doxygen-users — List for users of doxygen

[Doxygen-users] Doxygen style sheet

[Robert D. <rcd...@gm...>]

I like the style that the official Doxygen documentation uses. Can
someone provide directions on how to extract it and use it for my own
personally generated documentation? I tried just copying the CSS but
there are also some content files it needs, but I'd have to keep those
locally and change the references to them. Not entirely sure how to
properly set it up.

Thanks in advance.

[Doxygen-users] FULL_PATH_NAMES does not prepend path names to files

[didje <dia...@pd...>]

I have two libraries libA and libB. 

libA contains a file Action.h

libB contains a file action.h

I want to generate doxygen documentation in the same output directory for
both libraries. This directory is to be used in Windows, for which
action.html and Action.html are unfortunately considered to be the same
file. To prevent this clash, I wish to render the generated files unique by
prepending their path names to them.

Therefore, I set FULL_PATH_NAMES to YES.

I expect to see something like libA_Action.html and libB_action.html when I
generate the documentation, but I don't! I still see Action.html and
action.html. Its as if the FULL_PATH_NAMES parameter does nothing at all. Do
I also need to set some other parameter in the Doxyfile to make the
FULL_PATH_NAMES parameter work correctly?




--
View this message in context: http://doxygen.10944.n7.nabble.com/FULL-PATH-NAMES-does-not-prepend-path-names-to-files-tp6510.html
Sent from the Doxygen - Users mailing list archive at Nabble.com.

[Doxygen-users] Starting with 1.8.4 I get a blank output

[Timo K. <TKu...@gm...>]

Hi,

I'm getting a blank HTML/CHM output with any doxygen version newer than
1.8.3.1. I use this config file:

www.timosoft-software.de/stuff/public_docs_CHM.cfg


And basically this file is the only source file:

www.timosoft-software.de/stuff/AnimU.idl


The error message I get is:

"doxygen reached end of file while inside a (null) block!"

But I cannot find any block that gets opened, but not closed.
So is this a bug in doxygen? Or am I using it wrong?

Regards
Timo
-- 
www.TimoSoft-Software.de - Unicode controls for VB6
"Those who sacrifice freedom for safety deserve neither."
"Demokratie ist per Definition unsicher. Ihr Schutz entsteht aus der
Überzeugung, dass die demokratischen Kräfte überwiegen und sich – auf
demokratischem Wege – durchsetzen."

Re: [Doxygen-users] Best way to 'relocate' links to html

[Dimitri v. H. <do...@gm...>]

Hi Edward,

On 17 Feb 2014, at 16:32 , Edward Moyse <edw...@ce...> wrote:

> Hi all,
> 
> I'm just trying to understand why our doxygen builds take such an insane amount of time. I think we're not currently building tags first then making the html after, as recommended here:
> http://stackoverflow.com/questions/8247189/doxygen-is-slow/8247993#8247993
> and so I'm investigating changing this...
> 
> But also we seem to have an incredibly slow part of the process where the html is parsed to turn links like:
> file:///my/local/path/doc/xAODMuon/html/classSG_1_1AuxElement.html
> 
> into 
> html://a/web/location/doc/xAODMuon/html/classSG_1_1AuxElement.html
> 
> (The reason for this is we build the documentation on a local machine, then once it's done it's copied to a web visible location)
> 
> Presumably there is a better way to do this? I'd have thought it should be possible to specify that the html will eventually live elsewhere, but I had a look at the documentation and couldn't see it?

Doxygen normally doesn't create any absolute URLs (with file:// or http:// in them), so the URL renaming process is not part of doxygen and shouldn't be needed at all. Just copy the HTML files to the web location in the end.

Regards,
  Dimitri

Re: [Doxygen-users] xapian library building from source

[Dimitri v. H. <do...@gm...>]

Hi Chris,

On 17 Feb 2014, at 11:12 , Chris Chalmers <cch...@fo...> wrote:

> I am trying to build doxygen 1.8.5 from source on a windows machine.  I have the following installed
>  
> Flex, bison, python and QTtools
>  
> When I compile it under Visual Studio Express 2008 I get the following errors.  It seems like doxysearch needs
>  The xapian.h.  Any idea how to resolve this issue.  I would have thought it would have been bundled.
>  
> Any help you can give me would be very much appreciated

You can find the Xapian headers and libaries for Windows here:
ftp://ftp.stack.nl/pub/users/dimitri/xapian_doxygen_win.zip

Regards,
  Dimitri

[Doxygen-users] Best way to 'relocate' links to html

[Edward M. <Edw...@ce...>]

Hi all,

I'm just trying to understand why our doxygen builds take such an insane amount of time. I think we're not currently building tags first then making the html after, as recommended here:
http://stackoverflow.com/questions/8247189/doxygen-is-slow/8247993#8247993
and so I'm investigating changing this...

But also we seem to have an incredibly slow part of the process where the html is parsed to turn links like:
file:///my/local/path/doc/xAODMuon/html/classSG_1_1AuxElement.html

into 
html://a/web/location/doc/xAODMuon/html/classSG_1_1AuxElement.html

(The reason for this is we build the documentation on a local machine, then once it's done it's copied to a web visible location)

Presumably there is a better way to do this? I'd have thought it should be possible to specify that the html will eventually live elsewhere, but I had a look at the documentation and couldn't see it?

Cheers,

Ed

[Doxygen-users] xapian library building from source

[Chris C. <cch...@fo...>]

I am trying to build doxygen 1.8.5 from source on a windows machine.  I have the following installed

Flex, bison, python and QTtools

When I compile it under Visual Studio Express 2008 I get the following errors.  It seems like doxysearch needs
 The xapian.h.  Any idea how to resolve this issue.  I would have thought it would have been bundled.

Any help you can give me would be very much appreciated

Chris

>------ Build started: Project: doxysearch, Configuration: Debug Win32 ------
1>Compiling...
1>doxysearch.cpp
2>------ Build started: Project: doxywizard, Configuration: Debug Win32 ------
2>RCC ..\addon\doxywizard\doxywizard.qrc
2>The system cannot find the path specified.
2>Project : error PRJ0019: A tool returned an error code from "RCC ..\addon\doxywizard\doxywizard.qrc"
2>Project : warning PRJ0018 : The following environment variables were not found:
2>$(QT_DIR)
2>Build log was saved at "file://c:\Documents and Settings\cchalmers\Desktop\doxygen-1.8.5\winbuild\obj\BuildLog.htm<file:///c:\Documents%20and%20Settings\cchalmers\Desktop\doxygen-1.8.5\winbuild\obj\BuildLog.htm>"
2>doxywizard - 1 error(s), 0 warning(s)
1>c:\documents and settings\cchalmers\desktop\doxygen-1.8.5\addon\doxysearch\doxysearch.cpp(28) : fatal error C1083: Cannot open include file: 'xapian.h': No such file or directory
1>Build log was saved at "file://c:\Documents and Settings\cchalmers\Desktop\doxygen-1.8.5\winbuild\Debug\BuildLog.htm<file:///c:\Documents%20and%20Settings\cchalmers\Desktop\doxygen-1.8.5\winbuild\Debug\BuildLog.htm>"
1>doxysearch - 1 error(s), 0 warning(s)
3>------ Build started: Project: doxyindexer, Configuration: Debug Win32 ------
3>Compiling...
3>doxyindexer.cpp
4>------ Build started: Project: Doxygen, Configuration: Debug Win32 ------
4>Doxygen : warning PRJ0009 : Build log could not be opened for writing.
4>Make sure that the file is not open by another process and is not write-protected.
4>Performing Pre-Build Event...
3>c:\documents and settings\cchalmers\desktop\doxygen-1.8.5\addon\doxysearch\doxyindexer.cpp(31) : fatal error C1083: Cannot open include file: 'xapian.h': No such file or directory
3>Project : warning PRJ0018 : The following environment variables were not found:
3>$(XAPIAN_DIR)
3>Build log was saved at "file://c:\Documents and Settings\cchalmers\Desktop\doxygen-1.8.5\winbuild\Debug\BuildLog.htm<file:///c:\Documents%20and%20Settings\cchalmers\Desktop\doxygen-1.8.5\winbuild\Debug\BuildLog.htm>"
3>doxyindexer - 1 error(s), 0 warning(s)
4>'/' is not recognized as an internal or external command,
4>operable program or batch file.
4>'/' is not recognized as an internal or external command,
4>operable program or batch file.
4>Executing Lex for c:\Documents and Settings\cchalmers\Desktop\doxygen-1.8.5\src\/scanner.cpp ...
4>flex: can't open and
4>Project : error PRJ0019: A tool returned an error code from "Executing Lex for c:\Documents and Settings\cchalmers\Desktop\doxygen-1.8.5\src\/scanner.cpp ..."
4>Build log was saved at "file://c:\Documents and Settings\cchalmers\Desktop\doxygen-1.8.5\winbuild\Debug\BuildLog.htm<file:///c:\Documents%20and%20Settings\cchalmers\Desktop\doxygen-1.8.5\winbuild\Debug\BuildLog.htm>"
4>Doxygen - 1 error(s), 1 warning(s)
========== Build: 0 succeeded, 4 failed, 1 up-to-date, 0 skipped ==========

Re: [Doxygen-users] Unwanted page break before parameter table in Latex output

[John C. M. <Joh...@sm...>]

Hi Dimitri,

Thanks for the reply. 

On Saturday, February 15, 2014 02:50:35 PM Dimitri van Heesch wrote:

> If you are not using the latest doxygen release, try that first.

I am using 1.8.6 currently on Kubuntu 13.10.

> 
> If that doesn't help you could see if emptying the content of the
> ensurespace command in doxygen.sty helps, i.e. try:
> 
> \newcommand{\ensurespace}[1]{}
> 
> In general it is pretty hard to make LaTeX generate the right output
> automatically. It has the tendency to either leave too much whitespace or
> to put the content into the page margin :-(
> 
> Regards,
>   Dimitri

I tried removing the content of the ensurespace command as you suggested, but 
the results are the same.  

I moved things around in the code changing the order of the function 
declarations to try to avoid the problem, but still find it occurring in in two 
or three cases.  it does not seem to matter if the subsection header is the 
item to fall on the top of the page.  Having any one or two lines of paragraph 
text fall on the top of the page followed by the parameter table produces the 
same results. On the few cases where it occurs I get the paragraph line or 
lines (maybe a brief description line or the last line or two of a longer 
paragraph that started on the previous page), and then the page breaks and the 
parameter table appears at the top of the next page.  If I insert extra filler 
text in the paragraph before the parameters it places the table correctly, but 
creating an extra two lines of junk text to control the spacing behavior is 
not desirable.

>From what I can tell reading the doxygen.sty file, the DoxyParams envrionment 
is what is being processed when the page break occurs.  It does not seem to 
reference back to the ensurespace command, but does it?  Is there something in 
the table formatting that could be adjusted?

Getting Latex to cooperate seems to be more difficult than the linear algebra 
routines that I have to program.  Thanks for your help, it is most 
appreciated.

Best,
John

Re: [Doxygen-users] Unwanted page break before parameter table in Latex output

[Dimitri v. H. <do...@gm...>]

Hi John,

On 12 Feb 2014, at 16:38 , John C. Martin <Joh...@sm...> wrote:

> Hello,
> 
> I have been using Doxygen to generate a reference manual for some C++ code 
> using the Latex output to generate a pdf.  In the class documentation section 
> down in the member function documentation I am getting an unwanted page break 
> in the middle of the documentation for a function in some few cases and I 
> can't figure out what to adjust to make it work.
> 
> Specifically, when the sub section header for the function documentation 
> happens to fall on the top of a page, if there is no text or only a line or 
> two of text before starting the parameter table (I have listed all the 
> parameters using the \param commands) a page break is thrown and the parameter 
> table starts on the next page.  The parameter table in each case is only a few 
> entries long, so the table is not in need of a new page.  Originally I was 
> putting the parameter list after the brief description in the documentation 
> and this was occurring more frequently.  I rearranged the docs to move longer 
> description text up above the parameter table (which I would rather not do) 
> and it seems to have helped, but it still occurs if the longer description 
> text does not result in enough lines in the output.  I am guessing that there 
> is some threshold in the style for the parameter table that could be adjusted 
> to keep the page break from occurring.
> 
> I do not know enough about Latex to figure out how to adjust the output to keep 
> this from happening.  Google has not turned up anything useful so far, nor has 
> StackOverflow.  Can anyone give me some pointers on how to keep this from 
> happening?

If you are not using the latest doxygen release, try that first.

If that doesn't help you could see if emptying the content of the ensurespace
command in doxygen.sty helps, i.e. try:

\newcommand{\ensurespace}[1]{}

In general it is pretty hard to make LaTeX generate the right output automatically.
It has the tendency to either leave too much whitespace or to put the content into 
the page margin :-(

Regards,
  Dimitri

[Doxygen-users] Problem parsing dashes.

[Scott F. <sco...@gm...>]

http://doxygen.10944.n7.nabble.com/Escaping-dash-td5790.html
User talks about having to make changes to code to handle two dashes ("--").
I ran into a variation of the same issue.

I was assigning to a variable a string value that contained a search 
pattern.

label_dbg_comment = "%s+--%s+"

Doxygen would stop process the file at this line.

I ended up having to change the string and print separate char values to 
satisfy both code execution AND doxygen output.

Does Doxygen not recognize the dashes are in quotes and should not 
process the string?
Is there something else that should be done in this instance?

[Doxygen-users] Unwanted page break before parameter table in Latex output

[John C. M. <Joh...@sm...>]

Hello,

I have been using Doxygen to generate a reference manual for some C++ code 
using the Latex output to generate a pdf.  In the class documentation section 
down in the member function documentation I am getting an unwanted page break 
in the middle of the documentation for a function in some few cases and I 
can't figure out what to adjust to make it work.

Specifically, when the sub section header for the function documentation 
happens to fall on the top of a page, if there is no text or only a line or 
two of text before starting the parameter table (I have listed all the 
parameters using the \param commands) a page break is thrown and the parameter 
table starts on the next page.  The parameter table in each case is only a few 
entries long, so the table is not in need of a new page.  Originally I was 
putting the parameter list after the brief description in the documentation 
and this was occurring more frequently.  I rearranged the docs to move longer 
description text up above the parameter table (which I would rather not do) 
and it seems to have helped, but it still occurs if the longer description 
text does not result in enough lines in the output.  I am guessing that there 
is some threshold in the style for the parameter table that could be adjusted 
to keep the page break from occurring.

I do not know enough about Latex to figure out how to adjust the output to keep 
this from happening.  Google has not turned up anything useful so far, nor has 
StackOverflow.  Can anyone give me some pointers on how to keep this from 
happening?

Thanks

-- 
John C. Martin
Small Bear Technologies, Inc.
865.309.4LSA - Office
865.309.4007 - Direct
http://www.SmallBearTechnologies.com

[Doxygen-users] File reference page missing class methods

[Zoanie <zo...@gm...>]

Hello,

    I recently converted a C project to C++.  I'm using the Eclipse plugin. 
After converting the project I noticed that all the functions I converted to
class methods are not showing up in the source 'Files' reference web page. 
Only those functions not incorporated into a class remain.  The converted
class methods do appear in the 'Classes' tab.  I suppose my software team
and I are partial to the 'C' method of referencing functions to a specific
source file.  Is there an option I can use to have the 'Files' reference web
page include class methods?  Is this something I can resolve by upgrading to
the latest version?

    WinXP
    Doxygen 1.7.4
    Eclipse Galileo
    Eclox 0.8.0, Doxygen frontend plugin for Eclipse


    Many thanks for your replies,

Carlos



--
View this message in context: http://doxygen.10944.n7.nabble.com/File-reference-page-missing-class-methods-tp6500.html
Sent from the Doxygen - Users mailing list archive at Nabble.com.

[Doxygen-users] Issue with static functions if two header files are used

[skraxberger <ste...@nx...>]

Hi,

I have two different header files, one containing the public and on the
internal function prototypes. There are several functions in the "internal"
header file containing some documentation. In the source file there is some
additional documentation for the same functions. But the documentation from
the source file is ignored.

I also used \fn to explicitly mark the documentation in the header file to
assign it to the correct function.

The functions which have this issue are static. I have other static methods
with the same separation of the documentation (but only with one header and
one source file) where it works. EXTRACT_STATIC is therefore working
properly. If I remove the static keyword for the function it works.

What am I missing?

Best Regards,

Stefan Kraxberger




--
View this message in context: http://doxygen.10944.n7.nabble.com/Issue-with-static-functions-if-two-header-files-are-used-tp6499.html
Sent from the Doxygen - Users mailing list archive at Nabble.com.

Re: [Doxygen-users] Namespace page not generated

[didje <dia...@pd...>]

The library where this problem occurs (Library A) includes a tagfile from
another library (Library B) which contains an identically named namespace
("ABC"). When I remove the tagfile from the makefile.doxygen of library A,
then the namespace page "ABC" is indeed generated. However, the problem is
that I need to include that tagfile.






--
View this message in context: http://doxygen.10944.n7.nabble.com/Namespace-page-not-generated-tp6497p6498.html
Sent from the Doxygen - Users mailing list archive at Nabble.com.

[Doxygen-users] Namespace page not generated

[didje <dia...@pd...>]

I have a number of namespaces in a library. When I generate the documentation
for this library and look at the "Classes" section and click on "Class
List", I see the list of namespaces in the library, as well as a description
of each namespace.

However, for some of these namespaces, no namespace page (e.g.
namespaceABC.html) is created (Case A), whereas for some others, namespace
pages are created (Case B). Despite going through the code several time
comparing each case, I have been unable to discover what it is that causes a
namespace page to be generated for Case A, but not for Case B.




--
View this message in context: http://doxygen.10944.n7.nabble.com/Namespace-page-not-generated-tp6497.html
Sent from the Doxygen - Users mailing list archive at Nabble.com.

[Doxygen-users] References relation not resolved by C functions

[Abdoul <abd...@gi...>]

I'm using Doxygen 1.8.4 to parse Java and C sources. 
I have the following case: 
although the tag "REFERENCES_RELATION" is set to YES in the config file, the
references (calls) are not resolved by the C functions or only some
functions but not all. 
By Java all references are resolved correctly.
Does someone has any advice for me?

Thanks in advance
Abdoul



--
View this message in context: http://doxygen.10944.n7.nabble.com/References-relation-not-resolved-by-C-functions-tp6495.html
Sent from the Doxygen - Users mailing list archive at Nabble.com.

[Doxygen-users] doxygen stops with segmentation fault when using tagfile

[Bernd S. <ber...@un...>]

Hi,
I use doxygen 1.8.6 and it stops with segmentation fault if I use a
particular tagfile (generated with the same version). The documentation
links to a total of three other project documentations through tagfiles,
but I can clearly identify one of them as being the problem.

This is the command that invokes doxygen (called through a make file;
the TAGFILE option is dynamically assigned):

cd .. && (cat Doxyfile ; echo
"TAGFILES=/home/bs/development/share/BSUtilities-0.6.2+/documentation/BSUtilities.tag=/home/bs/development/share/BSUtilities-0.6.2+/documentation/html
\"/home/bs/development/share/Quantities-1.2.2+/documentation/quantities.tag=/home/bs/development/share/Quantities-1.2.2+/documentation/html\"
\"/home/bs/development/share/SciDatVis-0.0.0+/documentation/sciDatVis.tag=/home/bs/development/share/SciDatVis-0.0.0+/documentation/html\"
") | /home/bs/software/bin/doxygen -

the tagfiles are found:

Reading tag file
`/home/bs/development/share/BSUtilities-0.6.2+/documentation/BSUtilities.tag',
location
`/home/bs/development/share/BSUtilities-0.6.2+/documentation/html'...
Reading tag file
`/home/bs/development/share/Quantities-1.2.2+/documentation/quantities.tag',
location
`/home/bs/development/share/Quantities-1.2.2+/documentation/html'...
Reading tag file
`/home/bs/development/share/SciDatVis-0.0.0+/documentation/sciDatVis.tag',
location `/home/bs/development/share/SciDatVis-0.0.0+/documentation/html'...

but later on, doxygen dies:

ASSERT: "m_indent<MAX_INDENT" in ftvhelp.cpp (680)
/bin/sh: line 1:  2183 Done                    ( cat Doxyfile; echo
"TAGFILES=/home/bs/development/share/BSUtilities-0.6.2+/documentation/BSUtilities.tag=/home/bs/development/share/BSUtilities-0.6.2+/documentation/html
\"/home/bs/development/share/Quantities-1.2.2+/documentation/quantities.tag=/home/bs/development/share/Quantities-1.2.2+/documentation/html\"
\"/home/bs/development/share/SciDatVis-0.0.0+/documentation/sciDatVis.tag=/home/bs/development/share/SciDatVis-0.0.0+/documentation/html\"
" )
      2184 Segmentation fault      | /home/bs/software/bin/doxygen -

I have compiled 1.8.6 after configuration with the -debug option and
this is the result of running gdb on the core file produced:

> gdb /home/bs/software/doxygen/ core
GNU gdb (GDB; openSUSE 13.1) 7.6.50.20130731-cvs
Copyright (C) 2013 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later
<http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "i586-suse-linux".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://bugs.opensuse.org/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word".
..
/home/bs/software/doxygen/: No such file or directory.
[New LWP 5126]
Missing separate debuginfo for the main executable file
Try: zypper install -C
"debuginfo(build-id)=7630226778a54abc020c7f51183fd32fe5d6b840"
Core was generated by `/home/bs/software/bin/doxygen -'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0  0x08379dde in ?? ()

Any ideas?
Best regards
Bernd

Re: [Doxygen-users] The #if defined-#endif block, I can't get the correct result...

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

At 01:51 22. 01. 2014, yulily wrote:
>I just started using doxygen to document a C project.And there is some
>questions.
>The program like this:
>#if defined  (ABC)
>#include "aaa.h"
>#include "bbb.h"
>#include "ccc.h"
>#include "ddd.h"
>/**
>* @param [in]   : null
>* @param [out] : null
>* @retval         : OK-success
>* @retval         : ERROR-fail
>*/
>int abcd(void)
>{
>    return (abcde());
>}
>#endif
>
>I would expect Doxygen would produce document within the #if defined-#endif
>block, however it ignores them.The documents produced only have a file and
>it can't identify the function and it's annotation...but if I delete the #if
>defined and #endif,i can get the correct result.

Where does ABC get defined if you compile your program normally?
It either needs to be defined in the sources e.g. some included header
or then it is given on the command line. In this case you need to adjust
the doxygen config and add ABC to the list of predefined macros. This
is the equivalent of -DABC for gcc.

bye  Fabi

[Doxygen-users] The #if defined-#endif block, I can't get the correct result...

[yulily <353...@qq...>]

I just started using doxygen to document a C project.And there is some
questions.
The program like this:
#if defined  (ABC)
#include "aaa.h"
#include "bbb.h"
#include "ccc.h"
#include "ddd.h"
/**
* @param [in]   : null
* @param [out] : null
* @retval         : OK-success
* @retval         : ERROR-fail
*/
int abcd(void)
{
    return (abcde());
}
#endif

I would expect Doxygen would produce document within the #if defined-#endif
block, however it ignores them.The documents produced only have a file and
it can't identify the function and it's annotation...but if I delete the #if
defined and #endif,i can get the correct result.

Thanks;




--
View this message in context: http://doxygen.10944.n7.nabble.com/The-if-defined-endif-block-I-can-t-get-the-correct-result-tp6492.html
Sent from the Doxygen - Users mailing list archive at Nabble.com.

[Doxygen-users] Documenting specialized member functions

[Matthias K. <mk...@gm...>]

Hi!

I stumbled upon some difficulties in documenting specialized member
functions in my project. For experiments with this issues I created some
small files named foo.h containing the declaration of a template member
function, foo_a.cpp and foo_b.cpp with specializations and main.cpp calling
these functions:

### foo.h ###

    /// \file
    /// template declaration and definition.

    #include <iostream>

    class A;
    class B;

    /// provides the specialized method.
    /// further explanation...
    struct C {
        /// foos the object
        template< typename T > void foo();
    };

### foo_a.cpp ###

    /// \file
    /// implementation of foo< A >
    #include "foo.h"

    /// specialization for type A.
    /// further explanation...
    template<>
    void C::foo< A >() { std::cout << "A\n"; }

    /// \file
    /// implementation of foo< B >
    #include "foo.h"

### foo_b.cpp ###

    /// specialization for type B.
    /// further explanation...
    template<>
    void C::foo< B >() { std::cout << "B\n"; }

### main.cpp

    /// \file
    /// implementation of main
    #include "foo.h"

    /// main function
    int main() {
        C c;
        c.foo< A >();
        c.foo< B >();
    }

This is my Doxyfile:

    JAVADOC_AUTOBRIEF      = YES
    WARN_NO_PARAMDOC       = YES
    GENERATE_LATEX         = NO
    CLANG_ASSISTED_PARSING = NO
    HAVE_DOT               = YES
    CALL_GRAPH             = YES
    CALLER_GRAPH           = YES

Now, the documentation is generated without warnings or error, but the
documentation for the specialized members look like

    template<>
    void C::foo()
    specialization for type B.
    ...

and are missing the information about the concrete specialization. Is there
any method to get doxygen to include the <B> in the generation description?

Additionally, I'd like to produce a graph showing the calls to C::foo() from
main(). Just setting CALL_GRAPH and CALLER_GRAPH does not do the trick here.
The include dependency graphs are generated correctly.

Regards,
Matthias

[Doxygen-users] Linking to external documentation (tagfiles) doesn't preserve subgroups

[Brian L. <lin...@ho...>]

When linking to external documentation (tagfiles), the nested grouping layout of that external documentation is not preserved, and all the groups appear at the top level. This is very undesirable behavior.
 		 	   		  

[Doxygen-users] Documentation for derived classes no longer has an "Additional Inherited Members" section after upgrading from Doxygen 1.8.3.1 to 1.8.6

[Geoff A. <gd...@us...>]

I've found that documentation for derived classes no longer has an 
"Additional Inherited Members" section after upgrading from Doxygen 
1.8.3.1 to 1.8.6.  Is this an intended change?  I didn't see any mention 
of it in the change log.  Is there an option I should set in my Doxyfile 
to tell Doxygen to generate  "Additional Inherited Members" sections for 
derived classes?

Thanks,
Geoff Alexander, Ph.D.
Software Engineer, Corporate Tools Development
IBM Corporation
RTP, NC

[Doxygen-users] INLINE_GROUPED_CLASSES breaks generated tag files

[Brian L. <lin...@ho...>]

When INLINE_GROUPED_CLASSES is set to YES, it breaks the generated Doxygen tag file.

Here's the error I'm getting when trying to bring in the broken tagfile into another project:

Reading and parsing tag files
Reading tag file `../project1/Doxygen.tag'...
C:/Users/bl4703/Documents/subversion/doxygen/project1/Doxygen.tag:11: warning: Unexpected tag `ancho
r' found

error: Fatal error at line 13 column 10: tag mismatch
error: Fatal error at line 13 column 10: error while parsing element
error: Fatal error at line 13 column 10: error while parsing content
error: Fatal error at line 13 column 10: error while parsing main element
Parsing files

Here are the files that were used to generate the error:

project1/doxygen.cfg:

PROJECT_NAME = project1

INLINE_GROUPED_CLASSES = YES

GENERATE_TAGFILE = Doxygen.tag

INPUT = source.h

project1/source.h:

/*! @defgroup Project1Group Project 1 */

//! Project1
/*! @ingroup Project1Group */
class Project1
{
};

project2/doxygen.cfg:

PROJECT_NAME = project2

INPUT = source.h

TAGFILES = ../project1/Doxygen.tag

project2/source.h:

/*! @defgroup Project2Group */

//! Project2
/*! @ingroup Project2Group */
class Project2 : public Project1
{
};

 		 	   		  

[Doxygen-users] [ENHANCEMENT] Remove path dependencies from graphviz

[Arthur S. <asc...@at...>]

Dmitri;


It might be possible to remove the current path issues with the use of graphviz. If the dot interface was encapsulated in a class, then when doxygen is run it can determine which path format is used by dot and which is native to the invoking shell/system. Using the appropriate class object then does the necessary conversion.

Suppose there are:

class superPath {
    virtual string convertPath(string path) = 0;

}

class unixTounix: superPath { }class unixToDOS : superPath { }
class DOSToUnix : superPath { }
class DOSToDOS  : superPath { }

then at runtime initialization create a tmpfile and retrieve its absolute path. That determines the path environment for doxygen. Call a dot method requiring a path. If it is rejected or not rejected determines the path used by graphviz. Making the class instance global allows conversion throughout doxygen.

I don't know the code of anything, but just as a thought:
int decision = (doxygen_path_unix)? 0: 1;
decision    |= ((dot(path))? decision: !decision) << 1;

switch (decision) {
case 00:
    globalThing = new unixTounix();
    break;
case 01:
    globalThing = new unixToDOS();
    break;
case 10:
    globalThing = new DOSTounix();
    break;
case 11:
    globalThing = new DOSToDOS();
    break;
}

Thereafter:  dot(globalThing->convertPath(path));


This is not to say that it should, could, or would be done or done this way. Just a thought

Maybe in some future release?


art

Re: [Doxygen-users] Unknown Error

[Stefan P. <ste...@gm...>]

Am 17.01.2014 19:03, schrieb Arthur Schwarz:
> Dmitri;
>
> The issue is as you expected. I have a cygwin/unix doxygen and a Windows
> graphviz (2.30). The absolute paths (/cygdrive/...) sent to graphviz are
> rejected. Since I can't download a linux version of graphviz to a Windows
> system, is there a Windows version of doxygen lying around?
>
>
There is a Windows binary setup available at the doxygen home page.


--
Stefan P.

Top-posting:
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?