Since 9a4eadf, TOC in HTML output appears in the opposite order to that intended, when using --nosort

👋 just a friendly reminder that the release source archive did not get uploaded https://rfsber.home.xs4all.nl/Robo/archives/, not quite sure if that is intentional or not. Thanks!

Hello, both using automake or autoconf, here the error:

automake: warning: autoconf input should be named 'configure.ac', not 'configure.in'
automake: warning: 'configure.ac' and 'configure.in' both present.
automake: proceeding with 'configure.ac'
configure.ac: error: no proper invocation of AM_INIT_AUTOMAKE was found.
configure.ac: You should verify that configure.ac invokes AM_INIT_AUTOMAKE,
configure.ac: that aclocal.m4 is present in the top-level directory,
configure.ac: and that aclocal.m4 was recently regenerated (using aclocal)
configure.ac:32: error: required file './config.guess' not found
configure.ac:32: 'automake --add-missing' can install 'config.guess'
configure.ac:32: error: required file './config.sub' not found
configure.ac:32: 'automake --add-missing' can install 'config.sub'
Docs/makefile.am:5: error: FINK does not appear in AM_CONDITIONAL
Docs/makefile.am:17: error: FINK does not appear in AM_CONDITIONAL
Source/makefile.am: error: required file './depcomp' not found
Source/makefile.am: 'automake --add-missing' can install 'depcomp'
/usr/share/automake-1.14/am/depend2.am: error: am__fastdepCC does not appear in AM_CONDITIONAL
/usr/share/automake-1.14/am/depend2.am: The usual way to define 'am__fastdepCC' is to add 'AC_PROG_CC'
/usr/share/automake-1.14/am/depend2.am: to 'configure.ac' and run 'aclocal' and 'autoconf' again
/usr/share/automake-1.14/am/depend2.am: error: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.14/am/depend2.am: The usual way to define 'AMDEP' is to add one of the compiler tests
/usr/share/automake-1.14/am/depend2.am: AC_PROG_CC, AC_PROG_CXX, AC_PROG_OBJC, AC_PROG_OBJCXX,
/usr/share/automake-1.14/am/depend2.am: AM_PROG_AS, AM_PROG_GCJ, AM_PROG_UPC
/usr/share/automake-1.14/am/depend2.am: to 'configure.ac' and run 'aclocal' and 'autoconf' again
automake: warning: autoconf input should be named 'configure.ac', not 'configure.in'
automake: warning: 'configure.ac' and 'configure.in' both present.
automake: proceeding with 'configure.ac'
makefile.am:25: error: FINK does not appear in AM_CONDITIONAL
makefile.am:44: error: FINK does not appear in AM_CONDITIONAL

Hello! I am Anderson, a Nixpkgs commmiter.

When I was writing the expression, the compiler terminated with an error message related to "warnings treated as errors". I circumvented it by loosening the compiler constraints.

Here is the output of compilation with format-security enabled:

build flags: SHELL=/nix/store/cwnwyy82wrq53820z6yg7869z8dl5s7g-bash-4.4-p23/bin/bash
Making all in Source
make[1]: Entering directory '/mnt/sources/z-sandbox/0-descartavel/source/Source'
make  all-am
make[2]: Entering directory '/mnt/sources/z-sandbox/0-descartavel/source/Source'
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT optioncheck.o -MD -MP -MF .deps/optioncheck.Tpo -c -o optioncheck.o optioncheck.c
mv -f .deps/optioncheck.Tpo .deps/optioncheck.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT analyser.o -MD -MP -MF .deps/analyser.Tpo -c -o analyser.o analyser.c
mv -f .deps/analyser.Tpo .deps/analyser.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT generator.o -MD -MP -MF .deps/generator.Tpo -c -o generator.o generator.c
generator.c: In function ‘Generate_Item’:
generator.c:2051:13: warning: ignoring return value of ‘system’ declared with attribute ‘warn_unused_result’ [-Wunused-result]
 2051 |             system( line );
      |             ^~~~~~~~~~~~~~
mv -f .deps/generator.Tpo .deps/generator.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT headers.o -MD -MP -MF .deps/headers.Tpo -c -o headers.o headers.c
mv -f .deps/headers.Tpo .deps/headers.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT headertypes.o -MD -MP -MF .deps/headertypes.Tpo -c -o headertypes.o headertypes.c
mv -f .deps/headertypes.Tpo .deps/headertypes.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT items.o -MD -MP -MF .deps/items.Tpo -c -o items.o items.c
mv -f .deps/items.Tpo .deps/items.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT links.o -MD -MP -MF .deps/links.Tpo -c -o links.o links.c
mv -f .deps/links.Tpo .deps/links.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT globals.o -MD -MP -MF .deps/globals.Tpo -c -o globals.o globals.c
mv -f .deps/globals.Tpo .deps/globals.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT robodoc.o -MD -MP -MF .deps/robodoc.Tpo -c -o robodoc.o robodoc.c
mv -f .deps/robodoc.Tpo .deps/robodoc.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT util.o -MD -MP -MF .deps/util.Tpo -c -o util.o util.c
util.c: In function ‘RB_ReadWholeLine’:
util.c:1037:9: warning: ignoring return value of ‘fgets’ declared with attribute ‘warn_unused_result’ [-Wunused-result]
 1037 |         fgets( buf, MAX_LINE_LEN, file );
      |         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
util.c: In function ‘RB_Change_Back_To_CWD’:
util.c:1199:9: warning: ignoring return value of ‘chdir’ declared with attribute ‘warn_unused_result’ [-Wunused-result]
 1199 |         chdir( cwd );
      |         ^~~~~~~~~~~~
util.c: In function ‘RB_Change_To_Docdir’:
util.c:1227:5: warning: ignoring return value of ‘getcwd’ declared with attribute ‘warn_unused_result’ [-Wunused-result]
 1227 |     getcwd( tmp, sizeof( tmp ) );
      |     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
util.c:1234:5: warning: ignoring return value of ‘chdir’ declared with attribute ‘warn_unused_result’ [-Wunused-result]
 1234 |     chdir( tmp );
      |     ^~~~~~~~~~~~
mv -f .deps/util.Tpo .deps/util.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT html_generator.o -MD -MP -MF .deps/html_generator.Tpo -c -o html_generator.o html_generator.c
mv -f .deps/html_generator.Tpo .deps/html_generator.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT xmldocbook_generator.o -MD -MP -MF .deps/xmldocbook_generator.Tpo -c -o xmldocbook_generator.o xmldocbook_generator.c
mv -f .deps/xmldocbook_generator.Tpo .deps/xmldocbook_generator.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT latex_generator.o -MD -MP -MF .deps/latex_generator.Tpo -c -o latex_generator.o latex_generator.c
mv -f .deps/latex_generator.Tpo .deps/latex_generator.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT rtf_generator.o -MD -MP -MF .deps/rtf_generator.Tpo -c -o rtf_generator.o rtf_generator.c
mv -f .deps/rtf_generator.Tpo .deps/rtf_generator.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT test_generator.o -MD -MP -MF .deps/test_generator.Tpo -c -o test_generator.o test_generator.c
mv -f .deps/test_generator.Tpo .deps/test_generator.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT ascii_generator.o -MD -MP -MF .deps/ascii_generator.Tpo -c -o ascii_generator.o ascii_generator.c
mv -f .deps/ascii_generator.Tpo .deps/ascii_generator.Po
gcc -DHAVE_CONFIG_H -I.     -g -O2 -MT troff_generator.o -MD -MP -MF .deps/troff_generator.Tpo -c -o troff_generator.o troff_generator.c
troff_generator.c: In function ‘RB_TROFF_Generate_Item_Line_Number’:
troff_generator.c:563:5: error: format not a string literal and no format arguments [-Werror=format-security]
  563 |     fprintf( dest_doc, line_number_string );
      |     ^~~~~~~
troff_generator.c: In function ‘RB_TROFF_Generate_Header_Start’:
troff_generator.c:226:13: warning: ignoring return value of ‘symlink’ declared with attribute ‘warn_unused_result’ [-Wunused-result]
  226 |             symlink( basename( manpage ), buf );
      |             ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1: some warnings being treated as errors
make[2]: *** [makefile:461: troff_generator.o] Error 1
make[2]: Leaving directory '/mnt/sources/z-sandbox/0-descartavel/source/Source'
make[1]: *** [makefile:323: all] Error 2
make[1]: Leaving directory '/mnt/sources/z-sandbox/0-descartavel/source/Source'
make: *** [makefile:540: all-recursive] Error 1

Disabling this flag the project runs fine.

Systemverilog has a "begin" keyword.
One of my variables is named "be", When Robodoc parses it, it autocompletes it to begin for some weird reason.

Hello,
after install ROBODoc on my virtual machine with Ubuntu 12.04, I have tried the special tool robohdrs, but when I have ran it on a c test file it returns "execlp failed no tags".
After some research I found that I didn't have ctags installed.
I'm a kinda of newbie and I didn't know ctags, but I wonder if there's a chance to add a test when run ./configure.

Luca

Currently, all signs and keywords use the same css classes span.sign and span.keyword, respectively. I'd like to have finer control to give code elements more specific styling.

A possible solution could be to specify the class to use for a specific code element in the config file…

# robodoc.rc

# syntax: <element(s)> <css_classname>
signs:
  {} sign_braces
  () sign_parens
  =  sign_assignment

# syntax: <keyword(s)> <css_classname>
keywords:
  new               keyword_new
  return            keyword_return
  if, elseif, else  keyword_conditional
  try, catch, throw keyword_exception

We could then customize those css classes…

/* stylesheet */
span.sign_braces{color:red}
span.sign_parens{color:blue}
span.sign_assignment{color:#FF00FF}

span.keyword_new{color:#00FFFF}
span.keyword_return{green}
span.keyword_conditional{color:yellow}
span.keyword_exception{color:red}

This is #69 from the sourceforge page.

In FORTRAN I can overload a function and have multiple interfaces for the same function, for example:

INTERFACE write_data
MODULE PROCEDURE write_int
MODULE PROCEDURE write_real
END INTERFACE

SUBROUTINE write_int( InVar )
INTEGER :: Invar
END SUBROUTINE write_int
SUBROUTINE write_real( InVar )
REAL :: Invar
END SUBROUTINE write_int

Now I want to provide using the SOURCE keyword the interface for write_data, and not the source for write_int and write_real subroutines. The problem is the source interface does not exists for write_data.

It would be nice if I could do something like

!****s* H5A write_data
!
! NAME
! write_data
!
! INPUT
! Myvalue - Print value
!
! SOURCE
! SUBROUTINE write_data(Myvalue)
!*****

and have items in the SOURCE block written as source code without the "!"

I'd like to be able to write jade pipes…

* |jade
* #myDiv.myclass div content
*     p(style='border:1px solid red') paragraph inside div
* |jade end

which would output…

<div id="myDiv" class="myClass">
    <p style="border:1px solid red">paragraph inside div</p>
</div>

see jade-lang.com

I have an issue though with the way it (doesn't) analyze namespaces. In my Systemverilog code, I have 2 packages that have a class named driver. Moreover as they both inherit from uvm_driver, they hold the same functions: build_phase, connect_phase.

At first I gave the section names the same hierarchy as the code namespace. So I had
my_project/event_package
my_project/host_package
event_package/driver
host_package/driver

But this caused some confusion when building the TOC, as event_package/driver held driver/constraint_x but in the TOC, it would be under host_package/driver.

So I renamed the sections which is a shame, but I devided as long as the section NAME would hold the actual name of the class this would be OK.

So now, none of the code/comments hyperlinks driver which I can live with. But some of the code obviously references the functions and again I have the same issue: Both calls to build_phase hyperlinks to the same class. Although they should reference separate classes.

The "table of contents" page ("toc_index.html", generated via --toc) lists the top-level items in reversed alphabetical order, which is a bit awkward.

I get the following error when I try and run ROBODoc after our system was upgraded from redhat to CentOS.

-bash: /robodoc/Source/robodoc: /lib/ld-linux.so.2: bad ELF interpreter: No such file or directory

I'd like to be able to do something like…

* |html
* arbitrary html...
* arbitrary html...
* arbitrary html...
* arbitrary html...
* |html end

Hello,
the ROBODoc build does not find the "sys/unistd.h" header in a AIX environment.

"troff_generator.c", line 33.10: 1506-296 (S) #include file <sys/unistd.h> not found.
"troff_generator.c", line 225.13: 1506-1421 (W) Implicit function declaration for function "unlink".

I modified "Source/troff_generator.c" to fix the problem therefore

#include <sys/param.h>
#if defined(_AIX)
#include <unistd.h>
#else
#include <sys/unistd.h>
#endif

Based on the content of "sys/unistd.h" (in a Linux environment) I suggest to use "unistd.h" in general,
like in "Source/document.c" or "Source/robodoc.c", "Source/robohdrs.c".
#include <sys/unistd.h>
Should changed to
#include <unistd.h>

See also Issue #22 Does not build on macOS as unistd.h is not included

With regards,
Lutz

p.s.
The used build steps based on the github sources.
autoreconf -fi
./configure CC="c99 -D_XOPEN_SOURCE"
make

I need to link to the generated HTML docs for specific API functions from a manually edited web page. As described by the user manual, these links need the escaped version of the whole header name, for example, Analyser2fRB5fToBeAdded for Analyser/RB_ToBeAdded. When creating the links from my web page, I have to type these in manually, but I would find it much easier if the generated documentation itself would contain links which use this label type, because then I could right click on the link and copy the link address.

As far as I know, there is no option to use these labels in the generated documentation links, but I think it would be very useful.

When using LaTeX output (i.e. --latex), the option --sectionnameonly has no effect.
It can very easily be implemented though, by replacing in Source/latex_generator.c:

cur_header->name

with

course_of_action.do_sectionnameonly ? cur_header->function_name : cur_header->name

The combination --latex --sections also has no effect, but that bothers me less.

Should fail the same way on many other POSIX systems as unlink and symlink are documented to be declared in unistd.h which isn't ever included.

troff_generator.c:225:13: error: implicit declaration of function 'unlink' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
            unlink( buf );
            ^

troff_generator.c:226:13: error: implicit declaration of function 'symlink' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
            symlink( basename( manpage ), buf );
            ^

See POSIX reference to confirm that those functions are supposed to be declared by unistd.h

Also see man page:

SYNOPSIS
     #include <unistd.h>

     int
     unlink(const char *path);

Source :man 2 unlink

Note that sys/unistd.h is no official header, neither by C nor by POSIX standard, you should never be required to include that one. Same holds true for sys/param.h. These headers are automatically included for you if you included the correct standard headers.

I'm not found ./configure file to compilate source code.

Where's file please

Best Regards

I was thinking a good feature would be this: to have a new block type that allows you to put sourcecode within a commented portion of the header, and still have it be syntax highlighted as though it were not commented. Here is an example:

!*****************************************************************************************
!****f* json_module/json_create_array
!
!  NAME
!    json_create_array
!
!  DESCRIPTION
!    Allocate a json_value pointer and make it an array variable.
!    The pointer should not already be allocated.
!
!  EXAMPLE
!    type(json_value),pointer :: p
!    call json_create(p,'arrayname')
!
!  SOURCE

    subroutine json_create_array(me,name)

    implicit none

    type(json_value),pointer            :: me
    character(kind=CK,len=*),intent(in) :: name

    call json_value_create(me)
    call to_array(me,name)

    end subroutine json_create_array
!*****************************************************************************************

Currently, the output looks like this

What I would like is to be able to indicate that the EXAMPLE block should be syntax highlighted, excluding the first comment character. Note that I can't put it in the source items block, because it would highlight each line as a comment. It might also be useful to be able to specify some different settings for this type of block (for example, perhaps I wouldn't want to show the line numbers for these, but I do for the main SOURCE).

no Headers directory in repo as indicated in README