Conversation
so that this values do not extend above 360 (or bellow -360). This was resulting in some e2e test flakiness when a 360 deg would sometimes be interpreted as 360, and other times 0. With this change, it will be way more predictable and consistent
Small improvement : wrap camera angles
When changing the config (opacity, visibility) of an external layer (KML, GPX, WMS or WMTS) and then use the browser history navigation to revert the changes, it triggered a reload of the layer and set its name back to the default until the layer has been reloaded. We can simply avoid this by updating the current layer object instead of creating a new one.
PB-322: Avoid reloading external layer upon browser history navigation
and deconstruct its content when called, should shorten most calls to constructors that were using many default/blank/null values to reach one last argument and set it to something else Adding the updateDelay param to GeoJson layer (it will be relevant only relevant for this type of layers)
calculated once and for all at layer creation
we had once some attributes that were getThisId or someUrl, and other that used getThisID and someURL... I thought it would be easier to navigate all this if we sticked to a single way of naming thing, hence the move to baseUrl instead of baseURL
hopefully it is a bit clearer what this is about with this name
checking that what has been deconstructed from the single Object entry is sufficient to go on building the instance. Also moving the timeConfig directly in GeoAdminLayer, so that it may be set to any kind of layer (will be useful in the future) and add layer config parsing of hasLegend
OL features getId was caught by my IDE as I hadn't enabled case sensitivity... So this reverts all .getId() being replaced by .id on OL features A couple tests also slipped by when renaming baseURL to baseUrl And finally there was a little mistake in how kmlMetadata was handled when loading a legacy KML
set default attribution to false across the board, and either fail if none is given (for GeoAdmin layers) or create a default one based on the host for externals So, in essence, attributions are mandatory for GeoAdmin layers, and optional for externals change LayerLegend and GroupOfLayers to also be based on an args constructor
PB-207 preparation work
In the legacy viewer the layer title was part of the URL and some layer could contain a coma in its title which broke the layers parsing as the layers in URL are separated by coma.
In the WMS we can add a group of layer form the base of the capabilities.
Legacy external WMS layer name could contain comma, therefore added a comma to the unit tests. On the other hand new url don't contains anymore the layer name in the url. Also the legacy WMTS url did not contain the layer name therefore remove it from the test
The base layer was not parsed, so added a unit test to cover this.
PB-102: Fix legacy external layers with coma in layer tilte
When loading the menu as async componenent, it adds a kind of race condition when toggling in 3d mode for the first time making the application crash. So avoid this race condition by keeping the menu module within the index chunk. NOTE: I did not track down the race condition to understand what was really the issue.PB.2559: Fix toggling in 3d race condition
PB-259: Manual chunks for better startup performance
This is a 1 to 1 rewrite of the infobox module, switching it to composition API. Since we're going to change some logical lines within this file, it stands to reason to change it. The 'scrollTo' doesn't seem to work, but it's the same as in the current implementation.
upgrade to OL v9.0.0 and update other libs
and waiting on the animation of the menu to finish before giving the caller the control back (should alleviate some race condition we had with menu open/close and action triggered just after the call to the util)
…test_utils Replace the single clickOnMenuIfMobile by two open/close utils
Avoid loading multiple times the WMS/WMTS capabilities and avoid doing multiple dispatch. This reduce the startup time especially with multiple layers of the same source we can gain several seconds.
Make sure that that the layers coming from different urls and when adding more than one layer from a provider that the get capabilities is not called several times. Also make sure that the correct layer config is applied.
The layer id and also URL could contains specials characters ;,@| that are used by the parsing. So we need to encode them. They are encoded using percent encode. Other special characters are encoded by vue router.
The test now test the initial loading of several layers and make sure that the Get Capabilities are called only once
External layer that did not specified its opacity in URL could not be displayed as they default opacity was set to undefined. This has been broken with PR #677
The list of layer was inverted in the test.
When exiting the drawing, we could have a race condition between the closing save and pending debounced save. Sometimes a pending debounce save, not started was started after the closing and closing save, which saved an empty kml because the drawing layer would have been removed. This was making some test failing time to time and could also have been an issue for a real user if he would close the drawing too quickly.
Fix save drawing race condition
The logic has two issues, one is that for group of layers we counted always one index too much, and that for the on top of all layers we did not took the group of layers into account. This solve the issues in the e2e tests for the external layers.
PB-339: Fix layer z-index computation
Passing run #954 ↗︎Details:
Review all test suite changes for PR #692 ↗︎ |
|||||||||||||||
Apparently the debouncing of the feature title and description in drawing together with the save drawing debouncing creates a race condition when editing the title and/or description and closing the drawing mode. In this case the save kml triggered by the title/description update debounced happens only when the drawing mode is closed and when the drawing layer has removed all its feature which result to saving an empty drawing, clearing the existing drawing. To avoid this we speed up debouncing, this is not a perfect fix but it works and I could not find a way of correctly cleaning up the debouncing for the closing drawing menu event.
PB-341: Fix race condition in drawing e2e tests
…typing-issue PB-244: Fix typing issues of props
…css-styling-and-text-selectability PB-244: Coordinate display css styling and text selectability
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Test link