Take the 2-minute tour ×
TeX - LaTeX Stack Exchange is a question and answer site for users of TeX, LaTeX, ConTeXt, and related typesetting systems. It's 100% free, no registration required.

Is there any way how to convert pt to mm or cm in plain TeX?

share|improve this question
2  
Does this help? tex.stackexchange.com/questions/8260/… –  Torbjørn T. Dec 6 '11 at 17:08
    
Sorry, the above was was LaTeX, not sure if any of it works in plain. –  Torbjørn T. Dec 6 '11 at 17:14
    
@TorbjørnT. Well, the accepted answer there only requires a few basic LaTeX macros that are easy enough to copy from latex.ltx. –  Joseph Wright Dec 6 '11 at 17:15
    
@TorbjørnT. It is not difficult to modify Philippe Goutet's solution to use eTeX only. –  Leo Liu Dec 6 '11 at 17:16
1  
@Peter: Are we allowed to use the e-TeX extensions or does it have to be TeX82? –  Joseph Wright Dec 6 '11 at 17:16
show 5 more comments

4 Answers

To convert from a dimension given in points, save the dimension in a counter first, then do the conversion to what you need. When TeX saves the dimension in a counter it converts to sp units.

\catcode`@=11
\@tempdimb25.4mm 
\@tempcnta=\@tempdimb
\the\@tempcnta 
\divide\@tempcnta by65536\par 
\the\@tempcnta pt
\bye

Conversion from any unit to a point is a bit more involved:

\catcode`@=11
\newdimen\@tempdimb
\newcount\@tempcnta
\def\topoint#1#2{%
\@tempdimb=#1
\@tempcnta=\@tempdimb
\multiply\@tempcnta by10
\divide\@tempcnta by18647 \advance\@tempcnta by1
\multiply\@tempcnta by72 \divide\@tempcnta by2540
\expandafter\def\expandafter#2\expandafter{\the\@tempcnta}}

\topoint{25.4mm}{\test}

\test pt
\bye
share|improve this answer
add comment

Expandable solution without eTeX, using Heiko's bigintcalc package. Really, this is a mess, and you should use one of the solutions which make use of the eTeX extensions.

\input bigintcalc.sty
\catcode`@=11

% Step 1: expand the number.
\def\pttocm#1{\romannumeral\expandafter\pttocm@i\romannumeral-`0#1pt;;;;}
\def\pttocm@i#1{\pttocm@clean\ifcat\relax#1\the\fi#1}
\def\pttocm@clean{\expandafter\pttocm@clean@\romannumeral-`0}
\def\pttocm@clean@#1{%
  \ifcase\ifnum`#1<"2B 1\else\ifnum`#1>"39 1\else 2\fi\fi
  \or\expandafter\pttocm@clean@end
  \or\expandafter\pttocm@clean@do\fi #1}
\def\pttocm@clean@do#1#2;#3;{\pttocm@clean#2;#3#1;}

% Step 2: convert the decimal number to an integer/10^n
\def\pttocm@clean@end#1;#2;{\pttocm@clean@end@#2.;#2;.;!}
\def\pttocm@clean@end@#1.#2;#3.#4;#5!{\pttocm@decimal#4.;;#1#4}
\def\pttocm@decimal#1#2;{\ifx.#1\pttocm@decimal@\fi\pttocm@decimal#2;0}

% Step 3: convert from pt to nm (25400000nm=7227pt)
\def\pttocm@decimal@#1;0#2;#3;{\fi
  \expandafter\expandafter\expandafter\pttocm@result
  \bigintcalcDiv{\bigintcalcMul{25400000}{#3}}{7227#2}.;}

% Step 4: check the sign
\def\pttocm@result#1{%
  \ifx-#1\expandafter\pttocm@result@neg
  \else\expandafter\pttocm@result@pos\fi#1}
\def\pttocm@result@neg-{\expandafter\pttocm@rmstop
  \expandafter-\romannumeral\pttocm@result@pos}
\chardef\pttocm@rmstop=0

% Step 5: shift the period 5 steps to the left, by reversing twice.
\def\pttocm@result@pos#1;{%
  \pttocm@reverse 00000#1%
  {?\pttocm@result@shift;}?;%
  {?\pttocm@result@print;}?;}
\def\pttocm@reverse#1#2;{%
  \pttocm@gobble#1%
  \pttocm@reverse#2;#1}
\def\pttocm@gobble#1{}
\def\pttocm@result@shift#1;#2;#3;#4.#5#6#7#8#9%
  {\pttocm@reverse #4#5#6#7#8#9.}
\def\pttocm@result@print#1;#2;#3;#4;{\pttocm@trimzeros{#4}}

% Step 6: remove spurious zeros.
\def\pttocm@trimzeros#1{\pttocm@trim@#1;\pttocm@trim@ 0;\pttocm@trim@@}
\def\pttocm@trim@#10;#2{#2#1;}
\def\pttocm@trim@@#1;#2;{\pttocm@trim@@@#1}
\def\pttocm@trim@@@#1{%
  \ifx0#1\else\expandafter\pttocm@trim@end\expandafter#1\fi
  \pttocm@trim@@@}
\def\pttocm@trim@end#1#2{%
  \ifx.#1\expandafter\pttocm@rmstop\expandafter0%
  \else\expandafter\pttocm@rmstop\fi #1}
\catcode`@=12

\pttocm{-12.1232pt}

\pttocm{283000000000}

\def\foo{23}
\pttocm{12\foo.1\foo2pt}

\newdimen\testdimen
\testdimen=1cm
\pttocm{\testdimen}

\testdimen=-1mm
\pttocm{\testdimen}

\bye
share|improve this answer
add comment

With help of eTeX's \scantokens, \strip@pt can be also:

\def\strip#1.#2pt{#1\ifnum#2>0.#2\fi}
\def\convertto#1#2{%
  \expandafter\expandafter\expandafter\strip
  \expandafter\expandafter\scantokens
  \expandafter{\the\dimexpr#2*65536/\number\dimexpr1#1\relax\relax\endinput}}

\convertto{pt}{1pc}

\bye
share|improve this answer
add comment

Here is a combination of Yiannis' question from Stripping the pt from a dimension and Philippe Goutet's answer from What are the various units (ex, in, pt, etc.) expressed in mm?:

\catcode`@=11
\begingroup
  \catcode `P=12  % digits and punct. catcode
  \catcode `T=12  % digits and punct. catcode
  \lowercase{%
  \def\x{\def\rem@pt##1.##2PT{##1\ifnum##2>\z@.##2\fi}}}
     \expandafter\endgroup\x%
\def\strip@pt{\expandafter\rem@pt\the}
\def\convertto#1#2{\strip@pt\dimexpr #2*65536/\number\dimexpr 1#1\relax\relax}
\catcode`@=12
\convertto{mm}{10pt}
\bye

which requires e-TeX extensions.

share|improve this answer
1  
And what about something without e-TeX extensions? –  Peter Dec 6 '11 at 17:32
3  
You should terminate both \dimexpr with a \relax. Otherwise any dimen expression afterwards, like '* 10' will be taken as part of the inner \dimexpr. The \relax is removed by \dimexpr so you don't have to worry about expandable contexts. –  Martin Scharrer Dec 6 '11 at 17:52
2  
@Peter: \convertto{mm}{\the\hsize} will work just fine. The catcode changes are only valid during the definition of the macro, not during the execution. –  Martin Scharrer Dec 6 '11 at 17:53
1  
@MarcoDaniel: Why do you need Windows? Why do you need Linux? Why do you need Peugeot? Why do you need... >:-/ –  morbusg Dec 6 '11 at 18:16
1  
@Stephen: I only know one distribution which doesn't come with e-TeX, that's the commercial Scientific Workbench. Are there more? The majority of users should have e-TeX. –  Martin Scharrer Dec 6 '11 at 20:52
show 7 more comments

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.