Style "guidelines, not hard rules"
I'd like formal clarification about how we enforce our style guide. On https://bugs.webkit.org/show_bug.cgi?id=31806#c5 , Yong Li was notified that one of his patches violates a WebKit style rule (no {} on one-line conditional bodies). He objected to the rule, and George Staikos commented that the style guide rules were "guidelines and not hard rules". I'm not interested in having debates on individual bugs when people don't like particular style guide rules. I don't happen to like this rule either, but my impression is that we simply do not accept willful style guide violations. Can someone with more standing in the project than me confirm or deny that? PK
On 2009-12-01, at 21:43, Peter Kasting wrote:
I'd like formal clarification about how we enforce our style guide.
On https://bugs.webkit.org/show_bug.cgi?id=31806#c5 , Yong Li was notified that one of his patches violates a WebKit style rule (no {} on one-line conditional bodies). He objected to the rule, and George Staikos commented that the style guide rules were "guidelines and not hard rules".
I'm not interested in having debates on individual bugs when people don't like particular style guide rules. I don't happen to like this rule either, but my impression is that we simply do not accept willful style guide violations. Can someone with more standing in the project than me confirm or deny that?
If individual contributors could pick and choose which rules they applied based on their personal preference there would be little point in having the project-wide guidelines. They exist so that personal preference doesn't need to be considered from patch to patch, only when adopting new style guidelines for areas not previously covered. - Mark
On 2009-12-01, at 21:48, Mark Rowe wrote:
On 2009-12-01, at 21:43, Peter Kasting wrote:
I'd like formal clarification about how we enforce our style guide.
On https://bugs.webkit.org/show_bug.cgi?id=31806#c5 , Yong Li was notified that one of his patches violates a WebKit style rule (no {} on one-line conditional bodies). He objected to the rule, and George Staikos commented that the style guide rules were "guidelines and not hard rules".
I'm not interested in having debates on individual bugs when people don't like particular style guide rules. I don't happen to like this rule either, but my impression is that we simply do not accept willful style guide violations. Can someone with more standing in the project than me confirm or deny that?
If individual contributors could pick and choose which rules they applied based on their personal preference there would be little point in having the project-wide guidelines. They exist so that personal preference doesn't need to be considered from patch to patch, only when adopting new style guidelines for areas not previously covered.
And to quote from <http://webkit.org/coding/contributing.html>:
In order for your patch to be landed, it's necessary that it comply to the code style guidelines.
There isn't much room for interpretation in that :-) - Mark
On Dec 1, 2009, at 9:43 PM, Peter Kasting wrote:
I'd like formal clarification about how we enforce our style guide.
On https://bugs.webkit.org/show_bug.cgi?id=31806#c5 , Yong Li was notified that one of his patches violates a WebKit style rule (no {} on one-line conditional bodies). He objected to the rule, and George Staikos commented that the style guide rules were "guidelines and not hard rules".
I'm not interested in having debates on individual bugs when people don't like particular style guide rules. I don't happen to like this rule either, but my impression is that we simply do not accept willful style guide violations. Can someone with more standing in the project than me confirm or deny that?
Some rules in the style guide may call for judgment and occasionally admit exceptions, like the naming rule. But this one seems pretty clear-cut. Note: if people don't like this rule we can discuss changing the style guide. But in generally people should not be picking and choosing which style rules to follow individually. The point is to make the code base look consistent. Regards, Maciej
participants (3)
-
Maciej Stachowiak
-
Mark Rowe
-
Peter Kasting