feat: Display vCard contact name in the message summary

[iequidoo]

No description provided.

[Hocuri]

I'm a bit concerned that this makes Viewtype::Vcard very special code-wise because set_file() changes the viewtype to Viewtype::File or sets a param for only this one viewtype, which could produce subtle bugs in the future.

Would it be feasible put the code calling try_set_vcard() into do_set_draft() and prepare_msg_blob() instead of set_file() and set_file_from_bytes() (and keep the code in mimeparser.rs)?

Note to myself, I need to test this on an Android device - for [vcards created with DC, vcards created with Thunderbird, vcards created with Thunderbird that contain too many and should be shown as a file]:

• send .vcf file (sender & receiver side)
• draft file
• forward file
• share file

[iequidoo]

I'm a bit concerned that this makes Viewtype::Vcard very special code-wise because set_file() changes the viewtype to Viewtype::File or sets a param for only this one viewtype, which could produce subtle bugs in the future.

Would it be feasible put the code calling try_set_vcard() into do_set_draft() and prepare_msg_blob() instead of set_file() and set_file_from_bytes() (and keep the code in mimeparser.rs)?

It also was a question for me where it's better to parse a vCard on the sender side (and fall back to File if it can't be parsed). Pros of doing this in set_file*() are:

• set_file_from_bytes() already has data in memory, no need to open the file later.
• test_get_summary_text() checks that the summary is correct already after set_file_from_bytes(), no need to make vCards a special case.

Otoh it can be unexpected that set_file*() may change the message viewtype. Btw, also if the file content changes after the set_file*() call, the summary will be wrong.

Wrt the code in mimeparser.rs, in fact i just moved the check for vCard validity from get_mime_type() to do_add_single_file_part(), this way we don't need to parse a vCard twice -- the first time to check its validity and the second to set the message summary.

[Hocuri]

set_file_from_bytes() already has data in memory, no need to open the file later.

Except for Webxdcs, set_file_from_bytes() is only used in tests where performance doesn't matter as much, though.

test_get_summary_text() checks that the summary is correct already after set_file_from_bytes(), no need to make vCards a special case.

I think it's better to special-case in a test than to special-case set_file*() to do something extra just for a single viewtype.

We even can remove StockMessage::Contact now, it should never be displayed

Decided not to deprecate it yet, it may be useful in the future, but added a comment that it's unused currently

I do think we should remove it, if we add stock strings just in case we may need it in the future, we will end up with a lot of unused stock strings at some point. And in most cases when one thinks "it may be useful in the future", one ends up never needing it.

[iequidoo]

I think it's better to special-case in a test than to special-case set_file*() to do something extra just for a single viewtype.

Moved updating the message summary to prepare_msg_blob() and do_set_draft().

I do think we should remove it, if we add stock strings just in case we may need it in the future, we will end up with a lot of unused stock strings at some point. And in most cases when one thinks "it may be useful in the future", one ends up never needing it.

Removed.

[iequidoo]

DC_STR_CONTACT can't be just removed so i deprecated it.

[Hocuri]

Nice! I have 3 small, non-blocking comments. The only thing left to approve is me testing that this works, as mentioned above.

[Hocuri]

Status of this PR: When testing on a real Android device, I found that this doesn't work on the sending side and the summary is only "👤". @iequidoo and me texted in private about this, but we don't know why it happens (and I still have somewhat limited capacity to debug it).

[iequidoo]

Status of this PR: When testing on a real Android device, I found that this doesn't work on the sending side and the summary is only "👤". @iequidoo and me texted in private about this, but we don't know why it happens (and I still have somewhat limited capacity to debug it).

I tried to debug it on my Android yesterday, but had a problem with the file system and unfortunately lost my build env :) But i have reproduced the problem before that. Now building it again.

[iequidoo]

I tried to debug it on my Android yesterday, but had a problem with the file system and unfortunately lost my build env :) But i have reproduced the problem before that. Now building it again.

Now it works on my Android, so the last commit is really a fix. EDIT: the test is coming. EDIT: Done.

[iequidoo]

I've fixed it for drafts, added the corresponding test and verified on my Android