related to #3185
* I have limited updated dom to one update
-> `updateDom()` is activated by a timer which resets when a new event
arrives (`CALENDAR_EVENTS`)
* I have set no speed to self update. I think it's not necessary
-> update it directly
If somebody can test and tell me result
In all case, I will patch my prod mirror for testing
Fixes#3227
and also
- removes unused code
- de-duplicates code fragments
- fixes typos
- inlines code
- adds more weather util tests
@martingron and @Justheretoreportanissue would you be so kind to check
if I didnt mess anything up?
I just see `electron` package used in develop branch have `1 high
severity vulnerability`
Detail is [there](https://github.com/advisories/GHSA-j7hp-h8jx-5ppr)
We can patch it with electron v26.2.2 (last version at this day) and
will correct it
(ChangeLog is not needed in this case)
---------
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: veeck <michael.veeck@nebenan.de>
Fixes https://github.com/MichMich/MagicMirror/issues/3182
Fixed issue with probability of precipitation not showing up on hourly
or daily forecast
---------
Co-authored-by: veeck <michael.veeck@nebenan.de>
Hi,
Just because, i never try to code a test
I purpose to supervise this work
After, perhaps it is not necessary to integrate it in develop branch.
It's up to you to decide
Hello,
This pull request allows you to add a class to the tr of the event
sought in customEvents. You must enter the class with the "eventClass"
option.
---------
Co-authored-by: TeddyStarinvest <teddy.payet@starinvest.com>
PR: #3113
I see this bugs:
AnimateCSS merge hide() and show() animated css class when we do
multiple call
--> result it will stay en hide state
I think event listener (is animateCSS file) is not a proper solution
I correct it with like traditional code with timer
Fix too: AnimateIn on first start
... in `config.js`.
Solves #3174
With getting width/heigt from
`electron.screen.getPrimaryDisplay().workAreaSize` introduced with
https://github.com/MichMich/MagicMirror/pull/3161 the solution was to
remove the `setFullscreen` line.
So per default the fullscreen resolution is used but when someone now
uses `electronOptions.width`/`electronOptions.height` in `config.js`
these parameters are used and so "no fullscreen" is possible.
related to #2649
I was able to move to internal fetch and all tests seems fine so far.
But we have one problem with the calendar module. In the docs we have
several authentication methods and one of them is `digest`. For this we
used `digest-fetch` which needs `node-fetch` (this is not so clear from
code but I was not able to get it working).
So we have 3 options:
- remove `digest` as authentication method for calendar module (this is
what this PR does at the moment)
- find an alternative npm package or implement the digest stuff
ourselves
- use `digest-fetch` and `node-fetch` for calendar module (so they would
remain as dependencies in `package.json`)
Opinions? @KristjanESPERANTO @rejas @sdetweil @MichMich
Update calendarfetcherutils.js to force recurrence date time to be the
same as event datetime
I found an issue with one of my calendars displaying the wrong time for
certain recurring events. Each event was set up by someone in a
different timezone (Central European) than my own (Eastern US). I traced
the issue back to the `Rrule.between()` method generating odd time
portions under certain circumstances.
The fix I found was to set the UTC time portion of the recurrence
datetime to be the same as the UTC time portion of the event start date.
This resolved the issues with the maladjusted event times, and had no
effect on other event times. While there may be edge cases that are
affected, I have been unable to locate any.
---------
Co-authored-by: Veeck <github@veeck.de>
Hi,
This is my testing code for AnimateCSS for `show()`, `hide()`,
`updateDom()`
Naturally, we have to do better !
I voluntarily modify `newsfeed` and `compliments` in order to test
Note: I will correct checks later... it's a test...
---------
Co-authored-by: bugsounet <bugsounet@bugsounet.fr>
Co-authored-by: veeck <michael.veeck@nebenan.de>
Fixes#3126
Added the option `allowOverrideNotification` to `weather.js`. This
allows the module to receive the `CURRENT_WEATHER_OVERRIDE`
notification. The expected payload for this notification is a
full/partial `weatherObject` that is used to supplement/replace the
existing `weatherObject` returned by whichever weather provider is in
use.
No visual changes.
First time contributing - let me know if I've missed something
🙂
---------
Co-authored-by: veeck <michael.veeck@nebenan.de>
This PR uses `electron.screen.getPrimaryDisplay().workAreaSize` under
electron to get the screen size.
If this fails the current defaults (800x600) are used.
This solves some problems with xrandr under bullseye where the sreen
comes up with 800x600 instead of fullscreen, e.g. described
[here](https://khassel.gitlab.io/magicmirror/pi-modules/).
Tested this on my pi setup.
By the way, the Issue #1919 is back on my side (not related to this PR)
...
Takes https://github.com/MichMich/MagicMirror/pull/3153 (why does that
always base against master despite the yaml configuring it against
develop?) and updates othe rminor dependencies (not prettier v3 that
will take a little time)
---------
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
there are no "all" files in the v5 versions of roboto-fonts anymore so I
used the content of their css files in `roboto.css`.
~~PR is set to draft because @rejas wants to wait until next release is
done.~~
