After switching from the august2006 version of ConTeXt to 18-1-2007 suddenly an error crept up in a text not giving this before. The error was triggered by: \switchtobodyfont[24pt] and the culprit turned out to be the following statement in one of my own type files: \starttypescript [all] [lucidayy] [size] \definebodyfont [4pt,5pt,6pt,7pt,8pt,9pt,10pt,11pt,12pt,14.4pt,17.3pt] [rm,ss,tt,tts,cg,hw,fx,mm] [default] \stoptypescript There the 24pt had to be added: ...,17.3pt,24pt] My point is the phrasing of the error message: ! Math formula deleted: Insufficient symbol fonts. \dontleavehmode ...point \everymath \emptytoks $ $ }\unhbox \@@dlhbox \fi \fi \dodowithnextbox ->\noindentation \dontleavehmode \hbox to \localhsize {\ifc... l.6 \midaligned{\tfa\bf PRACTICUMBOEK CRYPTOGRAFIE} That is somewhat misleading, if I may permit myself an understatement. In the source of the midaligned there is not a math formula in sight. Now I know enough TeX to see the relevance of the two $'s further in the error message. But my point is that the text seems to point in the direction of missing (math) fonts, whereas it is just a missing fontsize. I have two questions: 1. Is it advisible (if possible at all) to make the message text more specific and directly understandable by newcomers: one may suppose they will not easily find the connection of \switchtobodyfont [something-big] to a missing size in the typescript. 2. I am curious why this error message only now pops up, whereas previously the typesetting proceeded without a hitch. Is this caused by more stringent testing in the ConTeXt code or is it just a default behaviour that is no longer valid? Hans van der Meer
Hans van der Meer wrote:
2. I am curious why this error message only now pops up, whereas previously the typesetting proceeded without a hitch. Is this caused by more stringent testing in the ConTeXt code or is it just a default behaviour that is no longer valid?
There is a bit of a mixup. Assuming you run pdfetex, the wrong definition of \dontleavehmode is now actually used. It should have been like this: \let\dontleavehmode\quitvmode For Hans: yes, the \pdf... prefix is missing from pdftex 1.40's \quitvmode. In the non-pdftex, this weird error message is currently almost unavoidable, I'm afraid. Perhaps a solution would be to issue an error or warning in the case of \switchtobodyfont[<unkownsize>] Best, Taco Taco
On 1/24/07, Taco Hoekwater wrote:
Perhaps a solution would be to issue an error or warning in the case of \switchtobodyfont[<unkownsize>]
But I still don't understand: why does \switchtobodyfont[33pt] sometimes work and sometimes not? You already told me once that I have to define that size (when using cow fonts for example), but is defining a size explicitely really necessary? Sometimes it works out of the box and sometimes it just causes problems. Mojca
Mojca Miklavec wrote:
On 1/24/07, Taco Hoekwater wrote:
Perhaps a solution would be to issue an error or warning in the case of \switchtobodyfont[<unkownsize>]
But I still don't understand: why does \switchtobodyfont[33pt] sometimes work and sometimes not? You already told me once that I have
It is not supposed to work at all. But sometimes, if your text is very simple, you get 'lucky' and it does not actually raise an error. But wat appears to work today, may be broken tomorrow, or the other way around. Best, Taco
Mojca Miklavec wrote:
On 1/24/07, Taco Hoekwater wrote:
Perhaps a solution would be to issue an error or warning in the case of \switchtobodyfont[<unkownsize>]
But I still don't understand: why does \switchtobodyfont[33pt] sometimes work and sometimes not? You already told me once that I have to define that size (when using cow fonts for example), but is defining a size explicitely really necessary? Sometimes it works out of the box and sometimes it just causes problems.
you need a matching body font environment; in th epast (long ago) i did generate such an environment bug it's confusing: slow, many redundant defs etc esp when used grouped; just define a matching bodyfont environment at the outer level and it should work ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
On Jan 24, 2007, at 20:34, Taco Hoekwater wrote:
Hans van der Meer wrote:
2. I am curious why this error message only now pops up, whereas previously the typesetting proceeded without a hitch. Is this caused by more stringent testing in the ConTeXt code or is it just a default behaviour that is no longer valid?
There is a bit of a mixup. Assuming you run pdfetex, the wrong
Well, if I did everything correct, that should not be the case. I had pdftex 1.40.1 made from the source and that made pdfetex equal to pdftex (as the INSTALL document tells me). Furthermore in TeXShop I explicitely in the Preferences changed pdfetex to pdftex.
definition of \dontleavehmode is now actually used. It should have been like this:
\let\dontleavehmode\quitvmode
For Hans: yes, the \pdf... prefix is missing from pdftex 1.40's \quitvmode.
In the non-pdftex, this weird error message is currently almost unavoidable, I'm afraid. Perhaps a solution would be to issue an error or warning in the case of \switchtobodyfont[<unkownsize>]
Hans van der Meer
Hans van der Meer wrote:
On Jan 24, 2007, at 20:34, Taco Hoekwater wrote:
Hans van der Meer wrote:
2. I am curious why this error message only now pops up, whereas previously the typesetting proceeded without a hitch. Is this caused by more stringent testing in the ConTeXt code or is it just a default behaviour that is no longer valid?
There is a bit of a mixup. Assuming you run pdfetex, the wrong
Well, if I did everything correct, that should not be the case.
It is not your mixup. ;-) Taco
participants (4)
-
Hans Hagen -
Hans van der Meer -
Mojca Miklavec -
Taco Hoekwater