Feature request: optional function to remove leftover pdftocgen documents and rename new document

[atanasj]

I think it would be great to have an optional function that removes the recipe.toml and original PDF, and renames the newly created document with the _out.pdf suffix to the original named document (pdf-filename).

My elisp is novice at best, however I think this function could be written purely in elisp.

[dalanicolai]

Thanks for the suggestion. The package has been designed to not remove the recipe.toml file, as the user might decide to 'change/improve' some previously extracted TOC. Personally, I find it no problem to have the recipe file in a book's directory. Indeed, if you are storing multiple books in a single directory, then the recipe.toml file will be reused for separate books. I will see if I find time to improve that behavior on some future update (it has no priority for now).

Regarding the _out.pdf file, that behavior is probably related to issue https://github.com/dalanicolai/doc-tools-toc/issues/1, as when using the package as instructed, the new pdf file replaces the old file by default (and the _out.pdf file should be only a temporary file).

[atanasj]

Thanks for your response, and I didn't know this feature was already there.... I too could live with the recipe.toml if the other part was working, and indeed it may be related to issue #1, as when I run the C-c C-c as outlined in the usage instructions, I get the error reported in issue #1 and outputted file that has the original filename with the _out.pdf suffix. Am I doing something wrong?

The debug gives me the following:

Debugger entered--Lisp error: (void-variable pdf-filename)
  doc-toc--pdftocgen-add-to-pdf()
  funcall-interactively(doc-toc--pdftocgen-add-to-pdf)
  command-execute(doc-toc--pdftocgen-add-to-pdf)

Any help in further tracking down the cause of this issue would be greatly appreciated.

[dalanicolai]

Did you do it all in one go? Can you see what the value of pdf-filename is in the toc buffer, i.e. after you used doc-toc-extract-with-pdf-tocgen?

If you'd like to track down the cause, then I would advise you to read about edebug, and then edebug the functions doc-toc-extract-with-pdf-tocgen and doc-toc--pdftocgen-add-to-pdf. Those are very simple functions, you might be able to find the culprit.

[atanasj]

By do it all in one go, do you mean just follow the process as outlined by your help docs? I think so.

The variable pdf-filename does not seem to exist:

Thanks for the pointer regarding the edebug... ~I'll give it a go when I get a bit more time and report back.~

So, I gave it a go, and it stopped on the below lines... Not sure what this is telling me. Any ideas?

[dalanicolai]

From the image, it is not clear to me at which line it stopped? Also, did you try to step through the code?

[atanasj]

It stopped on 325... Sorry, I did not try and step through the code. Very much learning to swim here, so I appreciate your patience. How do I "step through the code"?

[dalanicolai]

For vanilla Emacs that is explained [here]. However, it looks like you might be using Doom emacs, for which I can not quickly find any info about edebug keybindings. However, it might be possible Doom just uses evil-collection, in which case s steps through the code...

[atanasj]

Thanks for the tip. Okay, I have stepped through it, but I cannot make any more sense of it... Do you see anything in the gif?

[dalanicolai]

I see that here pdf-filename seems to be defined. Otherwise I do not see anything interesting (also no errors). But in the gif you do not complete stepping through doc-toc--pdftocgen-add-to-pdf... you should only stop when you see the error come up (or otherwise when successfully stepped through the function). Also what do those keypresses mean? C-c and C-s?

[atanasj]

Figured this out finally. Was a dependency and config issue on my end. Thanks!