add attachment infos to csv export

[R-Zwi]

resolves #17

[R-Zwi]

This is my first proposal.
But actually the master branch doesn't build in my environment so I developed and tested this on the portable branch and then cherry picked to master.

[R-Zwi]

FLAG_FILENAME_ID is also needed in cmd-export-messages.c now. I moved all of those FLAG_* to keep them together.

[R-Zwi]

Ok, if FLAG_FILENAME_ID is given, we get the ID twice (one in the filename and one here).
I think I should change this to not repeat it here when it is already in the filename.
Do you think so, too?

[tbvdm]

I think you are making too many changes. Please revert the changes to the attachment file logic. I'm not sure that's the right thing to do. Keep only the TAILQ_FOREACH() part in csv_write_message() and use the same text format as the original text_write_attachment_field().

Also, if you can, try keeping to the coding style. See https://man.openbsd.org/style.9 for details. But don't worry about it too much; I can tweak your changes if necessary.

[R-Zwi]

Thanks for your reply!

I can do that but first I'd like to diskuss if it would be the right thing to do.
The purpose for doing the attachment info different to the text export was to align it with the attachment export. So that one can find the correct file in the exported attachments.

The background for that is that I want to be able to create HTML pages which also show the attachments. That is not possible with the existing attachment info in the text export.
So it comes down to the question if the (new) attachment info in the CSV export should match the existing text export or the upcoming HTML export...

[tbvdm]

Ah, I see. How is your HTML export going to work? Will it be a new feature in sigbak or a separate program that processes the CSV output?

[R-Zwi]

Initially I planned to do a Python script to process the CSV output but it turned out to be easy to implement it as a new feature in sigbak, so I'm going this way.
I keep the HTML as simple as possible and leave the formatting to styles.css wherever possible.
So this will be quite basic but sufficient.

Other people (with more knowledge about web designing) might create more fancy output on base of the CSV...

[tbvdm]

Interesting! But then why do you want to make changes to the CSV format? How is that relevant for an HTML export feature?

[R-Zwi]

Actually that was required for my first approach with the Python script.
I didn't revert it because it seemd useful to me.
You are right that it isn't required for the HTML export.

[R-Zwi]

I uploaded my latest changes to my fork so you can have a closer look if you want.