Message120883
| Author |
doerwalter |
| Recipients |
belopolsky, doerwalter, ncoghlan, pitrou, vstinner |
| Date |
2010年11月09日.18:04:48 |
| SpamBayes Score |
2.6481835e-08 |
| Marked as misclassified |
No |
| Message-id |
<4CD98D13.1020908@livinglogic.de> |
| In-reply-to |
<1289213297.51.0.299688121399.issue10329@psf.upfronthosting.co.za> |
| Content |
> STINNER Victor <victor.stinner@haypocalc.com> added the comment:
>
>> ... it complicates handling of the output of trace.py.
>> For each file you have to do the encoding detection dance again ...
>
> What? You just have to call one function! tokenize.open() :-) Well, ok,
> it's not commited yet, but it looks like most people agree: #10335.
The problem is that the script that downloads and builds the Python
source and generates the HTML for http://coverage.livinglogic.de/ isn't
ported to Python 3 yet (and can't be ported easily). However *running*
the test suite of course uses the current Python checkout, so an option
that lets me specify which encoding trace.py/regrtest.py should output
would be helpful. |
|