W3C home > Mailing lists > Public > uri@w3.org > February 2008

Re: uriparser 0.6.2 released

From: Michael A. Puls II <shadow2531@gmail.com>
Date: Sat, 9 Feb 2008 16:49:07 -0500
Message-ID: <6b9c91b20802091349k7e491aa9medd977b338ab86c3@mail.gmail.com>
To: "Sebastian Pipping" <webmaster@hartwork.org>
Cc: uriparser-devel@lists.sourceforge.net, uri@w3.org

On 2/9/08, Sebastian Pipping <webmaster@hartwork.org> wrote:
> Michael A. Puls II wrote:
> > After converting the indents to tabs for the attachment, the makefile
> > does indeed work in the win32/mingw directory.
>
> Hm, could it be you copy-pasted from the mail instead of
> saving the attachment? The file did contain tabs for indentation.

Gmail showed it inline, so who knows what it did to the tabs. No biggy
though. I can sort that out.

> > However, would it be
> > better to have the .o files written somewhere else like in
> > "mingw/.objs/lib" similar to how code blocks does it when you load and
> > build the uriparser project.?
>
> Would be an option.
>
>
> > Also, having the makefile build the tests or a separate makefile build
> > the tests would be cool too.
>
> I agree.
>
>
> > Also, I did try to use MSYS and do ./configure, but I got a sed error.
> > Is it supposed to work with MSYS?
>
> No sure, I don't use MSYS.

I'll will try to investigate this for myself and if I come up with
anything, I'll show how I did it.

> > Also, I noticed that when building UriNormalize.c, there are some
> > warnings in uriFixPercentEncodingInplaceA. Are those O.K. You should
> > see the warnings when you build?
>
> Just lack of time, will take care of that later.

O.K. cool

> > Also, with the mingw makefile, did you check to see if the produced
> > liburiparser.a is correct? Does it have to be done statically using
> > "ar" like the following instead?
>
> I did not try, no. I let Code::Blocks or libtool take care of that
> usually. I think ar should be the one, for now you just have to try.

O.K.

> > Should the makefile give an option to produce release and debug
> > versions and shared and static versions etc.?
>
> Yes, why not.

:)

> The thing is I have to manually sync file
> lists at several places (Makefile.am, Code::Blocks, Visual Studio)
> So I'm not really looking for an increase of that list.
> Also I use Automake to not have to write Makefiles by hand.
> So that's why I originally asked you to either use Code::Blocks
> or help with maintenance.

Understood. I am able to build the lib with codeblocks at least.
Haven't tried building the test. Will investigate the other ways and
show you if I come up with anything.

> > Also, what flags can be passed when compiling? Any unicode flags?
>
> Building uriparser should produce Ansi and Unicode code at the
> same time regardless of any Unicode defines.

cool.

> >> Are you interested in adding support for make check
> >> and co-maintain the Makefile in the future? If so please
> >> tell me your nick on SourceForge so I can add you to the
> >> project.
> >
> > If I can figure things out, will let you know. I want to mess with the
> > parser first. I don't know how to use it at all yet.
>
> Sorry, tutorials on usage are still to come. For now
> looking at the test suite or in libSpiff Code is your
> best source besides talking to me.

O.K. No rush or anything. Just want to check the parser out.

> > For starters, I want to figure out how to just pass in a URI and get
> > the different parts. Have an example of that? (I see the doc
> > directory. Do I have to build the documentation too?)
>
> You don't necessarily have to build it as it's online at
> http://uriparser.sourceforge.net/doc/html/

O.K., cool.

> Currently to parse a URI you have code like this:
>
>     UriParserStateA state;
>     UriUriA uri;
>     state.uri = &uri;
>     if (uriParseUriA(&state, "file:///home/user/song.mp3") != URI_SUCCESS) {
>         /* FAILURE */
>     }
>     /* DO STUFF */
>     uriFreeUriMembersA(&uri);

Will try this out.

Thanks

-- 
Michael
Received on Saturday, 9 February 2008 21:49:17 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:25:12 UTC