in some cases the modulename.js may need to detect running in test mode
(compliments pr #3630)
window.name is not set web mode
add a new field to the index.html
window.intest
and use the server_function to replace the hard coded string like we do
for window.mmversion=#VERSION#
then change the two test helpers to set the env variable
app.js detects and sets global.intest=true
server func replace with value of global.intest
then module can use if(window.intest)
Besides updating cspell and handling spelling issues, the important
change is adding the spelling check to the GitHub workflow.
I'm not sure if it will bother us too much when people create PRs. But I
wanted to give it a try. Or do you have any other ideas on how we can
run the spelling check on a regular basis?
This will fix#3576
@FrankBlackMG:
I don't use `*env.unique_id` because some others modules can use pm2 too
for starting a service and unique_id is the same (this will make
confusion)
So I check `name` and `pm_id` for found it
fixes#3597
Changes:
- electron tests: add mocking to `electron.js` for mocking the server
side, before only the browser side was mocked
- publish "/tests/configs" and "/tests/mocks" always in `server.js`,
this reverts a change done with latest release, we need this for
debugging (otherwise you get on the screen that your config has errors
but config check is successful ...)
- revert hotfix in
`tests/configs/modules/calendar/show-duplicates-in-calendar.js`
- fix `tests/configs/modules/calendar/custom.js` to allow events in the
past (~~I don't know how this could work before~~ when testing css
classes `yesterday` and `dayBeforeYesterday` --> it worked before
because the server side did not mock and therefore was not in the past)
since API 3.0 is default, weatherEndpoint should be set to "/onecall"
Fixes#3574
ATTENTION: since lat / lon defaults to 0 / 0, the weather plugins works
after this patch, but shows the weather from
https://de.wikipedia.org/wiki/Null_Island if lat / lon is not manually
set.
---------
Co-authored-by: Karsten Hassel <hassel@gmx.de>
Co-authored-by: Pedro Lamas <pedrolamas@gmail.com>
When a library is missing on an 3rd party module, MM² stop loading and
display a black screen. (I'm sure it's happened to everyone.)
So, I have added a try/catch block and it's avoid black screen, display
errors and allow continue loading with next module
The old `stale.yaml` seems not to work anymore, so I set up the same
content in a new github workflow.
I think we should use it again to get rid of old issues.
I discover this bug:
When pm2 `sh` script is on MagicMirror root folder, updatenotification
is not able to detect pm2 using
```
0|MagicMirror | [2024-10-02 17:23:09.215] [DEBUG] Version Compare: 2.30.0-develop 2.30.0-develop --> true
0|MagicMirror | [2024-10-02 17:23:09.216] [DEBUG] Status: online
0|MagicMirror | [2024-10-02 17:23:09.216] [DEBUG] PM2 MagicMirror starting from Path: /home/bugsounet/MagicMirror-dev
0|MagicMirror | [2024-10-02 17:23:09.216] [DEBUG] MagicMirror Path /home/bugsounet/MagicMirror-dev/
0|MagicMirror | [2024-10-02 17:23:09.216] [DEBUG] Compare: false
0|MagicMirror | [2024-10-02 17:23:09.216] [INFO] updatenotification: [PM2] You are not using pm2
```
Related to #3573
I think it's better to keep new library node-libgpiod for testing
(library to manage RPI gpio for pir sensor management) and delete others
Only one test is better
after adding check:config to the MM startup process, #3450, we
accidentally discover module positions more than once, and write the
file each time..
add a check to see if we have done this work already
# Update electron-rebuild.yaml
* remove onoff library: Not updated since 3 years, don't work with last
rpi Os
* add node-libgpiod library in replacement
Since PR #3551 was not yet complete, I made my own attempt.
1. Update to ESLint v9.
2. Replace deprecated `.eslintrc.json` and `.eslintignore` by flat
config `eslint.config.mjs`.
3. Adapt `check_config.js` to use flat config.
4. Since `eslint-plugin-import` still doesn't support ESLint v9 I
removed it. We can add it back when it does support v9.
5. Run tests `npm run check:js` and `npm run config:check`.
6. In order not to overload this PR, I have not yet activated more
additional rules - there are some useful ones in the new plugin
`@eslint/js`.
@bugsounet, please don't take it as an offence that I have created a
competing PR. The migration to ESLint v9 has been burning under my nails
for some time.
Update of package*.json for minimal node verion requirement (v20.9.0)
* it's an addition to #3556
* According to changelog v2.28.0: `⚠️ This release needs nodejs version
>= v20.9.0`
Beside testing against node version `v20.x` and `v22.x` we should also
test against the minimal node version, currently `v20.9.0`.
This is for seeing changes in dependencies which needs higher node
version as with the July-24-release, where we wrote `node >=20` but
shipped an `eslint` version which required `node>=20.9.0`.
Only details changes. No functional changes.
- remove superfluous colors in Log.error
- invert negative if
- update ESLint env
- use camel case variable name
- optimize Log strings
- [x] Base your pull requests against the `develop` branch.
- [x] Include these infos in the description:
> - Does the pull request solve a **related** issue?
Yes - solves #3534
> - If so, can you reference the issue like this `Fixes
#<issue_number>`?
Fixes#3534 (also mentioned in commit message)
> - What does the pull request accomplish? Use a list if needed.
> > - Adds a new notification (`DOM_OBJECTS_UPDATED`) when the DOM is
updated via `updateDom`
- [x] Please run `npm run lint:prettier` before submitting
- [x] Don't forget to add an entry about your changes to the
CHANGELOG.md file.
More info can be found in #3534, but as a summary:
The `updateDom` function is not synchronous - there is an undetermined
amount of time between when it completes and when the DOM has actually
been re-rendered and is ready for interaction. The existing notification
(`MODULE_DOM_CREATED`) only fires once on the initial DOM render. This
PR solves the issue of subsequent re-renders by adding a new
notification that fires whenever the DOM is ready after an update. This
notification falls within expected lifecycle notifications (very similar
to other libraries that provide DOM lifecycle notifications).
… for setting these things from outside (and overriding corresponding
config.js properties `config.foreignModulesDir` and `customCss`)
- remove elements from index.html when loading script or stylesheet
files fails
- removed `config.paths.vendor` (could never work because `vendor` is
hardcoded in `index.html`) and renamed `config.paths.modules` to
`config.foreignModulesDir`. The `config.paths. ...` properties were
implemented in the initial commit in `js/defaults.js` but were never
functional.
- fixes `app.js` which didn't respect `config.paths.modules` before
- as `modules/defaults` is directly set in many places in the source
code restrict `config.paths.modules` to foreign modules (it has never
worked for default modules), now renamed to `config.foreignModulesDir`
- adds new `/env` section in `server.js` for getting the new env vars in
the browser
- fixes TODO in `server.js` so test directories are now only published
when running tests
These changes allow changing some main paths from outside mm with the
new env vars. You now **can** put all user stuff into one directory,
e.g. the `config` dir:
- `config.js` as before
- `custom.css`
- foreign modules
This would simplify other setups e.g. the docker setup. At the moment we
have to deal with 3 directories where 2 of them (`modules` and `css`)
contains mixed stuff, which means mm owned files and user files. This
can now simplified and leads to cleaner setups (if wanted).
I felt like adding a spell checker, but it's okay if you find it
superfluous. At least then we could fix the found spell issues.
What is still missing is an automatic integration so that the spell
checker does not have to be called manually. Would it perhaps make sense
to always do it before a release?
sliceMultiDayEvents occasionally gets the number of events wrong and
produces too many rows
Math.ceil() rounds up over 1.04 so we get an abnormal count
then the calcs for the midnight loop control used different moment()
functions, producing different results
fixes#3542
