mcptask-rails-runner 0.3.21 → 0.3.23
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/CHANGELOG.md +92 -0
- data/lib/mcptask_rails_runner/version.rb +1 -1
- metadata +1 -1
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 6d2a73c2de4ee862b3ce4354468099e515f0cbc59a14d2b910c18173b13e7a59
|
|
4
|
+
data.tar.gz: 8c89bceed9444902402d06f427a19fbf9bac18539713909f453ad0b424649285
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: f470628c32c59d11b1ae94e14391d46616fbbc11e1c1802d9ef93b794b9b41f75580b882b625dfe607d5d31b34f34497d8b698e9fcb2694805f5753c95939467
|
|
7
|
+
data.tar.gz: 0db755a5d77e70fb516581913998e8ef7ea8f916ec2f7d2bfbceb69f81a47309f536c49a337a9b6af5558546f6e835e5148e14323360121e141e592a334ae2cd
|
data/CHANGELOG.md
CHANGED
|
@@ -5,6 +5,98 @@ entries below describe wrapper changes only. Binary changes are listed in the
|
|
|
5
5
|
[mcptask-releases](https://github.com/jchsoft/mcptask-releases/releases)
|
|
6
6
|
release notes for the same tag.
|
|
7
7
|
|
|
8
|
+
## 0.3.23
|
|
9
|
+
|
|
10
|
+
No wrapper changes. Full notes in
|
|
11
|
+
[mcptask-releases v0.3.23](https://github.com/jchsoft/mcptask-releases/releases/tag/v0.3.23).
|
|
12
|
+
|
|
13
|
+
**Upgrading is the usual one command** — `mcptask_runner update --self`. The
|
|
14
|
+
skills-and-helpers data pack did not change in this version, so a follow-up
|
|
15
|
+
bare `update` is not required; the changes below live in the binary itself.
|
|
16
|
+
|
|
17
|
+
Binary changes carried by this version:
|
|
18
|
+
|
|
19
|
+
- **`--ignore-quota` now means ignore, everywhere.** The daily mode's
|
|
20
|
+
day-level scheduler used to keep asking mcptask.online about the budget even
|
|
21
|
+
with the flag set, so a REST outage could park an unattended daily host
|
|
22
|
+
until the next business day. With the flag set the runner now performs no
|
|
23
|
+
budget queries at all; the end-of-workday clock still applies — it is a time
|
|
24
|
+
rule, not a quota rule (task #11891).
|
|
25
|
+
- **The work window keeps its minutes.** `mcptask_runner init --until 18:30`
|
|
26
|
+
now stores `work_window.end_of_workday_minute` next to the existing hour key
|
|
27
|
+
and the loop stops picking up new work at 18:30, not 18:00. The key is
|
|
28
|
+
additive: a config without it means `:00`, so existing hosts change nothing.
|
|
29
|
+
A value the config cannot hold is refused loudly, never rounded — `--until`
|
|
30
|
+
used to keep only the hour and say nothing (tasks #11892, #11896).
|
|
31
|
+
- **`update` protects locally modified helper scripts** the way it has always
|
|
32
|
+
protected skills: content hashes in a manifest, one of five outcomes per
|
|
33
|
+
file (added / up-to-date / updated / conflict-skipped / force-updated), and
|
|
34
|
+
a `.bak` next to anything `FORCE=1` overwrites. On the first `update` after
|
|
35
|
+
this release, a helper edited by hand reports `conflict-skipped` instead of
|
|
36
|
+
being silently overwritten — that is the fix working, not a failure
|
|
37
|
+
(task #11894).
|
|
38
|
+
- The Ruby-gem conformance baseline now runs nightly in CI rather than on
|
|
39
|
+
every push, and the docs stopped contradicting each other about its role
|
|
40
|
+
(task #11893). Stale comments aligned (task #11895) and a Windows-only
|
|
41
|
+
test flake pinned down (task #11897).
|
|
42
|
+
|
|
43
|
+
## 0.3.22
|
|
44
|
+
|
|
45
|
+
No wrapper changes. Full notes in
|
|
46
|
+
[mcptask-releases v0.3.22](https://github.com/jchsoft/mcptask-releases/releases/tag/v0.3.22).
|
|
47
|
+
|
|
48
|
+
**Upgrading takes two commands this time, not one.** The usual
|
|
49
|
+
`mcptask_runner update --self` replaces the binary, and that is enough for the
|
|
50
|
+
loop changes below. It is *not* enough for the skills and helpers: those are a
|
|
51
|
+
data pack the binary carries, and they are only written into a project by
|
|
52
|
+
|
|
53
|
+
```
|
|
54
|
+
mcptask_runner update
|
|
55
|
+
```
|
|
56
|
+
|
|
57
|
+
Run it in each host project after updating the binary. Skipping it leaves the
|
|
58
|
+
old `/test-runner` and `/ci-runner` in place, and on a project that is not
|
|
59
|
+
Rails those name commands it cannot run — which is the very thing this version
|
|
60
|
+
fixes. A host that has locally edited a skill keeps its edit; `FORCE=1` takes
|
|
61
|
+
the new one and leaves a `.bak`.
|
|
62
|
+
|
|
63
|
+
Binary changes carried by this version:
|
|
64
|
+
|
|
65
|
+
- **The bundled CI and test skills stop naming a framework.** `/test-runner`
|
|
66
|
+
and `/ci-runner` used to spell out `bin/rails test`, `test:system` and
|
|
67
|
+
RuboCop, and `check_test_lock` recognised a running suite by six regexes,
|
|
68
|
+
five of which were Ruby. On a Django, Node or Swift host the commands did
|
|
69
|
+
not run and the concurrency guard waved every real test command through —
|
|
70
|
+
blind on exactly the projects a universal runner exists for. Both now read
|
|
71
|
+
what the project declares in `.claude/test-commands.json`, which
|
|
72
|
+
`mcptask_runner init` writes and `update` backfills; when the file is absent
|
|
73
|
+
the toolchain is detected from `go.mod`, `Gemfile`, `package.json`,
|
|
74
|
+
`manage.py`, `pyproject.toml` or `Package.swift`, and when nothing is
|
|
75
|
+
recognised the runner says so instead of guessing a command.
|
|
76
|
+
|
|
77
|
+
- **Two more places the prompt spoke Rails to everyone.** The auto-squash
|
|
78
|
+
workflow told every project to run `bin/rails assets:precompile
|
|
79
|
+
RAILS_ENV=test` between its unit and system suites, and the manual
|
|
80
|
+
screenshot step named a Rails class. Neither is measured by what it breaks —
|
|
81
|
+
the first costs a turn or invites a plausible substitute, the second is
|
|
82
|
+
silently ignored — but both contradict the claim the product makes.
|
|
83
|
+
|
|
84
|
+
- **A failed task now waits an hour, not the rest of the day.** A task that
|
|
85
|
+
fails is set aside and the runner moves on; the day no longer ends over one
|
|
86
|
+
piece.
|
|
87
|
+
|
|
88
|
+
- **The dashboard can take a piece off the skip list.** A piece set aside
|
|
89
|
+
earlier in the day can be released from the card, and the runner picks it up
|
|
90
|
+
again without a restart.
|
|
91
|
+
|
|
92
|
+
- **A closed socket is described by whoever actually saw it close.** Two paths
|
|
93
|
+
reported a dropped dashboard connection — the read loop, which knows the
|
|
94
|
+
close code the server sent, and a failed write, which knows only that the
|
|
95
|
+
socket was gone. Whichever finished first wrote the notice, so a close with
|
|
96
|
+
a code could be reported as a bare transport error. The reader's account now
|
|
97
|
+
wins, which is what makes "the server closed it, code 1001" trustworthy when
|
|
98
|
+
you are deciding whether to go and look at the server.
|
|
99
|
+
|
|
8
100
|
## 0.3.21
|
|
9
101
|
|
|
10
102
|
No wrapper changes. Full notes in
|
|
@@ -10,5 +10,5 @@ module McptaskRailsRunner
|
|
|
10
10
|
# Binary::INSTALL_DIR. So the pin is per project and per foreground rake
|
|
11
11
|
# invocation; the installed copy is per machine, and the newest binary any of
|
|
12
12
|
# those projects offers is the one that stays.
|
|
13
|
-
VERSION = "0.3.
|
|
13
|
+
VERSION = "0.3.23"
|
|
14
14
|
end
|