On Sat, 12 Nov 2011, Paul Menzel wrote:
just now I thought of the following and I am wondering if there exists already a solution.
Not exactly for wikipedia, but I have an experimental module that pulls information from the web. I use it get images from sites like yuml.me an dwebsequencediagrams.com. https://github.com/adityam/context-webfilter See test/ directory for examples.
Writing a text which includes people I want to add information about these peoples as footnotes. The first sentence in a Wikipedia article is most of the time good enough for that.
A macro `\infofromwikipedia{Donald Knuth}` would be nice which gets the first sentence of the article and puts an item into the bibliography.
This actually requires a more detailed spec. What happens if there is more than one person with the same name: http://en.wikipedia.org/wiki/Wolfgang_Schuster
There is even an API to access articles [2]. Besides coding that up I see the following problems.
1. The output [3] needs to be converted to ConTeXt.
I don't see anything in the API specs that returns the contents of the page. My guess is that simply downloading the html page and scraping the main paragraph might be easier. Once the data is retreived, using ConTeXt to typeset HTML is fairly easy. Another option is to just use one of the existing scripts to scrap the first paragraph/first line from Wikipedia, e.g., http://stackoverflow.com/questions/1565347/get-first-lines-of-wikipedia-arti... http://query7.com/scrape-the-first-paragraph-image-from-a-wikipedia-entry and use the filter module to call them. Aditya