sci-biology/imagej: bump version imagej sources and plugins

[IAHM-COL]

• ImageJ version bump 1.53t to 1.54g from NIH sources
• plugins version bump 153 to 154

[IAHM-COL]

@AndrewAmmerlaan

This continues the imagej version bumping. There are a few circumstances to be mentioned.

1. All ebuilds tested for compilation, and pass ok, in my box
2. Plugins are now available for version 154, so that gets updated here, fetching from NIH
   • see https://wsr.imagej.net/distros/cross-platform/
3. The latest version available both github and from NIH is 1.54h,
   • see https://imagej.net/ij/download/src/
   • see https://github.com/imagej/ImageJ/tags
4. However two ebuilds can't have same name! Therefore
   • I preserved 1.54h from github, same file as 9999
   • I updated the ebuild from NIH sources to one version below 1.54g
5. Although it would have made technical sense to make 1.54g stable, 1.54h ~amd64 and 999 ** keywords, as we have discussed before, the policy for sci as overlay is to keep ~amd64 and ** for all ebuilds, as a consequence,
   pkgcheck scan still results on

sci-biology/imagej
      RedundantVersion: version 1.54g: slot(0) keywords are overshadowed by version: 1.54h

With all this in consideration, I think the version bump is clear for this merge request.

[Nowa-Ammerlaan]

However two ebuilds can't have same name! Therefore

* I preserved 1.54h from github, same file as 9999

* I updated the ebuild from NIH sources to one version below 1.54g

What is the reason we need an ebuild for both locations? Is there some difference in the source files?

[IAHM-COL]

What is the reason we need an ebuild for both locations? Is there some difference in the source files?

I don't know the answer.

I presume, downloading from both sites and running a SHA sum can answer whether the github releases and the source zips on the NIH site are actually mirrors. If you want I can test this for 1.54h. it's too involved to check for many versions.

The NIH site kept the source dir outdated for a significant amount of time, and the releases appeared as tags on github. But they appear to be caught up now.

[IAHM-COL]

I presume, downloading from both sites and running a SHA sum can answer whether the github releases and the source zips on the NIH site are actually mirrors. If you want I can test this for 1.54h. it's too involved to check for many versions.

~/test $ ls
ij154h-src.zip  v1.54h.zip
~/test $ sha512sum ij154h-src.zip 
01fa9d3c52aa7a4ba9b2407d6157f75f18c8a3e9ef5ea0e30414671b6fb421574016159fc82718904cfe774c851f50a1da1460ebadd0cdc4b54f9e62691563eb  ij154h-src.zip
~/test $ sha512sum v1.54h.zip 
308db44defcf8672c565bf5481869ce24645bdd716dd70cd282f97f91ab6da17e509244793a5506461a713708e733040067d88358b4da0e9233b5b995044c31a  v1.54h.zip

That's the sha512s I am getting after downloading the zips from each source

[IAHM-COL]

After unzipping, I run the diff, for evaluating what's different. FYI ImageJ-1.54h is after unzipping the github, while source is after unzipping the NIH

~/test $ diff ImageJ-1.54h source
Only in ImageJ-1.54h: applet.html
Only in source: aREADME.txt
Only in source: build2.xml
Only in source: build copy.xml
diff ImageJ-1.54h/build.xml source/build.xml
29a30
>     <copy file="functions.html" todir="${build.dir}" />
Only in ImageJ-1.54h: compile.bat
Only in ImageJ-1.54h: .github
Only in ImageJ-1.54h: .gitignore
Common subdirectories: ImageJ-1.54h/ij and source/ij
Common subdirectories: ImageJ-1.54h/images and source/images
Only in ImageJ-1.54h: LICENSE.txt
Common subdirectories: ImageJ-1.54h/macros and source/macros
Only in ImageJ-1.54h: .mailmap
Only in ImageJ-1.54h: nbproject
Common subdirectories: ImageJ-1.54h/plugins and source/plugins
Only in ImageJ-1.54h: policy
Only in ImageJ-1.54h: pom.xml
Only in ImageJ-1.54h: README.md
diff ImageJ-1.54h/release-notes.html source/release-notes.html
8c8
< <li> <u>1.54h  15 December 2023</u>
---
> <li> <u>1.54h  14 December 2023</u>
11c11
< the RGB to indexed color conversion done by the
---
> the RGB to indexed color conversion done by
20,21c20,21
< and Dialog.getYesNoCancel macro functions
< (<a href="http://wsr.imagej.net/macros/YesNoCancelDialog.ijm">example</a>).
---
> and Dialog.getYesNoCancel macro functions DOCUMENT
> (<a href="http://wsr.imagej.net/macros/YesNoCancelDialog.ijm">example</a>,
28c28
< function not working in batch mode.
---
> functio not working in batch mode.
30c30
< "Composite" mode images.
---
> COMPOSITE mode images.
Only in ImageJ-1.54h: run_appletviewer.bat
Only in ImageJ-1.54h: tests

[Nowa-Ammerlaan]

Only in ImageJ-1.54h: tests

I think we can stick to one version, let's use this one that has the tests.

[IAHM-COL]

Only in ImageJ-1.54h: tests

I think we can stick to one version, let's use this one that has the tests.

I am ok with your call here. Let me update this pull request maintaining only the github sources.

[IAHM-COL]

Let me update this pull request maintaining only the github sources.

bf5e74c

[Nowa-Ammerlaan]

Thanks 👍

[IAHM-COL]

👍