Handle tag inheritance

[gscano]

This is an attempt at solving #24 based on the org-use-tag-inheritance and org-tags-exclude-from-inheritance variables.

Inherited tags are recursively listed up to the #+:FILETAGS keyword before being filtered according to those two variables, while headline tags are always added. The order in which the tags appear in the CSV column respects the hierarchy, with file level tags first and headline level tags last.

Additional coverage tests are provided.

Thanks in advance for your review!

[Wheest]

I had lost this PR in my inbox, but I recently returned to org-csv and experienced this issue again

I'm glad to see that you got something working, thanks!

I have tried running this code in a standalone machine, and found that it exhibited the expected behaviour.

Reviewing the code, everything LGTM. I believe it needs updated to fix the merge conflict, I will review if this is done.

[dapritchard]

Hi all, thanks for all your efforts regarding this incredibly useful package.

I have a branch at https://github.com/dapritchard/org-clock-csv/tree/merge-issue/24 that performs the merge to bring this PR into the master branch. Is there any way that my branch could be used to enable the merge in this repository? What would be the right protocol for me to do here, perhaps submit a PR to @gscano's branch?

As an aside, I also added a commit to add the STARTED TODO keyword to the issue-23.org file. It appears that in my version of Emacs (v27.2) STARTED is not one of the default keywords, and this was causing the test-issue-23 test to fail for me. I'm happy to raise this as a separate issue, if desired.

[gscano]

Hi David,

thank you for taking the time to solve this merge conflict! No need for a PR, I pushed your commit directly.

I also experienced the problem with the TODO keyword in master and you solved it. Since the tests do not pass on issue/24 I would suggest a PR on master to first solve the tests there.

[dapritchard]

Excellent! As per the suggestion, I've created a new issue #31 and associated PR #32 to handle the test-issue-23 TODO keywords issue.