Bumps .gitea/infra-pin from d637d0b to 0c04488, the head of admin/goodwe release/1.0. This clears the infra-pin / fresh STALE status, so release 0.14.0 gets tested against the firmware the boards now run.
What changed in admin/goodwe (d637d0b..0c04488)
FW-05 (#62: 9995c2b, 20b1d97, merge 676a6cd): adds a board status LED on GPIO4. This adds a light: block and the led_cmd_w global to master/goodwe-master-stage4-failsafe.yaml, plus master/status_led.h. The LED has no name, so HA gets no new entity. sim/mkhost.py drops light: from the host build, with a guard. Also changes master/test_failsafe.py, master/README.md and DECISIONS.
WITNESS-08 (#63: ebecd98, 530cd07, merge 0c04488): adds bus B alarms to ha/goodwe-witness-alert.yaml and a new tools/witness_alert_bus_check.py. Also changes tools/ux08_text_check.py, docs (alerts, commissioning-b, DECISIONS), and admin/goodwe's own CI: py3-jinja2 in its Dockerfile, image tag :2, and checks.yml.
Impact on the add-on
None expected. The add-on suite uses only sim/hwsim.py from admin/goodwe (via test_freeze.py), and that file is the same at both commits. The only sim change is sim/mkhost.py. It builds the host master for the sim rig and is not used by the add-on suite. The firmware's Modbus and API behaviour is unchanged: the LED is internal, and its only hook in the tick is one store after the emit. The add-on does not read the witness alert package or the witness entities. The add-on CI image (goodwe-addon-checks:1) stays as it is. The :2 tag applies only to admin/goodwe's own CI.
Proof
Locally: the full goodwe_controller/test_*.py suite in goodwe-addon-checks:1, with GOODWE_REQUIRE_HWSIM=1 and admin/goodwe checked out at 0c04488 next to the add-on (same layout as CI). Result: 60/60 PASS. The first run was cut off at 49/60 by my own timeout 1500 wrapper (exit 124, no FAIL). The other 11 were then run on their own: all PASS. test_freeze part 3 ran against the rig hwsim at /w/goodwe/sim/hwsim.py. CI run 1139 graded the same thing: infra: admin/goodwe @ 0c04488, 60/60 PASS.
On this PR: checks / unit (pull_request) and infra-pin / fresh must both be green.
Bumps `.gitea/infra-pin` from `d637d0b` to `0c04488`, the head of `admin/goodwe` release/1.0. This clears the `infra-pin / fresh` STALE status, so release 0.14.0 gets tested against the firmware the boards now run.
## What changed in admin/goodwe (d637d0b..0c04488)
- **FW-05 (#62: 9995c2b, 20b1d97, merge 676a6cd):** adds a board status LED on GPIO4. This adds a `light:` block and the `led_cmd_w` global to `master/goodwe-master-stage4-failsafe.yaml`, plus `master/status_led.h`. The LED has no name, so HA gets no new entity. `sim/mkhost.py` drops `light:` from the host build, with a guard. Also changes `master/test_failsafe.py`, `master/README.md` and DECISIONS.
- **WITNESS-08 (#63: ebecd98, 530cd07, merge 0c04488):** adds bus B alarms to `ha/goodwe-witness-alert.yaml` and a new `tools/witness_alert_bus_check.py`. Also changes `tools/ux08_text_check.py`, docs (alerts, commissioning-b, DECISIONS), and admin/goodwe's own CI: `py3-jinja2` in its Dockerfile, image tag `:2`, and checks.yml.
## Impact on the add-on
None expected. The add-on suite uses only `sim/hwsim.py` from admin/goodwe (via `test_freeze.py`), and that file is the same at both commits. The only sim change is `sim/mkhost.py`. It builds the host master for the sim rig and is not used by the add-on suite. The firmware's Modbus and API behaviour is unchanged: the LED is internal, and its only hook in the tick is one store after the emit. The add-on does not read the witness alert package or the witness entities. The add-on CI image (`goodwe-addon-checks:1`) stays as it is. The `:2` tag applies only to admin/goodwe's own CI.
## Proof
- Locally: the full `goodwe_controller/test_*.py` suite in `goodwe-addon-checks:1`, with `GOODWE_REQUIRE_HWSIM=1` and admin/goodwe checked out at 0c04488 next to the add-on (same layout as CI). Result: 60/60 PASS. The first run was cut off at 49/60 by my own `timeout 1500` wrapper (exit 124, no FAIL). The other 11 were then run on their own: all PASS. `test_freeze` part 3 ran against the rig hwsim at `/w/goodwe/sim/hwsim.py`. CI run 1139 graded the same thing: `infra: admin/goodwe @ 0c04488`, 60/60 PASS.
- On this PR: `checks / unit (pull_request)` and `infra-pin / fresh` must both be green.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
https://claude.ai/code/session_019DBUfvndgHwaV339Fvd2gr
CI-13 review: no blocking findings (Gitea will not let the shared account approve its own PR, so this is a comment).
The diff is only the pin line in .gitea/infra-pin, set to the full SHA 0c04488a9194039c2bf851b680a8ffc53beff671. That is admin/goodwe release/1.0 HEAD (ls-remote), as the README ("CI and the infra pin") asks.
Impact: git diff d637d0b 0c04488 -- sim/hwsim.py is empty. The add-on reads only ../goodwe/sim/ from the checkout: test_freeze.py uses hwsim.py, and test_tool_inputs.py loads tools/alloc_sim.py, which imports inject and sitesim. In sim/, only mkhost.py changed, and the add-on does not use it. Nothing in the add-on reads master/, ha/, tools/ or the witness files. The firmware YAML change only adds lines.
Run 1139 (job 2282) ran on head 69dc1f0 merged onto 858f77c. Log: infra: admin/goodwe @ 0c04488, 60 test file(s) to run, 60 PASS, test_freeze part 3 against /w/goodwe/sim/hwsim.py, Job succeeded. infra-pin / fresh is green.
Merging with head_commit_id pinned.
CI-13 review: no blocking findings (Gitea will not let the shared account approve its own PR, so this is a comment).
1. The diff is only the pin line in `.gitea/infra-pin`, set to the full SHA 0c04488a9194039c2bf851b680a8ffc53beff671. That is admin/goodwe release/1.0 HEAD (ls-remote), as the README ("CI and the infra pin") asks.
2. Impact: `git diff d637d0b 0c04488 -- sim/hwsim.py` is empty. The add-on reads only `../goodwe/sim/` from the checkout: test_freeze.py uses hwsim.py, and test_tool_inputs.py loads tools/alloc_sim.py, which imports inject and sitesim. In sim/, only mkhost.py changed, and the add-on does not use it. Nothing in the add-on reads master/, ha/, tools/ or the witness files. The firmware YAML change only adds lines.
3. Run 1139 (job 2282) ran on head 69dc1f0 merged onto 858f77c. Log: `infra: admin/goodwe @ 0c04488`, `60 test file(s) to run`, 60 PASS, test_freeze part 3 against `/w/goodwe/sim/hwsim.py`, Job succeeded. infra-pin / fresh is green.
Merging with head_commit_id pinned.
admin
merged commit e51e3e02b2 into release/1.02026-10-03 10:27:03 +02:00
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Bumps
.gitea/infra-pinfromd637d0bto0c04488, the head ofadmin/goodwerelease/1.0. This clears theinfra-pin / freshSTALE status, so release 0.14.0 gets tested against the firmware the boards now run.What changed in admin/goodwe (d637d0b..0c04488)
light:block and theled_cmd_wglobal tomaster/goodwe-master-stage4-failsafe.yaml, plusmaster/status_led.h. The LED has no name, so HA gets no new entity.sim/mkhost.pydropslight:from the host build, with a guard. Also changesmaster/test_failsafe.py,master/README.mdand DECISIONS.ha/goodwe-witness-alert.yamland a newtools/witness_alert_bus_check.py. Also changestools/ux08_text_check.py, docs (alerts, commissioning-b, DECISIONS), and admin/goodwe's own CI:py3-jinja2in its Dockerfile, image tag:2, and checks.yml.Impact on the add-on
None expected. The add-on suite uses only
sim/hwsim.pyfrom admin/goodwe (viatest_freeze.py), and that file is the same at both commits. The only sim change issim/mkhost.py. It builds the host master for the sim rig and is not used by the add-on suite. The firmware's Modbus and API behaviour is unchanged: the LED is internal, and its only hook in the tick is one store after the emit. The add-on does not read the witness alert package or the witness entities. The add-on CI image (goodwe-addon-checks:1) stays as it is. The:2tag applies only to admin/goodwe's own CI.Proof
goodwe_controller/test_*.pysuite ingoodwe-addon-checks:1, withGOODWE_REQUIRE_HWSIM=1and admin/goodwe checked out at 0c04488 next to the add-on (same layout as CI). Result: 60/60 PASS. The first run was cut off at 49/60 by my owntimeout 1500wrapper (exit 124, no FAIL). The other 11 were then run on their own: all PASS.test_freezepart 3 ran against the rig hwsim at/w/goodwe/sim/hwsim.py. CI run 1139 graded the same thing:infra: admin/goodwe @ 0c04488, 60/60 PASS.checks / unit (pull_request)andinfra-pin / freshmust both be green.🤖 Generated with Claude Code
https://claude.ai/code/session_019DBUfvndgHwaV339Fvd2gr
CI-13 review: no blocking findings (Gitea will not let the shared account approve its own PR, so this is a comment).
.gitea/infra-pin, set to the full SHA 0c04488a9194039c2bf851b680a8ffc53beff671. That is admin/goodwe release/1.0 HEAD (ls-remote), as the README ("CI and the infra pin") asks.git diff d637d0b 0c04488 -- sim/hwsim.pyis empty. The add-on reads only../goodwe/sim/from the checkout: test_freeze.py uses hwsim.py, and test_tool_inputs.py loads tools/alloc_sim.py, which imports inject and sitesim. In sim/, only mkhost.py changed, and the add-on does not use it. Nothing in the add-on reads master/, ha/, tools/ or the witness files. The firmware YAML change only adds lines.69dc1f0merged onto858f77c. Log:infra: admin/goodwe @ 0c04488,60 test file(s) to run, 60 PASS, test_freeze part 3 against/w/goodwe/sim/hwsim.py, Job succeeded. infra-pin / fresh is green.Merging with head_commit_id pinned.