Re: [dev-context] cont-en.xml

2010-09-11 Thread Marius
On 2 July 2010 10:41, Wolfgang Schuster
schuster.wolfg...@googlemail.com wrote:
 Am 30.06.10 12:57, schrieb Marius:

 Hello,

 I think that the best way is to put command syntax description just
 before its definition in the context file. This way it will be easier
 to keep it always in sync with the code. And most important point here
 is that it must be in sync always. If it is, then we can just generate
 cont-en.xml and other cont-xx.xml files, but also have a syntax
 checker!

 Just imagine:

 context --check-syntax myfile.tex

 shows you which options or commands are experimental, deprecated, and
 which are unrecognized, and suggests the command or option to fix
 that.

 What do you think?


 The idea is not bad but who is going to do this? Hans implemented
 in mkiv the function to check assignments list of valid keys but
 this function isn't used because nobody writes the necessary lists
 which contexts requires to do these checks and the same will happen
 with the system you suggest.

Please, point me to the function you are talking about.

Thanks,
Marius
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-07-06 Thread Wolfgang Schuster

Am 05.07.10 12:26, schrieb Taco Hoekwater:

On 07/05/2010 11:44 AM, Marius wrote:
On Fri, Jul 2, 2010 at 1:21 PM, Taco Hoekwatert...@elvenkind.com  
wrote:

I think this will be a wasted effort.  I predict that any future
version of context that will be able to do syntax checking is going
to be based on lua for the verification process, not tex macros.


The 'key' checker for \getparameters in mkiv is lua based.

There is already an example implementation of comments reading.

http://gitorious.org/context/context/blobs/master/scripts/context/lua/mtx-modules.lua 



cd tex/context/base
mtxrun --script modules --process page-txt.mkiv

see the resulting page-txt-mkiv.ted and page-txt-mkiv.pdf files.

Yes, this functionality has been around for ages, and is the basis of
luigi's typeset source documentation project:

  http://foundry.supelec.fr/gf/project/modules

And a similar mechanism is available to document lua files but
it use xml markup rather than tex (and this since the beginning).

One possible extension would be:

%X cd:command name=showbodyfont
%X cd:sequence
%X cd:string value=showbodyfont/
%X /cd:sequence
%X cd:arguments
%X cd:keywords n=1 optional=yes list=yes
%X cd:inherit name=setupbodyfont n=1/
%X /cd:keywords
%X /cd:arguments
%X /cd:command

But that is rather verbose and needs processing to create useful
solution. Another possibility using actual lua could be:

%X showbodyfont = {
%X   seq = { string = showbodyfont },
%X   arg = { keywords = {
%X  n=1,list=true, optional=true,
%X  inherit={ name = setupbodyfont, n = 1

Essentially the same information, but shorter, and it could
be loaded into lua directly using 'loadstring'. Or there could
be a more TeX-like approach looking like this:

%X \cmd{showbodyfont}([...,...,...])-#1:\inherited{setupbodyfont}

to be parsed by a bit of lua code.

It all depends on what Hans is most comfortable with in keying in,
because if Hans doesn't like it the chances of anything like this
remaining up-to-date are about zero.

I'm against storing the command definition in the relevant files because
the definitions are meant for documentation and it's nonsense in this way
because you need sometimes references to other commands and who should
you show a command for spacing which is written in a font file.

Another problem is that big macros like \section or \setuphead are split
over different files (strc-sec.mkiv, strc-doc.mkiv, strc-doc.lua and 
strc-ren.mkiv)

and when you store the \section definition now in strc-sec.mkiv but make
a change in strc-ren.mkiv you will forget to add the change in the 
definition

file are too high.

From your two different variants i prefer the lua solution because it's
closer to the current xml version and it's easier to add variants of
a command (e.g. \setupinterlinespace[small] and 
\setupinterlinespace[line=...])


Wolfgang

___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-07-06 Thread taco

Wolfgang Schuster wrote:


I'm against storing the command definition in the relevant files because
the definitions are meant for documentation and it's nonsense in this way
because you need sometimes references to other commands and who should
you show a command for spacing which is written in a font file.


Well, they could be extracted automatically to a normal
lua file, but you had a good point about the \section etc
later on, so a separate file would probably be better
indeed.

And I also prefer the lua style, much less verbose than xml
while still containing all needed info. The Tex-like
stuff was meant as an example of what *could* be done,
but getting a good syntax for that that is still extensible
could be quite hard and why should we bother if the lua
code is readable enough.

However, we *do* have to write a formal description of the
lua syntax to avoid confusion, if this is indeed the way
to go. At this point I think we should ask Hans' how he feels
about this stuff, but unf. he has hardware problems and is
not as reachable as could be. Should be solved in a day or so.


Best wishes,
Taco
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-07-06 Thread Hans Hagen

On 6-7-2010 8:23, taco wrote:

sorry for delays in responding ... i'm still in the process of fixing 
hardware / installations etc



Wolfgang Schuster wrote:


I'm against storing the command definition in the relevant files because
the definitions are meant for documentation and it's nonsense in this way
because you need sometimes references to other commands and who should
you show a command for spacing which is written in a font file.


fyi: long ago it started like that: a \startsetup... definition as part 
of the tex files but it was a pain togenerate the lists that way



Well, they could be extracted automatically to a normal
lua file, but you had a good point about the \section etc
later on, so a separate file would probably be better
indeed.


i'm not so sure if i want to abandon the xml file (yet) as theres 
nothing wrong with xml per se ... it's no big deal to generate lua from 
it if needed (both are as unreadable although with somenamespacing we 
could probably share more in lua but at the cost of readablity (and i'm 
really sensitive to formating of such files -)


needs more thinking as changes in that area should not happen too often

we can consider splitting up things but as already mentioned it's hard 
to come up with a devision based on the tex files ... ok, in due time we 
might get more granularity; an option is to make files per block: page, 
strc, core, ...


concerning the documentation in the lua files (not that much and 
currently in xml) ... it could be just context code as it then permits 
embedding examples


Hans


-
  Hans Hagen | PRAGMA ADE
  Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com
 | www.pragma-pod.nl
-
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-07-05 Thread Marius
On Fri, Jul 2, 2010 at 1:21 PM, Taco Hoekwater t...@elvenkind.com wrote:

 I think this will be a wasted effort.  I predict that any future
 version of context that will be able to do syntax checking is going
 to be based on lua for the verification process, not tex macros.

There is already an example implementation of comments reading.

http://gitorious.org/context/context/blobs/master/scripts/context/lua/mtx-modules.lua

cd tex/context/base
mtxrun --script modules --process page-txt.mkiv

see the resulting page-txt-mkiv.ted and page-txt-mkiv.pdf files.

 Possibly the internal structure could also be read in from structured
 comments in the actual source files as opposed to the current external
 xml files (not a bad idea as it increased the chance of everything staying
 up-to-date).

So, reading of comments could be implemented based on this example.

 But even in that case, those comments are much
 more likely to be lua or compressed ascii code than actual tex syntax
 because interpretation will be done mostly by lua code.

What do you mean by saying to be lua or compressed ascii code? I
can't imagine how it may look like, as I am new to the lua. Could you,
please, give an example?

I will try to code a rough draft, but at first it needs to be decided
what syntax we will use for these comments.

Regards,
Marius
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-07-05 Thread Taco Hoekwater

On 07/05/2010 11:44 AM, Marius wrote:

On Fri, Jul 2, 2010 at 1:21 PM, Taco Hoekwatert...@elvenkind.com  wrote:


I think this will be a wasted effort.  I predict that any future
version of context that will be able to do syntax checking is going
to be based on lua for the verification process, not tex macros.


There is already an example implementation of comments reading.

http://gitorious.org/context/context/blobs/master/scripts/context/lua/mtx-modules.lua

cd tex/context/base
mtxrun --script modules --process page-txt.mkiv

see the resulting page-txt-mkiv.ted and page-txt-mkiv.pdf files.


Yes, this functionality has been around for ages, and is the basis of
luigi's typeset source documentation project:

  http://foundry.supelec.fr/gf/project/modules

One possible extension would be:

%X  cd:command name=showbodyfont
%X  cd:sequence
%X  cd:string value=showbodyfont/
%X/cd:sequence
%Xcd:arguments
%X  cd:keywords n=1 optional=yes list=yes
%Xcd:inherit name=setupbodyfont n=1/
%X  /cd:keywords
%X/cd:arguments
%X  /cd:command

But that is rather verbose and needs processing to create useful
solution. Another possibility using actual lua could be:

%X showbodyfont = {
%X   seq = { string = showbodyfont },
%X   arg = { keywords = {
%X  n=1,list=true, optional=true,
%X  inherit={ name = setupbodyfont, n = 1

Essentially the same information, but shorter, and it could
be loaded into lua directly using 'loadstring'. Or there could
be a more TeX-like approach looking like this:

%X \cmd{showbodyfont}([...,...,...])-#1:\inherited{setupbodyfont}

to be parsed by a bit of lua code.

It all depends on what Hans is most comfortable with in keying in,
because if Hans doesn't like it the chances of anything like this
remaining up-to-date are about zero.


Best wishes,
Taco
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-07-02 Thread Wolfgang Schuster

Am 30.06.10 12:57, schrieb Marius:

Hello,

I think that the best way is to put command syntax description just
before its definition in the context file. This way it will be easier
to keep it always in sync with the code. And most important point here
is that it must be in sync always. If it is, then we can just generate
cont-en.xml and other cont-xx.xml files, but also have a syntax
checker!

Just imagine:

context --check-syntax myfile.tex

shows you which options or commands are experimental, deprecated, and
which are unrecognized, and suggests the command or option to fix
that.

What do you think?
   


The idea is not bad but who is going to do this? Hans implemented
in mkiv the function to check assignments list of valid keys but
this function isn't used because nobody writes the necessary lists
which contexts requires to do these checks and the same will happen
with the system you suggest.

Wolfgang

___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-07-02 Thread Peter Münster
On Fri, Jul 02 2010, Wolfgang Schuster wrote:

 The idea is not bad but who is going to do this? Hans implemented
 in mkiv the function to check assignments list of valid keys but
 this function isn't used because nobody writes the necessary lists
 which contexts requires to do these checks and the same will happen
 with the system you suggest.

That's why I'm eventually going to do this:
http://archive.contextgarden.net/message/20100508.204345.b5003312.en.html

Perhaps using tex-files like this:
http://pmrb.free.fr/tmp/context-commands/

Perhaps in this directory:
http://foundry.supelec.fr/svn/contextman/context-commands
So that everybody can contribute via svn.

But before doing that effort, I still need a bit more feedback...

Cheers, Peter

-- 
Contact information: http://pmrb.free.fr/contact/


___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-07-02 Thread Taco Hoekwater

On 07/02/2010 11:28 AM, Peter Münster wrote:

On Fri, Jul 02 2010, Wolfgang Schuster wrote:


The idea is not bad but who is going to do this? Hans implemented
in mkiv the function to check assignments list of valid keys but
this function isn't used because nobody writes the necessary lists
which contexts requires to do these checks and the same will happen
with the system you suggest.


That's why I'm eventually going to do this:
http://archive.contextgarden.net/message/20100508.204345.b5003312.en.html

Perhaps using tex-files like this:
http://pmrb.free.fr/tmp/context-commands/

Perhaps in this directory:
http://foundry.supelec.fr/svn/contextman/context-commands
So that everybody can contribute via svn.

But before doing that effort, I still need a bit more feedback...


I think this will be a wasted effort.  I predict that any future
version of context that will be able to do syntax checking is going
to be based on lua for the verification process, not tex macros.

Possibly the internal structure could also be read in from structured
comments in the actual source files as opposed to the current external
xml files (not a bad idea as it increased the chance of everything 
staying up-to-date). But even in that case, those comments are much

more likely to be lua or compressed ascii code than actual tex syntax
because interpretation will be done mostly by lua code.

Best wishes,
Taco
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-06-23 Thread Peter Münster
On Wed, Jun 16 2010, Wolfgang Schuster wrote:

 i plan to update cont-en.xml with changes in mkiv (new keys and commands) 

Hello Wolfgang,

Since Hans said documentation about tex can best be done in tex, I've put
my ideas into some files here:

http://pmrb.free.fr/tmp/context-commands/

What do you think about that?

Cheers, Peter

-- 
Contact information: http://pmrb.free.fr/contact/


___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-06-23 Thread Aditya Mahajan

On Wed, 16 Jun 2010, Wolfgang Schuster wrote:


Hi,

i plan to update cont-en.xml with changes in mkiv (new keys and commands) but 
i'm unsure if it's good to mix mkii and mkiv. I think it's to make a separate 
definition file for mkiv which contains the extra keys for the commands (e.g. 
sectionsegments for \setuphead) and use the current one for mkii only.


Is there a XML schema for the commands. The cont-en.xml file mentions 
http://www.pragma-ade.com/commands but that does not exist.


(I am trying to experiement with using YAML instead of XML...)

Aditya
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-06-18 Thread Hans Hagen

On 17-6-2010 3:11, Mojca Miklavec wrote:

On Wed, Jun 16, 2010 at 19:56, Wolfgang Schuster
schuster.wolfg...@googlemail.com  wrote:

Hi,

i plan to update cont-en.xml with changes in mkiv (new keys and commands)
but i'm unsure if it's good to mix mkii and mkiv. I think it's to make a
separate definition file for mkiv which contains the extra keys for the
commands (e.g. sectionsegments for \setuphead) and use the current one for
mkii only.


My opinion (though at the end you may do it either way): I would put a
special key to either each command or each option that would clarify
mkii only or mkiv only or both. If command is completely
different, for example
\usetypescript[name][ec] in mkii vs. \usetypescript[name] in mkiv
one could define two commands and label one as mkii only and the
second one as mkiv only.


there is already something like that but not applied yet to all commands


There might be also another thing worth keeping in mind. Some commands
may be valid, but useless in mkiv. It might make sense to label those
as such.

Even though the fact that mkii is frozen might be true, most of
valid options are still missing in those xml files.


well, 'most' is not fair

Hans

-
  Hans Hagen | PRAGMA ADE
  Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com
 | www.pragma-pod.nl
-
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-06-18 Thread Mojca Miklavec
On Fri, Jun 18, 2010 at 10:58, Hans Hagen wrote:
 On 17-6-2010 3:11, Mojca Miklavec wrote:

 Even though the fact that mkii is frozen might be true, most of
 valid options are still missing in those xml files.

 well, 'most' is not fair

I'm sorry, I probably wanted to say that most of undocumented options
are missing in those xml files :) :) :)

There is no doubt that the great majority and in particular all the
most important commands are present.

Mojca
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-06-17 Thread Hans Hagen

On 16-6-2010 7:56, Wolfgang Schuster wrote:

Hi,

i plan to update cont-en.xml with changes in mkiv (new keys and
commands) but i'm unsure if it's good to mix mkii and mkiv. I think it's
to make a separate definition file for mkiv which contains the extra
keys for the commands (e.g. sectionsegments for \setuphead) and use the
current one for mkii only.


we can have several files:

cont-en.xml (common stuff)
cont-en-mkiv.xml
cont-en-mkii.xml

the mkii and mkiv files can include cont-en.xml

-
  Hans Hagen | PRAGMA ADE
  Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com
 | www.pragma-pod.nl
-
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-06-17 Thread Hans Hagen

On 16-6-2010 7:56, Wolfgang Schuster wrote:

Hi,

i plan to update cont-en.xml with changes in mkiv (new keys and
commands) but i'm unsure if it's good to mix mkii and mkiv. I think it's
to make a separate definition file for mkiv which contains the extra
keys for the commands (e.g. sectionsegments for \setuphead) and use the
current one for mkii only.


if we consider mkii frozen it might indeed be better to completely 
separate file


-
  Hans Hagen | PRAGMA ADE
  Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com
 | www.pragma-pod.nl
-
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-06-17 Thread Mojca Miklavec
On Wed, Jun 16, 2010 at 19:56, Wolfgang Schuster
schuster.wolfg...@googlemail.com wrote:
 Hi,

 i plan to update cont-en.xml with changes in mkiv (new keys and commands)
 but i'm unsure if it's good to mix mkii and mkiv. I think it's to make a
 separate definition file for mkiv which contains the extra keys for the
 commands (e.g. sectionsegments for \setuphead) and use the current one for
 mkii only.

My opinion (though at the end you may do it either way): I would put a
special key to either each command or each option that would clarify
mkii only or mkiv only or both. If command is completely
different, for example
   \usetypescript[name][ec] in mkii vs. \usetypescript[name] in mkiv
one could define two commands and label one as mkii only and the
second one as mkiv only.

There might be also another thing worth keeping in mind. Some commands
may be valid, but useless in mkiv. It might make sense to label those
as such.

Even though the fact that mkii is frozen might be true, most of
valid options are still missing in those xml files.

Again: I would keep a single file with a new key that would specify in
which mark commands works and/or makes sense. That way it would be a
tiny bit easier to maintain both versions synchronized (assume that
you will want to document a new command that's also present in MKII -
are you going to remember that you also need to update MKII; and then
one will see a command in MKIV, but will have no idea whether the
absence in MKII means that somebody forgot that command or that it's
simply not present/working there).

But since you are planning to work on it, it's up to you to decide.
That was only my opinion.

Mojca
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml

2010-06-17 Thread Wolfgang Schuster

Am 17.06.10 15:11, schrieb Mojca Miklavec:

On Wed, Jun 16, 2010 at 19:56, Wolfgang Schuster
schuster.wolfg...@googlemail.com  wrote:
   

Hi,

i plan to update cont-en.xml with changes in mkiv (new keys and commands)
but i'm unsure if it's good to mix mkii and mkiv. I think it's to make a
separate definition file for mkiv which contains the extra keys for the
commands (e.g. sectionsegments for \setuphead) and use the current one for
mkii only.
 

My opinion (though at the end you may do it either way): I would put a
special key to either each command or each option that would clarify
mkii only or mkiv only or both. If command is completely
different, for example
\usetypescript[name][ec] in mkii vs. \usetypescript[name] in mkiv
one could define two commands and label one as mkii only and the
second one as mkiv only.

There might be also another thing worth keeping in mind. Some commands
may be valid, but useless in mkiv. It might make sense to label those
as such.

Even though the fact that mkii is frozen might be true, most of
valid options are still missing in those xml files.

Again: I would keep a single file with a new key that would specify in
which mark commands works and/or makes sense. That way it would be a
tiny bit easier to maintain both versions synchronized (assume that
you will want to document a new command that's also present in MKII -
are you going to remember that you also need to update MKII; and then
one will see a command in MKIV, but will have no idea whether the
absence in MKII means that somebody forgot that command or that it's
simply not present/working there).

But since you are planning to work on it, it's up to you to decide.
That was only my opinion.
   


For the start i'll go only through the mkiv files and use a separate 
file for them,
maybe later i can compare the mkiv with the mkii options and use hans 
approach

because tagging each key and command is a lot of work.

Wolfgang

___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml / unique keys

2008-12-23 Thread Taco Hoekwater


Patrick Gundlach wrote:
 Hi,
 
 (see setupinterlinespace2 for example, it has variant 2 but there is no
 setupinterlinespace2 with variant 1). 

IIRC, that trailing 2 in setupinterlinespace2 was just added due to my
misunderstanding of how \showsetup works.

Best wishes,
Taco
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml / unique keys

2008-12-21 Thread Hans Hagen

Patrick Gundlach wrote:

Hi,

could we add some unique key to the commands? That way we could have a
more reliable way to know which command is which. It is useful when you
use cd:inherit and refer to a command, such as color (which can be an
environment or a standalone command) or commands with variants. It even
removes the need for the variant=.. which is not perfect (see
setupinterlinespace2 for example, it has variant 2 but there is no
setupinterlinespace2 with variant 1). 


example:

  cd:command id=d1e311 name=setuplanguage file=lang-ini.tex category=lan


why not using something

cd:command id=cd:number

which makes it 'easier' to refer to them when discussing them (i hate 
funny keys -)


btw 1: i made a simple variation of texshow using texlua as webserver, 
just for local usage in editors like i do in scite)


btw 2: we can use the id also for related info as the one users can 
enter on the wiki; i can even imagine that we use such id's in the wiki 
examples (so that we can filter wikisection command=cd:number... )


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
-
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml and alike

2008-04-23 Thread Hans Hagen
Patrick Gundlach wrote:
 Hi,
 
 could this:
 
 symalign references symalign, which is not defined.

  cd:parameter name=symalign
 cd:resolve name=symalign/
   /cd:parameter
 
 in setupitemgroup, line 6072.
 
 and
 
 
cd:keywords optional=yes
  cd:constant name=lefttext/
  cd:constant name=middletext/
  cd:constant name=righttext/
/cd:keywords
 
 (substitute name with type) 
 
 be fixed? Or is this already done?

just fix what needs fixing and send me the new file unless you don't 
know what to fix -)


-
   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
-
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml and alike

2008-04-20 Thread Patrick Gundlach
Hi,

 So you also use the advanced feature with the cd:define keys? I
 recently came across a problem there, but I can't remember which it
 was, I'll try and find out again.

symalign references symalign, which is not defined.

 cd:parameter name=symalign
cd:resolve name=symalign/
  /cd:parameter

in setupitemgroup, line 6072.


How do you handle such cases in the references? In texshow.pl?

Has anybody created a schematron rule for that yet?

Patrick
-- 
ConTeXt wiki and more: http://contextgarden.net
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml and alike

2008-04-20 Thread Taco Hoekwater
Patrick Gundlach wrote:
 Hi,
 
 So you also use the advanced feature with the cd:define keys? I
 recently came across a problem there, but I can't remember which it
 was, I'll try and find out again.
 
 symalign references symalign, which is not defined.
 
  cd:parameter name=symalign
 cd:resolve name=symalign/
   /cd:parameter
 
 in setupitemgroup, line 6072.
 
 
 How do you handle such cases in the references? In texshow.pl?

In texshow.pl, this generates a display error:

  align=inner|outer|left|right|flushleft|flushright|middle
  symalign=,
  indentnext=yes|no]


 Has anybody created a schematron rule for that yet?
 
 Patrick

___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context


Re: [dev-context] cont-en.xml and alike

2008-04-19 Thread Hans Hagen
Patrick Gundlach wrote:
 Hi,
 
 how is cont-en.xml and the related files currently maintained? Is this
 file created of some other file or is it updated manually?

manual (it would make sense to merge it in the sources but ..)

 What are the current applications that use the xml-files?

reference list, manuals, wiki

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
-
___
dev-context mailing list
dev-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/dev-context