doxygen-users — List for users of doxygen

Re: [Doxygen-users] Forward declaration of enums

[<br...@sh...>]

Quoting Stephane Routelous <ste...@sy...>:

> Hi,
> 
> here is my problem :
> 
> I'm using forward declaration of enums and classes in my classes.
> With the enum, I've got a warning :
> E:/tmp/dox/class.h:8 Warning: Member MyEnum of file class.h is not
> documented.
> But I have no warnings for forward declarations of classes
> Is that normal ?
> Is that a bug ?

Yes... In your code, and possibly your compiler. :-)

In general, you don't want to forward-declare enums. I'm not sure off the top 
of my head whether the syntax is even valid; even if it is, you shouldn't use 
it. C++ allows the size of an enumerant to depend on the range of values the 
enumeration must be able to hold. To derive this information, the compiler must 
be able to see the definition of the enumeration in each translation unit where 
the enumeration is used.

Braden

Re: [Doxygen-users] inheritance problem

[Stephane R. <ste...@sy...>]

Hi,

thanks for your answer.
But I cannot make a doxygen project with the third party product.
I try once to build the doc for the Third party library, it takes me =
more than one night !!!
And the classes are not documented with doxygen tags.

Stephane
  ----- Original Message -----=20
  From: Glenn Maxey=20
  To: Stephane Routelous ; dox...@li...=20
  Sent: Thursday, January 17, 2002 5:28 PM
  Subject: RE: [Doxygen-users] inheritance problem


  I could be way off base, so take this advice with a grain of salt.

  I would create multiple doxygen projects, at least one for the TP and =
one for your stuff. Configure the projects so that they create tag =
files. (Assuming the inheritance is just one way), create a reference to =
the tag file from TP in the doxygen project for your stuff.

  It is best to use multiple directories for multiple projects.=20

  Coming soon to a download near you are some Perl tools that I =
developed that will allow you to wrap the output of multiple doxygen =
projects (in addition to FM/Mif2Go or FM/WWP output) into a single HTML =
system.

  Glenn
    -----Original Message-----
    From: Stephane Routelous [mailto:ste...@sy...]
    Sent: Thursday, January 17, 2002 3:16 PM
    To: dox...@li...
    Subject: [Doxygen-users] inheritance problem


    Hi,

    I'm using a third party library in my code.
    My code is in E:\MyProject\inc and E:\MyProject\src
    The third party includes are in E:\ThirdParty\inc

    In my config file,I have :
    INPUT                  =3D E:\MyProject\src E:\MyProject\inc
    INCLUDE_PATH           =3D E:\MyProject\inc E:\ThirdParty\inc=20

    I have some classes which inherits from some third party classes

    I have a Third party class (TP)

    my code :

    class A :public TP
    class B: public A
    class C : public TP
    class D : public C

    I only have :

    A
    +B
    C
    +D

    I would like to have=20
    TP
    +A
        +B
    +C
        +D

    What am I doing wrong ?

    Thanks,

    Stephane

RE: [Doxygen-users] inheritance problem

[Glenn M. <gle...@vo...>]

I could be way off base, so take this advice with a grain of salt.
=20
I would create multiple doxygen projects, at least one for the TP and
one for your stuff. Configure the projects so that they create tag
files. (Assuming the inheritance is just one way), create a reference to
the tag file from TP in the doxygen project for your stuff.
=20
It is best to use multiple directories for multiple projects.=20
=20
Coming soon to a download near you are some Perl tools that I developed
that will allow you to wrap the output of multiple doxygen projects (in
addition to FM/Mif2Go or FM/WWP output) into a single HTML system.
=20
Glenn

-----Original Message-----
From: Stephane Routelous [mailto:ste...@sy...]
Sent: Thursday, January 17, 2002 3:16 PM
To: dox...@li...
Subject: [Doxygen-users] inheritance problem


Hi,
=20
I'm using a third party library in my code.
My code is in E:\MyProject\inc and E:\MyProject\src
The third party includes are in E:\ThirdParty\inc
=20

In my config file,I have :
INPUT                  =3D E:\MyProject\src E:\MyProject\inc
INCLUDE_PATH           =3D E:\MyProject\inc E:\ThirdParty\inc=20
=20
I have some classes which inherits from some third party classes
=20
I have a Third party class (TP)
=20
my code :
=20
class A :public TP
class B: public A
class C : public TP
class D : public C
=20
I only have :
=20
A
+B
C
+D
=20
I would like to have=20
TP
+A
    +B
+C
    +D
=20
What am I doing wrong ?
=20
Thanks,
=20
Stephane

[Doxygen-users] inheritance problem

[Stephane R. <ste...@sy...>]

Hi,

I'm using a third party library in my code.
My code is in E:\MyProject\inc and E:\MyProject\src
The third party includes are in E:\ThirdParty\inc

In my config file,I have :
INPUT                  =3D E:\MyProject\src E:\MyProject\inc
INCLUDE_PATH           =3D E:\MyProject\inc E:\ThirdParty\inc=20

I have some classes which inherits from some third party classes

I have a Third party class (TP)

my code :

class A :public TP
class B: public A
class C : public TP
class D : public C

I only have :

A
+B
C
+D

I would like to have=20
TP
+A
    +B
+C
    +D

What am I doing wrong ?

Thanks,

Stephane

RE: [Doxygen-users] multiline /// comments

[Glenn M. <gle...@vo...>]

Hi Gedalia,

I realize that you probably grabbed the code out of context. In case you
didn't, I've included a couple of statements that makes your stuff more
reliable to others.

1) I initialize the $comment_count variable to 0 at the beginning.
Better safe than sorry that garbage was in memory and become the
starting point.

2) Don't forget the default fall-through which ends up printing the
modified $_ input line to the standard output. I purposely did not
include the print statement within this little "if" branch, because I
needed to have the ability to do other processing to the very same line,
like when I change from @bug to my user-defined @lim alias.

Glenn

> -----Original Message-----
> From: Gedalia Pasternak [mailto:ge...@tu...]
> Sent: Thursday, January 17, 2002 10:34 AM
> To: Glenn Maxey
> Cc: dox...@li...
> Subject: RE: [Doxygen-users] multiline /// comments

> Hey thanks glenn, it seems pretty trivial to convert it over:

=20
BEGIN {
   local ($comment_count) =3D 0;  # count if you're in the comments
somewhere.
}
=20
> NEW_LINE: while (<>) {
>=20
>    #########=20
>    # Begin replacement of /// comments with /** ... **/ comments
>    # This purposely does not do NEW_LINE or printing.
>    #########=20
>    if (/\/\/\//) {
>       if ($comment_count =3D=3D 0){
>          # first line of a comment block
>          $comment_count++;
>          $_ =3D~ s/\/\/\//\/\*\*/;
>       } else {
>          # Some line in the middle of a comment block.
>          $comment_count++;
>          ###
>          # Changed to have middle stuff with no asterix.
>          # $_ =3D~ s/\/\/\// \*\*/;
>          ###
>          $_ =3D~ s/\/\/\///;
>       }
>    } elsif ($comment_count > 0){
>       # We were in a comment block; need to terminate it.
>       $comment_count =3D 0;
>       $_ =3D " \*\*\/\n" . $_ ;
>    }
>    ######### End Comment style change.
 =20
   #########=20
   # Default Fall through
   #########=20
   print $_;

} # end of the NEW_LINE while loop  =20

[Doxygen-users] Forward declaration of enums

[Stephane R. <ste...@sy...>]

Hi,

here is my problem :

I'm using forward declaration of enums and classes in my classes.
With the enum, I've got a warning :
E:/tmp/dox/class.h:8 Warning: Member MyEnum of file class.h is not
documented.
But I have no warnings for forward declarations of classes
Is that normal ?
Is that a bug ?

TIA,

Stephane
////////////////////////////////////////////////////////////////////////////
/**
* \file class.h
* \brief Header file for the class class
*/
#ifndef _class_HeaderFile
#define _class_HeaderFile

enum MyEnum;
class MyOtherClass;

/**
* \class class
* \brief qqqqq
*/
class MyClass
{
public:

/**
 * \fn MyClass()
 * \brief ctor
 */
 MyClass()
 {};

/**
 * \fn ~MyClass()
 * \brief dtor
 */
 ~MyClass()
 {};

/**
 * \fn SetEnum(const enum MyEnum anEnum)
 * \brief set memeber
 * \param anEnum const enum MyEnum
 */
 void SetEnum(const enum MyEnum anEnum)
 {m_enum = anEnum;};


/**
 * \fn Enum()
 * \brief return member
 * \return MyEnum
 */
 MyEnum Enum()
 {return m_enum;};
private:
 /** the member */
 MyEnum m_enum;
};

#endif
////////////////////////////////////////////////////////////////////

Re: [Doxygen-users] Re: Doxygen-users digest, Vol 1 #248 - 12 msgs

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

On Thu, Jan 17, 2002 at 11:13:15AM -0600, Scot Wilcoxon wrote:
> > /*! \file
> >  *  Docs for this file.
> >  */
> > 
> > in each .c file? If so send me a bug report, if not, reread FAQ #3
> 
> I had not tried \file because I thought it was only for data files.  I had not 
> tried what is in FAQ #3 and page 7 of Intro because I wasn't yet trying to 
> get "functions" documented, I was starting by trying to get the file with my 
> main() working.

If you think you can improve the documentation please do! Look at 
the .doc files in the doc dir of the source distribution and send me 
the changes.

Regards,
  Dimitri

RE: [Doxygen-users] multiline /// comments

[Gedalia P. <ge...@tu...>]

Hey thanks glenn, it seems pretty trivial to convert it over:
NEW_LINE: while (<>) {

   #########=20
   # Begin replacement of /// comments with /** ... **/ comments
   # This purposely does not do NEW_LINE or printing.
   #########=20
   if (/\/\/\//) {
      if ($comment_count =3D=3D 0){
         # first line of a comment block
         $comment_count++;
         $_ =3D~ s/\/\/\//\/\*\*/;
      } else {
         # Some line in the middle of a comment block.
         $comment_count++;
         ###
         # Changed to have middle stuff with no asterix.
         # $_ =3D~ s/\/\/\// \*\*/;
         ###
         $_ =3D~ s/\/\/\///;
      }
   } elsif ($comment_count > 0){
      # We were in a comment block; need to terminate it.
      $comment_count =3D 0;
      $_ =3D " \*\*\/\n" . $_ ;
   }
   ######### End Comment style change.
   =20
-Gedalia

RE: [Doxygen-users] multiline /// comments

[Darren B. <db...@ga...>]

FYI, last I checked, multiline /// comments get cut off in unpredictable
ways for most comment types.  (This isn't a bug, as I recall that the
documentation doesn't claim to support such a multiline comment.)  You won't
get errors or warnings, so you have to examine the output manually.  Only
for describing a structure field does it seem to consistently work:

///
/// I don't trust this
/// comment here.
///
struct xxx {
    /// This works
    /// without problems.
    int y;
};

I've been avoiding /// comments in other contexts for several Doxygen
versions now.  Perhaps this has been fixed since, but I'd be on the lookout.

--Darren

-----Original Message-----
From: Gedalia Pasternak [mailto:ge...@tu...]
Sent: Thursday, January 17, 2002 8:24 AM
To: dox...@li...
Subject: [Doxygen-users] multiline /// comments



Hello all, 
  After poking around the documentation for a while, I found that doxygen
will concatate a multi-line /// comment into the brief description. I was
wondering if there is any way to get doxygen to generate both the brief and
the "more..." comments from multi-line comments. Like it would for /**
comments.  Our project leads really don't like the /** style comments, but
I'm not thrilled with shoving all comments into the brief info block.
  
  Thanks, it's an awesome tool.

-Gedalia Pasternak

---------------------------------------------------------------
Graphics Engineer - www.TurbineGames.com	  
AIM: gedaliap
781.407.4428
---------------------------------------------------------------
Fight Entropy!!!  Fight Entropy!!!  Figth Etnropy! !
iFgth Etnrop!y ! giFth tErno!py !  giFt htrEno!p y! --- Well maybe
not...

_______________________________________________
Doxygen-users mailing list
Dox...@li...
https://lists.sourceforge.net/lists/listinfo/doxygen-users

[Doxygen-users] Re: Doxygen-users digest, Vol 1 #248 - 12 msgs

[Scot W. <sc...@wi...>]

> /*! \file
>  *  Docs for this file.
>  */
> 
> in each .c file? If so send me a bug report, if not, reread FAQ #3

I had not tried \file because I thought it was only for data files.  I had not 
tried what is in FAQ #3 and page 7 of Intro because I wasn't yet trying to 
get "functions" documented, I was starting by trying to get the file with my 
main() working.

RE: [Doxygen-users] multiline /// comments

[Glenn M. <gle...@vo...>]

I use input filters. The FILTER_SOURCE_FILES tag and the INPUT_FILTER
tag in the project file can be used to specify a program that should be
invoked to filter for each input file. Doxygen uses the output that the
filter program writes to standard output.=20

I use use input filters to:
- Change comment styles from //! into /**...**/.=20
- Change class definitions into a format that is more standard for
Doxygen reporting.
- Change the @bug Doxygen command into @lim command that is defined in
the project file in the ALIASES section.=20
- Change program files (e.g., Perl, IVE) into a format that is more or
less recognized by Doxygen.Refer to the Doxygen documentation.

Attached is a simple Perl program that changes from //! comments to /**
... **/ comments. You'll easily be able to modify it to support ///
comments.

FWIW, /** ... **/ comments are more reliable because they are treated as
a block. This is important when the comments appear at the beginning of
a file and contain @file. When not in a block, Doxygen can sometimes
pick off individual //! comments from the file header and attach them to
the first valid code item it sees, like a #define.

The input filters mean than I can appease SW Engineering whims about
comment styles.

More tools to follow. The draft documentation is complete, but I need to
get it reviewed and approved.

Glenn Maxey
Technical Writer
Voyant Technologies, Inc.
1765 West 121st Avenue
Westminster, CO 80234-2301
Tel. +1 303.223.5164
Fax. +1 303.223.5275
gle...@vo...


> -----Original Message-----
> From: Gedalia Pasternak [mailto:ge...@tu...]
> Sent: Thursday, January 17, 2002 9:24 AM
> To: dox...@li...
> Subject: [Doxygen-users] multiline /// comments
>=20
>=20
>=20
> Hello all,=20
>   After poking around the documentation for a while, I found=20
> that doxygen will concatate a multi-line /// comment into the=20
> brief description. I was wondering if there is any way to get=20
> doxygen to generate both the brief and the "more..." comments=20
> from multi-line comments. Like it would for /** comments. =20
> Our project leads really don't like the /** style comments,=20
> but I'm not thrilled with shoving all comments into the brief=20
> info block.
>  =20
>   Thanks, it's an awesome tool.
>=20
> -Gedalia Pasternak
>=20
> ---------------------------------------------------------------
> Graphics Engineer - www.TurbineGames.com	 =20
> AIM: gedaliap
> 781.407.4428
> ---------------------------------------------------------------
> Fight Entropy!!!  Fight Entropy!!!  Figth Etnropy! !
> iFgth Etnrop!y ! giFth tErno!py !  giFt htrEno!p y! --- Well maybe
> not...
>=20
> _______________________________________________
> Doxygen-users mailing list
> Dox...@li...
> https://lists.sourceforge.net/lists/listinfo/doxygen-users
>=20

[Doxygen-users] multiline /// comments

[Gedalia P. <ge...@tu...>]

Hello all,=20
  After poking around the documentation for a while, I found that =
doxygen will concatate a multi-line /// comment into the brief =
description. I was wondering if there is any way to get doxygen to =
generate both the brief and the "more..." comments from multi-line =
comments. Like it would for /** comments.  Our project leads really =
don't like the /** style comments, but I'm not thrilled with shoving all =
comments into the brief info block.
 =20
  Thanks, it's an awesome tool.

-Gedalia Pasternak

---------------------------------------------------------------
Graphics Engineer - www.TurbineGames.com	 =20
AIM: gedaliap
781.407.4428
---------------------------------------------------------------
Fight Entropy!!!  Fight Entropy!!!  Figth Etnropy! !
iFgth Etnrop!y ! giFth tErno!py !  giFt htrEno!p y! --- Well maybe
not...

[Doxygen-users] Linking to external documentation / Excluding symbolic links automatically

[Alexander L. <li...@we...>]

Hello,

i have two questions concerning the output of doxygen:

1.)  i would like to link to several external packages which are used in
our project. i generated tagfiles and included them in the doxygen
configuration for our project. this works fine - doxygen creates links
to all used external symbols at the individual places in the source
code.
now i would like to link to the complete standalone doxygen
documentation for that external packages from the mainpage of our
project (or from some other point). (later i would like to integrate a
link to our
CVS repository using CVSWeb). i tried to achieve this by adding a
htmlonly <A HREF=PATH_TO_PACKAGE/html/index.html /A> on our mainpage.
the link is created but following that link will create
a second treeview on my page, so i have one for our project and a second
for the external package.
is there a way of linking to that page without creating a new treeview?

2.) since we develop on unix we are using symbolic links. it seems that
doxygen gets confused when following a symbolic link to generate docs
for it. at least it will be added twice to the file list and generating
"man" output will end up with an error that doxygen is unable to open a
special man-file for writing.
example: if i configured ./project/man as output path for man docs.
doxygen appends the relative path of the link to the man output path ...

is there a way to prevent doxygen from following symbolic links besides
using the EXCLUDE_* options in the config file? i do not necessarily
want to add all exclusions manually ...

thanks in advance
alex

RE: [Doxygen-users] @see and polymorphic functions

[Adam T. <ad...@fi...>]

place the types inside the function:  example:

foo(int)

-----Original Message-----
From: C. Williams [mailto:chr...@ch...]
Sent: Wednesday, January 16, 2002 3:37 PM
To: dox...@li...
Subject: [Doxygen-users] @see and polymorphic functions


Does anyone know how to point a "see" reference to a particular 
polymorphic function in a C++ class when commenting?

Chris


_______________________________________________
Doxygen-users mailing list
Dox...@li...
https://lists.sourceforge.net/lists/listinfo/doxygen-users

[Doxygen-users] @see and polymorphic functions

[C. W. <chr...@ch...>]

Does anyone know how to point a "see" reference to a particular 
polymorphic function in a C++ class when commenting?

Chris

[Doxygen-users] class __declspec(dllimport) or __declspec(dllexport) AND namespace hides classes from generated HTML help

[Hofi <ho...@ma...>]

Hi All!

I'd like to use namespaces with classes and __declspec(dllimport),
__declspec(dllexport).

first case:
//----------------------------------------------------------
//! \namespace NS Doc for namespace NS.
namespace NS {
    //! Doc for C1 class.
    class C1 {
    .
    .
    .
    };
}
works fine NS::C1 class documented correctly.
//----------------------------------------------------------

second case:
//----------------------------------------------------------
//! \namespace NS Doc for namespace NS.
namespace NS {
    //! Doc for C1 class.
    class __declspec(dllexport) C1 {
    .
    .
    .
    };
}
does not work :(( NS::C1 not documented
//----------------------------------------------------------

third case: // no namespace
//----------------------------------------------------------
//! Doc for C1 class.
class __declspec(dllexport) C1 {
.
.
.
};
works fine (of course) C1 class documented correctly
//----------------------------------------------------------

What am i doing wrong?

    TIA
        Hofi

[Doxygen-users] Documenting IDL

[Francois St-A. <fst...@do...>]

I am in the process of documenting an IDL file using Doxygen.

Are there any examples available out there that could help me in my =
task?
I'm looking tips on documenting an IDL file using JavaDoc-style =
comments as
opposed to a Java file in straight JavaDoc.

I'm also looking to better control the HTML output: how do I get text =
to
appear int the "main page" section? how can I get rid of sections =
(namespace
list, alphabetical list, compound list, etc.? I'm asking because, in my =
IDL,
there is only one interface that I want to document and I think that =
the
different would confuse a reader more than anything.

Fran=E7ois St-Arnaud (Eng.), Senior Software Developer
585 Charest Blvd E., 7th Floor, Quebec, QC, Canada, G1K 3J2
Tel: 418 681-8022/0006 x243  Fax: 681-8015

Visit our new website and see the benefits of our
PowerBus(tm) Technology @  www.domosys.com=20

[Doxygen-users] Re: Documenting C

[Scot W. <sc...@wi...>]

Thanks to those that pointed out that my .c file needs a "\file" definition.  
Once this was pointed out I was able to find the documentation which 
mentions "repeat that, because it is often overlooked".  Yeah.

I think the need to use "\file" for non-data files should be stated at the 
start of the "Documenting the Code" info, not obtusely in page 7.

At least I can mention it here for the archives, even if there is no search 
capability for the group.

> I am having the same problem.  *.c files create no output with
> EXTRACT_ALL=NO.  
> My .h file does create output, so doxygen is indeed running and looking
> at 
> files.  I tried various comments within a .c file with no reaction.
> > FROM: Bernd Brandstetter
> > DATE: 12/05/2001 09:49:53
> > SUBJECT:  [Doxygen-users] Documenting C 
> >
> > While I'm very satisfied with doxygen's output for C++ files I didn't
> > manage to get a reasonable result for C code. In fact, doxygen doesn't
> > document anything besides the file list as long as I don't set 
> > EXTRACT_ALL to YES in which case it really documents everything -- 
> > including all structures and functions which I don't want to be
> documented 
> > and therefore didn't spend a doxygen comment.
> > 
> > This is independent of the settings of EXTRACT_PRIVATE,
> EXTRACT_STATIC, 
> > HIDE_UNDOC_MEMBERS and OPTIMIZE_OUTPUT_FOR_C.

Re: [Doxygen-users] Re: Documenting C

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

On Wed, Jan 16, 2002 at 10:46:41AM -0600, Scot Wilcoxon wrote:
> I am having the same problem.  *.c files create no output with EXTRACT_ALL=NO.  
> My .h file does create output, so doxygen is indeed running and looking at 
> files.  I tried various comments within a .c file with no reaction.

And you also tried to put a comment like

/*! \file
 *  Docs for this file.
 */

in each .c file? If so send me a bug report, if not, reread FAQ #3

Regards,
  Dimitri

> Doxygen 1.2.13.1 on Solaris 8 SPARC.
> 
> > FROM: Bernd Brandstetter
> > DATE: 12/05/2001 09:49:53
> > SUBJECT:  [Doxygen-users] Documenting C 
> > 
> > While I'm very satisfied with doxygen's output for C++ files I didn't 
> > manage to get a reasonable result for C code. In fact, doxygen doesn't 
> > document anything besides the file list as long as I don't set 
> > EXTRACT_ALL to YES in which case it really documents everything -- 
> > including all structures and functions which I don't want to be documented 
> > and therefore didn't spend a doxygen comment.
> > 
> > This is independent of the settings of EXTRACT_PRIVATE, EXTRACT_STATIC, 
> > HIDE_UNDOC_MEMBERS and OPTIMIZE_OUTPUT_FOR_C.
> 

[Doxygen-users] Can not get a link for an example file

[Hofi <ho...@fw...>]

Hi All!

    First of all I'd like to say thank you to Dimitri for this great tool.

My first problem is that, when i generate HTML output, i can't get a working
link (auto or manual with \link and \endlink) for an example file. The
example files appear correctly on the 'Examples' page, auto or manual links
correctly generated for any other files in the project.

Finally i found a way to get it work, i included the example file among the
project files (e.g. at the INPUT section, added the path of the example
files), but i have two problems with this solution:
    1. the example file itself included twice into the project, and the
biggest problem
    2. the example file included into the 'File List' page too. :((

Is it a bug, or i'm missing something?

TIA
    bye
        Hofi

p.s.: Sorry for my poor english.

RE: [Doxygen-users] (no subject)

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

it's not binding when you declare/define the functions inside the class
either.

I don't make the rules.. I just report them

-----Original Message-----
From: Paul Beardsley [mailto:pee...@ya...]
Sent: Wednesday, 2002 January 16 12:09
To: dox...@li...
Subject: RE: [Doxygen-users] (no subject)


Kris, Victor,

Thanks, but I don't know if I agree with what has been

said.  In the first case (below), I assume we all
agree
that even though I did not explicitly say 'inline',
the compiler might choose to inline a method.

But my second case is very similar - the definitions
are still in the header file, they are simple placed
outside the main class declaration.  So why should the
compiler not choose to inline them anyway?  After all
all the information is still there, in the header
file, 
at compile time.

Now there may be a reason it doesn't do this, which
is what my question is about.  But I cannot see that
my use of the 'inline' keyword is going to
help.  This is just a suggestion, after all, not
binding on the compiler.  Is there a hard rule
somewhere that if the method definition is separate
from the method declaration, but still in the include 
file, it will not be inlined unless 'inline' is used? 

Thanks,
Paul.


----------FIRST CASE-----------------------
class A
{
public:
  int method1 () const {return simple_thing;}
  void method2 () {do_something_simple ();}
};
----------END FIRST CASE-----------------


----------SECOND CASE--------------------
class A
{
public:
  int method1 ();
  void method2 ();
};

/** 
 * Blah
 * @return blah
*/
int A::method1 () const {return simple_thing;}

/**
 * Blah2
 */
void A::method2 () {do_something_simple ();}

---------END SECOND CASE-----------------------



__________________________________________________
Do You Yahoo!?
Send FREE video emails in Yahoo! Mail!
http://promo.yahoo.com/videomail/

_______________________________________________
Doxygen-users mailing list
Dox...@li...
https://lists.sourceforge.net/lists/listinfo/doxygen-users

RE: [Doxygen-users] (no subject)

[Kris T. <kri...@cs...>]

Hi Paul,
>
> Kris, Victor,
>
> Thanks, but I don't know if I agree with what has been
>
> said.

Maybe you should read a good C++ book then! This is not a doxygen question
by the way, so I don't think we should be discussion this further on this
list (after this email of course!)

> In the first case (below), I assume we all
> agree
> that even though I did not explicitly say 'inline',
> the compiler might choose to inline a method.
>

As far as syntax goes, in the first case you can put inline in there or not,
syntactically it means exactly the same: it's an 'inline' method.

This doesn't say anything at all about what code the compiler will generate
(even with an explicit inline, it doesn't have to inline it).
It does say something about how the compiler should handle linking etc. In
particular, a function/method declared as inline can occur in multiple
object files, while a non-inline (and non-static) one cannot.

> But my second case is very similar - the definitions
> are still in the header file, they are simple placed
> outside the main class declaration.  So why should the
> compiler not choose to inline them anyway?  After all
> all the information is still there, in the header
> file,
> at compile time.
>

You have to remember that the actual compiler actually never sees your
header file, it only sees the preprocessed thing, which has the header files
and your .cpp file merged into 1 huge text file.

> Now there may be a reason it doesn't do this, which
> is what my question is about.  But I cannot see that
> my use of the 'inline' keyword is going to
> help.  This is just a suggestion, after all, not
> binding on the compiler.  Is there a hard rule
> somewhere that if the method definition is separate
> from the method declaration, but still in the include
> file, it will not be inlined unless 'inline' is used?
>

Depends what you mean. Syntactically, the only ways to get it the status of
an inline function are
- put it in the class definition
- add inline to its function declaration

If at some point, the compiler needs to fill in the function call, it can
choose to inline or not, depending on your suggestion and what it thinks is
best. That's independent of the above. (For example, you might have a
non-inline function which you call in the same object file as its
definition. The compiler might choose to inline that function call).

I guess here's the difference. You can have an inline function, or the
compiler can inline a function call. They're largely (but hopefully not
completely) independent of each other.


I hope this clarifies the issue. If not, I'd be willing to discuss this off
the list, although I'm not sure what other words I can use. You might be
better off in alt.languages.C++ or so.

All the best,

Kris

RE: [Doxygen-users] (no subject)

[Paul B. <pee...@ya...>]

Kris, Victor,

Thanks, but I don't know if I agree with what has been

said.  In the first case (below), I assume we all
agree
that even though I did not explicitly say 'inline',
the compiler might choose to inline a method.

But my second case is very similar - the definitions
are still in the header file, they are simple placed
outside the main class declaration.  So why should the
compiler not choose to inline them anyway?  After all
all the information is still there, in the header
file, 
at compile time.

Now there may be a reason it doesn't do this, which
is what my question is about.  But I cannot see that
my use of the 'inline' keyword is going to
help.  This is just a suggestion, after all, not
binding on the compiler.  Is there a hard rule
somewhere that if the method definition is separate
from the method declaration, but still in the include 
file, it will not be inlined unless 'inline' is used? 

Thanks,
Paul.


----------FIRST CASE-----------------------
class A
{
public:
  int method1 () const {return simple_thing;}
  void method2 () {do_something_simple ();}
};
----------END FIRST CASE-----------------


----------SECOND CASE--------------------
class A
{
public:
  int method1 ();
  void method2 ();
};

/** 
 * Blah
 * @return blah
*/
int A::method1 () const {return simple_thing;}

/**
 * Blah2
 */
void A::method2 () {do_something_simple ();}

---------END SECOND CASE-----------------------



__________________________________________________
Do You Yahoo!?
Send FREE video emails in Yahoo! Mail!
http://promo.yahoo.com/videomail/

RE: [Doxygen-users] (no subject)

[Kris T. <kri...@cs...>]

fantastic. I forgot to put the 'inline' in there..

You HAVE to say
class A
{
 public:
   inline int method1 ();
   inline void method2 ();
};


Sorry

Kris

[Doxygen-users] Re: Documenting C

[Scot W. <sc...@wi...>]

I am having the same problem.  *.c files create no output with EXTRACT_ALL=NO.  
My .h file does create output, so doxygen is indeed running and looking at 
files.  I tried various comments within a .c file with no reaction.

Doxygen 1.2.13.1 on Solaris 8 SPARC.

> FROM: Bernd Brandstetter
> DATE: 12/05/2001 09:49:53
> SUBJECT:  [Doxygen-users] Documenting C 
> 
> While I'm very satisfied with doxygen's output for C++ files I didn't 
> manage to get a reasonable result for C code. In fact, doxygen doesn't 
> document anything besides the file list as long as I don't set 
> EXTRACT_ALL to YES in which case it really documents everything -- 
> including all structures and functions which I don't want to be documented 
> and therefore didn't spend a doxygen comment.
> 
> This is independent of the settings of EXTRACT_PRIVATE, EXTRACT_STATIC, 
> HIDE_UNDOC_MEMBERS and OPTIMIZE_OUTPUT_FOR_C.