Revive plugin for latexml 0.8.4

[physikerwelt]

If I run the test I get some erros

$ make
Skip blib/lib/LaTeXML/Util/TestMath/Report.pm (unchanged)
Skip blib/lib/LaTeXML/Util/MathLexicon.pm (unchanged)
Skip blib/lib/LaTeXML/MathSyntax.pm (unchanged)
Skip blib/lib/LaTeXML/MathAST.pm (unchanged)
Skip blib/lib/LaTeXML/Util/TestMath.pm (unchanged)
$ make test
PERL_DL_NONLAZY=1 /usr/bin/perl "-MExtUtils::Command::MM" "-e" "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t
t/01_basic.t ...................... ok
t/02_simple_math.t ................ 1/?
#   Failed test 'Syntax tree match (1 parse):
#  (a+b)^2
# '
#   at /srv/LaTeXML/LaTeXML-Plugin-MathSyntax/blib/lib/LaTeXML/Util/TestMath.pm line 134.
#

#   Failed test 'Syntax tree match (1 parse):
#  (a+b)^2 = a^2 + 2ab + b^2
# '
#   at /srv/LaTeXML/LaTeXML-Plugin-MathSyntax/blib/lib/LaTeXML/Util/TestMath.pm line 134.
#
t/02_simple_math.t ................ 10/?
#   Failed test 'Syntax tree match (1 parse):
#  f(x)g(x)
# '
#   at /srv/LaTeXML/LaTeXML-Plugin-MathSyntax/blib/lib/LaTeXML/Util/TestMath.pm line 134.
#

#   Failed test 'Syntax tree match (1 parse):
#  (f \circ g) x
# '
#   at /srv/LaTeXML/LaTeXML-Plugin-MathSyntax/blib/lib/LaTeXML/Util/TestMath.pm line 134.
#
# Looks like you failed 4 tests of 12.
t/02_simple_math.t ................ Dubious, test returned 4 (wstat 1024, 0x400)
Failed 4/12 subtests
t/0511_DLMF_Special_Notation.t .... 8/?
#   Failed test 'Syntax tree match (1 parse):
#  \langle f,g \rangle
# '
#   at /srv/LaTeXML/LaTeXML-Plugin-MathSyntax/blib/lib/LaTeXML/Util/TestMath.pm line 134.
#
t/0511_DLMF_Special_Notation.t .... 16/? # Looks like you failed 1 test of 16.
t/0511_DLMF_Special_Notation.t .... Dubious, test returned 1 (wstat 256, 0x100)
Failed 1/16 subtests
t/0512_DLMF_Elementary_Algebra.t .. 2/?
#   Failed test 'Syntax tree match (1 parse):
#  \binom{n}{k}=\frac{n!}{(n-k)!k!}=\binom{n}{n-k}.
# '
#   at /srv/LaTeXML/LaTeXML-Plugin-MathSyntax/blib/lib/LaTeXML/Util/TestMath.pm line 134.
#

#   Failed test 'Syntax tree match (1 parse):
#  \binom{n}{0}+\binom{n}{1}+\dots+\binom{n}{n}=2^{n}.
# '
#   at /srv/LaTeXML/LaTeXML-Plugin-MathSyntax/blib/lib/LaTeXML/Util/TestMath.pm line 134.
#

#   Failed test 'Syntax tree match (1 parse):
#  \binom{n}{0}-\binom{n}{1}+\dots+(-1)^{n}\binom{n}{n}=0
# '
#   at /srv/LaTeXML/LaTeXML-Plugin-MathSyntax/blib/lib/LaTeXML/Util/TestMath.pm line 134.
#
t/0512_DLMF_Elementary_Algebra.t .. 5/?
#   Failed test 'Syntax tree match (1 parse):
#  \binom{n}{0}+\binom{n}{2}+\binom{n}{4}+\dots+\binom{n}{\ell}=2^{{n-1}},
# '
#   at /srv/LaTeXML/LaTeXML-Plugin-MathSyntax/blib/lib/LaTeXML/Util/TestMath.pm line 134.
#

#   Failed test 'Syntax tree match (1 parse):
#  \binom{n}{k}=\frac{n(n-1)\cdots(n-k+1)}{k!}%
#   = \frac{(-1)^{k}\left(-n\right)_{{k}}}{k!}%
#   = (-1)^{k}\binom{k-n-1}{k}.
# '
#   at /srv/LaTeXML/LaTeXML-Plugin-MathSyntax/blib/lib/LaTeXML/Util/TestMath.pm line 134.
#

#   Failed test 'Syntax tree match (1 parse):
#  \binom{n+1}{k}=\binom{n}{k}+\binom{n}{k-1}.
# '
#   at /srv/LaTeXML/LaTeXML-Plugin-MathSyntax/blib/lib/LaTeXML/Util/TestMath.pm line 134.
#
t/0512_DLMF_Elementary_Algebra.t .. 8/?
#   Failed test 'Syntax tree match (1 parse):
#  \sum^{m}_{{k=0}}\binom{n+k}{k}=\binom{n+m+1}{m}.
# '
#   at /srv/LaTeXML/LaTeXML-Plugin-MathSyntax/blib/lib/LaTeXML/Util/TestMath.pm line 134.
#

#   Failed test 'Syntax tree match (1 parse):
#  a+ax+ax^{2}+\dots+ax^{{n-1}}=\frac{a(1-x^{n})}{1-x},
# '
#   at /srv/LaTeXML/LaTeXML-Plugin-MathSyntax/blib/lib/LaTeXML/Util/TestMath.pm line 134.
#

#   Failed test 'Syntax tree match (1 parse):
#  a+ax+ax^{2}+\dots+ax^{{n-1}}=\frac{a(1-x^{n})}{1-x}, x\ne 1
# '
#   at /srv/LaTeXML/LaTeXML-Plugin-MathSyntax/blib/lib/LaTeXML/Util/TestMath.pm line 134.
#
# Looks like you failed 9 tests of 10.
t/0512_DLMF_Elementary_Algebra.t .. Dubious, test returned 9 (wstat 2304, 0x900)
Failed 9/10 subtests
t/0513_DLMF_Determinants.t ........ skipped: Testbed under construction...
t/06_RecDescent_parity.t .......... skipped: Testbed under construction...

Test Summary Report
-------------------
t/02_simple_math.t              (Wstat: 1024 Tests: 12 Failed: 4)
  Failed tests:  7-8, 11-12
  Non-zero exit status: 4
t/0511_DLMF_Special_Notation.t  (Wstat: 256 Tests: 16 Failed: 1)
  Failed test:  11
  Non-zero exit status: 1
t/0512_DLMF_Elementary_Algebra.t (Wstat: 2304 Tests: 10 Failed: 9)
  Failed tests:  2-10
  Non-zero exit status: 9
Files=6, Tests=40, 44 wallclock secs ( 0.09 usr  0.03 sys + 40.05 cusr  3.22 csys = 43.39 CPU)
Result: FAIL
Failed 3/6 test programs. 14/40 subtests failed.
make: *** [test_dynamic] Error 255

[dginev]

Wow, thanks for looking into this Moritz. It's been stale for 3 years! I've been thinking on and off and chatting with Bruce about reviving this line of work, but it may need certain pieces falling together first.

Shame that it regressed, I will take a quick look over the weekend if time permits. Thanks again.

[dginev]

3+2 = 5 years, how time flies.

I did some refresher reading here, and it looks like the very capability that made this plugin possible has been removed from latexml -- I think silently/implicitly when we made the move to github, and maybe reintroducing it got stalled at least a couple of times in the process of working on a more advanced customization system? (i struggle to remember...)

The mechanism I remember having implemented (cc @brucemiller for feedback) was a --parse option that modernizes the --noparse disable switch into allowing both disabling --parse=no and using a different grammar backend --parse=LaTeXML::MathSyntax in the case of this plugin.

That was then passed into MathParser.pm and the custom backend was instantiated in place of RecDescent for the internalparser. I wonder if this remained on some old latexml branch, or just got erased... Happy to reintroduce it and revive this plugin if it still seems like the right customization technique to Bruce. What's worse is that the option remains documented in latexmlc in the latexml master, and is indeed serviced via LaTeXML::Common::Config, but simply gets silently ignored in the Core.pm. I will try to file a small PR in latexml that gets the plugin back to working...

[dginev]

I should mention why I am looking into this plugin again. I am working on finalizing a more robust test suite for the MathGrammar in the latexml master, and would like to slowly approach it from the Marpa angle, to see if we can get ambiguous parses enabled, while also maintaining the prior grammar coverage (where meaningful)