- From: Rik Cabanier <cabanier@gmail.com>
- Date: Mon, 23 Apr 2012 09:14:14 -0700
- To: Aryeh Gregor <ayg@aryeh.name>
- Cc: Dirk Schulze <dschulze@adobe.com>, "public-fx@w3.org" <public-fx@w3.org>
- Message-ID: <CAGN7qDB0nd7StHJh-27EmnXg-XtuO2y4v_A_EmphfinMjJF7-A@mail.gmail.com>
On Mon, Apr 23, 2012 at 2:08 AM, Aryeh Gregor <ayg@aryeh.name> wrote: > On Mon, Apr 23, 2012 at 2:51 AM, Rik Cabanier <cabanier@gmail.com> wrote: > > Does it matter if it's code that only the author understands? It's > obviously > > working well enough for Wikipedia to create formulas and they have much > more > > complex math than us. > > It seems that we would have to do quite a bit of work ourselves > (including > > learning LaTex) instead of just using texvc. > > I think you're misunderstanding what texvc does -- it basically does > nothing but parse the input, make sure it looks like valid LaTeX with > all commands on its internal whitelist, wraps it in some boilerplate, > and then shells out to latex. Pretty much any valid texvc input is > already valid LaTeX, give or take some macros that texvc recognizes > and missing boilerplate that it adds. Everything you're entering into > Wikipedia's <math> tags *is* LaTeX. So there's not much gained by not > just using LaTeX. The effort of trying to get texvc set up in a > non-MediaWiki environment would likely be more than just learning how > to add the extra boilerplate LaTeX will want. > > Also, if all you want is a few PNGs, you can just enter the code in > Wikipedia, preview, and save the PNG. That's the easiest way. > I was actually planning on doing that :-) It would be a bit cleaner if I could just use the markup and the w3c server would do the magic. > > (Former MediaWiki developer here, BTW. :) ) >
Received on Monday, 23 April 2012 16:15:22 UTC