Re: TeX output via TeXForm: variable name translation
- To: mathgroup at smc.vnet.net
- Subject: [mg101031] Re: TeX output via TeXForm: variable name translation
- From: Jens-Peer Kuska <kuska at informatik.uni-leipzig.de>
- Date: Sun, 21 Jun 2009 07:07:08 -0400 (EDT)
- References: <h1hbfh$mq$1@smc.vnet.net>
Hi, set up the boxes for you symbol in a proper way and Mathematica does the rest, i.e. e1 /: MakeBoxes[e1, fmt_: StandardForm] := SubscriptBox["\[Epsilon]", "1"] and TeXForm[e1] gives \epsilon _1 Regards Jens dnquark wrote: > I am trying to find a good way to interoperate between Mathematica and > LaTeX (or LyX). > Typically, I am workign with multi-character variable names, which > TeXForm interprets as text: > > In[65]:= TeXForm[e1] > Out[65]//TeXForm= > \text{e1} > > To get around this, I am trying to create a translation table of > sorts: e.g. > > TranslationTable1 = {{e1, "\\epsilon_1"}, {e0, "\\epsilon_0"}, {kz0t, > "\\tilde{k^{(0)}_z}"}, {kz1t, "\\tilde{k^{(1)}_z}"}}; > > I would like, if possible, to perform those replacements in > Mathematica, without writing and invoking external scripts. The first > thing I tried was this: > > TeXReplacements := (Rule[#[[1]], ToExpression[#[[2]], TeXForm, > HoldForm]] & /@ TranslationTable1) > > In[67]:= (e1*kz0t) /. TeXReplacements // TeXForm > > Out[67]//TeXForm= > \epsilon _1 \tilde{k_z^0} > > (This simply does the following: > In[69]:= (e1*kz0t) /. {e1 -> ToExpression["\\epsilon_1", TeXForm, > HoldForm], > kz0t -> ToExpression["\\tilde{k^{(0)}_z}", TeXForm, HoldForm]} // > TeXForm > > Out[69]//TeXForm= > \epsilon _1 \tilde{k_z^0} > ) > > Notice how this doesn't work: Take a look at the translation table > above and notice parentheses around in k_z^{(0)}. Mathematica > discards these parentheses. > > So the question to the group is: (1) is it possible to preserve the > formatting exactly how I specified it in the translation table before > converting to TeXForm or (2) is it possible to do a replacement using > my translation table after I had converted to TeXForm or (3) is there > a better solution altogether? > > Thanks! >