On May 30, John H Palmieri wrote:
Thanks, that was my impression, I looked and looked and found tons of long lines and I don't remember one example of a broken output line (that would render ugly but still pass a doctest). Perhaps this could be added to the developer's guide. I started working on vim with the little annoyance of having two different colorcolumns (at 72 and 80) just because of that PEP, but I suspect that no one cares about the different line lenghts (admittedly this is just an unfounded hunch). Best, R.Just for kicks, I counted the number of .py and .pyx files in the Sage library, and then counted the ones with lines longer than 80 characters in their docstrings/doctests. 84% of them had such lines.
On Saturday, May 30, 2020 at 9:14:59 AM UTC-7, John H Palmieri wrote: Lines should be shorter than 80 characters when possible. If it isn't possible because it will cause confusion, break a doctest, make a doctest unhelpful, etc., then you can make an exception. There are plenty of exceptions in the Sage library already, for example [1]https://git.sagemath.org/sage.git/tree/src/sage/homology/ chain_complex.py#n225(I just picked a file at random in the Sage library and found an example. Iwould guess that this is typical.)Ellipses in doctest output should be used for parts of the output that are random or are too long (as in many lines long) to be useful. The details ofthe traceback from an error is a typical use case for this. On Saturday, May 30, 2020 at 6:31:34 AM UTC-7, Reimundo Heluani wrote: On May 30, Michael Orlitzky wrote: >On 5/30/20 8:51 AM, 'Reimundo Heluani' via sage-devel wrote: >> >> I've looked through the code and found numerous instances of long times in >> examples and tests blocks. So my question is: is there a policy about these >> things? My guess is to leave the long lines of output without wrapping. >> > >You can usually add parentheses and continue your doctest with a "....:" >on the next line so that the test retains its meaning and the HTML >output remains correct. For example, here's a line that's too long: >>> sage: from mjo.eja.eja_algebra import >> QuaternionMatrixEuclideanJordanAlgebra > >Instead of forcing a line break with (say) a backslash, you can do > > sage: from mjo.eja.eja_algebra import ( > ....: QuaternionMatrixEuclideanJordanAlgebra ) > >The same trick allows you to break sums, products, list comprehensions, >etc. over multiple lines. > Thanks, my question is mainly about output strings, how do you break them with ellipsis? R. > >-- >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, >sendan email to sage-...@googlegroups.com.>To view this discussion on the web visit >[2]https://groups.google.com/d/msgid/sage-devel/9cbbc443-97e8-deec-d29d-cab5976f3cfd%40orlitzky.com. -- 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 emailto [3]sage-devel+unsubscr...@googlegroups.com.To view this discussion on the web visit [4]https://groups.google.com/d/msgid/sage-devel/9cd7f65c-c9bc-41de-a93b-cbd45639dcf1%40googlegroups.com. References:[1] https://git.sagemath.org/sage.git/tree/src/sage/homology/chain_complex.py#n225 [2] https://groups.google.com/d/msgid/sage-devel/9cbbc443-97e8-deec-d29d-cab5976f3cfd%40orlitzky.com[3] mailto:sage-devel+unsubscr...@googlegroups.com[4] https://groups.google.com/d/msgid/sage-devel/9cd7f65c-c9bc-41de-a93b-cbd45639dcf1%40googlegroups.com?utm_medium=email&utm_source=footer
-- 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 view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/20200530164750.GA53062%40vertex.
signature.asc
Description: PGP signature
