It took a while to get used to the 80 limit, but now I like it for the most part. Like Milosz, I especially like that it facilitates side-by-side comparisons without the need to scroll horizontally (or the use of a size 4 font).
-Jeremy On Mon, Jun 8, 2009 at 10:54 AM, Donald Woods <[email protected]> wrote: > Yep, same here. The biggest complaint I have is because of Eclipse and its > weird behavior in how it does/doesn't wrap lines sometimes. > > Now that we have checkstyle setup and it's easy to discover problems during > a build (without having to rely on Eclipse), I can live with the 80 char > limit. > > > -Donald > > > > Miłosz Tylenda wrote: > >> Hi! >> >> I used to hate the 80-char limit but then I realized I quite like it when >> comparing files side-by-side. Now I am neutral. >> >> Greetings, >> Milosz >> >> >> Starting a separate thread to discuss the merit of 80 / 100 / 120 / {your >>> favorite length here} line lengths as a code convention. >>> >>> Since the project started we've adhered to the Sun code conventions [1] >>> and >>> recently introduced changes that enforce this length for source and test >>> code. This "archaic" limitation has been problematic in many areas, ie >>> generated metamodel classes. >>> >>> I think we can all agree that having some limit to line length is a >>> benefit >>> and the real contention is about the specific width in question. >>> >>> Sun's conventions justify 80 characters as being required for some >>> terminals >>> and tools. The debate has been taken up on several other sites [2],[3], >>> with >>> other justifications listed : >>> * dispersed development team (some people still have 13" laptop screens) >>> * it's easier to have multiple files open >>> * going past 80 characters indicates a problem with the code. (ed. >>> comment >>> nice) >>> * printer friendly >>> * plenty more justifications mainly from the comments sections. >>> >>> I am not terribly bothered about where we set the limit so long as we >>> decide >>> on one and stick to it. Regarding this limit I think Aaron Rubin [4] said >>> it >>> best on the python-ideas list (he was talking about how to break lines >>> though) : >>> >>> * >> This is an unsupported, and IMHO largely incorrect, assumption. >>>> >>> *>* >> Several correspondents have noted that they most often overrun >>> their >>> *>* >> intended line length by one or two characters. Just as there's >>> *>* >> nothing magical about the number "80", there's nothing magical >>> about >>> *>* >> "81" or "82" either. In a regime of 90-character lines, the limit >>> *>* >> will most often be exceeded by one or two characters. The same >>> will >>> *>* >> happen in a regime of 100-character lines, etc. We'll still need >>> to >>> *>* >> break lines, and wrapping them in parentheses will still be the >>> best >>> *>* >> way to do that. >>> * >>> >>> We started with 80 character columns and I don't think there's a >>> compelling >>> reason to change. I may be in the minority though, and if other devs feel >>> strongly about this issue I'd encourage them to reply to this thread. If >>> there's sufficient interest in a wider margin we can then start a [VOTE] >>> thread. >>> >>> [1] http://java.sun.com/docs/codeconv/html/CodeConventions.doc3.html >>> [2] >>> >>> http://richarddingwall.name/2008/05/31/is-the-80-character-line-limit-still-relevant/ >>> [3] >>> >>> http://stackoverflow.com/questions/758545/maximum-line-length-of-your-ide-checkstyle >>> [4] http://mail.python.org/pipermail/python-ideas/2009-May/004855.html >>> >>> -mike >>> >>> >>
