Forum:Accessing Plaintext Description of Article

Forums → Ten Forward → Accessing Plaintext Description of Article (reply • watch)

Greetings, is there any way at all of generating a simple plain-text return of the first section of a Memory-Alpha article?

This is as close a construction as I've been able to come: http://memory-alpha.org/index.php?title=USS_Enterprise_(NCC-1701-A)&action=raw&ctype=text/css&prop=sections&section=0

but that generates: (last-line description linebreaks added to avoid horizontal scrollbar of death)

{{sidebar starship|
        image= USS Enterprise-A quarter.jpg
|        Name= USS ''Enterprise''
|       Class= {{ShipClass|Constitution}} ([[refit]])
|    Registry= NCC-1701-A
| Affiliation= [[United Federation of Planets|Federation]] [[Starfleet]]
|    Launched= [[2286]]
|      Speed=Warp 8, Emergency Speed Warp 9.15
|    Status= Decommissioned
|  Datestatus= 2293
}}
The '''USS ''Enterprise'' (NCC-1701-A)''' was the second {{ShipClass|Constitution}}
 [[starship]] commissioned by the [[Federation]] with the name ''Enterprise''.
 She was constructed at the [[San Francisco Fleet Yards]]
 and commissioned at the [[Earth Spacedock]] facility in [[2286]].

When I'd prefer to have the plain-text that meta-content "description" generates of the regular page header. Which is:

The USS Enterprise (NCC-1701-A) was the second Constitution-class
 starship commissioned by the Federation with the name Enterprise.
 She was constructed at the San Francisco Fleet Yards and commissioned
 at the Earth Spacedock facility in 2286.

Yes, I am "lazy" and don't want to have to build a wiki-formatting stripper/parser (keep Constitution but throwout ShipClass, keep possessive apostrophes but not the rest, etc). All I want is that first smidgen of the article in plain-text but my research into the API and index.php?action=raw don't seem to indicate a method of achieving that.

For those curious. I would like a utility that does a lookup of a Memory-Alpha article and returns a plain-text brief description (usually the first paragraph fits the bill) and a link to the full article.

Help is appreciated, thanks. -- Eipy 09:22, September 14, 2010 (UTC)

If the meta-content description contains what you need, then why don't you just have your utility request the regular version of the article and grab the value of exactly that attribute? :) -- Cid Highwind 12:30, September 14, 2010 (UTC)

Very fair question! I know, right, especially since my program can already do that!? But... it's not for me it's for James T. Kirk, it's for Worf, it's for Riker & Picard too (not to mention ‎Star Trek parodies and pop culture references (television) (and all the rest). It seems inefficient and wasteful to download all of Worf's 91.97 KB page, when all I want is 0.85 KB (or less than 870 bytes). Quite piggy of me to ask for everything when all I want is a sample.

My application is still pretty fast but why be 108% greedy? And well, it could be that my server has been lagging on the DL a bit. -- Eipy 17:10, September 14, 2010 (UTC)

OK, that's sensible. I found the additional parameter "templates=expand" in the index.php documentation - this at least partially resolves the template issue: http://memory-alpha.org/index.php?title=USS_Enterprise_(NCC-1701-A)&action=raw&templates=expand&ctype=text/css&prop=sections&section=0 -- Cid Highwind 18:15, September 14, 2010 (UTC)

Hm, that's different but still not quite down to plain-text. Though, I realize, maybe stripping the wiki-text formatting mightn't be that hard (famous last words). It seems I could strip all non-single apostrophes, HTML tags, square brackets and anything in the first parameter, all content in curly brackets. I'll wait to see if there are other suggestions before I try and mock that up. Eipy 18:39, September 14, 2010 (UTC)