doxygen-users — List for users of doxygen

Re: [Doxygen-users] Function parameter type

[Dean P. <po...@ds...>]

>Hello,
>
>For the function parameters, we can use \param and \retval.
>Is it possible to have a flag to say that the parameter is "in-out"
>The name inout is perhaps not the best one, but I have no idea what to u=
se.
>

We tend to do this in our function documentation (Note: Javadoc syntax):

/**
 * A brief description of the function
 *
 * @param a [in] An input parameter
 * @param b [out] An out parameter
 * @param c [inout] An in-out parameter
 *
 * @return ...
 */

Which formats nicely in doxygen.

Cheers.



-- =

Dean Povey,         | e-m: po...@ds... | JCSI: Java Crypto Toolkit =

Research Scientist  | ph:  +61 7 3864 5120   | uPKI: C PKI toolkit for em=
bedded
Security Unit, DSTC | fax: +61 7 3864 1282   |       systems
Brisbane, Australia | www: security.dstc.com | =

[Doxygen-users] Function parameter type

[Stephane R. <rou...@ca...>]

Hello,

For the function parameters, we can use \param and \retval.
Is it possible to have a flag to say that the parameter is "in-out"
The name inout is perhaps not the best one, but I have no idea what to use.

If you think it is interessant to have, can somebody integrate it ( I don't
have the source and never read them )?

BTW, I'm using OpenCASCADE ( www.opencascade.org ), and they have a CDL
language to generate the C++ header and implementation files.
They are specifying interessant attribute for the parameters (in , out,
inout, mutable, immutable , etc ... ).
What a pity that these informations are not used in C++ :-((

Thanks,

Stephane

ex :

/** \fn void SetValue(const int anInteger);
 * \param anInteger const int : the value to set
*/
void SetValue(const int anInteger)
{    myValue = anInteger; };

/** \fn void Value(int& anInteger) const
* \retval anInteger int& : the value
*/
void Value(int& anInteger) const
{anInteger =  myValue;};

/** \fn void Multiply(int& anInteger,const int aMultiplicator)
 * \inout anInteger the value to muliply and returns the multiplied value
* \param aMultiplicator
*/
void Multiply(int& anInteger,const int aMultiplicator)
{
    anInteger = anInteger*aMultiplicator;
};
=================================================
Stephane ROUTELOUS
C A D / C A M  -  C o n c e p t   G m b H
Albert-Einstein-Straße 14 . D-12489 Berlin
rou...@ca...    http://www.cad-cam-concept.de
phone  : +49 (0)30 63.92.65.38      fax : +49 (0)30 63.92.63.19
=================================================
"It is ridiculous to claim that video games influence children.
For instance, if PacMan affected kids born in the eighties, we
should by now have a bunch of teenagers who run around in darkened
rooms eating pills while listening to monotonous electronic music."

Re: [Doxygen-users] Japanese SJIS

[Tom E. <tr...@ba...>]

Haruyuki Ohtani writes:
> After I downloaded the source code of doxygen, I understand the
> current doxygen embeds EUC Kanji code in the source and does not
> support generationg Japanese RTF which is typically Shift JIS.  I
> guess the most easiet way to support Shift JIS is to embed Shift JIS
> Kanji code instead of EUC in the source, to return the literal string
> "Shift_JIS" instead of "euc-jp" in QCString::idLanguageCharset(), and
> then to recompile all source code. But Maybe it is not good idea
> because of limiting extensibility.

Changing everything to use ShiftJIS (or, more appropriately on
Windows, CP932) is just changing the nature of the problem. Perhaps a
better solution would be to use iconv() to transcode the EUC-JP text
to CP932, at least when certain options are set. Doing this would
eliminate the duplication of messages that differ only in their
encoding.

I haven't looked at the Doxygen source in a long while: I don't know
if changing the Japanese encoding would effect the lexer at all... I
would expect not.

In any event, this could be used too to support different Russian
encodings.

    -tree

-- 
Tom Emerson                                          Basis Technology Corp.
Sr. Sinostringologist                              http://www.basistech.com
  "Beware the lollipop of mediocrity: lick it once and you suck forever"

Re: [Doxygen-users] Japanese SJIS

[Haruyuki O. <oh...@is...>]

Hello all.

After I downloaded the source code of doxygen, I understand the
current doxygen embeds EUC Kanji code in the source and does not
support generationg Japanese RTF which is typically Shift JIS.  I
guess the most easiet way to support Shift JIS is to embed Shift JIS
Kanji code instead of EUC in the source, to return the literal string
"Shift_JIS" instead of "euc-jp" in QCString::idLanguageCharset(), and
then to recompile all source code. But Maybe it is not good idea
because of limiting extensibility.

> Hello. I'm new to doxygen.
> 
> I'm using doxygen on Windows. Unfortunatelly Japanese SJIS code set is
> mainly used on Windows because of historical reason. EUC code set
> comes from Unix culture. Does anyone successfully use SJIS code with
> doxygen ? I have some problem when inlining sources and producing rtf.
> 
> Regards,
> ---
> Haru (Haruyuki Ohtani)

Regards,
---
Haru (Haruyuki Ohtani)

Re: [Doxygen-users] Segfault

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

On Tue, Jul 03, 2001 at 08:58:03AM +0200, John Sturton wrote:
> > 
> > doxygen -d debug_option [ -d debug_option ... ]
> > 
> > can be used for this, where "debug_option" is one of
> > the things mentioned in the DebugMask enum in src/debug.h
> > 
> 
> I also spent a couple of hours tracking down this bug, and in the
> process discovered another one: - if you do run doxygen with "-d
> FindMembers" then the Debug::print() function itself crashes as it tries
> to interpret NULL as a string :-( (this is Doxygen compiled with Sun's
> Solaris C++ compiler). 

Debug::print() just uses vfprintf of the libc and its implementation 
on my GNU/Linux box can handle NULL strings, so it is good enough for me ;-)

Regards,
  Dimitri

Re: [Doxygen-users] robodoc/ autodoc syntax

[Bogdan I. <bo...@ne...>]

For me, the fact that doxygen supports the javadoc syntax is an important
advantage. I want to have the same syntax for commenting both Java and C++
code. While I've seen C++ code documented in countless ways, I've yet to
see Java code documented using something other than javadoc. Also, there
are several tools and development environments that let you edit javadoc
comments in a friendly editor, with HTML syntax highlighting, automatic
generation of the leading asterisks and of most tags, and so on. This can
make writing comments much easier.

I agree that robodoc's syntax makes the source comments more readable. I'm
not sure how flexible it is about additional formatting of the comments:
for example, can it generate formatted lists or tables, other than the
default ones? I haven't really used it, so this is only a comment, not a
critique. You'll have to decide the tradeoffs by yourself.

Cheers,
Bogdan



----- Original Message -----
From: "Stephan Stapel" <st...@sc...>
To: <dox...@li...>
Sent: Wednesday, July 04, 2001 7:39 AM
Subject: [Doxygen-users] robodoc/ autodoc syntax


> Hi everyone!
>
> As I'm still examining different documentation tools, I discovered a  --
at
> least for me -- huge advantage of Robodoc/ Autodoc: The syntax in which
you
> format your source code documentation is far better human-readable than
the
> doxygen/ javadoc syntax.
> So I would like to know how difficult it would be to introduce this
syntax
> to doxygen as well. Has anyone any comments on advantages/ disadvantages
of
> Robodoc/ Autodoc in comparison to Doxygen?
>
> Thanks,
>
> Stephan
>
>
>
> _______________________________________________
> Doxygen-users mailing list
> Dox...@li...
> http://lists.sourceforge.net/lists/listinfo/doxygen-users

[Doxygen-users] robodoc/ autodoc syntax

[Stephan S. <st...@sc...>]

Hi everyone!

As I'm still examining different documentation tools, I discovered a  -- at
least for me -- huge advantage of Robodoc/ Autodoc: The syntax in which you
format your source code documentation is far better human-readable than the
doxygen/ javadoc syntax.
So I would like to know how difficult it would be to introduce this syntax
to doxygen as well. Has anyone any comments on advantages/ disadvantages of
Robodoc/ Autodoc in comparison to Doxygen?

Thanks,

Stephan

[Doxygen-users] Japanese SJIS

[Haruyuki O. <oh...@is...>]

Hello. I'm new to doxygen.

I'm using doxygen on Windows. Unfortunatelly Japanese SJIS code set is
mainly used on Windows because of historical reason. EUC code set
comes from Unix culture. Does anyone successfully use SJIS code with
doxygen ? I have some problem when inlining sources and producing rtf.

Regards,
---
Haru (Haruyuki Ohtani)

Re: [Doxygen-users] Fortran code

[John S. <joh...@se...>]

Stephan Stapel wrote:
> 
> Hi everyone!
> 
> Has anyone ever tried to use doxygen in combination with Fortran? Is it
> possible?
> 

Yes, we use this a lot. Our solution is based around a few awk scripts,
defined with the INPUT_FILTER option, which basically massages the
FORTRAN subroutine/function declarations to look like C++ code. If I
remember correctly the GNU FORTRAN compiler has a similar utility to
create C-style header files from FORTRAN source code - you could maybe
take this as a starting point and modify it to extract the Doxygen
comments as well.

Good luck!

Cheers,

John

Re: [Doxygen-users] Segfault

[John S. <joh...@se...>]

> >
> > As workaround I put an if clause around the cl=fd->getUsedClasses() in
> > function findClassDefinition in doxygen.cpp:
> >
> > if ( fd ) {
> >     cl = fd->getUsedClasses();
> > }
> >
> > After that it works for me, and the resulting Documentation seems to be ok.
> >
> > Any hint where I should look in the docu what could be missing?
> >
> 
> Your fix is correct. This is a bug.
> 
> ...
>
> > ---
> >
> > PPS: How can I activate the debugging output of Doxygen? I converted all
> > Debug::print commands that interested me to printf to find the error ...
> 
> doxygen -d debug_option [ -d debug_option ... ]
> 
> can be used for this, where "debug_option" is one of
> the things mentioned in the DebugMask enum in src/debug.h
> 

I also spent a couple of hours tracking down this bug, and in the
process discovered another one: - if you do run doxygen with "-d
FindMembers" then the Debug::print() function itself crashes as it tries
to interpret NULL as a string :-( (this is Doxygen compiled with Sun's
Solaris C++ compiler). 

Cheers,

John

[Doxygen-users] Fortran code

[Stephan S. <ste...@we...>]

Hi everyone!

Has anyone ever tried to use doxygen in combination with Fortran? Is it
possible?

Thanks in advance,

Stephan

[Doxygen-users] stuck in .ext file

[A R <och...@ya...>]

Resend the message, had some problems with the
subscription to the mailing list.

> 
> Same message, with a subject now to capture your
> attention...
> 
> Hi.
> 
> I have some troubles using doxygen. Please let me
> know
> what could be done.
> 
> The problem appears when a function XXX() with the
> implementation in a source file has also its
> prototype
> declared in a header file (.ext in my case, but
> probably the same thing would be with .h). 
> 
> I'm browsing the source of a .c file and I click on
> the link for XXX() - I go to the documentation of
> the
> .ext header file where I see only the prototype (and
> the 'referenced by' thing) and no link from there -
> stuck in the .ext, I have to go back again to the
> old
> plain editor and grep. Or I can figure out what's
> going on and look myself for the .c file where the
> XXX() implementation is, but this requires a lot of
> clicking here and there and defeats the purpose of
> doxygen, to ease code browsing.
> 
> So, is there a way to tell doxygen to send me to the
> documentation of the .c file where XXX() is
> implemented?
> 
> I'm using doxygen 1.2.0 and 1.2.8.
> 
> Thanks,
> Ochiosul


__________________________________________________
Do You Yahoo!?
Get personalized email addresses from Yahoo! Mail
http://personal.mail.yahoo.com/

[Doxygen-users] stuck in .ext file

[A R <och...@ya...>]

Same message, with a subject now to capture your
attention...

Hi.

I have some troubles using doxygen. Please let me know
what could be done.

The problem appears when a function XXX() with the
implementation in a source file has also its prototype
declared in a header file (.ext in my case, but
probably the same thing would be with .h). 

I'm browsing the source of a .c file and I click on
the link for XXX() - I go to the documentation of the
.ext header file where I see only the prototype (and
the 'referenced by' thing) and no link from there -
stuck in the .ext, I have to go back again to the old
plain editor and grep. Or I can figure out what's
going on and look myself for the .c file where the
XXX() implementation is, but this requires a lot of
clicking here and there and defeats the purpose of
doxygen, to ease code browsing.

So, is there a way to tell doxygen to send me to the
documentation of the .c file where XXX() is
implemented?

I'm using doxygen 1.2.0 and 1.2.8.

Thanks,
Ochiosul

__________________________________________________
Do You Yahoo!?
Get personalized email addresses from Yahoo! Mail
http://personal.mail.yahoo.com/

[Doxygen-users] (no subject)

[A R <och...@ya...>]

Hi.

I have some troubles using doxygen. Please let me know
what could be done.

The problem appears when a function XXX() with the
implementation in a source file has also its prototype
declared in a header file (.ext in my case, but
probably the same thing would be with .h). 

I'm browsing the source of a .c file and I click on
the link for XXX() - I go to the documentation of the
.ext header file where I see only the prototype (and
the 'referenced by' thing) and no link from there -
stuck in the .ext, I have to go back again to the old
plain editor and grep. Or I can figure out what's
going on and look myself for the .c file where the
XXX() implementation is, but this requires a lot of
clicking here and there and defeats the purpose of
doxygen, to ease code browsing.

So, is there a way to tell doxygen to send me to the
documentation of the .c file where XXX() is
implemented?

I'm using doxygen 1.2.0 and 1.2.8.

Thanks,
Ochiosul

__________________________________________________
Do You Yahoo!?
Get personalized email addresses from Yahoo! Mail
http://personal.mail.yahoo.com/

[Doxygen-users] Re: Doxygen-users -- confirmation of subscription -- request 897715

[A R <och...@ya...>]

--- dox...@li... wrote:
> Doxygen-users -- confirmation of subscription --
> request 897715
> 
> We have received a request from 64.229.133.124 for
> subscription of
> your email address, <och...@ya...>, to the
> dox...@li... mailing list. 
> To confirm the
> request, please send a message to
> dox...@li..., and
> either:
> 
> - maintain the subject line as is (the reply's
> additional "Re:" is
> ok),
> 
> - or include the following line - and only the
> following line - in the
> message body: 
> 
> confirm 897715
> 
> (Simply sending a 'reply' to this message should
> work from most email
> interfaces, since that usually leaves the subject
> line in the right
> form.)
> 
> If you do not wish to subscribe to this list, please
> simply disregard
> this message.  Send questions to
> dox...@li....


__________________________________________________
Do You Yahoo!?
Get personalized email addresses from Yahoo! Mail
http://personal.mail.yahoo.com/

Re: [Doxygen-users] Segfault

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

On Sat, Jun 30, 2001 at 08:46:13PM +0200, Berno Langer wrote:
> Hello!
>  
> I have a project that uses Qt. Therefore I create a Doxygen-Documentation of 
> the Qt-Classes and let Doxygen produce a "tagfile.qt".
>  
> If I create the docu of my project and use the "tagfile.qt", I get a segfault.
>  
> If I do the same, but don't use the "tagfile.qt", all runs as expected.
>  
> Ok, I admit that the Qt-Library isn't the "small example" that Dimitri always 
> requests to find the error :-)
>  
> The last words of Doxygen before the segfault are:
> --8<---------------------------------------------------------------------
> Computing class relations...
> Searching for enumerations...
> Searching for member function documentation...
>  
> [very much snipped]
>  
> findMember(root=0x8ffc840,funcDecl=`QFontDatabase::createDatabase()',related=`(null)',overload=0,isFunc=1 
> mGrpId=-1 tArgList=(nil)="(null)" mtArgList=(nil)="(null)" scopeSpec=(null) 
> memberSpec=(null) memSpec=0
> findMember() Parse results:
>   namespaceName=`(null)'
>   className=`QFontDatabasePrivate::QFontDatabase`
>   classTempList=`(null)'
>   funcType=`(null)'
>   funcName=`createDatabase'
>   funcArgs=`()'
>   funcTempList=`(null)'
>   funcDecl=`QFontDatabasePrivate::QFontDatabase::createDatabase'
>   related=`(null)'
>   exceptions=`(null)'
>   isRelated=0
>   isFriend=0
>   isFunc=1
>  
> 1. funcName=`createDatabase'
> 2. member name exists
> 3. member definition found scopeName=`QFontDatabasePrivate::QFontDatabase' 
> Member createDatabase (member scopeName=QFontDatabase) (this 
> scopeName=QFontDatabasePrivate::QFontDatabase) classTempList=(null)
> Speicherzugriffsfehler
> --8<---------------------------------------------------------------------
>  
> As workaround I put an if clause around the cl=fd->getUsedClasses() in 
> function findClassDefinition in doxygen.cpp:
>  
> if ( fd ) {
>     cl = fd->getUsedClasses();
> }
>  
> After that it works for me, and the resulting Documentation seems to be ok.
>  
> Any hint where I should look in the docu what could be missing?
>  

Your fix is correct. This is a bug.

> ---
> Gruß
>  
> Berno
>  
> PS: I use Doxygen 1.2.8-20010617 and the qt-copy of KDE
> 
> PPS: How can I activate the debugging output of Doxygen? I converted all 
> Debug::print commands that interested me to printf to find the error ...

doxygen -d debug_option [ -d debug_option ... ]

can be used for this, where "debug_option" is one of 
the things mentioned in the DebugMask enum in src/debug.h

I only use these for internal debugging, so it is likely that you do not
understand the output ;-)

Regards,
  Dimitri

[Doxygen-users] Segfault

[Berno L. <ber...@ep...>]

Hello!
 
I have a project that uses Qt. Therefore I create a Doxygen-Documentation of 
the Qt-Classes and let Doxygen produce a "tagfile.qt".
 
If I create the docu of my project and use the "tagfile.qt", I get a segfault.
 
If I do the same, but don't use the "tagfile.qt", all runs as expected.
 
Ok, I admit that the Qt-Library isn't the "small example" that Dimitri always 
requests to find the error :-)
 
The last words of Doxygen before the segfault are:
--8<---------------------------------------------------------------------
Computing class relations...
Searching for enumerations...
Searching for member function documentation...
 
[very much snipped]
 
findMember(root=0x8ffc840,funcDecl=`QFontDatabase::createDatabase()',related=`(null)',overload=0,isFunc=1 
mGrpId=-1 tArgList=(nil)="(null)" mtArgList=(nil)="(null)" scopeSpec=(null) 
memberSpec=(null) memSpec=0
findMember() Parse results:
  namespaceName=`(null)'
  className=`QFontDatabasePrivate::QFontDatabase`
  classTempList=`(null)'
  funcType=`(null)'
  funcName=`createDatabase'
  funcArgs=`()'
  funcTempList=`(null)'
  funcDecl=`QFontDatabasePrivate::QFontDatabase::createDatabase'
  related=`(null)'
  exceptions=`(null)'
  isRelated=0
  isFriend=0
  isFunc=1
 
1. funcName=`createDatabase'
2. member name exists
3. member definition found scopeName=`QFontDatabasePrivate::QFontDatabase' 
Member createDatabase (member scopeName=QFontDatabase) (this 
scopeName=QFontDatabasePrivate::QFontDatabase) classTempList=(null)
Speicherzugriffsfehler
--8<---------------------------------------------------------------------
 
As workaround I put an if clause around the cl=fd->getUsedClasses() in 
function findClassDefinition in doxygen.cpp:
 
if ( fd ) {
    cl = fd->getUsedClasses();
}
 
After that it works for me, and the resulting Documentation seems to be ok.
 
Any hint where I should look in the docu what could be missing?
 
---
Gruß
 
Berno
 
PS: I use Doxygen 1.2.8-20010617 and the qt-copy of KDE

PPS: How can I activate the debugging output of Doxygen? I converted all 
Debug::print commands that interested me to printf to find the error ...

[Doxygen-users] Group sorting

[James R. <ja...@sh...>]

Doxygen users,

	I noticed the group documentation is sometimes generated in reverse
order of declaration.  For the project I am documenting, the
documentation does not make sense if you read the last group first.  I
once saw something in the list archive about a patch for this issue, but
I am unable to find it now.  Does anyone have a patch or any news on
this?

Thanks,
James Roth <ja...@sh...>
Shugyo Design Technologies

Re: [Doxygen-users] New project, and, Can Doxygen understand functions at namespace level?

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

On Wed, Jun 27, 2001 at 06:04:27PM -0400, Phil Edwards wrote:
> Anyhow... I'm trying to comment the code for use with EXTRACT_ALL=NO,
> and am having problems with a template function at namespace level, using
> doxygen 1.2.6.  The code looks like this:
> 
>   namespace std {
> 
>     /**
>      *  ...some @-style tags in here...
>      *
>      *  The standard requires that the objects be passed by reference-to-const,
>      *  but LWG issue #181 says they should be passed by const value.
>      */
>     template <class _T1, class _T2>
>     #ifdef _GLIBCPP_RESOLVE_LIB_DEFECTS
>     //181.  make_pair() unintended behavior
>     inline pair<_T1, _T2> make_pair(_T1 __x, _T2 __y)
>     #else
>     inline pair<_T1, _T2> make_pair(const _T1& __x, const _T2& __y)
>     #endif
>     {
>       return pair<_T1, _T2>(__x, __y);
>     }
> 
>   }  // namespace std
> 
> Doxygen completely skips this function, no matter what I do.  However,
> if I move the code into the global namespace rather than namespace std,
> then make_pair correctly appears in the output.
> 
> What do I need to do for doxygen to see this function inside std:: correctly?

You need to document the namespace somewhere (a block with @namespace std
will do for instance), or put the function in a self defined group 
(see @defgroup, @ingroup, and friends).

Regards.
  Dimitri

Re: [Doxygen-users] HOW DO I iterate thru Doxygen::GroupList and GroupDef to get grou ped classes and methods?

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

On Fri, Jun 29, 2001 at 05:01:19PM +0200, mldb wrote:
> RC> From: Rodolfo Cazabon <rod...@au...>
> 
> RC> Hi everyone...
> [snip]
> RC> Second, I cheated, and declared these as PUBLIC...but when I do the
> RC> following : 
>  
> RC>         ClassListIterator cli(p_groupDef->classList);
> RC>         ClassDef *cd;
> RC>         for (cli.toFirst();(cd=cli.current());++cli)
> RC>         {
> RC>                 printf("%S \n", cd->name() );
> RC>         }
>  
> RC> I get a crash because that pointer is not valid!
> 
> I suspect, you missed the fact, that printf has a vararg signature
> (printf(const char *fmt,...)). The "const char *" conversion of
> QCString (this is what cd->name() is) is not used, since the compiler
> does not know *to what* you want it to be converted: '...' could be
> anything.
> 
>          printf("%s \n", (const char*)cd->name());

Or alternatively,

    printf("%s \n", cd->name().data());

which I prefer myself.

Regards,
  Dimitri

RE: [Doxygen-users] RE: Is it possible to customise the page layo ut for class members

[Wagner, V. <VW...@se...>]

while agreeing that the documentation needs to be useful, I'll point out
that what Doxygen currently outputs is LEGAL syntax for the compiler, your
proposal isn't.

-----Original Message-----
From: Joh...@ub... [mailto:Joh...@ub...]
Sent: Friday, 2001 June 29 11:27
To: dox...@li...
Subject: [Doxygen-users] RE: Is it possible to customise the page layout
for class members



Before I start there may be somnething about this on the sourceforge 
website but the search facility is broken and I get this message ......

-------------
htsearch detected an error. Please report this to the returned matches
webmaster of this site. The error message is:
unable to read configuration file '/bigassraid/htdig//conf/11668.conf'
-------------



Is it possible to customise the page layout for class members so that 
instead of a table like 

__________________________________
| RETURN VALUE | METHOD(ARGS)     |
|_________________________________|
| RETURN VALUE | METHOD(ARGS)     |
|_________________________________|


we have the alternative format ...

____________________________________
| METHOD  | (ARGS)   | RETURNVALUE |
|__________________________________|
| METHOD  | (ARGS)   | RETURNVALUE |
|__________________________________|

As this is a bit easier on the eye when looking to a particular method.

In my project out type names are pretty long and in the current table 
format the return value column takes up most of the width of the pane, 
leaving the method name and its ares all scrunched up.

I've tried the alternative format shown above in an html editor and it 
is much more readable.

I've attached examples of the orig and alternative formats.





This transmission may contain information that is privileged, confidential
and exempt from disclosure under applicable law.
If you are not the intended recipient, you are hereby notified that any
disclosure, copying, distribution, or use of the information contained
herein (including any reliance thereon) is STRICTLY PROHIBITED.
If you received this transmission in error, please immediately contact the
sender and destroy the material in its entirety, whether in electronic or
hard copy format.
Thank you

[Doxygen-users] RE: Is it possible to customise the page layout for class members

[<Joh...@ub...>]

Before I start there may be somnething about this on the sourceforge 
website but the search facility is broken and I get this message ......

-------------
htsearch detected an error. Please report this to the returned matches
webmaster of this site. The error message is:
unable to read configuration file '/bigassraid/htdig//conf/11668.conf'
-------------



Is it possible to customise the page layout for class members so that 
instead of a table like 

__________________________________
| RETURN VALUE | METHOD(ARGS)     |
|_________________________________|
| RETURN VALUE | METHOD(ARGS)     |
|_________________________________|


we have the alternative format ...

____________________________________
| METHOD  | (ARGS)   | RETURNVALUE |
|__________________________________|
| METHOD  | (ARGS)   | RETURNVALUE |
|__________________________________|

As this is a bit easier on the eye when looking to a particular method.

In my project out type names are pretty long and in the current table 
format the return value column takes up most of the width of the pane, 
leaving the method name and its ares all scrunched up.

I've tried the alternative format shown above in an html editor and it 
is much more readable.

I've attached examples of the orig and alternative formats.

Re: [Doxygen-users] HOW DO I iterate thru Doxygen::GroupList and GroupDef to get grou ped classes and methods?

[mldb <ml...@gm...>]

RC> From: Rodolfo Cazabon <rod...@au...>

RC> Hi everyone...
[snip]
RC> Second, I cheated, and declared these as PUBLIC...but when I do the
RC> following : 
 
RC>         ClassListIterator cli(p_groupDef->classList);
RC>         ClassDef *cd;
RC>         for (cli.toFirst();(cd=cli.current());++cli)
RC>         {
RC>                 printf("%S \n", cd->name() );
RC>         }
 
RC> I get a crash because that pointer is not valid!

I suspect, you missed the fact, that printf has a vararg signature
(printf(const char *fmt,...)). The "const char *" conversion of
QCString (this is what cd->name() is) is not used, since the compiler
does not know *to what* you want it to be converted: '...' could be
anything.

         printf("%s \n", (const char*)cd->name());
         
should work


-- 
Marco

[Doxygen-users] unsubscribe

[<li...@ca...>]

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D
|| Where there is a will there is a way
||--------------------------------------------
|| Hongtao LIAO                               =20
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D

[Doxygen-users] New project, and, Can Doxygen understand functions at namespace level?

[Phil E. <ped...@di...>]

First, I'd like to mention another project using Doxygen:  libstdc++-v3,
the GNU Standard C++ Library now shipping with GCC 3.0.  The documentation
is available via http://gcc.gnu.org/libstdc++/ .

Much of the code does not yet contain special comments; the version for
3.0 available above is the uncommented EXTRACT_ALL=YES version which lets
you see all the messy details.  :-)


Anyhow... I'm trying to comment the code for use with EXTRACT_ALL=NO,
and am having problems with a template function at namespace level, using
doxygen 1.2.6.  The code looks like this:

  namespace std {

    /**
     *  ...some @-style tags in here...
     *
     *  The standard requires that the objects be passed by reference-to-const,
     *  but LWG issue #181 says they should be passed by const value.
     */
    template <class _T1, class _T2>
    #ifdef _GLIBCPP_RESOLVE_LIB_DEFECTS
    //181.  make_pair() unintended behavior
    inline pair<_T1, _T2> make_pair(_T1 __x, _T2 __y)
    #else
    inline pair<_T1, _T2> make_pair(const _T1& __x, const _T2& __y)
    #endif
    {
      return pair<_T1, _T2>(__x, __y);
    }

  }  // namespace std

Doxygen completely skips this function, no matter what I do.  However,
if I move the code into the global namespace rather than namespace std,
then make_pair correctly appears in the output.

What do I need to do for doxygen to see this function inside std:: correctly?


Many thanks,
Phil

-- 
Would I had phrases that are not known, utterances that are strange, in
new language that has not been used, free from repetition, not an utterance
which has grown stale, which men of old have spoken.
                                     - anonymous Egyptian scribe, c.1700 BC