----- Forwarded message from Mohammed Sameer <[EMAIL PROTECTED]> -----
From: Mohammed Sameer <[EMAIL PROTECTED]>
To: Development Discussions <[EMAIL PROTECTED]>
Reply-To: Mohammed Sameer <[EMAIL PROTECTED]>,
Development Discussions <[EMAIL PROTECTED]>
Subject: Re: Developer Guide
Delivery-date: Sun, 11 Jan 2004 00:50:31 +0200
X-BeenThere: [EMAIL PROTECTED]
X-Mailman-Version: 2.1.3
List-Unsubscribe: <http://lists.arabeyes.org/mailman/listinfo/developer>,
<mailto:[EMAIL PROTECTED]>
List-Archive: </archives/developer>
List-Post: <mailto:[EMAIL PROTECTED]>
List-Help: <mailto:[EMAIL PROTECTED]>
List-Subscribe: <http://lists.arabeyes.org/mailman/listinfo/developer>,
<mailto:[EMAIL PROTECTED]>
On Sun, Dec 21, 2003 at 03:44:25AM -0800, Mohammed Elzubeir wrote:
> >
> > Makefile.cvs: why not use the ./autogen.sh script ?
> > Most of the projects i see use an autogen.sh script. I didn't find the
> > Makefile.cvs except in Qt apps.
> > and why run make -f Makefile.cvs when i can simply ./autogen.shh ?
> > with a proper autogen.sh you can pass arguments to the "configure"
> > script. That's one.
> > The second is that writing a proper Makefile.cvs forces the developer to
> > learn the syntax of Makefiles, That's bad.
> > and with the ./autogen script you can produce meaningfull errors and
> > instructions on where to get the missing packages from the autotools "I
> > think this can be done with the Makefile.cvs, but perhaps not that easy
> > ?"
> > I can't think of more reasons as I'm typing this in a hurry!
>
> Again, the requirement is to have a Makefile.cvs present. Having an
> autogen.sh boostrap.sh or whatever you would like to call it is entirely
> up to you.
>
And why don't we use the autogen.sh ? Most of the projects are using it
already.
> Learning how to write Makefiles should not be a problem. If you are a
> developer and don't know how to write a Makefile then you have some
> serious problems ;)
We are not judging the development skills here, And we are not forcing
the developer to learn new things just because we want to
>
> > I don't see a reason for the RCS headers ? why is it required ??
>
> It just is. I can give reasons, but we can argue over it for years to
> come. Consistency, forcing the user to explain what the file does,
> making it easy for others (and the developer) to know what revision he's
> looking at.. are among a few reasons that come to mind.
>
I don't see these necessary, but it's Ok.
> > doxygen ? I don't think that's required if I'm not writing a library.
>
> The Doxygen requirement is only there to force commenting the code. The
> comments should be done using the Doxygen style comments.. which are
> fairly easy and flexible. No one is required to use all the features
> that come with Doxygen (they are pretty elaborate). Simply commenting
> each and every function at a minimum.
I have a suggestion here:
I'm using Gtk, I think using gtk-doc 'd be better than doxygen
Another developer is using Qt/Kde, then use their documentation
framework.
etc...
Only use doxygenn if No documentation framework is there for the toolkit
used/or if you are developing a console application.
>
> > ChangeLog, Comeon guys DON'T complicate things. It's important but I
> > don't think it's mandatory.
>
> It is mandatory if you intend to make a release. I don't remember when
> was the last time I grabbed a prgram that did not have a changelog file.
> If there is one that did not have a changelog file, I would not be
> taking it seriously.. since the author would imply that he is not
> tracking its progress and so he is not serious enough.
>
Does this have to be in the standard ChangeLog format specified by the
GNU project ??
> > even for stable and odd for development. Are we maintaining such large
> > projects ?
> > Please leave the versioning system for each project maintainer.
>
> This is one of the main reasons that prompted the whole developer guide
> to begin with. It most certainly must be made as a rule across all
> projects. A verioning scheme/system must exist. This is irrelevant of
> the size of the project. I most certainly would not back down from my
> position on that.
>
for individual projects, i think it doesn't make sence.
but when we are to release togeather, then we can enforce this.
> When I wrote that, I did make a lot of considerations. For example, I
> did not want to break how Katoob was being versioned, so I made it so
> alpha releases hold the dates as a suffix.
>
and sure the new guidelines'll be followed by me regardless of what i
think about them ;-)
--
----------------
-- Katoob Main Developer
Linux registered user #224950, ICQ #58475622
--
Don't send me any attachment in Micro$oft (.DOC, .PPT) format please
Read http://www.fsf.org/philosophy/no-word-attachments.html
Preferable attachments: .PDF, .HTML, .TXT
Thanx for adding this text to Your signature
--
-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GCM/IT d-(++)@ s+(++):->+++ a-- C+++$>++++ UL+++$>++++ P+++$>+++++
L+++(++++)$>+++++ E>+++ W++?>$ N+>+++ o? K-? !w++ !O !M !V !PS@ !PE@ Y+ PGP+++
t? 5? !X R? tv-- b+@ DI D+ G-- e++>+++ h-->++ !r y?
------END GEEK CODE BLOCK------
_______________________________________________
Developer mailing list
[EMAIL PROTECTED]
http://lists.arabeyes.org/mailman/listinfo/developer
----- End forwarded message -----
--
----------------
-- Katoob Main Developer
Linux registered user #224950, ICQ #58475622
--
Don't send me any attachment in Micro$oft (.DOC, .PPT) format please
Read http://www.fsf.org/philosophy/no-word-attachments.html
Preferable attachments: .PDF, .HTML, .TXT
Thanx for adding this text to Your signature
--
-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GCM/IT d-(++)@ s+(++):->+++ a-- C+++$>++++ UL+++$>++++ P+++$>+++++
L+++(++++)$>+++++ E>+++ W++?>$ N+>+++ o? K-? !w++ !O !M !V !PS@ !PE@ Y+ PGP=+++
t? 5? !X R? tv-- b+@ DI D+ G-- e++>+++ h-->++ !r y?
------END GEEK CODE BLOCK------
pgp00000.pgp
Description: PGP signature
_______________________________________________ Developer mailing list [EMAIL PROTECTED] http://lists.arabeyes.org/mailman/listinfo/developer
