With the attached test file, which uses MPinclusions for a
staticMPfigure, I get an "Isolated expression". I think it's a parser
bug somewhere, because deleting almost any word in the file removes the
error.
The file itself looks a bit crazy. It is a minimal example generated by
a program to help with debugging. The program randomly deletes a word
in the file, then tests whether the error still happens. If yes, it
deletes another word. If no, it restores the word, and tries again by
choosing another word to delete. That process reduced the file from 4KB
to 2KB.
Here's the relevant piece of the transcript. The same error happens at
live.contextgarden.net (so I don't think it's a Debian-specific issue).
! terminal: >> r_ladder
! Isolated expression.
<to be read again>
;
<*> r_ladder;
;
.
system > tex > error on line 73 in file test-mpfigures.tex:
terminal: >> r_ladder
! Isolated expression.
<to be read again>
;
<*> r_ladder;
;
...
63
64 \startMPinclusions
65 def r_ladder := draw unitsquare; enddef;
66 \stopMPinclusions
67
68 \startstaticMPfigure{resistiveladder}
69 r_ladder;
70 \stopstaticMPfigure
71
72 \usestaticMPfigure[resistiveladder]
73 >>
74 \stoptext
75
<inserted text> ...0;;]===], [===[;;]===], false)}
\meta_process_graphic ...!!es , \MPaskedfigure )}}
\egroup \placeMPgraphic
\m...
\meta_handle_reusable_graphic ...cess_graphic {#3}
}\setxvalue {\??mpgraphic
...
\meta_use_graphic ...aphic \currentMPgraphicname }
\empty
\meta_end_graphic_g...
<to be read again>
\fi
\dodoubleemptyonespaced #1#2->#1[{#2}][]
...
l.73
And the start of the output, showing the version information:
mtx-context | run 1: luatex
--fmt="/var/lib/texmf/luatex-cache/context/f919750633122078d706dd10f1cfd39b/formats/cont-en"
--lua="/var/lib/texmf/luatex-cache/context/f919750633122078d706dd10f1cfd39b/formats/cont-en.lui"
--backend="pdf" "./test-mpfigures.tex" \stoptext
This is LuaTeX, Version beta-0.70.2-2012070301 (TeX Live 2012/Debian)
\write18 enabled.
(test-mpfigures.tex
ConTeXt ver: 2012.05.30 11:26 MKIV fmt: 2013.4.11 int: english/english
system > cont-new.mkiv loaded
If you can think of further tests to run, I'll be happy to try them.
--
-Sanjoy
\definereferenceformat [inproblem] [text=Problem]
setups
\setuplayout[marking=on,location=middle,
backspace=1.25in,
leftmargindistance=0.125in,
width=4.75in,
rightmargindistance=0.25in,
headerdistance=24bp,
height=8.5in,
bottom=12bp]
[problem]
]
\starttext
many games flipping tedious.
computer instead simulate games pseudorandom numbers
as substitute coin. several runs produced by
with $1$ $2$ indicate player won the
then shows the resulted in win.
these iterations, each player times. reasonable
conclusion Each chance
try With games, one count the
number player scanning the beginnings.
computer the
Each run because program different
pseudorandom each produces different The
recent $68$: In words, player~$1$
and player~$2$ won times. probability
winning now seems closer $2/3$ $1/2$.
\externalfigure[fig/coin-flip.1]
value, diagram as tree.
contains and T, and one flip. game
ends the leaves, when heads. The boldface
show the the first wins, e.g. TTH,
TTTTH. probabilities of each way are,
The sum is
the the first
\placeformula\startformula
.
\stopformula
This summed using formula (see
However, more enjoyable which
formula in \inproblem[prob:explain-geometric-series]
noticing the presence of tree its one
down. is, the first tails, happens
the second player starts the game as
he were the first player. player wins
the game with probability $1/2$ times (the factor is from
probability tosses tails). Because one of
the two two winning and $p/2$
add unity. Therefore, $p=2/3$, from the
\startMPinclusions
def r_ladder := draw unitsquare; enddef;
\stopMPinclusions
\startstaticMPfigure{resistiveladder}
r_ladder;
\stopstaticMPfigure
\usestaticMPfigure[resistiveladder]
\stoptext
_______________________________________________
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context
