| Summary: | Port PrettyPatch.rb to Python | ||||||
|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Kent Tamura <tkent> | ||||
| Component: | Tools / Tests | Assignee: | Nobody <webkit-unassigned> | ||||
| Status: | RESOLVED WONTFIX | ||||||
| Severity: | Normal | CC: | aroben, cmarcelo, hamaji, mrowe | ||||
| Priority: | P2 | ||||||
| Version: | 528+ (Nightly build) | ||||||
| Hardware: | All | ||||||
| OS: | All | ||||||
| Attachments: |
|
||||||
|
Description
Kent Tamura
2010-08-06 04:12:34 PDT
Created attachment 63713 [details]
Patch
Interesting. Is Ruby not available on some systems that need to use PrettyPatch? I have no philosophical objection to rewriting in Python. I think Python's difflib.SequenceMatcher could replace your DiffBuilder. The WebKit project has a number of scripts that are written in Ruby. Why is this suddenly a problem? Ruby is used only for PrettyPatch on Chromium port. Ruby is available on all Chromium platforms, but we'd like to minimize dependency as possible. We're setting up Chromium buildbots for tests. The webkitpy-test step is failing because the buildbot machine has no ruby [1]. We can install ruby, but we'd like to avoid to install it if possible. > I think Python's difflib.SequenceMatcher could replace your DiffBuilder. Thanks. I'll try it after I confirm the Python version of PrettyPatch works. [1] http://build.webkit.org/builders/Chromium%20Linux%20Release%20(Tests)/builds/3340/steps/webkitpy-test/logs/stdio I would prefer that Ruby be installed, since several recently-added scripts have been written in Ruby and it’s not out of the question that they may be used by the Chromium build in the future. Rewriting in a different language for the sake of it seems like a bad idea to me. I won't work on it because of objections. |