Add an AI client module that integrates with OpenAI-compatible API
providers (OpenAI, Perplexity, and custom providers) to provide
AI-assisted responses within the profanity client.
The implementation includes:
- src/ai/ai_client.c/h: Core AI client with provider management,
session handling, and async HTTP request handling via libcurl.
Supports per-provider API keys stored in preferences, reference-
counted sessions, and conversation history tracking.
- src/ui/window.c/window_list.c: New AI window type (ProfAiWin) for
displaying AI conversations, with response streaming and error
display capabilities.
- Command integration: New `/ai` command (cmd_defs.c, cmd_funcs.c)
for creating sessions, sending prompts, and managing providers.
Provider autocomplete support in cmd_ac.c.
- Preferences integration: API keys for providers are persisted in
the preferences system (config/preferences.c).
- Unit tests: 472 lines of comprehensive tests covering provider
management, session lifecycle, JSON escaping, and autocomplete
(tests/unittests/test_ai_client.c).
Architecture decisions:
- Asynchronous design: HTTP requests run on a separate thread to
avoid blocking the main UI loop. Callbacks are invoked on the main
thread via direct function call (profanity uses ncurses, not GLib
main loop).
- Reference counting: Both AIProvider and AISession use ref counting
for safe shared ownership.
- Response size limit: 10MB cap on HTTP responses to prevent OOM.
f27fa98 commit introduced issue of scrolling seemingly randomly being stuck and the following message flood in the logs: "WRN: Ncurses Overflow..."
A pad is an off-screen buffer — larger than the visible terminal screen — that you can scroll, render parts of, or update selectively.
Unlike regular windows, a pad is not tied to screen size. You can create a pad that's 1,000 lines tall, even if your terminal is only 40 lines tall.
This commit addresses the issue by increasing PAD_SIZE to a reasonable number,
allowing messages from buffer to be displayed without overflowing NCurses' window size (maxy).
Minor change: add .cache folder to gitignore
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.
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.
Fixes https://github.com/profanity-im/profanity/issues/1794
Explanation
The problem is the download's identifier. Downloads are given an ID so they can be referenced later when their progress changes. Currently, the download's ID is the download's URL. When you download the same file twice, you have two downloads with the same ID. Download progress updates are shown on the first of both downloads with the same ID.
Solution
Change the download's ID from its URL to a random number. A random ID is generated when get_random_string() is called from cmd_funcs.c. Several other functions are updated to cope with the new ID format.
Fixes https://github.com/profanity-im/profanity/issues/1649
Type `>` then press tab or shift tab to autocomplete previous messages,
then type your reply and send message.
Newlines are replaced with newline followed by `> `.
A newline is added at the end so that the user can immediately type a
message without manually adding a new line.
- Add NetBSD as a recognized platform without -ldl.
- Allow building with NetBSD libcurses instead of ncurses.
- Portability to NetBSD sh - use POSIX '=' instead of '=='.
In aa3693daa211b36c78d136d5a1ee9f3258e21352 I renamed
`win_println_me_message()` -> `win_print_outgoing_muc_msg()`.
Now: `win_println_them_message()` -> `win_println_incoming_muc_msg()`
to be more consistent and descriptive.
