Merge back master into updated-latest-electron #1277

mauricioszabo · 2025-04-25T03:04:19Z

No description provided.

…for spellchecking on a per-language basis. This opens up things that aren't really possible with the `excludedScopes` setting. For instance: you can specify `source.python comment, source.python string, source.js comment`; this will enable spellchecking in Python files for both comments and strings, while enabling spellchecking in JavaScript files for _only_ comments. Closes #1135.

This PR also makes this compatibility check testable, and even expandable, by being able to provide a new platform if preferred.

Fix Linux trash error message

* Add comments to `WASMTreeSitterLanguageMode`… …and emit an event in `IndentResolver::suggestedIndentForBufferRow` that was missed last time. * [language-c] Fix highlighting of `&` in a reference declarator (Should've made it into last month’s!) * [language-c] Move `reference_declarator` declaration to C++-only file… …because it's not valid C. * Attempt to fix chronically failing spec I _think_ this is happening because we're not waiting for the watcher to start. `nsfw` goes async when the watcher is created _and_ when it's started.

…rm-compatibility-check [settings-view] Fix Package keymap compatibility check

…nularity [spell-check] Allow the user to whitelist sections of a buffer…

v1.124.0 Changelog

Need to fix a compilation error in CI which is resulting from missing X11-related headers. Some of these deps presumably have been removed or changed in the recent "ubuntu-latest means Ubuntu 24.04" change. See this tracking issue about the runner migration for details, maybe: github[dot]com/actions[slash]runner-images/issues/10636 So, adding all of these deps just to be sure our code can compile and be tested successfully in CI. These are the same deps as used above in the main "build" Linux part of the workflow file.

…st-and-upload-Linux-job CI: Add build dependencies for Linux 'test bins' job

v1.124.0 Release

Also, adjust the timeout duration to be shorter, and adjust the number of retries to be greater. I figure successful runs usually succeed quickly (~27 minutes or less?) So, get to the retrying faster! But also, let's actually retry, a bunch more times if we have to!

…ethod [core] Expose `dbPromise` in `StateStore`

* [spell-check] Change test so it works identically across platforms * [spell-check] Remove errant `console.log`… …and clean up specs. * [spell-check] Revert addition to `spell-check.grammars`… …of the `"source comment"` value. A user can add it back in if they want spell-checking of code comments, but it's probably a bit too opinionated to be the default.

…ls (#1185) * [pulsar-updater] Don't prompt to update on non-default release channels * [pulsar-updater] Fix spec

37 minutes is a bit long. 31 minutes should allow most runs that are *going* to finish to finish, without wasting more time than we need to waiting for runs that are hung and are never going to finish anyway. Considered 27, since *most* runs finish at 16 minutes, and all but one of my ~11 test runs were under 27 minutes, but *one* of my test runs was just under 31 minutes, so this should do the trick. Having to wait for an entire additional run by being over-eager to retry (too impatient) feels wasteful and not the best trade-off. Anyhow, this is closer to the right number, even if picking a better "exact ideal timeout" may be possible, this is close enough.

Seems unreliable and not the right tool when we have apt-get. Also, our "package tests" CI recently stopped working with an error about broken/unsatisfiable package dependencies. This fixes it. --- For those who are curious... Passing (older image) details: Image: ubuntu-20.04 Version: 20241215.1.0 Included Software: https://github.com/actions/runner-images/blob/ubuntu20/20241215.1/images/ubuntu/Ubuntu2004-Readme.md Image Release: https://github.com/actions/runner-images/releases/tag/ubuntu20%2F20241215.1 Vs failing (newer image) details: Image: ubuntu-20.04 Version: 20250105.1.0 Included Software: https://github.com/actions/runner-images/blob/ubuntu20/20250105.1/images/ubuntu/Ubuntu2004-Readme.md Image Release: https://github.com/actions/runner-images/releases/tag/ubuntu20%2F20250105.1 (But now the tests pass even on the newer runner image, as of this fix.)

Attempt to install dependencies on new Ubuntu CI images

v1.126.0 Release

* Remove run-tests scripts Pro: * Less code to maintain & update * run-package-tests was unused Con: * Lost the formatting * Drop support for JUnit XML output for tests Pro: * Less code to maintain * Introduce helper/jasmine-singleton to have reference only to the vendored jasmine * only the vendored version got used during the tests * don't need to require jasmine multiple times Clean up vendored jasmine to make it easier to upgrade Remove checks about global jasmine as we're not setting it anymore * Move jasmine-test-runner to be under spec/runners * Introduce jasmine2 test runner * Improve test runner switcher * Fix spec files running with jasmine 2 * Upgrade archive-view to use jasmine2 test runner * Add retry mechanism to jasmine2 test runner * Write a wrapper to eliminate done callbacks from async tests * Run flaky spec only on linux * Fix the non-headless spec runner Co-authored-by: Andrew Dupont <[email protected]> * Fix core specs after merging from `master` * Add ability to specify trailing comment pattern in `runGrammarTests` * Add list reporter for CI --------- Co-authored-by: Andrew Dupont <[email protected]>

Wayland pulsar script

* Upgrade to `web-tree-sitter` v0.25.3 This is a big upgrade! `web-tree-sitter` has been rewritten in TypeScript and has made some backwards-incompatible changes to its exports — all of which are good things. I've also updated the instructions for generating the `web-tree-sitter` bundle. * [language-python] Various folds fixes… * Make function parameters spanning multiple lines fold properly * Make `try`/`except` clauses fold properly, with `try` folding only up to the first `except` * Add `folding-spec.js` to test these cases and others and guard them against regressions Also fix a silly mistake in an indentation query. Fixes #1216. * [language-python] Convert specs to Jasmine 2 runner * Get grammar specs working on Jasmine 2 runner * Monkey-patch `Query::captures` more directly * [language-ruby] Properly highlight and indent new `case/in` syntax * [language-ruby] Properly fold `case` statements and `when/if` clauses * Add logging to compare/contrast CI output between Pulsar and PulsarNext * [language-javascript] Ensure `foo` in `this.#foo = 0` highlights properly * Purge a stale fold boundaries iterator… …instead of silently failing. For some reason, a fold cache wasn't updating in PHP in certain circumstances. We were trying to resolve a fold on a given row by inspecting a capture whose node was useless to inspect because its tree had long since been invalidated. This was only reproducible on startup, so maybe it has to do with non-root language layers not having their fold caches purged after tokenization? This isn't easy to reproduce in a way that allows us to write a spec, so I've written this code very conservatively so that it can't possibly regress — only do slightly more work than it used to. One way around this would be to extract the metadata we need when we create the boundaries iterator (so that we don't have to inspect the node later). That would theoretically increase the lifetime of the cache, but would involve doing more work up front. So it may or may not be worth it performance-wise. (It's not like the folds query is particularly costly, and it might be pointless to try to extend the cache's lifespan when code changes constantly and the _underlying data_ will be stale after a short while.) * [language-php] Fold multiline comments * Remove temporary logging

Includes the ppm Pull Request: - Update to read v3 (ppm PR 150)

ppm: Update ppm to commit a6f843f0381f64cb5865efc7

Co-authored-by: Andrew Dupont <[email protected]>

….127.0 Changelog: Update changelog for Pulsar 1.127.0

meta: Update editor version to 1.127.0 (Pulsar v1.127.0 Release)

v1.127.1 Hotfix Release

…tron

savetheclocktower

When you said you were going to make this PR a while back, I was a bit puzzled — I didn't think I had diverged from master in any meaningful way.

In fact, there are a couple places where I failed to pick up one CI change or another — but other than that, if I accepted this PR, it would undo a lot of work I've done to ensure specs pass and to get some important features working.

I think this PR could be cleaned up, but honestly I think the only legitimate changes worth picking up are in .cirrus.yml and build.yml, and even on those I'll defer to @DeeDeeG to decide if they're worth adopting.

So that should give you peace of mind — if you see any divergence from master in the main source code at this point, it's on purpose. Other than what's happened since 1.127.0, the updated-latest-electron branch is up to date with master, and tomorrow I'll bring in the changes since then and put out a PulsarNext 1.128.0.

savetheclocktower · 2025-05-03T07:08:25Z

.cirrus.yml

    APPLEID: ENCRYPTED[549ce052bd5666dba5245f4180bf93b74ed206fe5e6e7c8f67a8596d3767c1f682b84e347b326ac318c62a07c8844a57]
    APPLEID_PASSWORD: ENCRYPTED[774c3307fd3b62660ecf5beb8537a24498c76e8d90d7f28e5bc816742fd8954a34ffed13f9aa2d1faf66ce08b4496e6f]
    TEAM_ID: ENCRYPTED[11f3fedfbaf4aff1859bf6c105f0437ace23d84f5420a2c1cea884fbfa43b115b7834a463516d50cb276d4c4d9128b49]
-    ROLLING_UPLOAD_TOKEN: ENCRYPTED[41b279482c1b61c9129e410233cc5cd71bbfd60e33d03cefae451aff6bc9915d1706a6e8fefd0e0f9041a4bd287c84be] # <-- This is the Electron Next Bins token! Careful when resolving merge conflicts to not overrite to/from master. master branch keeps its token, updated-latest-electron branch keeps this one!


@mauricioszabo, can you revert the change on this line? Per the comment, this upload token should remain distinct on the updated-latest-electron branch.

savetheclocktower · 2025-05-03T07:12:17Z

spec/text-editor-component-spec.js

 describe('TextEditorComponent', () => {
  beforeEach(() => {
-    if (!window.customElements.get('text-editor-component-test-element')) {
+    if(!window.customElements.get('text-editor-component-test-element')) {


No! These if(s shall not stand! :)

savetheclocktower · 2025-05-03T07:13:05Z

spec/text-editor-component-spec.js

-
-      // Set `line-height` to an integer to save us headaches later. (We
-      // believe that Chromium used to guarantee fixed increments of line
-      // height, but doesn't anymore — so we must specify it as an integer
-      // value ourselves if we want the math to be simpler.)
-      element.style.lineHeight = '20px';
-      component.measureCharacterDimensions();
-
-      // Make the editor tall enough to fit only two lines at a time.


All the additions in this file are purposeful; they're what I had to add to get specs to pass in Electron 30. They need to stay.

savetheclocktower · 2025-05-03T07:13:33Z

spec/theme-manager-spec.js

 const fs = require('fs-plus');
 const temp = require('temp').track();

-const { conditionPromise: waitForCondition } = require('./helpers/async-spec-helpers');


All the changes in this spec file are on purpose and should stay.

savetheclocktower · 2025-05-03T07:18:03Z

spec/workspace-element-spec.js

 const path = require('path');
 const temp = require('temp').track();
-
-const { conditionPromise } = require('./helpers/async-spec-helpers');


All the changes in this spec file are on purpose and should stay.

savetheclocktower · 2025-05-03T07:18:40Z

spec/workspace-spec.js

        let pane4 = null;
        let editor;

-        await atom.workspace


All the changes in this spec file are on purpose and should stay.

savetheclocktower · 2025-05-03T07:19:23Z

src/main-process/atom-application.js

 const AtomWindow = require('./atom-window');
 const ApplicationMenu = require('./application-menu');
 const AtomProtocolHandler = require('./atom-protocol-handler');
-const { onDidChangeScrollbarStyle, getScrollbarStyle } = require('./scrollbar-style');


All the changes in this file are purposeful and should stay.

savetheclocktower · 2025-05-03T07:19:39Z

src/wasm-tree-sitter-grammar.js

    await parserInitPromise;
    if (!this._language) {
      try {
-        this._language = await WASMTreeSitterGrammar.loadLanguage(this.treeSitterGrammarPath);


This is a purposeful change and should stay.

savetheclocktower · 2025-05-03T07:21:56Z

packages/language-python/package.json

  "license": "MIT",
  "atomTestRunner": "runners/jasmine2-test-runner",
  "dependencies": {
+    "tree-sitter-python": "0.19.0"


Let's please not add back a Tree-sitter dependency.

DeeDeeG · 2025-05-13T20:40:14Z

Seems like a "process of how we work with these branches" PR as opposed to a strictly code-focused PR.

@savetheclocktower has been cherry-picking from master, this merges from master.

I dunno if it is necessary from a "syncing the code" perspective, but it might clarify the relationship of the branches from a revision history/version tracking (lower-level git-isms) perspective? Effectively marks how much diff we want to yeet vs master and what we want to confirm is desired, and let git understand such explicitly...

But the real thing is whether to be doing merges or cherry-picks, I think that should be down to the preferences of active contributors to this branch IMO. If there's, say, @savetheclocktower (preferring cherry-picks?) and @mauricioszabo (preferring merges?) that's a tie and I dunno.

But maybe a chat on the Discord could clear up how to handle history divergence (cherry-picks or merge commits) in the future for this work ??

(I don't think it makes sense to scrutinize this PR's diff closely without resolving that larger-picture question.)

savetheclocktower · 2025-05-13T21:26:29Z

Maybe this will put everyone's mind at ease: updated-latest-electron is never going to get PR’d against master.

When we're ready (or nearly ready) to ship Pulsar on Electron 30, I'll create a new branch off of master. I'll use the diff between updated-latest-electron and master as a reference and a target to reach, but I'll recapitulate the changes in a much more logical series of commits. (One commit for removing legacy Tree-sitter, one for updating pathwatcher, etc.) When that candidate branch has zero diff with updated-latest-electron, it'll be suitable for landing.

In the meantime, rest assured I'm keeping this branch up to date.

savetheclocktower and others added 30 commits December 1, 2024 13:42

Fix Linux trash error message

bc3ceae

Fix package-keymap-view compatibility check

bcb9a2d

This PR also makes this compatibility check testable, and even expandable, by being able to provide a new platform if preferred.

Rename function to match expectations. Flip all boolean logic

d624d26

Merge pull request #1151 from pulsar-edit/fix-error-msg

d754ed7

Fix Linux trash error message

Merge pull request #1161 from pulsar-edit/settings-view/keymap-platfo…

7ab0460

…rm-compatibility-check [settings-view] Fix Package keymap compatibility check

Update changelog notes

f695ec4

[spell-check] Add source comment to default grammars setting

8d6e677

Merge pull request #1147 from savetheclocktower/spell-check-scope-gra…

b04e997

…nularity [spell-check] Allow the user to whitelist sections of a buffer…

Add new PR to changelog

5e5889a

Merge pull request #1164 from pulsar-edit/1.124.0-changelog

d8c0585

v1.124.0 Changelog

Changelog: Add PR 1165 (deps fix for Linux CI)

b3a6042

Merge pull request #1165 from pulsar-edit/install-missing-deps-for-te…

82955a6

…st-and-upload-Linux-job CI: Add build dependencies for Linux 'test bins' job

Update Version

6675e9b

Add back build identifier

c971c1b

Merge pull request #1166 from pulsar-edit/v1.124.0-release

a5fccfc

v1.124.0 Release

Expose dbPromise in StateStore

bcf7f48

Add specs

4788055

Merge pull request #1171 from pulsar-edit/expose-database-dbPromise-m…

e46ba9a

…ethod [core] Expose `dbPromise` in `StateStore`

Only remove items from PATH on uninstall, not upgrade

1be24bb

[pulsar-updater] Don't prompt to update on non-default release channe…

64ec928

…ls (#1185) * [pulsar-updater] Don't prompt to update on non-default release channels * [pulsar-updater] Fix spec

Merge branch 'master' into fix-windows-path-addition

360dbb7

Attempt to install dependency on new Ubuntu CI images (#1192)

1ec04f6

Attempt to install dependencies on new Ubuntu CI images

Install dependencies for CI documentation job (#1193)

d90260e

DeeDeeG and others added 27 commits March 5, 2025 22:21

Merge pull request #1238 from pulsar-edit/v1.126.0-release

397d66b

v1.126.0 Release

GH Action Documentation

619ae10

Adding wayland parameters to pulsar.sh script

4ffe10d

Changelog update

76ffb12

Removed duplicated condition

90f12d9

Refactored to EXTRA_ARGS

eccee53

Merge pull request #1246 from pulsar-edit/wayland-pulsar-script

a74eda7

Wayland pulsar script

GH Action Documentation

194ce77

ppm: Update ppm to commit a6f843f0381f64cb5865efc7

e483c45

Includes the ppm Pull Request: - Update to read v3 (ppm PR 150)

Merge pull request #1252 from pulsar-edit/upate-ppm-late-march-2025

4939501

ppm: Update ppm to commit a6f843f0381f64cb5865efc7

Changelog: Update changelog for Pulsar 1.127.0

090c368

Changelog: Apply review suggestion

a852d09

Co-authored-by: Andrew Dupont <[email protected]>

Merge pull request #1254 from pulsar-edit/Changelog-Regular-release-1…

0b05270

….127.0 Changelog: Update changelog for Pulsar 1.127.0

meta: Update editor version to 1.127.0

7e52810

meta: Add back "-dev" to the version string

aa753c4

Revert "Wayland pulsar script" (#1261)

e0845d7

Merge pull request #1257 from pulsar-edit/v1.127.0-release

d981004

meta: Update editor version to 1.127.0 (Pulsar v1.127.0 Release)

Changelog: Add changelog entries for hotfix v1.127.1

c1f0cdc

meta: Update version to "v1.127.1" for hotfix release

62d16f6

meta: Add back "-dev" to version string

d0cded6

Merge pull request #1262 from pulsar-edit/v1.127.1-release

1bdcb9b

v1.127.1 Hotfix Release

Merge remote-tracking branch 'origin/master' into updated-latest-elec…

0c4ed41

…tron

Lockfile update

8a29334

Fixed merge

c6ee0c1

More merge problems fixed

ea36799

savetheclocktower requested changes May 3, 2025

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

Merge back master into updated-latest-electron #1277

Merge back master into updated-latest-electron #1277

Uh oh!

mauricioszabo commented Apr 25, 2025

Uh oh!

savetheclocktower left a comment

Uh oh!

savetheclocktower May 3, 2025

Uh oh!

savetheclocktower May 3, 2025

Uh oh!

savetheclocktower May 3, 2025

Uh oh!

savetheclocktower May 3, 2025

Uh oh!

savetheclocktower May 3, 2025

Uh oh!

savetheclocktower May 3, 2025

Uh oh!

savetheclocktower May 3, 2025

Uh oh!

savetheclocktower May 3, 2025

Uh oh!

savetheclocktower May 3, 2025

Uh oh!

DeeDeeG commented May 13, 2025 •

edited

Loading

Uh oh!

savetheclocktower commented May 13, 2025

Uh oh!

Uh oh!

Uh oh!

Merge back master into updated-latest-electron #1277

Are you sure you want to change the base?

Merge back master into updated-latest-electron #1277

Uh oh!

Conversation

mauricioszabo commented Apr 25, 2025

Uh oh!

savetheclocktower left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

DeeDeeG commented May 13, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

savetheclocktower commented May 13, 2025

Uh oh!

Uh oh!

DeeDeeG commented May 13, 2025 •

edited

Loading