Will do! On Monday, 30 September 2013 10:43:57 UTC+1, Volker Braun wrote: > > I agree that the current state is bad. Also, the sphinx error frequently > points you at the docstring of the wrong class/method in the file so you > need to hunt through the whole file content to find where it hurts. We > should definitely improve on that. Can somebody open at ticket? > > > > On Monday, September 30, 2013 9:29:50 AM UTC+1, Andrey Novoseltsev wrote: >> >> Hi Travis, >> >> On Sunday, 29 September 2013 20:12:48 UTC+1, Travis Scrimshaw wrote: >> >>> In care you are unaware, I would recommend doing a more narrow docbuild >>> with "reference/subsection" where subsection is for example combinat or >>> algebras. Also on the next time you're rebuilding the doc, it should only >>> print about what files were changed, so it shouldn't be too far to scroll. >>> IDK, I guess what I'm saying is I've never really thought that this was an >>> issue. >>> >>> I was aware, but forgot, given that it does not take too much time >> compared say to long testing a particular section. I'd still prefer an >> option to have only errors/warnings reported or repeated/summarized in the >> very end, even if it is just "There were warnings". >> >>> >>> As Volker stated, it's relative to the start of the docstring, but it >>> would be nice to have some absolute positioning in the file. IMO the best >>> method to figure out where a docstring problem is to look through the >>> output as they tend to make a portion be clearly misformatted. >>> >>> There is a big difference in time spent looking at line 1234 where the >> error was detected and looking at a compiled documentation with a hundred >> methods. In my case errors were not obvious to scrolling through: one was >> wrong indentation for seealso which made the link appear not on a yellow >> background, which is not particularly visible on my laptop screen anyway, >> unless I look at it at a particular angle. One can of course look at what >> the branch is changing and then search for those methods (which, of course, >> tend to be in a different order). But no matter what - reading the line >> number is MUCH faster than manually trying to figure it out! >> >> So if someone can translate line numbers to something meaningful or at >> least show a line of context, please implement it! >> >> Thank you, >> Andrey >> >> >
-- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at http://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/groups/opt_out.
