[Reproducible-builds] Bug#783882: jodconverter: please make the build reproducible

Reiner Herrmann reiner at reiner-h.de
Fri May 1 12:11:27 UTC 2015


On 05/01/2015 01:46 PM, Mattia Rizzolo wrote:
> [removing the bug from the recipients list...]
> 
> On Fri, May 1, 2015 at 1:36 PM, Reiner Herrmann <reiner at reiner-h.de> wrote:
>> On 05/01/2015 11:30 AM, Holger Levsen wrote:
>>> That said, according to
>>> https://wiki.debian.org/ReproducibleBuilds/TimestampsInDocumentationGeneratedByJavadoc
>>> strip-nondeterminism should work around the issue too, anyone got an idea why
>>> this doesnt happen?
>>
>> jodconverter uses neither dh, nor cdbs, so strip-nondeterminism is not called.
> 
> you said you think not many packages call javadoc directly, and
> javahelper now passes -notimestamps. Do we know how many affected
> packages strip-nondeterminism fixes? maybe is the case to remove that
> code from strip-nondeterminism and fix the remaining packages?
> 

Hm, that is a good question.
From the list of currently known unreproducible packages with the javadoc issue [1], 8 are calling
javadoc directly in debian/rules.

Other packages with this issue are using maven-debian-helper, which doesn't seem to be fixed yet.
This in turn uses libmaven-javadoc-plugin-java to call javadoc, but I haven't found the exact location
there yet, where javadoc is called (and how to pass it parameters).

The fixed javahelper is in unstable since a few days, so we could indeed remove the code from
strip-nondeterminism and rebuild the packages using javahelper. But I can't estimate how many packages
actually are using it and how many others would still require strip-nondeterminism.

[1]: https://reproducible.debian.net/issues/unstable/timestamps_in_documentation_generated_by_javadoc_issue.html

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 801 bytes
Desc: OpenPGP digital signature
URL: <http://lists.alioth.debian.org/pipermail/reproducible-builds/attachments/20150501/366c069d/attachment.sig>


More information about the Reproducible-builds mailing list