Final fixes for v.1.07.5 release

[djb-rwth]

No description provided.

[JanCBrammer]

The majority of this PR consists of comments that refer to code changes. However, the majority of these comments is not accompanied by any changes to the code. Have these changes been committed earlier, outside of this PR? If so, could you list the commits @djb-rwth ? We'd have to review those.

[djb-rwth]

Hi @JanCBrammer,

The majority of this PR consists of comments that refer to code changes. However, the majority of these comments is not accompanied by any changes to the code. Have these changes been committed earlier, outside of this PR? If so, could you list the commits @djb-rwth ? We'd have to review those.

Main impact fixes have already been submitted in the previous PR. Medium impact fixes are in the 80% of the case useless, but 15% of the remaining cases have been fixed (these are small changes but nevertheless important).
BTW, to determine the uselessness of many issues, thorough analyses had to be conducted, so I would not say that the notes addressing these issues are simple "comments".

The real problems arise with the spots marked as unresolved issue -- revision required, which were detected but could not be fixed now after series of debugging sessions. These are simply chunks of badly written code; we decided to leave them as they are currently.

Also, please have in mind that due to the inaccessibility of Coverity Scan resources, around 80 more issues should have been addressed or fixed.

[JanCBrammer]

to determine the uselessness of many issues, thorough analyses had to be conducted, so I would not say that the notes addressing these issues are simple "comments".

Thanks for clarifying. So when a comment says "addressed" it means "determined to be irrelevant"?

Could you include the result of the analyses in the comments where feasible (result not overly complex / lengthy)?

[nbehrnd]

A question: pending the release of InChI 1.07.5, were the copyright issues (cf #8) resolved? To me, they possibly were (part of) the reason why DebiChem did package InChI 1.07.3 (Debian's tracker page), however did not not opt-in for InChI 1.07.4.

If resolved / cured, this then equally would affect the next LTS 26.04 of Ubuntu (release schedule, for instance with a feature freeze scheduled for February 19th), too. For other distributions, see repology.org.

[djb-rwth]

Hi @nbehrnd,

A question: pending the release of InChI 1.07.5, were the copyright issues (cf #8) resolved? To me, they possibly were (part of) the reason why DebiChem did package InChI 1.07.3 (Debian's tracker page), however did not not opt-in for InChI 1.07.4.

If resolved / cured, this then equally would affect the next LTS 26.04 of Ubuntu (release schedule, for instance with a feature freeze scheduled for February 19th), too. For other distributions, see repology.org.

According to what you described, it looks like a very important question.
I am forwarding it to @gblanke02, @fbaensch-beilstein and @JanCBrammer as they are more acquainted with this matter.

[djb-rwth]

Hi @JanCBrammer,

Thanks for clarifying. So when a comment says "addressed" it means "determined to be irrelevant"?

Yes.

Could you include the result of the analyses in the comments where feasible (result not overly complex / lengthy)?

I tried to succinctly explain every issue having in mind that all references to Coverity Scan issue IDs will be accessible to any interested party through their webpage.
Reproducing and describing all these issues in full detail would require quite an effort, at least timewise. For example, some of these issues refer to several functions in a row, following a specific code flow.

Also, the Coverity Scan website is still down, making things even more difficult...

[fbaensch-beilstein]

@nbehrnd Thanks for pointing out this issue. We are completely under MIT License now. I have updated the FAQ and renamed a legacy txt file that was specific for v1.05.

[nbehrnd]

@fbaensch-beilstein Thank you for your check and subsequent confirmation. Then some / all of the three points of the list in #8 could be marked as resolved.

@merkys Can you please check if license wise InChI 1.07.5 now indeed is a shape good enough for your volunteering upload to DebiChem? At present I can't check this since the checkout to windows' git (git version 2.46.0.windows.1) is stuck

@DESKTOP-DEFD86M MINGW64 ~/Desktop/inchi_git
$ git clone https://github.com/IUPAC-InChI/InChI.git
Cloning into 'InChI'...
remote: Enumerating objects: 8110, done.
remote: Counting objects: 100% (65/65), done.
remote: Compressing objects: 100% (46/46), done.
remote: Total 8110 (delta 36), reused 41 (delta 19), pack-reused 8045 (from 1)
Receiving objects: 100% (8110/8110), 309.63 MiB | 12.55 MiB/s, done.
Resolving deltas: 100% (4510/4510), done.
error: unable to create file INCHI-1-DOC/FAQ/2019/FAQ/5. Composition and Connectivity/5.6. In InChIs of structures containing more than one component, is the ; separator necessary between contributions from components if one contribution is empty.md: Filename too long
Updating files: 100% (672/672), done.
fatal: unable to checkout working tree
warning: Clone succeeded, but checkout failed.
You can inspect what was checked out with 'git status'
and retry with 'git restore --source=HEAD :/'

-- which maybe just an issue on this local machine.

[merkys]

@merkys Can you please check if license wise InChI 1.07.5 now indeed is a shape good enough for your volunteering upload to DebiChem? At present I can't check this since the checkout to windows' git (git version 2.46.0.windows.1) is stuck

I gave quick look at the contents of this branch (core_dev) at d449b2a and it seems that all pointers to the old license (except from historical release notes) are now gone. Thus this PR closes #8.

[merkys]

Well, several files named inchi-faq.* are still outdated wrt InChI version and licensing, it would be nice to have them updated to reflect the current state better.

[nbehrnd]

@merkys This is great (in perspective of the licenses and to prevent a gap between InChI trust vis-a-vis DebiChem). My cloning issue is solved / bypassed (an additional -c core.longPaths=true overrides Windows' threshold of maximal 260 characters in the absolute path).

[cm-beilstein]

@gblanke02 as I have mentioned before, can you go through the documentation in "INCHI-1-DOC" and mark or make a list of documents to keep (see #4). That way we can stay on top with updating versions. Thanks.