[texhax] Any crazy math formulas for testing a TeX language interpreter

[Douglas McKenna]

The bcs test mentioned by Jim Hefferon wouldn't run for me either (using TeXLive 2014).  Probably the same problem as others have discussed; I didn't investigate.

So I took the formula out of the first test, commented out the LaTeX-related multi-line commands, and it worked correctly.

Fortunately, Barbara Beeton kindly found in archives and sent me the entire bcs test ported to the plain format.  I pasted all of its source code into my monograph test file and, feeling like I was about to jump off a cliff, typeset it with both pdfTeX and JSBox.

Yeah, baby!  Chemical symbols, circuit layout, math, and all.

Well, $100\% - \epsilon$.  The only discrepancy I could find among all of the tests was a couple of missing symbols (one is the Wierstrauss P maybe), due (I think) to not including the "msbm10" and "ecsx1200" fonts in the collection of fonts visible to the client program.

I have gleaned from various sources that the otherwise opaque "msbm" stands for "Math Society Blackboard Bold Math" or similar, and that "ecsx" stands for "European Computer Modern Symbol eXtensions" or similar.

The glyphs in msbm10 are in the distribution tree (e.g., msbm10.pfb) but I don't understand where pdfTeX is finding/mapping the glyphs to use for the ecsx1200.tfm file.  The only file with "ecsx" in its name I can find in the tree is the MetaFont ".mf" source file.  I need a ".pfb" or other file that can be converted to ".otf".

Doug McKenna