aboutsummaryrefslogtreecommitdiff
path: root/cmd
Commit message (Collapse)AuthorAgeFilesLines
* the directory lock is the kernel's, not a pid we believeLukasz Kasprzak47 hours3-12/+58
| | | | | | | | | | | | | | | | | | | | | | | | | | flock(2) on the lock file's descriptor replaces "write my pid, and decide whether the pid in the file is still alive". The kernel drops the lock when the process ends, however it ends, so there is no stale krino lock to detect and no takeover to race over. What that fixes: - Two runs that both judged a lock stale could remove and recreate it and both believe they held it. Remove-then-create cannot be made atomic; there is nothing to make atomic now. Pinned by a test with eight callers over twenty rounds. - A pid reused after a crash made the lock live for ever, and the message named neither the file nor the pid, so there was nothing to act on. The message now names both. - Signal(0) reads EPERM as "not running", so a lock held by another user was taken over. There is no such judgement left to get wrong. A run that waits for a held lock now says so first. Waiting is what the spec asks for, but the wait has no timeout, and in silence it is indistinguishable from a hang - I spent two minutes on one myself today, waiting on a lock the window was holding. Tests that faked a held lock by writing a file now hold a real one.
* the module path is git.labunix.xyz/krinoLukasz Kasprzak48 hours21-47/+47
| | | | | | | | | | | So that go install can find it. cgit serves no go-import meta tag, so nginx answers a ?go-get=1 request for /NAME with one pointing at https://git.labunix.xyz/NAME.git; the alternative was carrying a .git suffix through every import line forever. One sed over the imports, both go.mod files, and the dependency gate's whitelist. Nothing else depends on the path. go install works from the next tag, the first release whose go.mod carries it.
* comments that explain the code, not how it was writtenLukasz Kasprzak2 days14-270/+260
| | | | | | | | | | | | | | | | | | About 340 comments cited the development process: task and plan numbers, fix waves, rulings, reviewers, and the author in the third person with a date. None of that exists outside the work itself, so to a reader it pointed at nothing. Each one now states the engineering reason it was standing in front of; where a comment was provenance and nothing else, it is gone. References to docs/design.md and docs/gui-design.md by section stay: both ship with the repository. The design documents lose their amendment diaries - CHANGELOG.md is that record - and the GUI's says plainly that the window has gone further than the document. Only comments changed. Every .go file was parsed and its code printed with comments stripped, before and after: the two hashes are identical across all 175 files.
* milestone 1 review: claims span the run, explain's chain is opt-in and its ↵Lukasz Kasprzak3 days1-0/+4
| | | | own, overrides keyed by clean path, splice and enum guards
* undo runs through the same sessionLukasz Kasprzak3 days2-41/+25
|
* the engine owns a run: lock, log, run id, claimsLukasz Kasprzak3 days1-73/+19
|
* an undecided (stop) rule ends the search for that fileLukasz Kasprzak4 days2-1/+27
|
* only where an earlier directory's files ended up stays claimedLukasz Kasprzak4 days2-1/+44
|
* review and undo headings wrap past the label column; format characters take ↵Lukasz Kasprzak4 days5-7/+68
| | | | a column
* an earlier directory's applied results stay claimed for later onesLukasz Kasprzak4 days2-2/+51
|
* krino log marks a run partly undone while reversible steps remainLukasz Kasprzak5 days1-4/+11
|
* tests: apply error exits 1 and w stops krino through the real command; undo ↵Lukasz Kasprzak5 days2-1/+74
| | | | projection occupied half
* output: wrap by terminal columns, names cannot fake step lines, -v lists ↵Lukasz Kasprzak5 days4-19/+143
| | | | unscanned destinations
* --json carries exclusions and matching warningsLukasz Kasprzak5 days2-1/+64
|
* a real run's later directories are not blocked by earlier claims; -n across ↵Lukasz Kasprzak5 days2-4/+46
| | | | directories documented
* explain: shows a duplicate's skipped delete, leaves the cache alone, walks ↵Lukasz Kasprzak5 days2-0/+9
| | | | only for duplicate tests
* tests run on OpenBSD; a tool's error message survives one large write; lock ↵Lukasz Kasprzak5 days1-1/+3
| | | | pids beyond 32 bits name no process
* plan 10 re-check: cut run column, damaged undo run, emptied directory ↵v0.0.7Lukasz Kasprzak5 days3-2/+12
| | | | cleanup, text turning binary, explain flags, interrupt docs
* plan 10: docs (-n, README version and scope, trash identity, --json U+FFFD, ↵Lukasz Kasprzak5 days1-1/+1
| | | | limitations, changelog)
* plan 10: missing and weak tests (per-step logging, trash path, fuzz oracle, ↵Lukasz Kasprzak5 days3-16/+57
| | | | w after error)
* plan 10: commands refuse global flags they do not useLukasz Kasprzak5 days7-0/+75
|
* plan 10: stderr messages cannot be split by quoted newlines; config paths ↵Lukasz Kasprzak5 days9-32/+115
| | | | escaped; krino new refuses control characters; independent terminal oracle
* plan 10: an interrupt stops between steps; nohup keeps ignoring hangupsLukasz Kasprzak5 days1-1/+7
|
* plan 9: undo review matches review, --min-age validated, future mtimes, rule ↵Lukasz Kasprzak5 days9-24/+132
| | | | names, conflict enum, dependency gate, absolute tool paths
* plan 9: errors, explain traces, JSON and log fields never reach the terminal rawLukasz Kasprzak5 days8-16/+150
|
* plan 9: an interrupted or failed undo can be finishedLukasz Kasprzak5 days2-2/+80
|
* plan 9: apply logs each step as it completes and stops a chain that landed ↵Lukasz Kasprzak5 days1-1/+1
| | | | elsewhere
* plan 8: enum values read from source must all be handledLukasz Kasprzak5 days1-0/+35
|
* plan 8: escape terminal controls in everything krino printsLukasz Kasprzak5 days9-26/+252
|
* krino: 0.0.6 — w applies and quits, choices echoed in redv0.0.6Lukasz Kasprzak5 days4-28/+141
|
* krino: 0.0.5 — keyword cache, t and d in reviewv0.0.5Lukasz Kasprzak5 days9-45/+234
|
* 0.0.4: max-size, exclude forms, --min-agev0.0.4Lukasz Kasprzak5 days7-1/+231
|
* krino: 0.0.3 — the plan as one block per file, wrapped, and -Pv0.0.3Lukasz Kasprzak5 days10-227/+461
| | | | | | | | Each file shows its steps, then the rule and the reason it matched, one field per line; on a terminal every line wraps to its width with continuation lines under their own column, and piped output is never wrapped. Choosing per file shows the same block. -P / --no-pager prints the plan without the pager. A duplicate's original is shown with ~.
* krino: coloured output, and --no-colorLukasz Kasprzak5 days11-106/+364
| | | | | | | | | One palette type styles the plan table, warnings, headers, outcome counts, prompt keys, undo's refused steps and krino log's (undone), from the 16-colour ANSI palette plus bold and faint only. Widths are measured on the plain text, so columns line up; with colour off the output is unchanged. --no-color works before or after any subcommand, as NO_COLOR does. The two search-and-replace colourings are gone.
* krino: duplicates are found, never deletedLukasz Kasprzak5 days2-26/+79
| | | | | | | | | A rule combining (duplicate) with a delete action is refused at load, and a file that is a duplicate under any duplicate scope its directory uses gets no delete step from any rule: the plan shows it skipped and the chain continues. A failed duplicate check blocks the delete too. Tests cover (matched), another rule's own condition, a test evaluation skipped, two scopes and a failed check, each checking every copy is still on disk.
* krino: release 0.0.1 — man pages, install, examples, cross and release, ↵v0.0.1Lukasz Kasprzak6 days5-3/+172
| | | | | | | | | README, changelog Also: undo removes the directories its run created; a hardlink is never a duplicate of its own other name; a flag written before "undo" is honoured; --version prints no leading v. Duplicate conditions with different scopes not sharing an original is documented as a known limitation.
* krino: acting — trash, journal, apply, lock, review, undoLukasz Kasprzak7 days12-50/+1863
|
* krino: planning — chains, placeholders, conflicts, JSONLukasz Kasprzak7 days4-103/+696
|
* krino: matching — scan, ignore, conditions, extraction, duplicates, ↵Lukasz Kasprzak7 days8-21/+749
| | | | explain, dry run
* krino: foundation — sexp reader, config language, init/new/checkLukasz Kasprzak8 days7-0/+437
generated by cgit v1.3 (git 2.53.0) at 2026-09-19 10:24:19 +0000