MrStradivarius added a subscriber: MrStradivarius.
MrStradivarius added a comment.

In https://phabricator.wikimedia.org/T71540#720099, @Anomie wrote:

> (In reply to Jackmcbarn from comment #10)
>
> > <hoo> I know... but this is supposed to be easy to use and I don't want to
>
> >  completely throw this over
>
> >  <hoo> also I'm not sure how that will work with content from Wikdiata
>
>
> Then the fetch-from-Wikidata should preprocess when necessary. Don't go 
> preprocessing everything that comes in from the parser, double-preprocessing 
> is bad.


I would also prefer to preprocess things only when necessary. As well as the 
points Jackmcbarn and Brad brought up, I have a few other concerns about 
possibly preprocessing things twice:

1. The performance hit. While a second preprocess likely won't take as much 
time as the first, it will still add to the parse time unnecessarily.
2. It would be inconsistent with Scribunto's current behaviour. If module 
authors have a legitimate reason to preprocess text inside Scribunto, it would 
be confusing to have to use frame:preprocess in most situations, but not have 
to do anything when using Capiunto.
3. It may introduce bugs in modules if we decide to alter Capiunto's 
preprocessing behaviour later. I can easily imagine a new module writer passing 
text in ref tags to Capiunto without realising that it needs to be 
preprocessed. If we later changed Capiunto to stop preprocessing such text then 
it would break the module, and the module writer may have no idea as to what 
went wrong. On the other hand, if we start out without preprocessing, module 
writers would be more likely to notice mistakes like these during testing.


TASK DETAIL
  https://phabricator.wikimedia.org/T71540

REPLY HANDLER ACTIONS
  Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign 
<username>.

EMAIL PREFERENCES
  https://phabricator.wikimedia.org/settings/panel/emailpreferences/

To: ori, MrStradivarius
Cc: greg, Anomie, JohnLewis, Lydia_Pintscher, hoo, ori, Jackmcbarn, 
Manybubbles, Wikidata-bugs, aaron, MrStradivarius



_______________________________________________
Wikidata-bugs mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs

Reply via email to