Last modified: 2014-08-08 21:04:20 UTC
I would like to return wikipedia summaries only (without having to download an entire wikipedia page) however none of these options include the links in the text,only html text formatting is included with the text.
So, for exmaple, /w/api.php?action=query&prop=extracts&format=json&exsentences=3&exlimit=10&exintro=&titles=Therion returns { "query": { "pages": { "1590770": { "pageid": 1590770, "ns": 0, "title": "Therion", "extract": "<p><b>Therion</b>, Greek for \"wild animal\" or \"beast\" (\u03b8\u03b7\u03c1\u03af\u03bf\u03bd), may refer to:</p>" } } } } where the word "Greek" is not a link, although it's a link on https://en.wikipedia.org/wiki/Therion It would be cool if there would be a switch to specify which HTML elements to return; e.g. links/no links. That would be a useful feature for "Word-sense disambiguation". Otherwise the action=parse has to be used.
Well, extracts are supposed to be extracts - ot much will be left to extract if we will preserve links too, so if you need full HTML, use action=parse.
(In reply to Max Semenik from comment #2) I recommended action=render of index.php -- is there another way to get just the introduction (with links)? It's about saving bandwidth.
