| Umar | |
| Dec 23 2016, 4:10 PM |
| F5143591: 8QGQXSPo.png | |
| Dec 23 2016, 4:31 PM |
| F5143390: mPMB2u8W.png | |
| Dec 23 2016, 4:16 PM |
Step to reproduce: open Special:UserRights (for a user with multiple groups) or write {{#property:P102}} in [[w:ce:Трамп, Дональд]].
Result: you get «АЦШ-н республикан парти, АЦШ-н республикан парти, АЦШ-н республикан парти» and «куьйгалхо, патруль а автопатруль» respectively; instead, every list item should be followed by an a (before the comma or after the last item).
Please upload attachments with https://phabricator.wikimedia.org/file/upload , not with random websites.
I don't understand: which screenshot shows the actual problem you're seeing? F5143591 or F5143390?
Are you saying the problem has something to do with the formatting of lists?
The last screenshot seems to be from w:ce:Трамп, Дональд. The "Парти" parameter is not filled locally and seems to come from https://www.wikidata.org/wiki/Q22686#P102 .
In [[w:ce:Трамп, Дональд]], if I write {{#property:P102}} I get "АЦШ-н республикан парти, АЦШ-н республикан парти, АЦШ-н республикан парти" . Is that prefix "АЦШ-н" correct?
UserRights currently says: «бакъо автопатруль, патруль а куьйгалхо → патруль а куьйгалхо», «бакъо куьйгалхо → куьйгалхо, патруль а автопатруль», «бакъо куьйгалхо → куьйгалхо а IP-блоктохаран юкъарадаккхар».
The listToText() and commaList() functions, which should be used here, use:
You can replace the comma "," with "а," in MediaWiki:Comma-separator/ce, if you want. Then we still have to add an "a" after the last item?
that's so right "АЦШ-н республикан парти А, АЦШ-н республикан парти А АЦШ-н республикан парти А"
We don't yet have a method to add something after the last item of the list, I think.
If this means that in Chechen, we should also add the word "and" after the last item of an array-like sentense, then why can't we create an *optional* "trailing-separator", and add it when necessary?
Or we can ask for suggestions from CLDR/AMP/IANA/IETF/...?
That seems like a sensible approach. The other alternative is to add a LanguageCe class that overrides the commaList function, which might be more appropriate if Chechen is the only language known to require this feature. (But I suspect that is not the case.)
Or we can ask for suggestions from CLDR/AMP/IANA/IETF/...?
If you look at what most of those organisations do, you'll see that they would not be interested in this issue.
Both approaches are acceptable, but adding another message for the lists (empty in English) also bears a little cognitive cost for translators of all the other languages. The more complex this area of i18n becomes, the more important it gets to document it correctly; at least a subsection of https://translatewiki.net/wiki/Translating:MediaWiki#Translation_notes is in order, I guess.
Here you yourself have written everything correctly - Chechen style for lists requires "a" after every list item.
Now so: Ангола, Луанда, Африка, Андорра, Андорра-ла-Велья, Европа, Антигуа, Сент-Джонс, Северная Америка, Аргентина, Буэнос-Айрес А Южная Америка
That's right: Ангола а, Луанда а, Африка а, Андорра а, Андорра-ла-Велья а, Европа а, Антигуа а, Сент-Джонс а, Северная Америка а, Аргентина а, Буэнос-Айрес а, Южная Америка а
In the absence of any evidence that other languages require this feature (@Umar, do you know of any other languages where lists of words behave in a similar way to Chechen?) it might be more straightforward to pursue this.