Bug#1010907: lintian: bad-jar-name check doesn't handle suffixes

2022-05-12 Thread Emmanuel Bourg
Package: lintian
Version: 2.114.0
Severity: normal

The bad-jar-name check flags many packages that do conform to the Java policy,
which allows a suffix after the base package name (packagename-extraname.jar).

See for example the batik package, /usr/share/java/batik-all.jar and its
siblings are wrongly reported.

Emmanuel Bourg



Re: Bug#1005762: marked as pending in lintian

2022-02-21 Thread Emmanuel Bourg

Le 15/02/2022 à 23:28, Felix Lechner a écrit :


Thank you for the advice. The value was adjusted back to the previous value:

 
https://salsa.debian.org/lintian/lintian/-/commit/11926263c63e9286339f49dbbee55dd45982b621


It looks good, thank you.

Emmanuel Bourg



Bug#1005762: marked as pending in lintian

2022-02-15 Thread Emmanuel Bourg

Le 15/02/2022 à 06:50, Felix Lechner a écrit :


Recognize Java 18 in unstable, and Java 19 as otherwise available. (Closes: 
#1005762)

With openjdk-18 version 18~32ea-1 in unstable, we also adjusted the highest
version available in Debian to bytecode version 62.


Thank you for the quick fix Felix.

I think max-bytecode-version should remain at 56, because Java 11 is 
still the default version. Once the transition to Java 17 is completed 
this could be changed to 61. Java 18 isn't a LTS release, so 62 will 
never be used.


Emmanuel Bourg



Bug#1005762: lintian: Update known Java version up to 19

2022-02-14 Thread Emmanuel Bourg
Package: lintian
Version: 2.114.0
Severity: normal
User: debian-j...@lists.debian.org
Usertags: default-java17

Dear Maintainer,

Lintian doesn't recognize bytecode generated by Java 17 yet (the 
unknown-java-class-version
warning is emitted). Could you please update the maximum known Java version to
the latest one please (i.e. Java 19, with bytecode version 63).

Thank you,

Emmanuel Bourg



Bug#963939: lintian: breakout-link wrongly reported against jar files

2020-07-04 Thread Emmanuel Bourg
Hi Felix,

Le 02/07/2020 à 16:15, Felix Lechner a écrit :

> Why did you place the links, which appear to be
> architecture-independent, in /usr/lib and not in /usr/share?

The src:eclipse-* packages install files in /usr/share/ and /usr/lib/.
The /usr/lib/ links were added to preserve the compatibility with the
layout of the old src:eclipse package (which dates back to 2002, it was
replaced two years ago by a set of smaller packages) :

  https://packages.debian.org/stretch/amd64/eclipse-platform/filelist
  https://packages.debian.org/stretch/amd64/eclipse-rcp/filelist

Emmanuel Bourg



Bug#963939: lintian: breakout-link wrongly reported against jar files

2020-06-28 Thread Emmanuel Bourg
Package: lintian
Version: 2.82.0
Severity: normal

Hi,

lintian is reporting breakout-link warnings on several Eclipse packages
(such as eclipse-platform-ui). This is due to links in /usr/lib/eclipse/plugins/
pointing to jar files in /usr/share/java/.

I don't think this warning applies to architecture independent jar files.

Emmanuel Bourg



Bug#962448: mailing-list-obsolete-in-debian-infrastructure: Please ignore the Debian Java Maintainers address

2020-06-08 Thread Emmanuel Bourg
Package: lintian
Version: 2.80.0
Severity: normal

Hi,

Lintian reports a mailing-list-obsolete-in-debian-infrastructure warning
for the packages maintained by the Java Team:

  mailing-list-obsolete-in-debian-infrastructure Debian Java Maintainers 


The address is valid and the team doesn't plan to migrate 1000+ packages
to a different address.

Could you please exclude this address from the ones reported by this tag?

Thank you,

Emmanuel Bourg



Bug#471537: fixed in lintian 2.42.0

2019-12-29 Thread Emmanuel Bourg
On 29/12/2019 17:50, Felix Lechner wrote:

> Done with:
> 
> 
> https://salsa.debian.org/lintian/lintian/commit/ca5adad9cb21805b871a9f2e6cdd30b8bdb0246c
> 
> Thanks for helping to make Lintian better.

Thank you for the quick fix!

Emmanuel Bourg



Bug#471537: fixed in lintian 2.42.0

2019-12-28 Thread Emmanuel Bourg
>  lintian (2.42.0) unstable; urgency=medium
>  .
>* Add new checks for when repackaged
>  sources are not properly advertised as such. (Closes: #471537)

I've just stumbled on this new warning and I disagree with the
recommendation. Sometimes we just filter out garbage from the upstream
sources and we don't want to add a +repack suffix.

This is a very common pattern in the Java team where .jar, GitHub CI and
IDE settings files are frequently removed and we don't want to fiddle
with the version to keep things clear and simple.

I suggest changing the severity to info or pedantic, and adjust the
description to explain the suffix is optional.

Emmanuel Bourg



Bug#789802: lintian: False positive source-contains-prebuilt-java-object reported against jar files without classes

2018-03-18 Thread Emmanuel Bourg
Control: tag -1 - moreinfo

On 24/12/2017 21:33, Chris Lamb wrote:

> Do you have any up-to-date false-positives? The ones you listed
> are now not showing up for me :)

cronometer is another example. Lintian reports two jar files that only
contain xml files and no compiled Java files:

P: cronometer source: source-contains-prebuilt-java-object lib/crdb_005.jar
P: cronometer source: source-contains-prebuilt-java-object lib/usda_sr24.jar



Bug#873211: lintian: does not warn about .class binaries

2017-09-25 Thread Emmanuel Bourg
On Wed, 20 Sep 2017 17:34:47 + Chris Lamb <la...@debian.org> wrote:
>  lintian (2.5.53) unstable; urgency=medium
>  .
>* Summary of tag changes:
>  + Added:
>- package-installs-java-bytecode

Hi,

Thanks for improving the Java support in Lintian, I'd suggest CCing
these topics to debian-j...@lists.debian.org to gather more feedback on
the proposed changes.

I got a look at the packages affected by the new
package-installs-java-bytecode tag [1], they mostly consist in
demos/examples, webapp classes or one-class programs not meant to be
re-used as libraries. For these cases installing .class files directly
in the binary package is legitimate I think. jar files are really
important for reusable libraries and large applications (since jar files
are more space efficient), but there is no harm shipping a few isolated
.class files. The Java Policy probably needs a clarification on this point.

As I understand Carnë was concerned about .class files in the upstream
tarballs not recompiled from source and installed as-is in the binary
packages. I was under the impression this case was already covered by
source-contains-prebuilt-java-object but it isn't. I agree it would be
nice to handle this.

So I suggest the following:
- modify source-contains-prebuilt-java-object to also detect .class files
- lower the severity of package-installs-java-bytecode to pendatic or info
- trigger package-installs-java-bytecode in non-library packages only
when the number of classes detected in the package exceeds 20.
- do not trigger package-installs-java-bytecode if the path contains
"WEB-INF", "demo", "doc", "example", "sample" or "test".
- strictly speaking a class file isn't raw bytecode instructions, so
maybe rename the tag to "package-installs-java-class-files".
- verify if the .class files are really Java class files (by checking
the 0xCAFEBABE header, this will avoid false positives like
apertium-eo-fr and grass-core).

Emmanuel Bourg

[1] https://lintian.debian.org/tags/package-installs-java-bytecode.html



Bug#857123: lintian: warning about missing classpath is confusing

2017-03-08 Thread Emmanuel Bourg
Le 8/03/2017 à 10:19, Markus Koschany a écrit :

> I suggest to remove this Lintian tag or lower the severity from
> warning to info.

+1 for lowering the severity to info.

Emmanuel Bourg



Bug#789802: lintian: False positive source-contains-prebuilt-java-object reported against jar files without classes

2015-10-22 Thread Emmanuel Bourg
apache-log4j2 is another example, it contains an IntellijSettings.jar
file with IDE settings (plain text files, no compiled classes).

https://sources.debian.net/src/apache-log4j2/2.2-1/src/ide/Intellij/13/



Bug#789802: lintian: False positive source-contains-prebuilt-java-object reported against jar files without classes

2015-07-09 Thread Emmanuel Bourg
maven-invoker-plugin is a good example of a package containing
empty jar files. The source tarball ships with 3 jar files used
by the integration tests, they contain only a MANIFEST.MF file
and no compiled class:

ebourg@icare:~/packaging/maven-invoker-plugin$ tar -tf 
../maven-invoker-plugin_1.5.orig.tar.gz | grep jar
maven-invoker-plugin-1.5/src/it/staging-dependencies/repo/org/apache/maven/its/minvoker-70/1.0-SNAPSHOT/minvoker-70-1.0-20081020.164906-1.jar
maven-invoker-plugin-1.5/src/it/staging-dependencies/repo/org/apache/maven/its/dep/1.0/dep-1.0-tests.jar
maven-invoker-plugin-1.5/src/it/staging-dependencies/repo/org/apache/maven/its/dep/2.0/dep-2.0.jar

ebourg@icare:~/packaging/maven-invoker-plugin$ jar -tf 
src/it/staging-dependencies/repo/org/apache/maven/its/dep/1.0/dep-1.0-tests.jar
META-INF/
META-INF/MANIFEST.MF


-- 
To UNSUBSCRIBE, email to debian-lint-maint-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/559e9fb0.5050...@apache.org



Bug#791552: lintian: [new check] verify that JAR filename complies with Debian Java Policy

2015-07-07 Thread Emmanuel Bourg
Le 06/07/2015 06:36, tony mancill a écrit :

 This warning is useful because
 some components of the Debian Java toolchain fail when JAR files don't
 comply with the naming policy.

Hi Tony,

For my understanding, what are the tools relying on this naming policy?
I'm asking because I'm pretty sure many packages don't strictly adhere
to the Java policy on this point (for example liblog4j1.2-java vs
/usr/share/java/log4j-1.2.jar) and I'm not under the impression it's
causing such a havoc.

Emmanuel Bourg


-- 
To UNSUBSCRIBE, email to debian-lint-maint-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/559bb4dc.2080...@apache.org



Bug#789802: lintian: False positive source-contains-prebuilt-java-object reported against jar files without classes

2015-06-24 Thread Emmanuel Bourg
Package: lintian
Version: 2.5.31
Severity: normal

Dear Maintainer,

Some packages contain jar files without classes which are improperly reported
as prebuilt Java objects. It would be nice to also check the content of the jar
files and ensure they do not contain any .class file before reporting this 
warning.

Thank you,

Emmanuel Bourg


-- 
To UNSUBSCRIBE, email to debian-lint-maint-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/20150624153133.4316.17930.report...@icare.ariane-software.com



Bug#786895: lintian: incompatible-java-bytecode-format warning needs update for Java 1.7

2015-05-27 Thread Emmanuel Bourg
Le 27/05/2015 15:41, Jan Henke a écrit :

 I think gcj serves one single purpose only at this point in time:
 Bootstrapping during the OpenJDK build.

This is no longer true with OpenJDK 8 unfortunately, Java 7 is now required.


-- 
To UNSUBSCRIBE, email to debian-lint-maint-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/5565cf4a.4040...@apache.org



Bug#786895: lintian: incompatible-java-bytecode-format warning needs update for Java 1.7

2015-05-26 Thread Emmanuel Bourg
Le 26/05/2015 16:52, Rene Engelhard a écrit :

 I think we should decide what our Java baseline is and how it affects
 release archs_before_ changing this.

The best we can do I think is to identify the applications that should
work with GCJ (Ant and LibreOffice for example) and ensure their
dependencies are still compatible with the Java 5 API. But as Niels
stated it's impossible to keep the Java 5 compatibility everywhere (Java
9 will even be unable to generate Java 5 bytecode [1]).

Emmanuel Bourg

[1] http://openjdk.java.net/jeps/182


-- 
To UNSUBSCRIBE, email to debian-lint-maint-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/5564952a.6030...@apache.org



Bug#786895: lintian: incompatible-java-bytecode-format warning needs update for Java 1.7

2015-05-26 Thread Emmanuel Bourg
Le 26/05/2015 15:58, Markus Koschany a écrit :

 I assume all members of the team agree with this change.

Yes that makes sense.

Emmanuel Bourg


-- 
To UNSUBSCRIBE, email to debian-lint-maint-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/55647e82.4080...@apache.org



Bug#757615: (no subject)

2014-09-07 Thread Emmanuel Bourg
It looks like this new lintian check gives false positives when the
License field contains or:

License: CDDL or GPL-2

  W: jenkins source: space-in-std-shortname-in-dep5-copyright cddl or
gpl-2 (paragraph at line 99)


-- 
To UNSUBSCRIBE, email to debian-lint-maint-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/540c531d.6050...@apache.org



Bug#743384: lintian: Update unknown-java-class-version to support Java 8 class version (52)

2014-04-02 Thread Emmanuel Bourg
Package: lintian
Version: 2.5.22.1
Severity: normal

Hi,

Could you please support the Java 8 class version in the
unknown-java-class-version check? OpenJDK 8 is being packaged and
lintian complains about the new class version (52 for Java 8, Java 7
used 51)

Thank you,

Emmanuel Bourg


-- 
To UNSUBSCRIBE, email to debian-lint-maint-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/533bcb3f.9030...@apache.org



Bug#743384: lintian: Update unknown-java-class-version to support Java 8 class version (52)

2014-04-02 Thread Emmanuel Bourg
Le 02/04/2014 13:26, Matthias Klose a écrit :

 and I think it should warn about it. We do not want Java 8 bytecode in
 the archive except for the JDK itself.

I think that's the purpose of incompatible-java-bytecode-format:

http://lintian.debian.org/tags/incompatible-java-bytecode-format.html

The package contains Java class files with a minimum requirement on the
listed Java version. This Java version is not supported by the default
JVM in Debian and is therefore likely to be a mistake.

Emmanuel Bourg


-- 
To UNSUBSCRIBE, email to debian-lint-maint-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/533bf7c1.6070...@apache.org