doxygen-develop Mailing List for Doxygen (Page 58)
Brought to you by:
dimitri
You can subscribe to this list here.
2001 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(4) |
Jun
(4) |
Jul
(29) |
Aug
(8) |
Sep
(8) |
Oct
(17) |
Nov
(34) |
Dec
(6) |
---|---|---|---|---|---|---|---|---|---|---|---|---|
2002 |
Jan
(20) |
Feb
(14) |
Mar
(11) |
Apr
(9) |
May
(8) |
Jun
(7) |
Jul
(25) |
Aug
(12) |
Sep
(12) |
Oct
(24) |
Nov
(27) |
Dec
(12) |
2003 |
Jan
(12) |
Feb
(14) |
Mar
(15) |
Apr
(11) |
May
(17) |
Jun
(20) |
Jul
(32) |
Aug
(13) |
Sep
(34) |
Oct
(12) |
Nov
(16) |
Dec
(33) |
2004 |
Jan
(20) |
Feb
(6) |
Mar
(20) |
Apr
(15) |
May
(16) |
Jun
(28) |
Jul
(7) |
Aug
(7) |
Sep
(17) |
Oct
(16) |
Nov
(17) |
Dec
(43) |
2005 |
Jan
(15) |
Feb
(5) |
Mar
(14) |
Apr
(4) |
May
(3) |
Jun
(8) |
Jul
(17) |
Aug
(16) |
Sep
(7) |
Oct
(17) |
Nov
(1) |
Dec
(7) |
2006 |
Jan
(7) |
Feb
(6) |
Mar
(10) |
Apr
(6) |
May
(3) |
Jun
(4) |
Jul
(3) |
Aug
(3) |
Sep
(18) |
Oct
(11) |
Nov
(10) |
Dec
(3) |
2007 |
Jan
(12) |
Feb
(12) |
Mar
(23) |
Apr
(5) |
May
(13) |
Jun
(6) |
Jul
(5) |
Aug
(4) |
Sep
(8) |
Oct
(10) |
Nov
(6) |
Dec
(7) |
2008 |
Jan
(7) |
Feb
(13) |
Mar
(35) |
Apr
(14) |
May
(13) |
Jun
(4) |
Jul
(9) |
Aug
(6) |
Sep
(12) |
Oct
(9) |
Nov
(6) |
Dec
(3) |
2009 |
Jan
(2) |
Feb
(2) |
Mar
(2) |
Apr
(15) |
May
(1) |
Jun
(2) |
Jul
(7) |
Aug
(3) |
Sep
(4) |
Oct
(1) |
Nov
(2) |
Dec
(1) |
2010 |
Jan
(4) |
Feb
|
Mar
(5) |
Apr
(1) |
May
(5) |
Jun
|
Jul
(2) |
Aug
(3) |
Sep
(11) |
Oct
(2) |
Nov
(1) |
Dec
(5) |
2011 |
Jan
(12) |
Feb
(3) |
Mar
(28) |
Apr
(4) |
May
(3) |
Jun
(4) |
Jul
(15) |
Aug
(12) |
Sep
(2) |
Oct
(3) |
Nov
(6) |
Dec
(3) |
2012 |
Jan
(1) |
Feb
(4) |
Mar
(9) |
Apr
(5) |
May
(6) |
Jun
(6) |
Jul
(3) |
Aug
(3) |
Sep
(4) |
Oct
(2) |
Nov
(9) |
Dec
(7) |
2013 |
Jan
(8) |
Feb
(14) |
Mar
(15) |
Apr
(21) |
May
(29) |
Jun
(34) |
Jul
(3) |
Aug
(7) |
Sep
(13) |
Oct
(1) |
Nov
(3) |
Dec
(5) |
2014 |
Jan
|
Feb
|
Mar
|
Apr
(10) |
May
(2) |
Jun
(4) |
Jul
(2) |
Aug
(2) |
Sep
(4) |
Oct
(4) |
Nov
(4) |
Dec
(2) |
2015 |
Jan
(7) |
Feb
(4) |
Mar
(3) |
Apr
(15) |
May
(4) |
Jun
(9) |
Jul
(1) |
Aug
(2) |
Sep
|
Oct
|
Nov
(3) |
Dec
(7) |
2016 |
Jan
(1) |
Feb
|
Mar
|
Apr
(1) |
May
(1) |
Jun
(1) |
Jul
|
Aug
(5) |
Sep
|
Oct
(1) |
Nov
(1) |
Dec
(1) |
2017 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(1) |
Jun
|
Jul
(9) |
Aug
|
Sep
|
Oct
|
Nov
(1) |
Dec
(5) |
2018 |
Jan
|
Feb
(2) |
Mar
(3) |
Apr
|
May
(7) |
Jun
(1) |
Jul
|
Aug
(1) |
Sep
|
Oct
|
Nov
|
Dec
|
2019 |
Jan
(4) |
Feb
|
Mar
(1) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(3) |
Oct
|
Nov
|
Dec
|
2020 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
(1) |
Aug
|
Sep
|
Oct
(1) |
Nov
(1) |
Dec
(1) |
2021 |
Jan
(2) |
Feb
|
Mar
(2) |
Apr
|
May
|
Jun
(3) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2022 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(1) |
Sep
|
Oct
|
Nov
(1) |
Dec
|
From: Ruben G. <ru...@ug...> - 2002-11-21 15:28:49
|
In the Mandrake distribution, number 9 or so, install has moved from=20 package fileutils to package coreutils. The line 290 of the configure script, echo -n " Checking for GNU install tool... " if test "$f_insttool" =3D NO; then install_names=3D"ginstall install" install_dirs=3D"/usr/bin /usr/local/bin /bin /sbin $bin_dirs" install_prog=3DNO install_found=3DNO for i in $install_names; do for j in $install_dirs; do if test -x "$j/$i"; then if test -n "`$j/$i --version 2>/dev/null | grep fileutils`";=20 then <---- This one here install_found=3DYES install_prog=3D"$j/$i" break 2 fi fi done done f_insttool=3D"$install_prog" Should be changed to =20 if test -n "`$j/$i --version 2>/dev/null | grep utils`"; then =20 or to whatever seems more apropiate, so that the install from package=20 coreutils gets accepted. This is the output from my install: #install --version install (coreutils) 4.5.3 Escrito por David MacKenzie. Copyright (C) 2002 Free Software Foundation, Inc. Esto es software libre; vea el c=C3=B3digo fuente para las condiciones de= copia. No hay NINGUNA garant=C3=ADa; ni siquiera de COMERCIABILIDAD o IDONEIDAD = PARA UN FIN DETERMINADO. |
From: Paco H. <do...@fi...> - 2002-11-20 15:44:01
|
I thought I'd post this in the developer list and get feedback. In 1.3-rc1, I noticed the following changes from 1.2.18. They're undesirable from my point of view because I have a huge source base that is documented (5000 pages of PDF) via doxygen, and this significantly changes how the output looks. 1. It used to be the case that you could use \section and \subsection commands inside the \mainpage. You can still do this in 1.3-rc1, but in the LaTeX output, they all get turned into subsection commands. I used to have Chapter 1, and then section 1.1 Overview, and section 1.2.2 Device drivers, etc. Now, everything is 1.0.1, 1.0.2, 1.0.3, etc. Regardless of whether I use \page, \section, or \subsection, they all end up at the same level on the \mainpage. In the "related pages" documents, everything behaves as expected. 2. As for the related pages, it used to be possible to have more than one \page command within a comment block. I.e. this used to be legit: /*! \page foo Glossary \section sterms Software Terms .. \page ascii ASCII Character Set blah blah blah */ This doesn't work. The second \page gets ignored. The "\page" gets absorbed so it isn't in the output, but it doesn't create a new section in the related pages. The workaround is to close and reopen the comment block: /*! \page foo Glossary .. */ /*! \page ascii ASCII Character set .. */ I'd like to have the old functionality back, if possible. On an unrelated note, if any of you ever try to create PDF from doxygen on the scale I'm talking about, you'll hit some serious limits in how many hyperlinks pdflatex can grok in a single document. I have solved this problem using the teTeX distribution and a customized texmf.cnf. I'd be happy to share this with anyone who has the same problems. Regards, Paco -- pa...@pa... http://paco.to/ |
From: Bjoern G. <bjo...@gm...> - 2002-11-19 06:45:03
|
Hi, As a re to Kilian Kiko, Johan Eriksson and Cormac Twomey here is a small XSL code that will merge all xml: <xsl:stylesheet xmlns:xsl=3D"http://www.w3.org/1999/XSL/Transform" version=3D"1.0"> <xsl:output method=3D"xml" version=3D"1.0" indent=3D"yes" standalone=3D"yes" /> <xsl:template match=3D"/"> <doxygen version=3D"{doxygen/@version}"> <!-- Load all doxgen generated xml files --> <xsl:for-each select=3D"doxygen/compound"> <xsl:copy-of select=3D"document( concat( @refid, '.xml' ) )/doxygen/*" /> </xsl:for-each> </doxygen> </xsl:template> </xsl:stylesheet> Use this with an external XSLT processor [1] on the doxygen generated index.xml et voila :) Hope this helps, Bj=F8rn. [1] I'm only aware of MS cmd line util: http://msdn.microsoft.com/downloads/default.asp?URL=3D/code/sample.asp?ur= l =3D/msdn-files/027/001/485/msdncompositedoc.xml |
From: Moshe K. <mo...@te...> - 2002-11-17 10:56:52
|
Hi, Dimitri and other listers I downloaded the latest version of Doxygen for Windows and tried it out on C# source files. Sources are developed in VISUAL .NET for C#. Comments to date were inserted using MS's C# XML tags. The output (plain HTML) was pretty rough. (A) 1. As expected, doxygen didn't handle the XML tags. 2. Among the sources were interface definitions for classes. Doxygen didn't identify these interfaces at all. 3. I checked GENERATE HTMLHELP in the HTML tab and the path to the desired output. I got nothing. ANY WORDS OF ENCURAGMENT? (B) In your email, you suggest using an input filter. I have seen this suggested by other listers, Glenn Maxely among them. 1. Do you have any suggestions how one would write this? 2. Does the input filter appear in the list of input files? What is the work flow with such a filter? (C) Does anyone have experience with including documentation (I think it has to be CHM) into the Microsoft Documentation Explorer (this allows you to read MSDN docs, as well as your own)? TIA for your responses. Moshe Kruger Freelance Technical Writer Tel/fax (972)-3-9607130 Mobile: (972)-57-569-093 ~~~~~~~~~~~~~~~~~~~~ APIs/SDKs * Online help * Training courseware * User manuals Moshe Kruger Freelance Technical Writer Tel/fax (972)-3-9607130 Mobile: (972)-57-569-093 ~~~~~~~~~~~~~~~~~~~~ APIs/SDKs * Online help * Training courseware * User manuals ************************************************************************************************** ** eSafe-IL scanned this outgoing email for viruses, vandals and malicious content ** ************************************************************************************************** |
From: Dimitri v. H. <di...@st...> - 2002-11-15 12:35:29
|
On Fri, Nov 15, 2002 at 10:03:03AM +0100, Carsten Zerbst wrote: > Am Don, 2002-11-14 um 19.02 schrieb Dimitri van Heesch: > > > { > > case ILinkedText::Kind_Text: // plain text > > result+=dynamic_cast<ILT_Text*>(lt)->text()->latin1(); break; > > case ILinkedText::Kind_Ref: // a link > > result+=dynamic_cast<ILT_Ref *>(lt)->text()->latin1(); break; > > } > > Hello Dimitri, > > thanks very much. As I wrote to much java the last years, > the dynamic_cast did not come to my mind. I assume that the same > solution will work to extract the documentation within > > IDocRoot* docRoot = myMethod->detailedDescription(); > IDocIterator* docit = docRoot->contents(); > IDoc doc; > for (docit->toFirst();(doc=docit->current());docit->toNext()) { > > }; > docit-release(); > > ? > > If yes, do you have an example available as well ? Yes, look at addon/doxmlparser/test/main.cpp in the source distribution. Regards, Dimitri |
From: Carsten Z. <car...@at...> - 2002-11-15 09:06:06
|
Am Don, 2002-11-14 um 19.02 schrieb Dimitri van Heesch: > { > case ILinkedText::Kind_Text: // plain text > result+=dynamic_cast<ILT_Text*>(lt)->text()->latin1(); break; > case ILinkedText::Kind_Ref: // a link > result+=dynamic_cast<ILT_Ref *>(lt)->text()->latin1(); break; > } Hello Dimitri, thanks very much. As I wrote to much java the last years, the dynamic_cast did not come to my mind. I assume that the same solution will work to extract the documentation within IDocRoot* docRoot = myMethod->detailedDescription(); IDocIterator* docit = docRoot->contents(); IDoc doc; for (docit->toFirst();(doc=docit->current());docit->toNext()) { }; docit-release(); ? If yes, do you have an example available as well ? Thanks, Carsten -- Carsten Zerbst <car...@at...> Atlantec Enterprise |
From: <har...@ya...> - 2002-11-15 08:38:36
|
On the page 'graphical class hierarchy' there's no link to the graph legend page. Shouldn't that link be there too like on the class pages below the class graph? __________________________________________________________________ Gesendet von Yahoo! Mail - http://mail.yahoo.de Möchten Sie mit einem Gruß antworten? http://grusskarten.yahoo.de |
From: Dimitri v. H. <di...@st...> - 2002-11-14 18:02:49
|
On Tue, Nov 12, 2002 at 10:36:09AM +0100, Carsten Zerbst wrote: > Hello, > I struggle how to extract the parameter type from the xml file using > doxmlparser. I get the parameter name from > > > printf("param decl %s\n", param->declarationName()->latin1()); > > but how to get the paramter type ? Using param->type() of course ;-) The result is not a string but can be converted to one using something like this: QString linkedTextToString(ILinkedTextIterator *ti) { QString result; ILinkedText *lt=0; for (ti->toFirst();(lt=ti->current());ti->toNext()) { switch (lt->kind()) { case ILinkedText::Kind_Text: // plain text result+=dynamic_cast<ILT_Text*>(lt)->text()->latin1(); break; case ILinkedText::Kind_Ref: // a link result+=dynamic_cast<ILT_Ref *>(lt)->text()->latin1(); break; } } return result; } Regards, Dimitri |
From: <Joh...@sy...> - 2002-11-14 14:27:45
|
Hi, If the GENERATE_XML option is set to YES and special characters like '>', '&', '$' etc. are used in the comment, doxygen comes up with error messages of the type: Error: Unexpected character `<' It removes the character in the xml output (where as in HTML it stays unchanged). Is this intended or is it a bug? It should rather transform the characters, to e.g. "<" , as it did in version 1.2.17. Regards Johannes Enter your application into the Symbian developer contest and win great prizes http://www.symbian.com/news/2002/apps-dvlprs-feat.html Closes December 1st! ********************************************************************** Symbian Ltd is a company registered in England and Wales with registered number 01796587 and registered office at 19 Harcourt Street, London, W1H 4HF, UK. This message is intended only for use by the named addressee and may contain privileged and/or confidential information. If you are not the named addressee you should not disseminate, copy or take any action in reliance on it. If you have received this message in error please notify pos...@sy... and delete the message and any attachments accompanying it immediately. Symbian does not accept liability for any corruption, interception, amendment, tampering or viruses occuring to this message in transit or for any message sent by its employees which is not in compliance with Symbian corporate policy. ********************************************************************** |
From: Cormac T. <ct...@ct...> - 2002-11-12 22:35:55
|
I did something like this using XML External entities. I have a file, called master.xml, which merges all of the files together. It looks like this: <?xml version="1.0" encoding="ISO-8859-1" standalone="no" ?> <!DOCTYPE doxygengroup [ <!ENTITY IInterface1 SYSTEM "doxygen/xml/interfaceIInterface1.xml"> <!ENTITY IInterface2 SYSTEM "doxygen/xml/interfaceIInterface2.xml"> <!ENTITY IInterface3 SYSTEM "doxygen/xml/interfaceIInterface3.xml"> <!ENTITY IInterface4 SYSTEM "doxygen/xml/interfaceIInterface4.xml"> <!ENTITY IInterface5 SYSTEM "doxygen/xml/interfaceIInterface5.xml"> ]> <doxygengroup> &IInterface1; &IInterface2; &IInterface3; &IInterface4; &IInterface5; </doxygengroup> At this point I would be able to transform master.xml using xsl stylesheets, except that each of the xml files generated by doxygen has an xml header of the form <?xml ... ?> which I don't think should appear in an xml external entity; regardless, it isn't accepted by my xsl transformer. To solve this problem, I looked at the doxygen configuration variables available to control the xml header, but unfortunately there isn't any to enable/disable the header. So I patched xmlgen.cpp to provide this for me (introduced new config var XML_GENERATEHEADER). I don't know if the doxygen developers want to include this option in the main tree, but here it is (patch is against the current HEAD version in cvs, 1.38) regardless: c:\doxygen>diff -c xmlgen.cpp xmlgen-patched.cpp *** xmlgen.cpp Wed Oct 30 12:57:52 2002 --- xmlgen-patched.cpp Tue Nov 12 14:28:08 2002 *************** *** 56,67 **** { QCString dtdName = Config_getString("XML_DTD"); QCString schemaName = Config_getString("XML_SCHEMA"); ! t << "<?xml version='1.0' encoding='ISO-8859-1' standalone='"; ! if (dtdName.isEmpty() && schemaName.isEmpty()) t << "yes"; else t << "no"; ! t << "'?>" << endl; ! if (!dtdName.isEmpty()) { ! t << "<!DOCTYPE doxygen SYSTEM \"doxygen.dtd\">" << endl; } t << "<doxygen "; if (!schemaName.isEmpty()) --- 56,70 ---- { QCString dtdName = Config_getString("XML_DTD"); QCString schemaName = Config_getString("XML_SCHEMA"); ! if (Config_getBool("XML_GENERATEHEADER")) { ! t << "<?xml version='1.0' encoding='ISO-8859-1' standalone='"; ! if (dtdName.isEmpty() && schemaName.isEmpty()) t << "yes"; else t << "no"; ! t << "'?>" << endl; ! if (!dtdName.isEmpty()) ! { ! t << "<!DOCTYPE doxygen SYSTEM \"doxygen.dtd\">" << endl; ! } } t << "<doxygen "; if (!schemaName.isEmpty()) Regards, --Cormac Twomey ----- Original Message ----- From: "Johan Eriksson" <joh...@ua...> To: "doxygen-develop" <dox...@li...> Sent: Monday, November 11, 2002 11:29 PM Subject: Re: [Doxygen-develop] XML Output > > Kilian Kiko wrote: > > Is it possible to get a single file output instead of a file per class? > This is something that I have wondered about too. I would like to have > the possibility to get it something like the RTF output, divided into > sections and subsections, etc. > > Or does it exist a seperate tool that compile all the XML files into one > large file, alternatively > keeping the XML files but the tool compile it into one document a' la > RTF while viewing or printing. > > /Regards, > Johan > > > So that all classes are just separated through their compounddef tag? > > > > > > > > > ------------------------------------------------------- > This sf.net email is sponsored by:ThinkGeek > Welcome to geek heaven. > http://thinkgeek.com/sf > _______________________________________________ > Doxygen-develop mailing list > Dox...@li... > https://lists.sourceforge.net/lists/listinfo/doxygen-develop > > |
From: Carsten Z. <car...@gr...> - 2002-11-12 09:37:58
|
Hello, I struggle how to extract the parameter type from the xml file using doxmlparser. I get the parameter name from printf("param decl %s\n", param->declarationName()->latin1()); but how to get the paramter type ? Thanks, Carsten |
From: Johan E. <joh...@ua...> - 2002-11-12 07:30:06
|
Kilian Kiko wrote: > Is it possible to get a single file output instead of a file per class? This is something that I have wondered about too. I would like to have the possibility to get it something like the RTF output, divided into sections and subsections, etc. Or does it exist a seperate tool that compile all the XML files into one large file, alternatively keeping the XML files but the tool compile it into one document a' la RTF while viewing or printing. /Regards, Johan > So that all classes are just separated through their compounddef tag? > > |
From: Kilian K. <ki...@ru...> - 2002-11-11 16:56:07
|
Is it possible to get a single file output instead of a file per class? So that all classes are just separated through their compounddef tag? -- Regards, Kilian |
From: Cormac T. <ct...@ct...> - 2002-11-07 01:55:53
|
I have noticed that @see is encoded into xml as: <simplesect kind=">see"> <para> <ref refid="blahblah_1a22" kindref="member">foo()</ref> </para> </simplesect> Now, it is clearly illegal xml to have the > symbol in the literal ">see" on the first line. Is it safe to assume that the greater than symbol squeaked in by accident? Here is the offending line of code, in xmldocvisitor.cpp: void XmlDocVisitor::visitPre(DocSimpleSect *s) { m_t << "<simplesect kind=\">"; // <-- Error switch(s->type()) . . The offending line needs to have the greater-than removed: m_t << "<simplesect kind=\""; --Cormac Twomey |
From: Phil E. <ph...@ja...> - 2002-11-01 22:36:26
|
I built 1.2.18-20021030 (providing the tarballs of the CVS repository is a great idea, by the way), and the output is correct again. On behalf of the libstdc++ maintainers, thanks! Phil -- I would therefore like to posit that computing's central challenge, viz. "How not to make a mess of it," has /not/ been met. - Edsger Dijkstra, 1930-2002 |
From: Dimitri v. H. <di...@st...> - 2002-11-01 08:25:39
|
On Fri, Nov 01, 2002 at 12:32:41AM -0500, Phil Edwards wrote: > trouble.html tells me I can report a bug to the developers mailing list. So > here we go... > > We have a large number of comment blocks that look like this: > > /** > * @if maint > * @brief Some short description. > * @param stuff goes here > * @return stuff goes here > * > * Longer description here. > * @endif > */ > void the_function (the params) {......} > > Note that there is no unconditional text at all. And in 1.2.16, nothing at > all would be generated when "maint" was not enabled. And that was excellent. > > I've just tried 1.2.18, and the function now shows up in the output, with the > special tags being emitted directly as HTML, but with none of the extra text. > The block above, for example, would have a text section consisting of > > @brief@param@return > > with "brief@param" linked as a mailto:. > > > What further information can I provide to help figure out what's going wrong? This bug has already been fixed. Just try the latest CVS release. Regards, Dimitri |
From: Phil E. <ph...@ja...> - 2002-11-01 05:32:56
|
trouble.html tells me I can report a bug to the developers mailing list. So here we go... We have a large number of comment blocks that look like this: /** * @if maint * @brief Some short description. * @param stuff goes here * @return stuff goes here * * Longer description here. * @endif */ void the_function (the params) {......} Note that there is no unconditional text at all. And in 1.2.16, nothing at all would be generated when "maint" was not enabled. And that was excellent. I've just tried 1.2.18, and the function now shows up in the output, with the special tags being emitted directly as HTML, but with none of the extra text. The block above, for example, would have a text section consisting of @brief@param@return with "brief@param" linked as a mailto:. What further information can I provide to help figure out what's going wrong? Phil -- I would therefore like to posit that computing's central challenge, viz. "How not to make a mess of it," has /not/ been met. - Edsger Dijkstra, 1930-2002 |
From: Biswapesh C. <bis...@tc...> - 2002-10-31 04:48:00
|
Hi First of all, thanks for the great tool. I'm using it to produce documentation for my MS project. I have a problem about the chapters layout. Basically, I'd like to divide the main page into multiple chapters. Currently, my PDF file looks like this: Repgen Reference Manual Chapter 1: Repgen Report Generator ---> Introduction ---> Rationale ---> Architecture ---> Design ---> References --->.... Chapter 2: Repgen Data Structures Chapter 3: Repgen File Index etc.. What I'd like to do is seperate Chapter 1 into multiple chapters, i.e. Chapter 1: Introduction Chapter 2: Architecture ... Is this possible ? If so how ? I went through the manuals but did not find anything relevant. Also, if this is not possible, are there any plans to implement this feature ? Thanks in advance. Rgds, Biswa. |
From: Johan E. <joh...@ua...> - 2002-10-29 17:34:57
|
Hi, if I understand the Dot graphs correctly and the possible configuration file options, there is no way today to distinguish the "uses" relation aggregated vs reference?? If this is true, how difficult would it be to make a Doxygen patch to support it? I've looked in the Doxygen sources but I get lost in some of the help classes (UsesClassDef, BaseClassDef, BaseInfo) vs the main classes (ClassDef, MemberDef) that Doxygen uses and the collaboration with the Dot classes. I think the functions findUsedClassesForClass(), addUsedClass(), buildGraph() should be changed some way e.g. by handling a new member variable(aggregated/reference). But in which class should such info go? I've also noticed the member function determineImplUsageRelation() in class ClassDef, which by the sound of it, maybe the intention here is something similar. If this is a feature for doxygen main stream, what would be the best way to patch doxygen? Some simple guidelines would be enough for the time being........ Thank you, Johan PS I know there probably would be a problem, if a class both aggregates and references the same class, but for simplicity I am currently pretending that problem doesn't exist. DS |
From: Olaf S. <rh...@po...> - 2002-10-29 14:34:12
|
On Tue 29 Oct 2002 at 12:08:49 +0100, Olaf Seibert wrote: > Dimitri wrote: > > Looks a bit like the compiler cannot be trusted. What gcc version are you > > using? did you compile the compiler yourself? > > "gcc version egcs-2.91.60 19981201 (egcs-1.1.1 release)". It's the > version that came with the system. By now it's getting rather old, on > the other hand, the system compilers always include some extra fixes. > > I'll try a system with a newer compiler. "gcc version egcs-2.91.66 19990314 (egcs-1.1.2 release)" still fails in the same way, but with "gcc version 2.95.3 20010315 (release) (NetBSD nb3)" it works. All this is for i386 by the way. I'll see if I can try another architecture too (Alpha). -Olaf. -- ___ Olaf 'Rhialto' Seibert - rhialto@ -- Woe betide the one who feels \X/ polderland.nl -- remorse without sin - Tom Poes, "Het boze oog", 4444. |
From: Olaf S. <rh...@po...> - 2002-10-29 11:09:04
|
Dimitri wrote: > Looks a bit like the compiler cannot be trusted. What gcc version are you > using? did you compile the compiler yourself? "gcc version egcs-2.91.60 19981201 (egcs-1.1.1 release)". It's the version that came with the system. By now it's getting rather old, on the other hand, the system compilers always include some extra fixes. I'll try a system with a newer compiler. > Regards, > Dimitri Cheers, -Olaf. -- ___ Olaf 'Rhialto' Seibert - rhialto@ -- Woe betide the one who feels \X/ polderland.nl -- remorse without sin - Tom Poes, "Het boze oog", 4444. |
From: Dimitri v. H. <di...@st...> - 2002-10-28 21:04:01
|
On Mon, Oct 28, 2002 at 05:50:41PM +0100, Olaf Seibert wrote: > I am using doxygen-1.2.18 on NetBSD 1.4.2. I installed it via the pkg > system (compiling from source). Since I wanted to see a bigger example > that the given examples, I wanted to run doxygen on its own sources. So > I did this: > [snip] > > If I add -O2 to the Makefile.libdoxygen then a different crash occurs: > [snip] > > Any ideas? Looks a bit like the compiler cannot be trusted. What gcc version are you using? did you compile the compiler yourself? Regards, Dimitri |
From: Olaf S. <rh...@po...> - 2002-10-28 16:50:51
|
I am using doxygen-1.2.18 on NetBSD 1.4.2. I installed it via the pkg system (compiling from source). Since I wanted to see a bigger example that the given examples, I wanted to run doxygen on its own sources. So I did this: $ cd doxygen-1.2.18/src $ doxygen -g config edited file "config" so that INPUT = ., and no other changes $ doxygen config this ran for some time, the last lines of output are this> Generating code for file util.h... Generating code for file version.h... Generating code for file xmldocvisitor.h... Generating code for file xmlgen.h... Generating file documentation... Generating docs for file defargs.cpp... Generating docs for file main.cpp... Generating docs for file outputlist.cpp... Generating docs for file translator.cpp... Generating class documentation... Generating annotated compound index... Generating hierarchical class index... Generating member index... Generating docs for compound Argument... Generating docs for compound ArgumentList... Generating docs for compound BaseClassDef... Generating docs for compound BaseClassList... Generating docs for compound BaseClassListIterator... Generating docs for compound BaseCodeDocInterface... Segmentation fault $ Examining this from gdb shows that the crash occurs here: Generating docs for compound BaseCodeDocInterface... Program received signal SIGSEGV, Segmentation fault. 0x2a1ca5 in TreeDiagram::drawConnectors (this=0x2cae600, t=@0xbfbfc6a0, image=0x0, doBase=false, bitmap=false, baseRows=1, superRows=4, cellWidth=0, cellHeight=0) at diagram.cpp:816 816 t << protToString(di->protection()) << endl; (gdb) bt #0 0x2a1ca5 in TreeDiagram::drawConnectors (this=0x2cae600, t=@0xbfbfc6a0, image=0x0, doBase=false, bitmap=false, baseRows=1, superRows=4, cellWidth=0, cellHeight=0) at diagram.cpp:816 #1 0x2a6519 in ClassDiagram::writeFigure (this=0xbfbfd068, output=@0x60ca5c, path=0x1486040 "/usr/tmp/devel/doxygen/work/doxygen-1.2.18/src/latex", fileName=0x4dbb360 "classBaseCodeDocInterface") at diagram.cpp:1233 #2 0x159c46 in LatexGenerator::endClassDiagram (this=0x60ca00, d=@0xbfbfd068, fileName=0x4dbb360 "classBaseCodeDocInterface") at latexgen.cpp:1420 #3 0x18c194 in OutputList::forall (this=0x6635a0, func={__delta = 0, __index = 173, __pfn_or_delta2 = {__pfn = (void ( OutputGenerator::*)( ClassDiagram &,char *,char *)) 303104, __delta2 = 0}}, a1=@0xbfbfd068, a2=0x4dbb360 "classBaseCodeDocInterface", a3=0x4d32420 "BaseCodeDocInterface") at outputlist.cpp:305 #4 0x18d57a in OutputList::endClassDiagram (this=0x6635a0, d=@0xbfbfd068, f=0x4dbb360 "classBaseCodeDocInterface", n=0x4d32420 "BaseCodeDocInterface") at outputlist.h:370 #5 0x24e963 in ClassDef::writeDocumentation (this=0x34dd000, ol=@0x6635a0) at classdef.cpp:1012 #6 0x2100d in generateClassList (classSDict=@0x4b02fc) at doxygen.cpp:5547 #7 0x212c6 in generateClassDocs () at doxygen.cpp:5588 #8 0x30784 in generateOutput () at doxygen.cpp:7770 #9 0x1957 in main (argc=2, argv=0xbfbfd5c8) at main.cpp:37 (gdb) list 811 protToMask(di->protection())); 812 } 813 } 814 else // draw pixels 815 { 816 t << protToString(di->protection()) << endl; 817 if (doBase) 818 { 819 t << "1 " << di->xPos()/(float)gridWidth << " " 820 << (di->yPos()/(float)gridHeight+superRows-1) << " in\n"; (gdb) print di $1 = (DiagramItem * ?) 0x2cae500 (gdb) print *di $2 = {children = 0x2cae540, parent = 0x4588940, x = 125, y = 100, num = 0, prot = Public, virt = Normal, templSpec = {<QArray<char>> = {<QGArray> = { shd = 0x4d4c330, _vptr$ = 0x2a85d0{8 vtable entries, 0x2a8408 <QCString::~QCString(void)>, 0x2a85b8 <QArray<char>::detach(void)>, 0x39f214 <QGArray::newData(void)>, 0x39f2b0 <QGArray::deleteData(QGArray::array_data *)>, 0x2a8afc <QArray<char> type_info function>, 0x2a8304 <QArray<char>::~QArray(void)>, 0x2a85b8 <QArray<char>::detach(void)>, 0x39f214 <QGArray::newData(void)>}}, <No data fields>}, <No data fields>}, inList = false, classDef = 0x34e0000} (gdb) info locals x = 0 y = 0 dil = (DiagramItemList * ?) 0x2cae540 parent = (DiagramItem * ?) 0x4588940 di = (DiagramItem * ?) 0x2cae500 this = (TreeDiagram * ?) 0x2cae600 dr = (DiagramRow * ?) 0x2cae5c0 done = false (gdb) If I add -O2 to the Makefile.libdoxygen then a different crash occurs: Generating code for file translator_ua.h... Generating code for file treeview.h... Generating code for file unistd.h... Generating code for file util.h... Generating code for file version.h... Generating code for file xmldocvisitor.h... Generating code for file xmlgen.h... Generating file documentation... Generating docs for file defargs.cpp... Program received signal SIGSEGV, Segmentation fault. 0x51cc8e in QGArray type_info function () (gdb) bt #0 0x51cc8e in QGArray type_info function () #1 0x51bd83 in QGArray::~QGArray () #2 0x2780c1 in MemberDef::isConstructor (this=0x11f9e00) at ../qtools/qarray.h:60 #3 0x268787 in MemberDef::isBriefSectionVisible (this=0x11f9e00) at memberdef.cpp:607 #4 0x27bf0e in MemberList::countDecMembers (this=0x7740d4) at memberlist.cpp:65 #5 0x27e404 in MemberList::writeDeclarations (this=0x7740d4, ol=@0x7d15a0, cd=0x0, nd=0x0, fd=0x774000, gd=0x0, title=0x4ef1af0 "Defines", subtitle=0x0) at memberlist.cpp:416 #6 0x821fb in FileDef::writeDocumentation (this=0x774000, ol=@0x7d15a0) at filedef.cpp:386 #7 0x42042 in generateFileDocs () at doxygen.cpp:5442 #8 0x611ef in generateOutput () at doxygen.cpp:7767 #9 0x1957 in main (argc=2, argv=0xbfbfd5c8) at main.cpp:37 (gdb) Any ideas? -Olaf. -- ___ Olaf 'Rhialto' Seibert - rhialto@ -- Woe betide the one who feels \X/ polderland.nl -- remorse without sin - Tom Poes, "Het boze oog", 4444. |
From: Cormac T. <ct...@ct...> - 2002-10-25 01:32:02
|
Hi, I noticed that xml generation doesn't pass function argument documentation through, unlike html generation. I patched xmlgen.cpp (v1.2.18) to support this (adds a <briefdescription> tag for the argument comment). Note that I looked at this code for the first time today, so I don't have 100% confidence that I'm doing the right thing(tm). But it works for me. The patch would need to be reviewed by someone else though before inclusion. At least it's tiny ;-) Thanks, --Cormac Twomey C:\Foo>diff -c v1.2.18\xmlgen.cpp xmlgen.cpp *** v1.2.18\xmlgen.cpp Sun Sep 15 12:22:50 2002 --- xmlgen.cpp Thu Oct 24 17:51:20 2002 *************** *** 1441,1446 **** --- 1441,1453 ---- linkifyText(TextGeneratorXMLImpl(t),scopeName,md->name(),a->defval); t << "</defval>" << endl; } + if (defArg && defArg->hasDocumentation()) + { + t << " <briefdescription>"; + writeXMLDocBlock(t,md->getDefFileName(),md->getDefLine(),scopeName,md,defArg->docs); + t << "</briefdescription>" << endl; + } t << " </param>" << endl; if (defArg) ++defAli; } |
From: Xavier O. <xav...@an...> - 2002-10-22 10:34:17
|
Here it is, Always late as a typical French. :) Greetings, Xavier -- #-------------------------------------------------------# | Artificial Anthill Project (*) | | http://aanthill.org | #-------------------------------------------------------# | D2SET Science and Technology Association | | http://d2set.org | #-------------------------------------------------------# | Group of exchange regarding imagery technics (French) | | http://fr.groups.yahoo.com/group/imagerie-tech/ | #-------------------------------------------------------# | Employement in imagery: offers and demands (French) | | http://fr.groups.yahoo.com/group/imagerie-emploi/ | #-------------------------------------------------------# (*) We are seaking for webmasters, graphic designer, biologists, biochimists, tranlators. |