doxygen-develop — List for discussing new features, submitting patches, etc.

[Doxygen-develop] re: copydoc patch

[Iain B. <ia...@pc...>]

I have been trying to patch doxygen to extend the copydoc special
command (to copy any item, eg enums).  I am a programmer, but I'm stuck!
(not having much experience with other peoples' code)

I found where I think I need to make some mods, but I would appreciate
some pointers from here on (I realise that asking someone to do it for
me would probably never happen, but thats not what I'm asking, unless
you're really bored!)

I followed CMD_COPYDOC to 'void DocCopy::parse()' in docparser.cpp, and
then found 'findDocsForMemberOrCompound' and had a look at that
function.  But thats where I got lost.  It looks like
'findDocsForMemberOrCompound' returns true if copydoc references a
namespace, group, page or file, and false otherwise.  Is this correct?=20
How might I extend it for an enum?

I would greatly appreciate some help here, whatever you may be able to
provide.  I'm sure it would only take a few minutes for someone
experienced with the doxygen code.

Many TIA,
--=20
Iain Buchanan <ia...@no...>
"Live or die, I'll make a million."
-- Reebus Kneebus, before his jump to the center of the earth, Firesign The=
ater

[Doxygen-develop] re: customising doxygen output

[Iain B. <ia...@pc...>]

I have written a perl script to extend the use of the ALIAS
configuration file tag.  What I want from this email is some response if
anyone thinks this may be useful for any reason, in which case I can
update the script a bit and release it.  If you're interested, keep
reading; if not, keep reading anyway, because it may do more than you
think.

History.
I started writing the perl script about 6 months ago (or more), before
doxygen version 1.2.10, before the doxygen xml output was used much, and
before the new perl (1.3-rc2) output was available.  I finished it for
internal use by our company, and forgot about it.  Its recently come up
again though as there is another major update on the way.

Reason.
Basically it is a very devious way of extending doxygen's "alias"
configuration file tag.  Essentially, I had a look at the doxygen source
code and thought my own perl script would be quicker and easier (now I'm
not so sure!)

The problem with the alias tag, is its a simple substitution.  For
example:
    ALIASES =3D "myalias=3D<b>"
  would cause
    @myalias hello world
  to be replaced with
    <b> hello world
  There is no option to add something at the end, to the effect of
    <b> hello world </b>

Usage.
The script works as a pre-filter for doxygen, and takes as an arguement
a filename which has a list of aliases in it, along with pre-html,
post-html, pre-latex and post-latex fields.  It can even add
@(end)latexonly or @(end)htmlonly pairs if necessary.  For example you
could put in the config file:
@texttt, "<code>", "</code>", "\texttt {", "}"
and when the prefilter came accross, say,=20
"@texttt foo bar baz"
in doxygen comment blocks, it would replace it with
"@htmlonly <code> foo bar baz </code> @endhtmlonly @latexonly \texttt
{foo bar baz} @endlatexonly"

Either html or latex output (or both, if you want to ignore some
aliases) can be excluded for any set.

I try and keep line numbers the same so that included sources won't be
stuffed up.

What now?
as I said, what I want from this email is some response if anyone thinks
this may be useful for any reason, in which case I can update the script
a bit (remove the comments that say "<coworker> is an idiot" ...!) and
release it.  I have found it to be _very_ powerful, and I can show some
examples that turn our doxygen output into latex tables inside sections
inside more tables...!

Of course I realise that with all the recent doxygen enhancements, this
may all be unnecessary as there might be easier ways of doing it now.

Many thanks for any comments,
--=20
Iain Buchanan <ia...@no...>
It is easier for a camel to pass through the eye of a needle if it is
lightly greased.
		-- Kehlog Albran, "The Profit"

Re: [Doxygen-develop] Controlling the output in the "Alphabetical List" regarding namespace

[Dimitri v. H. <di...@st...>]

On Mon, Jan 13, 2003 at 10:28:27AM +0100, Ren? Dencker Eriksen wrote:
> Hi,
> 
> I will hereby suggest an improvement to how the output on the 
> "Alphabetical List" (and the "Class Hierarchy") can be controlled.
> 
> At the moment I have for instance the following two classes, that shows 
> up on the "Alphabetical List" as this:
> 
> ipl::CHoughBase (ipl)
> ipl::CHoughBase::CBinaryImage (ipl)
> 
> Both in the namespace "ipl". The CBinaryImage is defined inside CHoughBase.
> 
> It would be nice to be able to remove the ipl:: in front of both lines 
> since it is redundant (ipl is also shown at the end of each line).

Has already been fixed in the latest CVS release.

> The only control of this in the setup-file is setting the value of 
> HIDE_SCOPE_NAMES. But this leads to the following:
> 
> CHoughBase (ipl)
> CBinaryImage (ipl)
> 
> The CHoughBase scope is also stripped from the CBinaryImage but is still 
> placed under "H" in the alphabetical ordered list. This is very 
> confusing. It would be nice with an option to remove ipl:: only, so the 
> output becomes:
> 
> CHoughBase (ipl)
> CHoughBase::CBinaryImage (ipl)

This should now be the default behaviour (it used to be until I introduced a 
bug).

Regards,
  Dimitri

[Doxygen-develop] XML (DTD and Schema Support)

[<Nas...@Sy...>]

I just installed the latest version of Doxygen 1.3rc2 and noticed the XML 
tab in Doxywizard has two options to add a XML DTD and schema. Has anyone 
tested these and are there any more details on what these options do?

Thanx.
______________________________________
Nasser Saleem Ahmed
Software Engineer

Symbian Ltd 
2-6 Boundary Row
London SE1 8HP
United Kingdom

Tel :   +44 (0) 207 563 2886
Fax :  +44 (0) 207 563 2083
Mob : +44 (0)  7949 245430
Email: nas...@sy...
http://www.symbian.com
______________________________________


**********************************************************************
Symbian Ltd is a company registered in England and Wales with registered number 01796587 and registered office at 19 Harcourt Street, London, W1H 4HF, UK.
This message is intended only for use by the named addressee and may contain privileged and/or confidential information. If you are not the named addressee you should not disseminate, copy or take any action in reliance on it. If you have received this message in error please notify pos...@sy... and delete the message and any attachments accompanying it immediately. Symbian does not accept liability for any corruption, interception, amendment, tampering or viruses occurring to this message in transit or for any message sent by its employees which is not in compliance with Symbian corporate policy.
**********************************************************************

[Doxygen-develop] Controlling the output in the "Alphabetical List" regarding namespace

[<ed...@mi...>]

Hi,

I will hereby suggest an improvement to how the output on the 
"Alphabetical List" (and the "Class Hierarchy") can be controlled.

At the moment I have for instance the following two classes, that shows 
up on the "Alphabetical List" as this:

ipl::CHoughBase (ipl)
ipl::CHoughBase::CBinaryImage (ipl)

Both in the namespace "ipl". The CBinaryImage is defined inside CHoughBase.

It would be nice to be able to remove the ipl:: in front of both lines 
since it is redundant (ipl is also shown at the end of each line).

The only control of this in the setup-file is setting the value of 
HIDE_SCOPE_NAMES. But this leads to the following:

CHoughBase (ipl)
CBinaryImage (ipl)

The CHoughBase scope is also stripped from the CBinaryImage but is still 
placed under "H" in the alphabetical ordered list. This is very 
confusing. It would be nice with an option to remove ipl:: only, so the 
output becomes:

CHoughBase (ipl)
CHoughBase::CBinaryImage (ipl)

best regards,
-- 
	René Dencker Eriksen  ed...@mi...
	Homepage:  http://www.mip.sdu.dk/~edr
	Tlf: Work: 65503548 Private: 66193416

[Doxygen-develop] Re: [Doxygen-users] Need to make some extra money every month??

[Dimitri v. H. <di...@st...>]

On Fri, Jan 10, 2003 at 02:38:19PM +0000, The Advantage internat system wrote:
> SPAM SPAM SPAM.

Hi, I've just limited posting to the mailing list to members-only in order
to avoid spammers flooding the list with their crap. Sorry for the 
inconvenience this may cause.

Regards,
  Dimitri

[Doxygen-develop] Need to make some extra money every month??

[The A. i. s. <The...@ww...>]

<html><head><title>email</title><meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1"></head><body bgcolor="#FFFFFF" text="#000000" link="#0000FF" vlink="#00CCFF"><table width="100%" border="0" cellspacing="0" cellpadding="5"><tr><td align="center"><table width="500" border="0" cellspacing="0" cellpadding="5"><tr><td align="center"><font face="Arial, Helvetica, sans-serif"><b><font color="#000000"><big>
<html>
<head>
<title>Untitled Document</title>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
</head>

<body text="#000000" bgcolor="#FFFFFF">
<table width="100%" border="0" cellspacing="0" cellpadding="5">
  <tr bgcolor="#003366" align="center"> 
    <td> 
      <table width="700" border="0" cellspacing="0" cellpadding="5">
        <tr> 
          <td align="center"><a href="http://www.a247.co.uk/sitepro"><img src="http://www.a247.co.uk/sitepro/images/advantagesys.jpg" width="450" height="69" border="0"></a></td>
        </tr>
      </table>
    </td>
  </tr>
  <tr> 
    <td valign="top" align="center"> 
      <table width="100%" border="0" cellspacing="0" cellpadding="5">
        <tr> 
          <td bgcolor="#FFFFFF"><font color="#FFFFFF" face="Arial, Helvetica, sans-serif" size="4"><b></b></font></td>
          <td bgcolor="#FFFFFF" align="center"> 
            <table width="360" border="0" cellspacing="0" cellpadding="5">
              <tr> 
                <td align="center"><a href="http://www.a247.co.uk/sitepro" target="_blank"><img src="http://www.a247.co.uk/sitepro/images/silversys.gif" width="115" height="20" border="0"></a></td>
                <td align="center"><a href="http://www.a247.co.uk/sitepro/purchase.htm" target="_blank"><img src="http://www.a247.co.uk/sitepro/images/silverpur.gif" width="115" height="20" border="0"></a></td>
                <td align="center"><a href="http://www.a247.co.uk/sitepro" target="_blank"><img src="http://www.a247.co.uk/sitepro/images/silverdemo.gif" width="115" height="20" border="0"></a></td>
              </tr>
            </table>
          </td>
          <td bgcolor="#FFFFFF">&nbsp;</td>
        </tr>
        <tr> 
          <td bgcolor="#003366" align="center" colspan="3"><font color="#FFFFFF" face="Arial, Helvetica, sans-serif" size="4"><b><i>the 
            system </i></b></font></td>
        </tr>
        <tr valign="top" align="left"> 
          <td colspan="3"> 
            <p><font face="Arial, Helvetica, sans-serif"><br>
              <b>It's probably everybodys dream to make or win enough money so 
              that they never have to work again. It never happens to most people, 
              however every now and again an opportunity comes along that can 
              take the strain off financially. That is exactly what <a href="http://www.a247.co.uk/sitepro">The 
              Advantage&#153; internet system does.</a> </b></font></p>
            <p><font face="Arial, Helvetica, sans-serif">We have spent the past 
              twelve months developing, testing and fine tuning a way of making 
              money while you carry on with your normal day. It is all based around 
              a new and rapidly growing concept that has been created by the internet.</font></p>
            <p><font face="Arial, Helvetica, sans-serif">Did you know that <b>every 
              single day</b> of the year last year it is estimated that more than 
              <b>&pound;100million</b> pounds was spent on the internet.. yes 
              everyday! This year it is expected to at least <b>double and probably 
              triple!</b> </font></p>
            <p><font face="Arial, Helvetica, sans-serif">If that doesn't come 
              as a surprise maybe the fact that more than 90% of those sales had 
              a commission attached to it might do. YES! Almost every single transaction 
              that takes place on the internet has a commission tagged aginst 
              it - even those from the big well known names like <b>Argos, Dixons, 
              Marks &amp; Spencer, Tesco etc.</b> But the most amazing thing is 
              that more than half of those are never claimed! </font></p>
            <p><font face="Arial, Helvetica, sans-serif">We have developed a system 
              that gives you the chance to get your hands on some of that money 
              this year.</font></p>
            <p align="center"><font face="Arial, Helvetica, sans-serif"><a href="http://www.a247.co.uk/sitepro">Click 
              here to find out more about the Advantage&#153; internet system</a></font></p>
            <p><font face="Arial, Helvetica, sans-serif"> </font></p>
            </td>
        </tr>
      </table>
      <p>&nbsp;</p>
    </td>
  </tr>
  <tr> 
    <td>&nbsp;</td>
  </tr>
</table>
</body>
</html>
</big></font></font></td></tr></table></td></tr><tr><td align="center"><table width="80%" border="0" cellspacing="0" cellpadding="5"><tr><td><p><font face="Arial, Helvetica, sans-serif" color="#000000">
<a href="" target="_blank"><font color="#0000FF"></font></a></font></p><p><font color="#000000"><font face="Arial, Helvetica, sans-serif">
<a href="" target="_blank"><font color="#0000FF"></font></a></font></font></p><p><font face="Arial, Helvetica, sans-serif" color="#000000"><a href="" target="_blank"><font color="#0000FF"></font></a>
</font></p></td></tr></table></td></tr><tr><td>&nbsp;</td></tr></table></body></html>


<center>If you wish to be removed from this mailing list please click below.<br><center> <a href="http://workinglunch.a247.co.uk/sitepro/remove.pl?dox...@li...&files/list8.txt"> CLICK HERE TO BE REMOVED</a></center><br>

[Doxygen-develop] Request for a feature

[neil. l. <nig...@ho...>]

I often forget to save my settings before I close the window, can someone 
write a few lines of code to prompt me with a messagebox if I've changed 
something and not saved?




_________________________________________________________________
MSN 8 with e-mail virus protection service: 2 months FREE* 
http://join.msn.com/?page=features/virus

[Doxygen-develop] subscribe

[Nguyen H. H. <hu...@io...>]

[Doxygen-develop] subscribe

[Nguyen H. H. <hu...@io...>]

[Doxygen-develop] Treeview v.4

[<mat...@vo...>]

Hi,

I have noticed that an earlier version of tree view is used by Doxygen. That
version needs very long time to display larger tree.

I have managed to convert the tree.js to the format that can be used by TreeView
v.4 using sed and the performance is much better.

I suggest that next versions of Doxygen use newer version of TreeView if it is
not prohibited due to licensing.

http://www.treeview.net

Best regards,

Matej

P.S.: I have attached new tree.html and a conversion script if you want to use
it with any version of Doxygen. The TreeView sources can be found on the site
mentioned above.

[Doxygen-develop] Marked-up word at end of list-item causes next list-item to be merged

[Mark T. <mth...@at...>]

I think I found a small bug where list items are
(incorrectly?) merged when the end of the previous
list item is marked-up (with \a, \c, or \b).

From a quick look at the code, it seems that
handleStyleArgument() is probably reading a
TK_LISTITEM token and dropping it without
processing?

An easy work-around is to add extra (non-whitespace)
stuff at the end of the line.  Don't know why, but
extra whitespace at the end doesn't avoid the issue.

I'm running:
Doxygen version 1.3-rc1
on Windows 2000

Below, I've included a small file that shows a few
cases where list items are (incorrectly?) merged.

regards,
markt

/** \file foo.c
*   \brief a test file
*/

//! Marked-up word at end of list-item causes next list-item to be merged
/**
*  - Here is a line with 'emphasis' mark-up at the \a end     
*  - This line should be a new list entry, but instead is merged with the line above
*
*  - Same thing with 'code' mark-up at the \c end     
*  - This line should be a new list entry, but instead is merged with the line above
*
*  - Same thing with 'bold' mark-up at the \b end     
*  - This line should be a new list entry, but instead is merged with the line above
*/
int my_function1() {}

Re: [Doxygen-develop] php and doxygen

[Dimitri v. H. <di...@st...>]

On Fri, Dec 06, 2002 at 05:59:55PM +0500, Julien Laumonier wrote:
> Hello,
> 
> I tried to generate documentation for a php 4 project, and doxygen does 
> not seem to parse .php4 in php but only .php and .inc.
> 
> to correct it, i have changed one line in scanner.l (line 396)
> 
> insidePHP  = fileName.right(4)==".php" || fileName.right(4)==".inc" || 
> fileName.right(5)==".php4";
> 
> Is there a simpler way to parse .php4 files in php or is it a bug (or 
> something else ?

Doxygen didn't recognise .php4 files as being PHP files. With you patch
it does, so I'll include it. If you know more typical php file 
extensions please let me know.

Regards,
  Dimitri

[Doxygen-develop] dot font problem

[Ryunosuke S. <pe...@mi...>]

Hi,

Dot tool use Helvetica font, but some windows has no Helvetica.
And I want to change Helvetica to Arial used by dot on windows.

Do you know famous linux distribution has Arial font?

Ryunosuke Satoh su...@ho...

[Doxygen-develop] dot font problem

[Ryunosuke S. <su...@ho...>]

Hi,

Dot tool use Helvetica font, but some windows has no Helvetica.
And I want to change Helvetica to Arial used by dot on windows.

Do you know famous linux distribution has Arial font?

Ryunosuke Satoh su...@ho...

[Doxygen-develop] php and doxygen

[Julien L. <jla...@fr...>]

Hello,

I tried to generate documentation for a php 4 project, and doxygen does 
not seem to parse .php4 in php but only .php and .inc.

to correct it, i have changed one line in scanner.l (line 396)

insidePHP  = fileName.right(4)==".php" || fileName.right(4)==".inc" || 
fileName.right(5)==".php4";

Is there a simpler way to parse .php4 files in php or is it a bug (or 
something else ?

Thanks

Julien Laumonier

[Doxygen-develop] Re: Doxygen bug reports and suggestions

[Dimitri v. H. <di...@st...>]

Hi Thorsten!

On Wed, Dec 04, 2002 at 03:40:43PM +0100, Thorsten Schleinzer wrote:
> since the mailing list does not seem to be very vivid
> I send you here a list of bugs and suggestions I
> Found over the last week while writing a generic
> Frontend parsing the 'doxygenerated' XML and
> Converting it to various formats.

I read your mails on the list but hadn't found the time to respond.

> I'm using the current release version (1.3.1 I think)

Probably 1.3-rc1

> 
> On Windows XP Pro.
> 
> Bugs:
> 
> 1)
> 
> @note creates a kind="bug" simplesect instead of a kind="note"

Yep, that's a typo in src/xmldocvisitor.cpp

> 
> 2)
> 
> The @example tag does not work as expected:
> 
> Using the following code does lead to doxygen generate no
> detailed description at all:
> 
> /**
>  * Some descry...
>  *
>  * @example MM3D/Exception.example.cs
> 
> while the following workaround is ok:
> 
> /**
>  * Some descry...
>  *
>  * @par Example
>  * @include MM3D/Exception.example.cs

I'll look into this

> Suggestions:
> 1)
> 
> Please convert all mandatory whitespaces in the XML
> Output to  
> 
> Because in the currently generated code there is a lot
> Of stuff like:
> 
> <highlight class="normal"> </highlight>
> 
> Although this single space is important for the
> Actual output, most commonly used XML-Parsers don't
> Parse it but tell the tag is empty so that you can't
> Reproduce it exactly.
> 
> So please implement a switch to make XML-Output
> Encode all (important) whitespaces.

Doesn't have to be a switch for me. This is very easy
to implement by changing the filtering in src/xmlgen.cpp: codify(),
but I'm not sure if I then also have to define the meaning
of &nbsp; somewhere since &xxx; are merely macro's as far as I understand
XML otherwise not all parser will accept it. An alternative would be
a <space/> tag.

> 2)
> please add a 'static' attribute to the xml
> output so that one can see if a compound
> is static just by this attribute (just as 'const', 'volatile',...)
> without the need to take the surrounding sectiondef
> into account.

Good point. Will do that.

> 3) 
> 
> And last but not least:
> 
> Don't change the XML output to much in future versions
> Since it would render my parser useless ;-)

I don't intend to make major changes, but until doxygen 2.0 I'm not
making promises.

Please keep reporting problems as you see then and feel free to help me 
with coding ;-)

Regards,
  Dimitri

[Doxygen-develop] 'Static' attribute in xml output

[Thorsten S. <ts...@mm...>]

Hi there,

=20

please add a 'static' attribute to the xml

output so that one can see if a compound

is static just by this attribute (just as 'const', 'volatile',...)

without the need to take the surrounding sectiondef

into account.

=20

Thanx and best regards

Thorsten

[Doxygen-develop] Keywords in output

[Fabian C. <Cen...@in...>]

Hi

We try to build a help system based on HTML files. So the readymade
HTML output of Doxygen is very suitable. The problem is that we don't want
to use the generated index file/javascript. The help system scans all HTML
files and builds its own index. For this it would be very convenient to have
the index keywords already in the page where they belong to. Is there a
possibility to achieve this? The words are already known as they are used
for the .hhk and .js so it shouldn't be that difficult to output them in the
HTML file. I don't know about the other output formats, if they can use this.

If this is not easy/possible, what about the XML output? Is there something
like keywords/index that could be used for this?

And last but not least :) What about a possibility to define own keywords
in the source code that get scanned and outputted in this list? Something
like

/keyword(s) bla

or

<keywords>bla morebla</keywords>

Thanks

Fabi

[Doxygen-develop] XML Output and whitespace

[Thorsten S. <ts...@mm...>]

Hi there,

=20

I would like to suggest a feature regarding XML output:

=20

Would it be possible to encode 'important' whitepspace

(that is: those in <codeline>-tags) with &nbsp; ?

=20

Because in the currently generated code there is a lot

Of stuff like:

=20

<highlight class=3D"normal"> </highlight>

=20

Although this single space is important for the

Actual output, most commonly used XML-Parsers don't

Parse it but tell the tag is empty so that you can't

Reproduce it exactly.

=20

So please implement a switch to make XML-Output

Encode all (important) whitespaces.

=20

Regards

Thorsten

[Doxygen-develop] Out of the office.

[<Joh...@sy...>]

I will be out of the office starting  28/11/2002 and will not return until
02/12/2002.







**********************************************************************
Symbian Ltd is a company registered in England and Wales with registered number 01796587 and registered office at 19 Harcourt Street, London, W1H 4HF, UK.
This message is intended only for use by the named addressee and may contain privileged and/or confidential information. If you are not the named addressee you should not disseminate, copy or take any action in reliance on it. If you have received this message in error please notify pos...@sy... and delete the message and any attachments accompanying it immediately. Symbian does not accept liability for any corruption, interception, amendment, tampering or viruses occuring to this message in transit or for any message sent by its employees which is not in compliance with Symbian corporate policy.
**********************************************************************

[Doxygen-develop] Italian translation update ...

[Alessandro F. <ale...@da...>]

In the attached archive you'll find the updated italian translation of
Doxygen messages.
I hope to be in time for 1.3 release.

--
Alessandro Falappa
web: http://www.falappa.net/alessandro

[Doxygen-develop] XML and Graphs

[Thorsten S. <ts...@mm...>]

Hi there,

=20

just wanted to suggest 2 features:

=20

1)       Implement 'one-xml-file' only functionality so that=20
one could write parsers/xsl=20

2)       Make the graphs UML conform. Don't use colors and
the appropriate arrowheads and so on

=20

Best Regrads

Thorsten

=20

[Doxygen-develop] doxygen: undefined behaviour

[Ruben G. <ru...@ug...>]

In the make process, a few lines like this are printed:
    ../qtools/qdict.h:85: warning: deleting `void*' is undefined

I think "free" should be used there, or a delete of the casting of the 
pointer to the appropiate type.
Can somebody have a look where the original pointer came from?

[Doxygen-develop] Re: Doxygen-develop digest, Vol 1 #160 - 1 msg

[Paco H. <pa...@pa...>]

Quoting Simon Goodwin <sim...@at...>:

> On the Doxygen-develop mailing list you wrote:
>> On an unrelated note, if any of you ever try to create PDF from doxygen on
>> the scale I'm talking about, you'll hit some serious limits in how many
>> hyperlinks pdflatex can grok in a single document.  I have solved this
>> problem using the teTeX distribution and a customized texmf.cnf.
>> I'd be happy to share this with anyone who has the same problems.
> 
> Yes please! 
> [... snip ...]
> 
> There is a file `texmf.cnf` in c:/cygwin/usr/share/texmf/web2c which sets a
> load of limits on TeX memory gobbling. These cause problems when
> handling big files,  and the partial cure is to increase values in there.
> Not being Knuth (and being lazy) I just bopped an extra zero onto the
> end of the following lines:
> 
> pool_size.context = 7500000
> pool_size = 1250000
> [...]
> % Maximum number of strings.
> max_strings.context = 550000
> max_strings = 150000
> 
> This worked for larger files but not for the documentation for our
> whole game engine - so if there's another limit you know how to fix
> that's be very useful.

As far as I can tell, the really really important one is this:
dest_names_size=300000

The reason I say it is important is that I got a little more slap-happy in
adjusting the memory limits in that file.  I changed all  kinds of things and
still ran out of "destinations."  As far as I can tell, the dest_names_size
starts of at 20000, and it has an impact on how many hyperlinks you can have in
a file.  As you know, doxygen generates an amazing amount of hyperlinks.  It
doesn't have a default value in that file, so it's kind of a stealth option.  It
doesn't appear on many searchable web pages, either, so finding it via google
was quite difficult.  Here are all the other values I changed from the defaults.

main_memory.context = 4000000
main_memory = 2000000 
extra_mem_top = 21000    
extra_mem_bot = 21000    
obj_tab_size = 256000
hash_extra.context = 45000
hash_extra = 40000
pool_size.context = 7500000
pool_size = 2250000             
string_vacancies.context = 65000
string_vacancies = 65000
max_strings.context = 85000
max_strings = 85000
pool_free.context = 100000
pool_free = 21000
dest_names_size=300000 
trie_size = 30000
save_size.context = 10000
save_size = 20000       
stack_size.context = 3000
stack_size = 3000       
ocp_buf_size = 40000    
ocp_stack_size = 40000  


> P.S. I've built 1.3 candidate 1, but it bombs with a zero dereference
> on my (Win2K) machine when I give it a substantial file to generate.

I'm building under Win98 (ugh).  I have cygwin 1.3.13-2 (according to 'cygcheck
-s').

> Have you seen that?

No.  I did:
./configure --english-only --prefix /usr/local -platform win32-g++

And it built out of the box and ran just fine the first time.

Paco
-- 
pa...@pa...
http://paco.to/