JDK-8339271 : giflib attribution correction
  • Type: Bug
  • Component: client-libs
  • Sub-Component: java.awt
  • Affected Version: 14,23,24
  • Priority: P4
  • Status: Resolved
  • Resolution: Fixed
  • OS: generic
  • CPU: generic
  • Submitted: 2024-08-29
  • Updated: 2024-09-30
  • Resolved: 2024-09-26
The Version table provides details related to the release that this issue/RFE will be addressed.

Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.

To download the current JDK release, click here.
JDK 24
24 b17Fixed
Related Reports
Relates :  
JDK-8230597 - Update GIFlib library to the 5.2.1
Relates :  
JDK-8225487 - giflib legal file is missing attribution for openbsd-reallocarray.c.
Relates :  
JDK-8328999 - Update GIFlib to 5.2.2
Description
The change to the giflib copyrights file was updated in JDK-8328999, and in that change introduced a tree/README and an updated Authors section. However the tree/README is an incorrect change, and the Authors section has an extra "There have been many other contributors; see the attributions in the
version-control history to learn more." text taken from the README file from the 5.2.2 update.

These lines should be removed from the giflib md file.
Comments
Changeset: 66f16398 Branch: master Author: Alisen Chung <achung@openjdk.org> Date: 2024-09-26 01:16:13 +0000 URL: https://git.openjdk.org/jdk/commit/66f1639846645f1d3b4096ef6d62f2b301cf7ed2
26-09-2024
I'm still listed as the reporter of this bug, but the summary and description no longer contain anything I actually wrote. Is it usual to completely rewrite a bug report like this? I agree the sentence about other contributors is vague, but that and the link to the VCS would appear more accurate than what is being proposed as a replacement. The new version implies there are only three authors which is simply untrue e.g. https://sourceforge.net/p/giflib/code/ci/368f28c0034ecfb6dd4b3412af4cc589a56e0611/ https://sourceforge.net/p/giflib/code/ci/d565f6fa04be8973425bd7cd3f169908ac9e95c2/
05-09-2024
the reviewed and approved content of changes actually applicable to JDK-8230597 changes as well (from JDK 14), so this is not a regression introduced in JDK 23. Also this sentence removed/reviewed do not include: “There have been many other contributors; see the attributions in the version-control history to learn more.“ because attribution md file is place for exact and precise list of licenses, copyrights and authors.
04-09-2024
A pull request was submitted for review. Branch: master URL: https://git.openjdk.org/jdk/pull/20859 Date: 2024-09-04 19:43:19 +0000
04-09-2024
It was brought to my attention that the list of AUTHORS states "There have been many other contributors; see the attributions in the version-control history to learn more." This seems vague, and thus problematic because it doesn't list the explicit list of authors. But, in the most recent version of giflib, no list of authors exists. This excerpt was pulled from the provided README in giflib, and the README file DOES have a small list of authors (including this line saying other contributors can be found in the version-control history). Looking through all of the other files in giflib and at their host site, I cannot find any exact list of authors other than the one in the README. The ChangeLog file has a list of changes accompanied by the person who made the change (in the form of a username or actual name), but this isn't the same either. Seems like giflib relies on sourceforge's history to see who else contributed to the project and just uses this line as a "catch-all". And this isn't realistic since this would require someone to manually go thru the history of the sourceforge project for EACH version release and manually add to a list of each different contributor.
30-08-2024
This removal of the "URL" looks correct. The "URL" is covered by the "tree/openbsd-reallocarray.c". However, "tree/README" does seem to be an unnecessary addition as well. My suggestion for an actual solution to this: 1. Remove the "tree/README" as it's unnecessary. 2. We can change "tree/openbsd-reallocarray.c" to "giflib/openbsd-reallocarray.c" 3. Move the "authors" block underneath the license 4. Add this header before "giflib/openbsd-reallocarray.c": --------------------------------- The below applies to the following file(s):
30-08-2024
[~dnguyen] please re-evaluate based on attribution tool input and some investigation we done yesterday
30-08-2024
[~achung] Please take a look at this issue. Concern related to removing URL for version control information seems valid. Are we storing this link somewhere else? Also if this needs to be worked upon, please discuss with [~andrew] and assign this issue to you. If we need to update giflib.md, it needs to be backported to appropriate streams.
30-08-2024