Last modified: 2014-04-28 19:01:52 UTC
It would be cool if you supported revisions in TextExtracts. The Query module specifies that one can ask for pages by title, pageid (both supported) and by revision ID using the revids parameter, e.g. revids=478198|54872|54894545 Although it states that "Most query modules will convert revision ID to the corresponding page ID. Only prop=revisions actually uses the revision ID itself." It this case it makes sense to get text from a revision. Stalled templates could break some text but the api caller can be aware of this.
Extraction requires parser output, obtaining which is a potentially very slow operation (in case of cache miss). I want to go away from on-demand generation of extracts and obtain them on page save instead. Because I don't want to store extracts for every revision, I don't think that implementing this feature request is viable, sorry.