Fix issues restoring cloud projects that have been closed.
On current develop, the frontend assumes cloud projects are still open from the last time they were open. If this is not the case, it tries to open WebSocket endpoints that no longer exist
# Important Notes
For a relatively easy way to test:
- Run `Enso.dmg` or `./ide run watch`
- Open a cloud project
- Refresh Electron to make sure restoration is working
- Go to `localhost:8080`
- Close the project
- Refresh Electron to make sure startup + restoration is working
Should also test restoring local projects to make sure that didn't break.
I was unable to properly test cloud projects (the one I tested with opened, but was blank with just the IDE background and no cursor and no nodes, for some reason)
When building the dev IDE version with `./run ide build`, it creates the engine distribution with version `2023.2.1-dev`. This causes the version check to fail preventing opening the projects.
- Closes#6730
- Changes config to allow environment variables to override server host and port
- Adds port scanning to Electron app to ensure the PM is started at a free port
# Important Notes
- `SERVER_PORT=abcd enso.AppImage` does NOT work. It would not be difficult to implement, but it probably needs discussion on how exactly it should be implemented - for example, `SERVER_PORT` is quite a generic name, should the Electron app pass though something like `ENSO_PM_SERVER_PORT` to the PM as `SERVER_PORT` instead?
⚠️ Port scanning is *only* implemented in the JS frontend. It is not implemented:
- In Scala, because the JS/Rust code calling it needs to know the port as well. There shouldn't be any problems with adding port scanning though, if that's desired
- In Rust, because I'm not sure parsing the host and port from a string is a good idea.
- (This also applies to JS, but it *must* work in JS, and port scanning is already a dependency there so it's quite a bit easier)
- QA *will* need a new PM (`sbt buildProjectManagerDistribution` or `./run backend sbt` -> `buildProjectManagerDistribution`), and the path must be supplied as: `-engine.project-manager-path=path/to/new/pm/here`
- Fixes#7603
- "Move All To Trash" option sometimes appears even when the list is empty
# Important Notes
This appears to be caused by multiple things:
- deleting a single item does not remove it from the set of selected keys
- waiting for `project/close` to make sure a project can be `project/delete`d, takes a while.
Both issues have been fixed, by adding a new `AssetListEvent` to let the assets table to immediately remove the asset's key (its id or placeholder id) from the set of selected keys
This commit introduces a new cursor shape, a double arrow, to be displayed during the hover over the resize grip in the graph editor's visualization component. The rotation of the double arrow depends on whether the cursor is to the left or right of the dividing line from the top left to bottom-right corner of the grip. This gives a more intuitive visual cue to users about the direction of resizing.
Looks like that
```
[info] [2023-08-23T13:12:58.119Z] [org.enso.projectmanager.boot.ProjectManager$] Starting Enso Project Manager
Version: 0.0.0-dev
Built with: scala-2.13.11 for GraalVM 17.0.7
Built from: wip/db/fix-dev-version-check* @ 52bc6b8fcf
Built on: Linux (amd64)
```
- Update list of groups to agreed list.
- Lower case `ALIAS` names to be consistent with function names.
- Add `GROUP` to methods.
- All constructors and functions have doc comments.
- Correct a few typos (e.g. `PRVIATE`).
- Mark some more things as `PRIVATE`.
- Use `ToDo:` and `Note:` consistently.
- Order tags in doc comment.
# Important Notes
We don't have all the doc comments on types and will want to add them in future,
- Closes#5159
- Now data downloaded from the database can keep the type much closer to the original type (like string length limits or smaller integer types).
- Cast also exposes these types.
- The integers are still all stored as 64-bit Java `long`s, we just check their bounds. Changing underlying storage for memory efficiency may come in the future: #6109
- Fixes#7565
- Fixes#7529 by checking for arithmetic overflow in in-memory integer arithmetic operations that could overflow. Adds a documentation note saying that the behaviour for Database backends is unspecified and depends on particular database.
close#7604
After moving the rename action to the dashboard, IDE is unaware of the new project name. PR implements a new `refactoring/projectRenamed` notification that is sent from the server to clients and informs them about the changed project name.
# Important Notes
https://github.com/enso-org/enso/assets/357683/7c62726d-217e-4e69-8e48-568e0b7b8c34
- Closes https://github.com/enso-org/cloud-v2/issues/610
- New user menu
- Remove "go to profile" action that does not currently have an action, and does not exist in new design
- Add placeholder icons for existing actions
- Re-style "change password" modal to fit in with the design
# Important Notes
There are many differences from the design - none are visual differences though:
- The list of actions is completely different - there are no menu entries in common between the design and the current
- This also means that *all* current icons are placeholders. There are no appropriate icons in the "icons" Figma tab either.
- The user icon is still a placeholder, as there is no backend support for user icons yet.
- The user menu entries are highlighted on hover (not specified in the design), but to make this look nice, some of the padding has been moved from the outer container to the individual menu entries.
- The menu entries use the same component as the context menu, so they *do* support shortcuts, and adding shortcuts to them will be very easy, *however* no shortcuts have been set for the new actions, because they are different from the actions in the Figma design (and so they don't have an official default shortcut)
- Closes https://github.com/enso-org/cloud-v2/issues/511
- Adds sorting. Currently supported columns: "Name" and "Last Modified"
# Important Notes
The sort arrows have a slightly thicker border (changed from 2px to 2.14px), to remove the (very small) internal hole in the Figma design. It is possible to make the shape more accurate to the original design by using a polygon (or a path) that traces around the original outline instead, but I figured it's not worth spending the extra time on a fix that may not be correct.
ℹ️ The comparison function for sorting is quite complicated. I think this is the least intrusive change for now, but it is worth considering changing `AssetsTable` to store items internally as a tree instead, and do a preorder traversal to flatten it into an array when needed.
- Closes https://github.com/enso-org/cloud-v2/issues/615
- Add category switcher
# Important Notes
None of the buttons currently do anything. This is intentional, as the corresponding backend functionality does not yet exist.
- Closes#7576
- Stop using icons for non-macOS platforms, because it is non-standard to use icons
- Increase width otherwise entries will line wrap
# Important Notes
None
New rules to recognize `"""` and `'''` at the end of line and color everything that's inside nested block of text as string. It properly _stops the string literal when nested block ends_.
# Important Notes
The [text mate grammar rules](https://macromates.com/manual/en/language_grammars) properly _end_ the text block when the block ends. For example in:
```ruby
main =
x = """
a text
y
```
`a text` is properly recognized as string.
- Closes#7536
- Top bar is now below fullscreen visualizations
Other changes:
- Fully hides mouse cursor in editor view
# Important Notes
There should be basically no logic changes - check the code to be extra sure.
The main things that should be tested are:
- cursor should be hidden on top bar in the editor, and visible when not in the editor
- cursor should be visible in the "change password" modal
Use the new Enso Font; also change the anti-aliasing logic to be based on device pixel ratio, rather than platform. This will improve the clarity of font rendering on Windows/Linux machines with high pixel densities.
Design reference:
Tested on various combinations of DPR/platform:
OS X, `devicePixelRatio` = 2 (should look similar to how we were already rendering *mplus1* on OS X):
<img width="1440" alt="Screenshot 2023-08-07 at 5 46 11 PM" src="https://github.com/enso-org/enso/assets/1047859/2fdf251a-ba5e-426f-b6c4-194347a9cee4">
Windows, `devicePixelRatio` = 1.25 (should look similar to how we were already rendering *mplus1* on this platform/DPR):
Linux, `devicePixelRatio` = 1 (should look similar to how we were already rendering *mplus1* on this platform/DPR):
# Important Notes
Style changes:
- Use the Enso Font for code in Rust, replacing the DejaVu fonts.
- Use the Enso Font in HTML: code in documentation, and error visualizations.
- Change SpanWidgets from Bold to Extra Bold, to match the design.
Implementation improvements:
- The new font download is cached (and Github-authenticated); this should eliminate a "rate limit" build failure I've
encountered in the past.
- Clean up DocSection HTML rendering a bit.
- Remove a CSS file that seems to have been superseded.
- Closes#7498
- Validation regex previously disallowed non-ASCII characters, now fixed
- Changes every password input to use the correct regex. For old/new passwords, it uses the regular input. For password confirmations, it uses the new password, converted to a regex. This makes it so that the password confirmation input's validation only succeeds when it exactly matches the new password.
# Important Notes
None
* Support arguments list in the doc parser
* Support new doc sections in documentation panel
* Remove headers
* Remove outer dom and place breadcrumbs
* Fix methods icon
* Use unordered list class in css
* Improve tags styles
* Remove virtual component groups docs
* Cleanup top-level css styles
* Small adjustments to headers
* Add styles for emphasized text
* Add bold font for arguments
* Self-review
* Remove redundant placeholder struct.
* Update outdated doc.
* Avoid allocation when comparing strings.
* Avoid empty paragraph.
* Reduce allocations.
* Update test to remove empty paragraph.
* Fix rebase issues.
* Improve padding and size handling in UI themes
Added padding_x and padding_y to hardcoded theme's breadcrumb settings to ensure consistent padding. Also, these padding settings and breadcrumb_height are now used directly in the Style structure, eliminating hardcoded values in the view documentation.
* Adjusted breadcrumb background dimensions calculation.
* Add support for improper arguments formatting in documenation comments
* Do not include Icon tag into the docs
* Fix documentation panel resizing
* enso-formatter
---------
Co-authored-by: Michael Mauderer <michael.mauderer@enso.org>
While investigating behavior of
```
sbt:std-benchmarks> withDebug benchOnly --dumpGraphs -- Vector_Operations.Max_Stat
```
in IGV I realized there is a deep chain of nodes when reading an element of `Vector` related to work with warnings. There is an invocation of `WarningsLibrary` on `this` - that's probably unnecessary as we know how it is going to resolve. This PR skips such one level of indirection by directly delegating to `this.storage`.
However I haven't seen any effect of this change on peak performance. The library overhead seems to disappear. Anyway I wanted to bring this finding to your attention and perform independent measurement on our CI server.
Fixes#7423
Refactored widget matching algorithm to allow creating wrapper widgets even in cases where the widget config override is present. That allowed the widgets to be reordered, such that the argument name ends up being inside the dropdown widget. That way clicking it opens the dropdown.
Added explicit manual layout for the dropdown arrow position. Now it is positioned on the center of a selected appropriate child widget. For prefix chains, the leftmost part of the prefix application (the method or constructor) is selected.
https://github.com/enso-org/enso/assets/919491/86678b9d-c04e-4545-8d11-2d1e85b9b951
Fix for `showLogs` when opening a second cloud project. Alternative to #7568.
This one avoids loading the cloud's `index.js.gz`, unless it is in the cloud - and in the cloud, it only loads the first `index.js.gz`.
