Using pageimages instead of using custom code to find imageless article

[PierreSelim]

Exemple of usage
https://fr.wikipedia.org/w/api.php?action=query&formatversion=2&prop=pageimages&titles=Nicolas%20Sarkozy

https://fr.wikipedia.org/w/api.php?action=query&formatversion=2&prop=pageimages&titles=Antoine%20Waechter

[PierreSelim]

So with mwclient

site.api('query', prop='pageimages', titles='Nicolas Sarkozy')

[PierreSelim]

Not totally convinced by this approach due to performance. It implies doing more query than needed to the API, apparently.

[JeanFred]

Can’t you do both queries at the same time?

[PierreSelim]

Thing is I believe (I might be wrong), that mwclient preloads the article text when accessing to Page object, sending a new request would be an overhead :)

[PierreSelim]

A pretty stupid implementation could go as this

def isthereanimage(article):
    """Returns whether there is an image in the article or not."""
    LOG.info("Analyzing: %s", article.name.encode('utf-8'))
    result = site.api('query', prop='pageimages', titles=article.name.encode('utf-8'))
    ...
    # test result length :-)

Something a bit more clever would be to try to do all the queries in one with something like

 result = site.api('query', prop='pageimages', titles='|'.join(articlenames))