)));
   XPUSHs(sv_2mortal(newSViv(length)));
                }
                free (data);
--------------
when call newSVpv(data, length) with length=0, that means 'auto detect length of string of data'.
But length=0 should treat as 'no data'.

If (const char *)data consist by other than '\0', it happens that no-needed bytes is returned.

That is happend with S300 scaner and gscan2pdf, scanimage-perl, and scanadf-perl utilities.

ProblemType: Bug
DistroRelease: Ubuntu 11.10
Package: libsane-perl 0.03-1build1
ProcVersionSignature: Ubuntu 3.0.0-15.26-generic 3.0.13
Uname: Linux 3.0.0-15-generic x86_64
NonfreeKernelModules: nvidia
ApportVersion: 1.23-0ubuntu4
Architecture: amd64
Date: Sat Feb 11 15:31:17 2012
InstallationMedia: Ubuntu 10.04 LTS "Lucid Lynx" - Release amd64 (20100429)
ProcEnviron:
 PATH=(custom, user)
 LANG=ja_JP.UTF-8
 SHELL=/bin/bash
SourcePackage: libsane-perl
UpgradeStatus: Upgraded to oneiric on 2011-12-24 (49 days ago)

[Hiroshi Miura (miurahr)]

An attached fix solves the problem. using newSVPvn(data, len): if len is zero, Perl will create a zero length string.

[Jeffrey Ratcliffe (jeffreyratcliffe)]

Thanks for all your work on this. Can you reproduce the problem with the test backend?

I really like to introduce add regression tests when fixing bugs, and it would be good to have one for this.

[Ubuntu Foundations Team Bug Bot (crichton)]

The attachment "libsane-perl-0.03-sane_read-length-0-fix.patch" of this bug report has been identified as being a patch. The ubuntu-reviewers team has been subscribed to the bug report so that they can review the patch. In the event that this is in fact not a patch you can resolve this situation by removing the tag 'patch' from the bug report and editing the attachment so that it is not flagged as a patch. Additionally, if you are member of the ubuntu-reviewers team please also unsubscribe the team from this bug report.

[This is an automated message performed by a Launchpad user owned by Brian Murray. Please contact him regarding any issues with the action taken in this bug report.]

[Hiroshi Miura (miurahr)]

It is not possile to reproduce the problem without modifing test backend.

I attach my suggestion for sane-backend.

I'm not sure it is good for sane standard...

[Hiroshi Miura (miurahr)]

this patch for sane-backend may be good enough for zero length test for data.

[Hiroshi Miura (miurahr)]

regression test code proposal

[Hiroshi Miura (miurahr)]

previous patch fails with options.t. fix it.

[Jeffrey Ratcliffe (jeffreyratcliffe)]

I have uploaded Sane-0.04, which includes this patch, to CPAN

[Launchpad Janitor (janitor)]

This bug was fixed in the package libsane-perl - 0.05-2

---------------
libsane-perl (0.05-2) unstable; urgency=low

  * Disable tests
    Closes: #626019 (FTBFS on various architectures (hanging tests?))

 -- Jeffrey Ratcliffe <email address hidden> Mon, 21 May 2012 21:52:44 +0200