Aditya Mahajan wrote:

Hans, if you agree with Wolfgang's fix, I think it should be included in core-mat.

i just wait till you send me a core-mat with fixes .. fixing whitespace is always a bit tricky ... test with/without whitespace as well as with grids Hans ----------------------------------------------------------------- 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 -----------------------------------------------------------------

Mikael Persson wrote:

1) I also tried the new context minimal with LuaTeX (to see if it worked better there) and found out that the fourier fonts are not included in the minimals anymore. Are there any plans to include them? Or to make some kind of package with these and other nice free fonts?

mojca will look into it (out of reach, mail etc now) since it came up last week

2) Will the problem from http://archive.contextgarden.net/message/20070830.154143.005528b8.en.html go away in the near future (of new versions)? From the answers to that question I did not really get if the problem was a big one or not.

depends on who wants to fix it -)

PS I am impressed by the speed of LuaTeX. It compiled my document in ~20s instead of 1-2 minutes.

Due to mplib or in general? Some parts of mkiv can still be sped up, e.g. xml processing, but there's not that much more to gain in font handling (that is ... we're playing with runtime mp fonts and there we can gain a lot when the new img related features show up). Also, some mechanisms will be reimplemented and as such be sped up. Hans ----------------------------------------------------------------- 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 -----------------------------------------------------------------

Mikael Persson wrote:

First, thank you Wolfgang, I tried to change it in the core-mat.tex and it looks great now!

...

...

1) I also tried the new context minimal with LuaTeX (to see if it worked better there) and found out that the fourier fonts are not included in the minimals anymore. Are there any plans to include them? Or to make some kind of package with these and other nice free fonts?

mojca will look into it (out of reach, mail etc now) since it came up last week

sounds great!

...

...

2) Will the problem from http://archive.contextgarden.net/message/20070830.154143.005528b8.en.html go away in the near future (of new versions)? From the answers to that question I did not really get if the problem was a big one or not.

depends on who wants to fix it -)

I don't know what to do (and I am not a programmer), but if there is some monkey-see-monkey-do work, I am happy to help.

...

...

PS I am impressed by the speed of LuaTeX. It compiled my document in ~20s instead of 1-2 minutes.

Due to mplib or in general? Some parts of mkiv can still be sped up, e.g. xml processing, but there's not that much more to gain in font handling (that is ... we're playing with runtime mp fonts and there we can gain a lot when the new img related features show up). Also, some mechanisms will be reimplemented and as such be sped up.

Hm, I hoped you would tell me why ;) The document is a math document with a lot of references, 5 different bibliographies. But no MetaPost graphics, just plain text + formulas. It is about 100 pages at the moment. Could it maybe be the fact that I am changing font? With LuaTeX I was using the lm fonts. I have a feeling it takes a bit longer to load other fonts, but that might just be a feeling from what is written in the terminal?

mkiv needs less fonts but they may be bigger (more glyphs) at the end of a run you see the statistics, just write them down occasionally so that we can see the improvements other reasons can be that some mem bugs have been removed over time adn that mkiv share more data, so that overall less memory is needed which in turn gives less garbage collection try adding \usemodule[timing] to the top of the doc and one or more \ShowUsage {} \ShowMemoryUsage{} \ShowNodeUsage {} at the end ... it generates some pictures that show what takes most time ----------------------------------------------------------------- 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 Mon, 28 Apr 2008, Mikael Persson wrote:

Hello again,

* snip *

...

I could reproduce the result and have also a fix but I don't know what the correct reults should be and other people like Aditya should decide if it makes sense to change the current behaviour.

\def\startdisplaymath {\ifgridsnapping \beforedisplayspace \snapmathtogrid\vbox

* snip *

...

Wolfgang

This fix seems not to solve the problem entirely. With this fix applied, If one now put a formula inside an enumeration one gets "double" space before the formula. This is a test file that shows the problem:

This is basically because in enumerations indenting is set to never. You can see the same affect with \setupindenting[no] \testtext One possible fix is to add indenting=(reset|next|first) %default is never to the \setupenumeration command. For example

\defineenumeration[example] [text=Example, location=serried, width=fit, indenting=reset, %%<<<---- Add This distance=0.5em, way=bysection, ]

Another is to change \startdisplaymath to \def\startdisplaymath {\ifgridsnapping \beforedisplayspace \snapmathtogrid\vbox \bgroup \informulatrue %\forgetall % breaks side floats \else \bgroup \parskip\formulaparskip % ! ! \informulatrue %\forgetall % otherwise backgrounds fail \ifdim\lastskip<\zeropoint\else \par \ifvmode \ifdim\parskip>\zeropoint\relax \whitespace \vskip-\parskip % kind of forces and cancels again \fi \fi \fi \doif\displaygridcorrection{-\v!top}{\kern-\strutht}% new, currently only option/default \beforedisplayspace \par \ifvmode \ifdim\parskip>\zeropoint\relax \else \ifindentfirstparagraph \verticalstrut \vskip-\struttotal \vskip-\baselineskip \fi\fi \fi \fi $$\setdisplaydimensions \setpredisplaysize\lastlinewidth \startinnermath} But this is getting ugly. Aditya

Thank you Aditya, but I don't get that to work. Both with

ConTeXt ver: 2007.04.03 13:01 MKII fmt: 2008.4.24 int: english/english (with the fix earlier in this thread) and ConTeXt ver: 2007.09.28 16:52 MKII fmt: 2008.4.24 int: english/english (which is the one from new ubuntu, without the fix)

Update your ConTeXt istallation, both systems use old versions. Wolfgang

On Wed, Apr 30, 2008 at 9:45 AM, Mikael Persson wrote:

On Wed, Apr 30, 2008 at 9:33 AM, Wolfgang Schuster wrote:

...

...

Thank you Aditya, but I don't get that to work. Both with

ConTeXt ver: 2007.04.03 13:01 MKII fmt: 2008.4.24 int: english/english (with the fix earlier in this thread) and ConTeXt ver: 2007.09.28 16:52 MKII fmt: 2008.4.24 int: english/english (which is the one from new ubuntu, without the fix)

Update your ConTeXt istallation, both systems use old versions.

Wolfgang

I wish I could. I cant even run the new version from ubuntu, since there is some problem with the fourier fonts, explained in an earlier mail in this thread (and also in an earlier post: http://archive.contextgarden.net/message/20070830.154143.005528b8.en.html)

No solution for this problem.

And new minimals seem not to include the fourier fonts.

If I set indenting=next (or first) in the enumeration, the file compiles...

Sorry for "complaining" about old versions of ConTeXt. I am finishing my PhD thesis, and have not found any nice combinations of text and math fonts in the new minimals. (I am not too fond of the lm fonts). If you have a tip about such fonts, please tell me :)

You don't have update your complete installation, cont-tmf.zip is enough for a new version but you also need the latest version of the latin modern fonts. Wolfgang