We were inadvertently calling g_object_unref on a wild pointer leading
to the following error message:
GLib-GObject-CRITICAL **: g_object_unref: assertion
`G_IS_OBJECT (object)' failed
Now, why glib doesn't abort on critical errors, I'll never understand.
I noticed that these functions would sometimes leave point on an
invisible character[*]. The problem would be that point would appear
to be on a particular message, but adding or removing a tag would
actually add/remove a tag from the *previous* message.
Fix the C-n and C-p keybindings at least to call the underlying
command and then advance to a visible character. We set this-command
in our overrides so that the temporary-goal-column feature still
works.
[*] The documentation says that command loop is supposed to move point
outside of any invisible region when a command exits. But apparently
not.
Clearly some recent code was very fragile, which I noticed in that the
space bar would no longer scroll a long message if it was the only
message in a thread.
This resulted in a lot of churn, but hopefully things are more robust
now, (for example by using new predicates like
notmuch-show-last-message-p rather than doing heuristics based on
(eobp) or (window-end)).
As usual, the presence of invisible characters complicates the task of
making this stuff robust.
With the recent change of showing the first unread message, we would
scroll down to the end of the buffer if all messages were already
read. This would confusingly show nothing visible in the window.
Instead, detect this case and move to the beginning of the buffer.
The magic of the space bar is all about unread messages, so there's no
reason for it to advance to messages that have already been read.
Similarly, we now remove any magic from (n)ext so that it simply
advances to the next message without marking anything read, (which
makes it symmetrical with (p)revious).
This allows for pleasant termination of the "show next thread" magic
in notmuch-show mode. Now, it will terminate and show the
notmuch-search results rather than continually displaying the last
thread over and over.
This is implemented by stashing away the parent notmuch-search buffer
into a variable within the notmuch-show buffer. Then, when magic space
bar triggers an archive of the current thread, it switches to the parent
search buffer and shows the next thread.
The approach here is to move the optimization from mark-read to the
more general remove-tag. Namely, don't call out to a "notmuch tag"
command to remove a tag that's not there already.
Currently this will either advance by screenfuls, or to the next
message if it's already within a screenful, and will mark each message
read as it is left.
It doesn't yet complete the magic by archiving the messages nor by
advancing to the next thread in the search.
Now, if the user has manually moved point to somewhere within a
message, executing the previous-message command onece will rewind
point only to the beginning of the current message. Previously this
would go back to the previous message, (which the user can now do
easily and naturally by simply executing the command one more time).
Before this just brought the current line to the top of the
window. Now it actually moves to the beginning of the current message.
This is built on a much more solid foundation now with a function to
move to the summary-line of the current message, and then moving from
there.
These now provide a summary of the most useful features/bindings
as well as a complete printout of the relevant mode maps to show
all available keybindings.
This is the command in notmuch-search mode and it's cer convenient
for it to advance to the next line there. (It would be even more
convenient if it didn't also take forever, but as mentioned before
that's an issue we'll need to fix in Xapian.)
We turn on the scroll-preserve-screen-position option which seems
like what's desired here, (though that's not what I normally use
when editing files---but I think scrolling through a list of email
threads is different).
I had put these in here since I had originally planned to copy
liberally from the body of the implementation of 'compile in order
to get process output into a buffer. But once I found call-process
in the documentation of emacs, that was all I needed.
And all the code I've written since has been entirely my own with
just the help of emacs documentation.
This is our first race-free implementation of archive-thread! It
acts only on the messages explcitly contained in the buffer, not
on an entire thread ID, so it's safe in the face of new messages
have been delivered for this thread since the view was made.
This optimization wouldn't be necessary if we had a nice fast "notmuch
tag" command. But since it's currently fairly slow, (see Xapian defect
250: http://trac.xapian.org/ticket/250), we're willing to take some
extra care to avoid calling "notmuch tag" unnecessarily.
I previously had a hack that special-cased the "unread" tag and
printed it on the same line as the message ID. But now that we are
printing all tags at the end of the one-line summary we don't need
this anymore. Get rid of it, and just read "unread" from the list of
tags just like any other tag.
This is in notmuch-show mode rather than in notmuch-search mode,
(where we had + and - working already). This gives the same visual
feedback as in notmuch-search-mode, (the tags are manipulated first in
the database and then the list of tags in the buffer is updated).
Also hide all markers.
From here, all we really need for legibility is the following:
* Hide away citations and signatures
* Call out the one-line summary some way, (larger font size?)
* Add nesting for replies
We were previously using things like "%message{" which were not
guaranteed to never appear in an email message. Using a control
character (^L or '\f' instead of '%') gives us better assurance that
our delimiter doesn't show up in an original email message.
This still isn't entirely safe since we're decoding encoded text in
the body of the email message so almost all bets are off really.
Almost starting to get usable now. Still need to make it mark messages
as they are read, (by removing the unread tag), and selectively hiding
the full header.
The notmuch.c main program now uses GMime directly rather than using
these functions, and I'd rather not export any functions unless we
have good evidence that the functions are necessary.
We had originally copied this function in at a time when notmuch
wasn't actually depending on the GMime library. Now that it does,
we might as well call the function that exists there rather
than having a private copy of it.
