On 08/03/2019 09:46, Jan Beulich wrote: >>>> Ian Jackson <[email protected]> 03/07/19 3:02 PM >>> >> Stefano Stabellini writes ("Re: [PATCH v11 5/9] xen/x86: use DECLARE_BOUNDS >> as required"): >>> On Wed, 6 Mar 2019, Jan Beulich wrote: >>>> Is the line wrapping really needed here? >>> >>> It would end at 80 characters exactly otherwise. I am happy to do as you >>> prefer. >> >> Certainly I prefer lines to end strictly less than 80 characters and >> preferably even shorter. My mailer/editor produces wrap damage for >> exactly-80-character lines. >> >> I think this wrapping was introduced by Stefano after a prompt from >> me. >> >> Jan, it is quite unfortunate that you are replying to Stefano to >> disagree with things that Stefano did because I suggested them, rather >> than replying to my patch comments. We must not put Stefano in the >> middle of a disagreement between different committers. > > I'm sorry, but I may have easily overlooked this earlier comment of yours. > Anyway, a later reply by Stefano suggests that the line would end up > being 79 chars, which is in line with ./CODING_STYLE. > > >> On this style question, while I have an opinion, I don't consider >> myself a maintainer, so the hypervisor maintainers' answer is >> definitive. >> >> >> Nevertheless, I will have one go at trying to convince Jan: >> >> Note that: >> >> - When code is turned into a patch, an extra character is added for >> the diff +/-. That means that 80-column code becomes 81 columns >> wide. >> >> - When a patch is quoted for review in email, two (usually) extra >> quoting characters are added ('> ') for each level of reply, >> so 80-column code becomes 83 or 85 (or more) columns wide. >> >> - One purpose of the line length limit is to fit within a >> conventional 80-column text terminal window (or at least, to >> minimise the number of lines which overflow such a window) >> >> - In an 80 column ssh session, simple representations are only >> capable of unambiguously displaying lines of up to 79 characters. >> >> - Therefore the total available code width which can be displayed >> unambiguously in an 80-column ssh session, within a singly quoted >> patch, is 76 characters. Longer lines produce wrap damage. >> >> To me would seem to imply that a *code* line length limit of 76 or 74 >> characters should be usual. Certainly it seems churlish to object to >> patches where the new code is wrapped to avoid lines >76. > > ./CODING_STYLE is pretty clear about it being "less than 80 characters". > If you want a lower limit, I think you'd have to propose a patch to that file > (which I'd likely try to prevent from going in). Wrapping in particular > for(;;) > (as was the case here iirc) is always weighing length vs readability. I for > one consider for(;;) easier to read when it's all on one line. Therefore I'd > prefer if no "early" wrapping was done. But as always - if a majority thinks > differently, so be it.
-2 for lowering the limit. Juergen _______________________________________________ Xen-devel mailing list [email protected] https://lists.xenproject.org/mailman/listinfo/xen-devel
