gmane.comp.web.rdf.redland.devel

Re: Bug in raptor_vsnprintf()

John Emmas — 2012-05-24T10:49:42


Yes Lauri, "pull" is what I needed.  Still getting used to the terminology.  BTW, regarding my previous issue with the "non Windows compatible" URIs.....


On 23 May 2012, at 16:43, Lauri Aalto wrote:



I tried your suggestion this morning and I think it should work.  However along the way, I realised that a slight change to 'raptor_parser_parse_file_stream()' might be beneficial :-

At line 535 (raptor_parse.c) instead of these two tests:-

  if(!stream || !base_uri)
    return 1;

I simplified the code to this:-

  if(!stream)
    return 1;

I made the change because 'base_uri' doesn't necessarily need to be non-zero for correct functionality.  A few lines further down, 'rapter_parser_parse_start()' will check the variable 'base_uri' for us and will return with an error status if a value was needed but NULL got supplied.  It's not a bug as such but it's helpful in our situation because we won't always be providing a base uri.

John_______________________________________________
redland-dev mailing list

Re: Bug in raptor_vsnprintf()

Lauri Aalto — 2012-05-24T09:31:45

If you check the GitHub web interface commit history
<https://github.com/dajobe/raptor/commits/master>, you'll see that
there are a lot of commits after April 13. Commits are visible to
everyone when they're pushed to GitHub, there's no delay.

I don't know what terminology your git client is using, but the
standard git terminology is "clone" for svn checkout like
functionality and "pull" i.e. "fetch + merge" for svn update. "Sync"
is not a standard git operation and the "ahead origin/master" message
suggests the operation is trying to "push" local commits to master
branch of a remote repo called origin, not pull in commits from a
remote to local repo.

Lauri

On Thu, May 24, 2012 at 10:08 AM, John Emmas <johne53-IWqWACnzNjwqdlJmJB21zg< at >public.gmane.org> wrote:

Re: Bug in raptor_vsnprintf()

John Emmas — 2012-05-24T07:08:31


Hmmm...  I'm confused.  This is my first time using Git so I could well be doing something wrong.  However, although I checked out from https://github.com/dajobe/raptor (i.e. originally - about a week ago) my Git client is telling me that there are no new commits.  I'm assuming that Git Sync is similar to svn update which I'm more used to using.  However, when I tried a new sync this morning I just got the message "nothing ahead origin/master".  The most recent update seems to have been 13th April.

Shouldn't I be seeing your commit by now?

John

Re: Bug in raptor_vsnprintf()

John Emmas — 2012-05-23T18:15:21


That sounds like a good compromise Lauri.  Thanks for the tip.

John

Re: Bug in raptor_vsnprintf()

Lauri Aalto — 2012-05-23T15:43:13

Hi John,

generally, Windows support is not too actively maintained so it may
contain various bugs and quirks. Patches are welcome.

As a workaround, you can open the FILE* yourself and use one of the
parse_file_handle functions to process it.

Lauri


On Wed, May 23, 2012 at 6:28 PM, John Emmas <johne53-IWqWACnzNjwqdlJmJB21zg< at >public.gmane.org> wrote:

Re: Bug in raptor_vsnprintf()

John Emmas — 2012-05-23T15:28:24


Thanks Lauri.  I'll update from GIT later today and in future, I'll report bugs to the right place.

I've found something else which might be a bug (well, an oversight really) but before I report it, let me describe it here in case I'm misunderstanding something.

I'm working on a cross-platform app which uses lrdf to open and parse some files.  'lrdf_read_file()' eventually defers to 'raptor_parser_parse_file()' to open the actual disk file.  'raptor_parser_parse_file()' accepts a URI to identify the file to be opened.  According to this document by Tim Berners-Lee:-

http://tools.ietf.org/html/rfc3986

the specification states that before conversion to URI, a character string must first get converted to UTF-8 (see the final paragraph of Section 2 - just before Section 3).

'raptor_parser_parse_file()' calls 'raptor_uri_uri_string_to_filename()' to produce a string that identifies the local file.  However, after decoding, that string will be in UTF-8 format.  UTF-8 format seems to work for Linux and OS-X

Re: Bug in raptor_vsnprintf()

Lauri Aalto — 2012-05-23T09:46:28

Hi John,

thanks for the report. The best place for bug reports is
<http://bugs.librdf.org>.

I fixed this particular issue in commit
<https://github.com/dajobe/raptor/commit/391e09a5acddaf3ed63005bec880804896ca860e>.

Normally variable initialization bugs like this are detected by clang
static analyzer ("make analyze" in src) but it only analyzes the
active configuration so this one went undetected.

Lauri

On Tue, May 22, 2012 at 2:47 PM, John Emmas <johne53-IWqWACnzNjwqdlJmJB21zg< at >public.gmane.org> wrote:

Bug in raptor_vsnprintf()

John Emmas — 2012-05-22T11:47:47

I've found a bug in raptor_vsnprintf2() (raptor2).  Where should I report it?  For anyone who's curious, here's the problem:-


int
raptor_vsnprintf2(char *buffer, size_t size,
                  const char *format, va_list arguments)
{
  size_t len;

  RAPTOR_ASSERT_OBJECT_POINTER_RETURN_VALUE(format, char*, 0);

#ifdef CHECK_VSNPRINTF_RUNTIME

  if(vsnprintf_is_c99())
    VSNPRINTF_C99_BLOCK(len, buffer, size, format, arguments) ;
  else
    VSNPRINTF_NOT_C99_BLOCK(len, buffer, size, format, arguments) ;

#else

#ifdef HAVE_C99_VSNPRINTF
  VSNPRINTF_C99_BLOCK(len, buffer, size, format, arguments) ;
#else
  VSNPRINTF_NOT_C99_BLOCK(len, buffer, size, format, arguments) ;
#endif

#endif
  
  return RAPTOR_BAD_CAST(int, len);
}

Note that 'len' is uninitialised when it gets passed to either of the macros (VSNPRINTF_NOT_C99_BLOCK, in my case).  In abbreviated form, the macro translates to this code:-

#define VSNPRINTF_NOT_C99_BLOCK(len, buffer, size, format, arguments)   \
  do {

[Rasqal RDF Query Library 0000506]: configure.ac hasbash -specific (non-POSIX) == syntax

Mantis Bug Tracker — 2012-05-19T20:53:30

The following issue has been SUBMITTED. 
====================================================================== 
http://bugs.librdf.org/mantis/view.php?id=506 
====================================================================== 
Reported By:                ssuominen
Assigned To:                
====================================================================== 
Project:                    Rasqal RDF Query Library
Issue ID:                   506
Category:                   portability
Reproducibility:            always
Severity:                   trivial
Priority:                   high
Status:                     new
Query Language:              
====================================================================== 
Date Submitted:             2012-05-19 20:53
Last Modified:              2012-05-19 20:53
====================================================================== 
Summary:                    configure.ac has bash -specific (non-POSIX) ==
syntax
Description: 
Line 1124 of configure.ac has

Re: First use of raptor2 (Windows)

John Emmas — 2012-05-15T10:19:01


Thanks Steve,

I've attached the project file and sln file for VC++8 in case Bob or Andras might want to try them.  I was going to test it myself by building the raptortest project but the source code seems to be out of date (raptortest.cpp is using an old struct which has changed since it was originally built).  As far as our end-user app is concerned though, my raptor library seems to be working, so far.

John








_______________________________________________
redland-dev mailing list
redland-dev-cunTk1MwBs+qEqLfThIrmQ< at >public.gmane.org
http://lists.librdf.org/mailman/listinfo/redland-dev

Re: First use of raptor2 (Windows)

Steve Harris — 2012-05-15T09:18:49


They should be generated by flex (turtle_lexer.c) and yacc/bison (turtle_parser.c), so would be in the tarball, but not in Git.

- Steve

Re: First use of raptor2 (Windows)

John Emmas — 2012-05-15T09:04:01



On 10 May 2012, at 23:28, Bob DuCharme wrote:


Hello Dave & Bob,

I've done those manual edits and I'm hoping that I'll have a new MSVC project file ready for libraptor in the next few hours (it'll only be for raptor itself - not for rasqal / librdf etc).  It'll be for VC++ 8 and will be similar to the original one I got from Andras.

Unfortunately, my mind's gone blank about some of the turtle modules....  I remember noticing that turtle_parser.c and turtle_lexer.c were missing from the sources that I downloaded from GIT - but I can't remember whether or not that was deliberate.  They were present in the 2.0.7 source tarball.  Do I need to add them back or have they been deliberately removed now?

John

[Raptor RDF Syntax Library 0000505]: Parsing certainescaped unicode strings in Turtle cases an error

Mantis Bug Tracker — 2012-05-14T12:57:52

The following issue has been SUBMITTED. 
====================================================================== 
http://bugs.librdf.org/mantis/view.php?id=505 
====================================================================== 
Reported By:                swh
Assigned To:                
====================================================================== 
Project:                    Raptor RDF Syntax Library
Issue ID:                   505
Category:                   api
Reproducibility:            have not tried
Severity:                   minor
Priority:                   normal
Status:                     new
Syntax Name:                Turtle 
====================================================================== 
Date Submitted:             2012-05-14 12:57
Last Modified:              2012-05-14 12:57
====================================================================== 
Summary:                    Parsing certain escaped unicode strings in Turtle
cases an error
Description: 
Trying to parse th

ANNOUNCEMENT: Rasqal RDF Query Library 0.9.29

Dave Beckett — 2012-05-14T03:53:51

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

   Rasqal RDF Query Library 0.9.29
      http://librdf.org/rasqal/

Rasqal is a free software / Open Source C library that handles Resource
Description Framework (RDF) query language syntaxes, query construction
and execution of queries returning results as bindings, boolean, RDF
graphs/triples or syntaxes. The supported query languages are:
  SPARQL 1.0[3], RDQL (will be removed in 0.9.30),
  SPARQL Query 1.1[4], SPARQL Update 1.1[5] (no executing) and
  Experimental SPARQL extensions (LAQRS).
Rasqal can write binding query results in the following formats:
  SPARQL XML, SPARQL JSON, CSV, TSV, HTML, ASCII tables, RDF/XML and
  Turtle / N3
and read them in:
  SPARQL XML, RDF/XML and Turtle / N3.

Rasqal was designed to work closely with the Redland librdf[1] RDF
library and the Raptor[2] RDF Syntax Library but is entirely separate
from both.

This is a beta quality library - the code is mature, the API is
mostly stable but changing slowly.  Rasqal supports a

Re: First use of raptor2 (Windows)

John Emmas — 2012-05-10T07:25:05


Thanks Dave!

Re: First use of raptor2 (Windows)

David Brooks — 2012-05-10T06:42:55

The link is on the main Raptor web page, http://librdf.org/raptor/  -- 
on LH side, near the bottom. Ditto for the other Redland components.


Regards,
Dave

On 10/05/12 6:30 PM, John Emmas wrote:

Re: First use of raptor2 (Windows)

John Emmas — 2012-05-10T06:39:20


Oops - one further question if i may....

What's the difference between rapper and librdf?  Do they have completely different uses or are they essentially just different ways of doing the same job?

John

Re: First use of raptor2 (Windows)

John Emmas — 2012-05-10T06:30:44

One more question about raptor2.....

I believe that the source code is available through git?  I only installed git yesterday so I haven't yet figured out how to find the URL for a git repository. Where do I need to point my git client so that it can check out the raptor2 source code?

John

Re: First use of raptor2 (Windows)

John Emmas — 2012-05-09T09:51:46

On 8 May 2012, at 15:04, Dave Beckett wrote:


We're still testing our build which has thrown up a few problems but once it's stable I'd be happy to share the project files and any modifications I need to make to the sources (don't think I've had to make any so far though - my modifications have been confined to the VC project).

Having said that...  I'm not sure how much use my VC project will be to anyone else.  VC projects are only really portable if they're pretty self-contained.  libraptor depends on external libs such as libcurl, libxml2, libxslt etc - so anybody else would need to adjust the project if they're using different versions from the ones I'm using (or if their folder structure differs from mine).  The adjustments should be quite trivial though.

John

Re: First use of raptor2 (Windows)

Dave Beckett — 2012-05-08T14:04:42

Those files were removed as part of 2.0.7 - expat support (alternative to
libxml) was removed, NFC checking was removed. The release notes for 2.0.7
describe this in more detail.


This is declared in raptor2.h which is included by raptor_uri.c at the top.


This is fixed in GIT already, after the 2.0.7 release.


Have you got the final project files, source, binaries you could share?  I
can host them.

Dave

Re: First use of raptor2 (Windows)

John Emmas — 2012-05-08T06:37:39

Many thanks for the advice, guys - especially to Andras.  Your modules were a huge help.  Thanks very much.

As it happens, I'm building a slightly newer version than Andras (2.0.7) so there were a few small differences although they're mostly documented here:-

http://librdf.org/raptor/api/raptor2-changes-2-0-6-to-2-0-7.html

I don't know if it's significant but I was unable to find some files from Andras's project, namely:-

raptor_expat.c
raptor_nfc.c
raptor_nfc_data.c

Have they been removed between 2.0.6 and 2.0.7?  If so, it might be handy to add them to that change document.  Apart from that, everything went very smoothly although I did notice a couple of minor issues in the file 'raptor_uri.c' so I thought I'd feed them back to the devs:-

1) The function 'raptor_uri_uri_string_to_counted_filename_fragment()' needs to be declared somewhere (I just declared it at the top of raptor_uri.c).  This is needed for the obvious reason (it gets called before its definition).

2) There's a problem in this block