Revert "Merge pull request #1943 from H3rnand3zzz/gmainloop
This reverts commit 609fde0998, reversing
changes made to 2ec94064ed.
Revert "Merge pull request #1948 from H3rnand3zzz/fix/rl-less-refreshes"
This reverts commit 11762fd2b0, reversing
changes made to 609fde0998.
We have got several issues, that we don't quite see how to solve, with
the merge of the gmainloop PR.
* Slashguard is broken (#1955) (though #1956 could fix that)
* One person reported problems with copy paste selection via mouse
* Some input buffer seems not to be cleared correctly
It happened that I was debugging profanity used `/connect` and typed
the password. I then debugged so long that a time out occurred, so
profanity disconnected. Then it printed "unknown command: $password".
There was something else that I forgot now.
Bottomline is: so far we didn't get it right so we will undo these
changes until someone proposes a working solution.
We got a slight performance increase (apparently noticable when
alt+mouse scrolling) but got too many issues with this change.
The redisplay function in keyboard handling event was introduced by @rodarima
in the following commit:
7eac636fc8
At the point first commit introduction it made sense, but later
the following commit added proper handling of input redisplay.
ccede06a65
This change was overlooked by me and introduced in #1943.
Potential solution for #1947.
Before there was a problem of overscrolling:
when messages longer than y axis of the terminal are fetched from the DB,
profanity scroll "jumps" to the top, skipping some messages.
It's resolved by keeping messages' starting and ending line in the
internal profanity buffer, which allows to track proper message positions
and to adjust window position accordingly.
Message size is now tracked as part of the buffer's record in `_line`
variable, which allows calculation of the total buffer size, which
might be a part of the improved solution for the "underscrolling" problem,
if we are going to limit profanity's buffer size by amount of lines as
opposed to the limitation based on the amount of message which is currently
used.
Before adding a limitation by amount of lines, careful consideration is
required, as some users don't use history and their temporary message
history can be cut to minimal limit because of 1 long received/sent message.
Underscrolling problem was fixed in a previous commit
d7e46d64fe
Short recap of the problem:
Despite user scrolling to top/bottom of history,
factual position is offset from the intended location
Another feature of this commit is a minor change which adds fetching
message stanza IDs from the DB. It allows correcting messages
fetched from history.
Fixes https://github.com/profanity-im/profanity/issues/1934
Use a singleton `Jid` inside the connection instead of always re-creating
a `Jid` from the same string.
Signed-off-by: Steffen Jaeckel <jaeckel-floss@eyet-services.de>
Before this patch the following scenario lead to a segfault:
1. open a window that sends a MAM request
2. fast enough close that window again before the MAM response was
processed
Once the MAM response is received we'd call `_mam_rsm_id_handler()`
from the `_iq_handler()` and `window` would point to a non-existant window
which leads to a segfault.
Signed-off-by: Steffen Jaeckel <jaeckel-floss@eyet-services.de>
The problem: if user scrolled to the edge of the window, the state that
edge is reached, transfers to other windows, rendering user unable to
scroll further up/down in other windows as well.
Make unique states for each window that show that edge of the DB is reached.
Later it might be improved with MAM states as well to decrease load on program
in cases when user keeps scrolling despite reaching the edge.
Bug discovered by @atomicwatch, reported through the developers' MUC
`SCROLL_INNER` naming suggestion by @jubalh
The problem is caused by ncurses library rendering only 1000 lines, while
it's possible that we have long messages in the buffer.
Current buffer limit is 200 messages, with just average size of message
of 10 lines, ncurses buffer is being overflown and we render only part
of the message. This causing a problem of "underscrolling": user is unable
to scroll to the top or bottom and he is stuck somewhere in the middle
of the history.
It's not a permanent solution, but rather a temporary fix before
introducing line-counting which will align Profanity's message buffer size
in **lines** with ncurses' limits. It might be even adjustable setting
in the future.
The problem was caused by a faulty logic that 1 message = 1 line and
it's a second try to address it, the try earlier was
ea39657b0a.
Return static tabmode as default,
separate previous change in `dynamic` mode.
Despite usefulness of the solution,
it was not approved to be a new default.
Vote (link below) amongst users to change default
has shown inconclusive result and it was not representative,
as it had low number of participants.
https://github.com/profanity-im/profanity/pull/1912#issuecomment-1816232546
This commit changes the current behaviour which displays always the first
tabs until `maxtabs` to display N tabs around the currently selected tab.
So if we are having a maxtab of 1 and the actively selected window is 2,
then 2 is displayed only.
So far we have only displayed `>` to indicate that there are more windows.
Since this PR shifts the range of tabs that are displayed we also add `<`
indicator now to indicate windows to the left of the currently displayed.
Fix https://github.com/profanity-im/profanity/issues/1283
Fix https://github.com/profanity-im/profanity/issues/1764
Let user print newline character using alt+enter keybind,
it allows great flexibility in communication,
now user will be able to write much better longer messages.
Ensure consistent invocation of `plugins_pre_chat_message_display`
for outgoing messages. Before the change, the function was not
called for sent messages upon sending, but only on fetching
sent messages from DB.
Fix https://github.com/profanity-im/profanity/issues/1917
Refactor function to enhance memory handling, addressing
temporary workaround introduced in 2e0adbd. Adjustments
ensure cleaner code and maintainability.
Part of the improvements suggested by @sjaeckel.
**Please, backup your DB before performing any testing.**
Introduce new DB structure and DB migration mechanism.
Index `timestamp`, `to_jid`, `from_jid` columns to improve performance.
Add trigger for `replaced_by_db_id` calculation by DB on message insert.
Now LMC messages are interconnected with original messages,
this way we have fast access to last (hence correct) applicable edits,
as well as reference to the original message from any edit (in case of chained edits).
Change the way LMC messages are being displayed. Now we check if we
can replace a message from current buffer. If we don't have a message in
the buffer, it might've been lost, but we can still display it as a
new message.
Further information available here:
https://github.com/profanity-im/profanity/pull/1893https://github.com/profanity-im/profanity/issues/1899https://github.com/profanity-im/profanity/pull/1902
Refactor getting char, introduce get_show_char() to follow DRY principle.
In theory, performance might be negatively affected after this change.
Ideally get_show_char should be optimized in the future.
Stop fetching message from DB when profanity reached start or end of the history
with the user. This allows to save resources on page up/down
when ceiling or bottom accordingly are reached.
Currently buffer is being redrawn on each DB fetch,
this causes a big waste of resources as on every page_up button click
we redraw 1200 messages (each message can be longer than 1 line),
commit will reduce buffer size to 200,
thus reducing amount of messages redrawn on each page up,
this is a temporary solution for a bigger problem.
Previously we printed a format string into a `GString` to then print that
result again into a `GString` and display that.
This patch removes one level of `GString` printing by forwarding the
`va_arg` to the newly added API `win_println_va()`.
Introducing `static win_println_va_internal()` also allowed refactoring
most of the `win_print*()` and `win_append*()` implementations and removing
all the redundant code.
Signed-off-by: Steffen Jaeckel <jaeckel-floss@eyet-services.de>
Thanks to @alarixnia fot the patch and bugreport.
```
Forwarding a bug reported using the NetBSD bug tracker ("Profanity crashes at login"): http://gnats.netbsd.org/57050
The code in question here handles messages by people not in the user's contact list, and what is probably happening is that the XMPP server is sending such a message before profanity has set up its window layout.
Setting roster.unsubscribed=false indeed makes profanity start without crashing, and the attached patch bails out early in the problematic code path if layout->subwin is NULL (Adrian verified for me that either of the two makes profanity work for him again). But I am not familiar enough with the profanity code to say if this has any issues.
Profanity compiled with ncurses doesn't crash because ncurses's wattron() has a guard against the user passing null pointers, but the X/Open curses library standard doesn't have this guarantee and there's no guard in the NetBSD curses implementation.
```
Fix https://github.com/profanity-im/profanity/issues/1769
