| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
The phrase "input line" has already been used once in the file.
|
|
|
|
|
|
|
|
|
| |
With IRCv3.2 echo-message, each successfully sent message would
move us to the front of the list used for chanuser autocomplete.
Such behaviour seems useless.
Also abandon the idea of bumping on other kinds of messages.
|
|
|
|
| |
This makes nick autocompletion start in a non-arbitrary state.
|
| |
|
| |
|
|
|
|
| |
Try not to commit, push and tag releases tired.
|
| |
|
|
|
|
| |
A recent addition of an N_ELEMENTS macro invocation broke it.
|
| |
|
| |
|
|
|
|
| |
Works for aliases as well. Resolves a TODO entry.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Replaces the inaccurate Rec. 709 luma we used to use before.
This is the first feature here that requires libm, which doesn't
seem to be a particularly great sacrifice.
Moreover, I've rectified that the input isn't linear in sRGB,
and then was even normalized wrong for the luma formula.
|
|
|
|
|
|
|
| |
It didn't make sense to have these unimplemented,
though perhaps += shouldn't enforce a set.
Sadly, autocomplete is fairly difficult for -= of multiple items.
|
|
|
|
| |
It's pretty annoying to type `/mode -o <user>`, for little reason.
|
|
|
|
| |
Colour parsing code taken from prime.lua, and modified to strip.
|
|
|
|
| |
I've almost forgotten that we use this for message spliting.
|
|
|
|
| |
This is somewhat similar to a nick change.
|
|
|
|
|
|
|
| |
So far it's only been mentioned in the NEWS file,
which is definitely not sufficient.
It would be good to move this kind of stuff out from README.adoc.
|
|
|
|
| |
Too much repeated, non-obvious code.
|
|
|
|
| |
Just set `tls_cert`, and add `sasl` to `capabilities`.
|
|
|
|
|
| |
It doesn't require much effort to cancel capabilities, plus with
the newer version we get the respective notification anyway.
|
|
|
|
|
| |
The list may later be requested manually, which shouldn't have
an unexpected side-effect.
|
|
|
|
| |
We can receive and display capability values now.
|
| |
|
|
|
|
| |
The bug has apparently been there since the beginning.
|
|
|
|
| |
Forgotten to do it when adding the support for it.
|
|
|
|
|
| |
We might just always set the highlighted bit on,
it would be consistent with PMs.
|
|
|
|
|
|
| |
Also fixed the possibility of eating a sequence of signals
as we reset the indicators /after/ we took action,
which creates a time window for races.
|
| |
|
| |
|
|
|
|
| |
Don't trust the IRCd to have them in a subset of UTF-8.
|
|
|
|
|
|
|
| |
It was super annoying to just slightly modify strings and
string arrays, now you can have existing values filled in.
complete_word() looks a bit cleaner now as well.
|
|
|
|
|
|
|
|
| |
When fancy-prompt.lua is enabled, tho prompt is two-lined
and a simple PageUp would skip one line of content.
It works slightly better than it should: when there's under
a page of content to scroll, there is no shift at all.
|
|
|
|
|
| |
A most unfortunate 06d3b3b regression, mostly stemming from
forgetting why the `break` was in place and not documenting it.
|
| |
|
| |
|
|
|
|
|
| |
It may theoretically bite us in the ass with non-UTF-8-compliant
IRC servers, and certainly with double-width characters.
|
|
|
|
| |
Partly by unifying the interface for prompt hooks to match GNU Readline.
|
|
|
|
|
|
|
| |
Now that the input to the backlog helper is wrapped the same way
as what we display. There's a slight issue always triggered by
fancy-prompt.lua where a multiline prompt/command line makes less(1)
go too high up but it's nothing too important.
|
|
|
|
|
|
| |
There's nothing experimental about this project anymore. It's stable.
Maybe we should add a photo of Hitler or something.
|
|
|
|
| |
A few things might have changed.
|
| |
|
|
|
|
| |
To some extent they duplicate the README but from a different angle.
|
|
|
|
| |
Regression from the last release.
|
| |
|
|
|
|
| |
Now our user just needs to be able to guess that it's a hex string.
|
|
|
|
| |
It's quite unlikely that this project will ever see compiled plugins.
|
|
|
|
|
|
|
|
|
|
|
|
| |
Install plugins to /usr/share rather than /usr/lib since they're
arch-independent. Many precedents can be found for scripted plugins
in /usr/share and fewer for /usr/lib.
Look for plugins in all XDG data directories and repurpose
the "plugin_dir" setting to override this behaviour.
This adds some complexity to the bot but unifies the project.
It might make sense to remove the "plugin_dir" setting.
|
| |
|