Problem: M._cache = cache_data captured the initial empty table reference
at module load time. After M.load() reassigns cache_data to the decoded
JSON, M._cache is permanently stale and returns the wrong table.
Solution: remove the field assignment and expose get_raw_cache() which
closes over cache_data and always returns the current table.
Problem: vim.loop is deprecated since Neovim 0.10 in favour of vim.uv.
Five call sites across scraper.lua, setup.lua, utils.lua, and health.lua
still referenced the old alias.
Solution: replace every vim.loop reference with vim.uv directly.
Also fix contest-change detection so URL open logic triggers when either platform or contest changes. This makes :CP next/:CP prev and problem jumps open the correct page when open_url is enabled.
Co-authored-by: Codex <noreply@openai.com>
Problem: after an install or update, the on-disk cache may contain data
written by an older version of the plugin whose format no longer matches
what the current code expects.
Solution: embed a CACHE_VERSION in every saved cache file. On load, if
the stored version is missing or differs from the current one, wipe the
cache and rewrite it. Corrupt (non-decodable) cache files are handled
the same way instead of only logging an error.
Add LuaCATS annotations to the env conversion helper and drop the table.sort call since ordering is not required by uv.spawn.
Co-authored-by: Codex <noreply@openai.com>
Neovim/libuv spawn expects env as a list of KEY=VALUE strings. Passing the map from vim.fn.environ() can fail process startup with ENOENT, which breaks NDJSON test scraping and surfaces as 'Failed to start scraper process'.\n\nConvert env map to a deterministic list before uv.spawn in the NDJSON scraper path.
Co-authored-by: Codex <noreply@openai.com>
Problem: setup_python_env() is called from check_required_runtime()
during config.setup(), which runs on the very first :CP command. The
uv sync and nix build calls use vim.system():wait(), blocking the
Neovim event loop. During the block the UI is frozen and
vim.schedule-based log messages never render, so the user sees an
unresponsive editor with no feedback.
Solution: remove setup_python_env() from check_required_runtime() so
config init is instant. Call it lazily from run_scraper() instead,
only when a scraper subprocess is actually needed. Use vim.notify +
vim.cmd.redraw() before blocking calls so the notification renders
immediately via a forced screen repaint, rather than being queued
behind vim.schedule.
Problem: with debug = true, there is not enough diagnostic output to
troubleshoot environment or execution issues. The resolved python path,
scraper commands, and compile/run shell commands are not logged.
Solution: add logger.log calls at key decision points: python env
resolution (nix vs uv vs discovery), uv sync stderr output, scraper
subprocess commands, and compile/run shell strings. All gated behind
the existing debug flag so they only appear when debug = true.
Problem: setup_python_env() skips uv sync when .venv/ exists. If a
previous sync was interrupted (e.g. network timeout), the directory
exists but is broken, and every subsequent session silently uses a
corrupt environment.
Solution: remove the isdirectory guard and always run uv sync. It is
idempotent and near-instant when dependencies are already installed,
so the only cost is one subprocess call per session.
Problem: tbl_deep_extend merges user platforms on top of defaults, so
all four default platforms survive even when the user only configures a
subset. The picker then shows platforms the user never intended to use.
Solution: before the deep merge, prune any default platform not present
in the user's platforms table. This preserves per-platform default
filling (the user doesn't have to re-specify every field) while ensuring
only explicitly configured platforms appear.
Problem: when required dependencies (GNU time/timeout, Python env) are
missing, config.setup() throws a raw error() that surfaces as a Lua
traceback. On macOS without coreutils the message is also redundant
("GNU time not found: GNU time not found") and offers no install hint.
Solution: wrap config.setup() in pcall inside ensure_initialized(),
strip the Lua source-location prefix, and emit a vim.notify at ERROR
level. Add Darwin-specific install guidance to the GNU time/timeout
not-found messages. Pass capability reasons directly instead of
wrapping them in a redundant outer message.
Problem: the deprecated vim.g.cp_config fallback was kept for
backwards compatibility after the rename to vim.g.cp in v0.7.6.
Solution: drop the shim entirely and update the setup() deprecation
target to v0.7.7.
Problem: when opening a contest for the first time (metadata not
cached), the setup_code hook fired before state.set_language() was
called, causing state.get_language() to return nil inside the hook.
Solution: call state.set_language(lang) before the hook in the
provisional-buffer branch of setup_contest(). The value is already
computed at that point and is identical to what setup_problem() sets
later, so the early write is idempotent.
