-
Notifications
You must be signed in to change notification settings - Fork 3.1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
shadow-dom: Reorganize the test structure #130
shadow-dom: Reorganize the test structure #130
Conversation
This script converts tests under the shadow-dom/ directory to separate HTML files with appropriate metadata format. The output is produced within the out/ directory by default to avoid conflicts with existing files, but I intend that the files under the out/ directory will take over the existing contents of the shadow-dom/ directory. The script file will be removed after the conversion.
This change reorganizes the tests under the shadow-dom/ directory. In the new format, each test has a dedicated, clean HTML with appropriate metadata. The new format provides a clear directory structure based on the section hierarchy of the specification. In addition, numerous issues with the existing tests are addressed, such as unnecessary files, messy metadata definitions, and so on. The new files are generated by convert.py, which is included in the previous commit. There are some issues I could not address using a script; most notably each test does not have a sensible name. These issues need to be addressed manually, so I left these as future work.
Just as a general note, I don't like having tests named with just a number rather than a descriptive name. OTOH, you're not making to worse. |
@Ms2ger: I'm aware that such style is bad. I did so because I was not able to extract a sensible name in an automatic way. I'll probably rename the tests on one-by-one basis in another iteration. |
shadow-dom: Reorganize the test structure
Awesome! I was about to do something similar as this was blocking coverage analysis. I agree there's value in keeping the section numbers around but that isn't consistent with the rest of how the repo is organized. It's also more brittle as section names don't change when new sections get added above them while section numbers do. I'd suggest removing those altogether. |
@tobie: Okay, I'll make a change that drops the section numbers, probably sometime tomorrow (I'm somewhat busy today). |
…nite_loop Fix potential infinite loop in AssertionError.get_stack(); r=Ms2ger
Switch to more compact results representation format.
dd57b9f0db chore(package): update dev deps 7f99dea07c chore(package): bump version number to 10.2.1 c357d4cb01 Add bug link and remove unnecessary quote tests cd1f7a95ba Add a comment about not using for (const ...) 39a957721c Tokenise a solitary '/' correctly 4efd252f5e Use "let" instead of "var" in for statement f940514f36 Add count() function. Use idiomatic string methods. affdfbdc2b Optimise tokenisation and whitespace skipping ba00d5d9ec refactor: small syntax changes (web-platform-tests#137) d5c4fbfdfc chore(CHANGELOG): regenerate 3d009b634a chore(CHANGELOG): regenerate 23bb9bb290 chore(CHANGELOG): regenerate 6db306af62 chore(package): bump version number to 10.2.0 ac7ef088f3 fix: solve conflicts from typing union types f422f2a6f5 feat: type on union types (web-platform-tests#135) bde0553b49 feat: add const-type for idlTypes (web-platform-tests#132) c03cd7e8f3 feat: add dictionary/typedef-type (web-platform-tests#133) 6eb1e7f4db feat: add argument/return type (web-platform-tests#134) d2cfdfd901 feat: add type: attribute-type on idlTypes (web-platform-tests#131) 1e29dcb71b Auto acquisition for parser result changes (web-platform-tests#130) 154eabfbb1 chore(package): update mocha, expect, bump version 36932debd7 Let error messages include the current definition name (web-platform-tests#129) 664f63b61e chore(package-lock): regenerate 536157bffa chore(package): bump version number to 10.0.0 d5b88179df chore(pacakge): upgrade expect dep 0c103b356e Maintain writer.js (web-platform-tests#122) e2d4467ea1 remove typeExtAttrs from docs 0226b76587 remove iterator documentation (web-platform-tests#123) e36ae6bf70 BREAKING CHANGE: remove deprecated iterator operation (web-platform-tests#121) 8e73c4ff4c use for-of on tests c56a921d9b docs(README): iterables ildType is always array 3f39cb152e chore(CHANGELOG): regenerate ae0060f859 chore(package): bump version number to 9.0.0 24669ed245 BREAKING CHANGE: consistent array type for iterable.idlType (web-platform-tests#117) f89b5803c7 Update package-lock.json f581ac63c3 Add myself to contributors in package.json 420ac52ac5 Revert "chore: drop Node 6 support (web-platform-tests#102)" 1c031ed86e chore(CHANGELOG): regenerate git-subtree-dir: resources/webidl2 git-subtree-split: dd57b9f0db1adbb5712f9fdd6a4e38533ff4ba4b
This pull request basically converts the tests under the shadow-dom/ directory into separate HTMLs, in order to have better alignment with other test suites in web-platform-tests.
This branch consists of three commits:
(1) a change adding a script named convert.py that does the conversion;
(2) the actual conversion (note: this change is huge, so it will take some time to load on GitHub UI);
(3) a change removing convert.py.
If you are interested in what is the actual change, please consult the content of convert.py. I did not make any manual changes in the commit (2), except for moving the content of the output directory of the script.