Mail Search

[franciscoxc]

Native Mail Search is almost useless.
I found this wonderful app and is super fast in locating attachments but I cannot search for emails.
Apple stores emails in .emlx format in the library.

Am I missing something??

Thanks for your awesome app

[ldm0]

Here? Try search with *.emlx

[franciscoxc]

Thank you, that worked but kind of useless if you can't even search by subject or let alone contents...
I was expecting searching by content, maybe it's beyond cardinal's capabilities??

[ldm0]

@franciscoxc Use *.emlx content:Good or *.emlx content:"Good morning" for content filtering. Select file and press space key to preview it's content.

[franciscoxc]

Oh wow!! yeah!! That's what I was looking for. It's a thousand times faster than mail's search function!

A highlight of the searched term would be nice ☺️

Thanks!

[ldm0]

@franciscoxc Would append it to README. :-D

[franciscoxc]

I just realized you are the developer! I am not worthy sir!!
So let me disturb your peaceful Saturday with requests lol

How about a column before the path that shows the found term in context? It would be nice right??

[ldm0]

@franciscoxc Interesting idea, add an optional column for it would be useful.

[franciscoxc]

Closing the loop on this one, since it is still open and someone searching for the same thing may land here.

The context column @ldm0 called useful above is implemented and open as #226. It shows the matching text from inside the file with the searched term highlighted, and it only appears while a content: search is running. The highlighted terms come from the query parser rather than from a string split, so a negated !content: does not highlight and content:"Bearer " keeps its trailing space.

Two things in that PR go beyond what we discussed here:

• type:email covers .eml, .emlx, .emlxpart, .msg and .mbox, so finding mail no longer means remembering *.emlx. It is also in a plain-language dropdown next to the search bar, which writes the filter into the query rather than hiding it.
• The snippet is extracted in the engine, reusing the same buffer and matcher content: already uses, so it costs one read of the files that are on screen rather than a second pass.

Since then I have also added the legacy Outlook for Mac formats, which keep every message as its own file — .olk15Message, .olk15MsgSource, .olk15MsgAttach and the .olk14 generation — plus .olm, .pst, .ost and .nst. That part is not in #226; it lives in my fork for now and I am happy to open it separately if there is interest.

Thanks again for the *.emlx content: tip back then — it is still a thousand times faster than Mail's own search.