Alex Shi <seakeel@xxxxxxxxx> 于2021年12月8日周三 14:16写道： > > On Wed, Dec 8, 2021 at 12:55 PM yanteng si <siyanteng01@xxxxxxxxx> wrote: > > > > Tang Yizhou <tangyizhou@xxxxxxxxxx> 于2021年12月8日周三 12:35写道： > > > > > > On 2021/12/8 11:16, Alex Shi wrote: > > > > On Tue, Dec 7, 2021 at 10:57 PM Jonathan Corbet <corbet@xxxxxxx> wrote: > > > >> > > > >> Alex Shi <seakeel@xxxxxxxxx> writes: > > > >> > > > >>> What the result of scripts/checkpatch.pl or make htmldocs say? > > > >>> Let's tame these tools and follow the regluar styles of kernel. > > > >> > > > >> checkpatch knows nothing about different languages, and "make htmldocs" > > > >> doesn't do that sort of stylistic checking. > > > > > > > > Thanks for the info! > > > > > > > >> > > > >> I don't know enough (i.e. anything :) to have an opinion on the best > > > >> length for lines of Chinese text. Surely there must be some conventions > > > >> out there? > > > > > > > > The only reason I know is to remove extra spaces char in 'rendered > > > > documentation' > > > > which is generated by line break. For this purpose, 40 or 60 line chars limit > > > > can't fulfill it obviously. > > > > > > > > Yizhou's suggestion has no line break unless a punctuation hitted, so the > > > > space will happen with the punctuation. This makes the rendered document look > > > > good, but will lead to toooo long and never alignment lines in the > > > > source document. > > > > It looks like we have no good way to resolve this issue by style > > > > change. Also the '60' > > > > number seems an arbitrary number? If we have to relief this problem from coding > > > > style change, why not follow the new rule after checkpatch's change: > > > > from 80 to 100, > > > > let's say about 50 Chinese chars plus breaking after a meaningful Chinese word? > > > > > > Agreed. > > > > > > Now I prefer 50 Chinese chars plus breaking after a meaningful Chinese word. > > All guys: > > > > ref <https://stackoverflow.com/questions/8550112/prevent-browser-converting-n-between-lines-into-space-for-chinese-characters> > > Nice finding! > > > > > I have tested that firefox has no space problem. Looks like some > > browser issues, But I still think it's because rst didn't consider > > Chinese support when it was originally designed. > > > > I'm sure there are many ways to solve the space problem elegantly, but > > changing the document source code is definitely not one of them. > > Right. > > Just since checkpatch has changed, so let's move on... > And of cause, 40 chars width is also acceptable if someone like it. OK! I agree with 50 chars. We need to take care of the kernel developers who don't have a GUI, and the doc developers who use vi. So I still insist that the Chinese document should be as neat as the original one（English doc). Thanks, Yanteng
