Migrating Pology to Python 3

[Adrian Chaves]

I think the number of characters may be OK. Before it must have been 
counting bytes, i.e. counting `É` as 2.

I will look into the word counting.

On 2022-10-08 13:13, Karl Ove Hufthammer wrote:

> Adrian Chaves skreiv 04.10.2022 12:49:
> 
>> So I would like every Pology user to test as soon as possible their 
>> usual workflows with pology with the python3 branch, and report to me 
>> any issue. I do expect there to be issues. So send me a command to 
>> reproduce the issue you get, and I will fix it.
> This is a minor bug, posieve stats seems to count the number of words 
> and characters in a different way now. Example:
> 
> msgid "One Selected File"
> msgid_plural "%1 Selected Files"
> msgstr[0] "Éi merkt fil"
> msgstr[1] "%1 merkte filer"
> 
> Python 2 'posieve stats' reports:
> 
> - msgmsg/totw-orw/tot-orw-trch-orch-tr
> translated      1    100.0%      3     100.0%      3      14      11
> 
> In other words, three English words (w-or) and 14 English characters 
> (ch-or).
> 
> While Python 3 'posieve stats' reports:
> 
> - msgmsg/totw-orw/tot-orw-trch-orch-tr
> translated      1    100.0%      2     100.0%      2      14      10
> 
> So 2 English words instead of 3, but the same number of English 
> characters (14). However, for the translations, the number of 
> characters also shrink, from 11 to 10 (and the number of words from 3 
> to 2).
> 
> I think both statistics are wrong. There are clearly 5 (or 6) English 
> words.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-i18n-doc/attachments/20221008/eda315e5/attachment-0001.htm>