[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Discuss-gnustep Digest, Vol 6, Issue 24
|
From: |
Dean Bittner |
|
Subject: |
Re: Discuss-gnustep Digest, Vol 6, Issue 24 |
|
Date: |
13 May 2003 09:08:12 -0600 |
Hi,
Are you available for a brief chat today (Tues) or tomorrow (Wed)?
Cheers,
Dean
On Tue, 2003-05-13 at 05:00, discuss-gnustep-request@gnu.org wrote:
> Send Discuss-gnustep mailing list submissions to
> discuss-gnustep@gnu.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://mail.gnu.org/mailman/listinfo/discuss-gnustep
> or, via email, send a message with subject or body 'help' to
> discuss-gnustep-request@gnu.org
>
> You can reach the person managing the list at
> discuss-gnustep-owner@gnu.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Discuss-gnustep digest..."
>
>
> Today's Topics:
>
> 1. Re: autogsdoc, once more (Chris B. Vetter)
> 2. Re: autogsdoc, once more (Richard Frith-Macdonald)
> 3. Re: autogsdoc, once more (Chris B. Vetter)
> 4. How does the GNU runtime encode method signatures?
> (Ziemowit Laski)
> 5. Re: Gorm and the GormImage class (Gregory Casamento)
> 6. Using frameworks for scriptign with StepTalk (Stefan Urbanek)
>
>
> ----------------------------------------------------------------------
>
> Date: Mon, 12 May 2003 10:57:23 -0700
> From: "Chris B. Vetter" <chrisv@web4inc.com>
> To: Richard Frith-Macdonald <richard@brainstorm.co.uk>
> Cc: discuss-gnustep@gnu.org
> Cc: bug-gnustep@gnu.org
> Subject: Re: autogsdoc, once more
> Message-ID: <20030512105723.6d5a5b75.chrisv@web4inc.com>
> In-Reply-To: <63AB42A1-7FAB-11D7-B4C4-00306544502E@brainstorm.co.uk>
> References: <20030505093401.2043f50f.chrisv@web4inc.com>
> <63AB42A1-7FAB-11D7-B4C4-00306544502E@brainstorm.co.uk>
> Content-Type: text/plain; charset=US-ASCII
> MIME-Version: 1.0
> Content-Transfer-Encoding: 7bit
> Precedence: list
> Message: 1
>
> On Tue, 6 May 2003 11:13:31 +0100
> Richard Frith-Macdonald <richard@brainstorm.co.uk> wrote:
> > On Monday, May 5, 2003, at 05:34 pm, Chris B. Vetter wrote:
> > It didn't understand difference between protocol definitions and
> > forward declarations ... I've fixed the CVS code to ignore the forward
> > declarations.
>
> Well, that works, but now I run into another problem, I'm pretty sure I
> didn't have before:
>
> To reference a protocol method, you would use [(Protocol)-method] and
> similarly, to reference only the protocol, you use [(Protocol)]. Both
> used to work fine.
>
> However, now the following line
>
> The [FOOBar] interface implements the [(FOOBar)] protocol.
>
> appears as
>
> The (link)FOOBar(/link) interface implements the protocol.
>
> Note the two spaces between "the" and "protocol" instead of a link.
> In addition, something like
>
> This category implements the [(FOOBar)] protocol for [NSArray].
>
> appears as
>
> This category implements the protocol for (link)NSArray(/link) .
>
> Again, just two spaces instead of a link to the protocol, plus two
> spaces after the link to NSArray.
> The same happens, when you have something like "foo bar [NSArray], foo
> bar" -- you'll get two spaces between "NSArray" and the comma.
>
> --
> Chris
>
>
> ------------------------------
>
> Date: Mon, 12 May 2003 20:25:51 +0100
> From: Richard Frith-Macdonald <richard@brainstorm.co.uk>
> To: "Chris B. Vetter" <chrisv@web4inc.com>
> Cc: discuss-gnustep@gnu.org
> Cc: bug-gnustep@gnu.org
> Subject: Re: autogsdoc, once more
> Message-ID: <8B2F2C36-84AF-11D7-A52E-00306544502E@brainstorm.co.uk>
> In-Reply-To: <20030512105723.6d5a5b75.chrisv@web4inc.com>
> Content-Type: text/plain; charset=US-ASCII; format=flowed
> MIME-Version: 1.0 (Apple Message framework v552)
> Content-Transfer-Encoding: 7bit
> Precedence: list
> Message: 2
>
>
> On Monday, May 12, 2003, at 06:57 pm, Chris B. Vetter wrote:
>
> > To reference a protocol method, you would use [(Protocol)-method] and
> > similarly, to reference only the protocol, you use [(Protocol)]. Both
> > used to work fine.
> >
> > However, now the following line
> >
> > The [FOOBar] interface implements the [(FOOBar)] protocol.
> >
> > appears as
> >
> > The (link)FOOBar(/link) interface implements the protocol.
>
> It's a libxml bug/feature ... libxml is changing '<' to '<' even when
> told not to substitute entities, so the text written to the html file
> contains <FOOBar> rather than <FOOBar> and the web browser
> treats it as an unknown htm tag and ignores it.
>
> I've hacked in a workaround for this libxml feature.
>
>
>
> ------------------------------
>
> Date: Mon, 12 May 2003 12:42:45 -0700
> From: "Chris B. Vetter" <chrisv@web4inc.com>
> To: discuss-gnustep@gnu.org
> Cc: bug-gnustep@gnu.org
> Subject: Re: autogsdoc, once more
> Message-ID: <20030512124245.28ef2ae3.chrisv@web4inc.com>
> In-Reply-To: <8B2F2C36-84AF-11D7-A52E-00306544502E@brainstorm.co.uk>
> References: <20030512105723.6d5a5b75.chrisv@web4inc.com>
> <8B2F2C36-84AF-11D7-A52E-00306544502E@brainstorm.co.uk>
> Content-Type: text/plain; charset=US-ASCII
> MIME-Version: 1.0
> Content-Transfer-Encoding: 7bit
> Precedence: list
> Message: 3
>
> On Mon, 12 May 2003 20:25:51 +0100
> Richard Frith-Macdonald <richard@brainstorm.co.uk> wrote:
> > It's a libxml bug/feature ... libxml is changing '<' to '<' even
> > when told not to substitute entities, so the text written to the html
> > file contains <FOOBar> rather than <FOOBar> and the web browser
> > treats it as an unknown htm tag and ignores it.
>
> I noticed that. I remember that this happened before.
>
> > I've hacked in a workaround for this libxml feature.
>
> Come to think of, the reference index lists Protocols as
>
> <link>(Protocol)</link>
>
> which looks kinda weird. Wouldn't it be possible to instead add
> references to Protocols as
>
> <<link>Protocol</link>>
>
> in general? After all, in some cases, you would get
>
> <link><Protocol></link>
>
> which looks kinda weird too.
>
> --
> Chris
>
>
> ------------------------------
>
> Date: Mon, 12 May 2003 15:39:23 -0700
> From: Ziemowit Laski <zlaski@apple.com>
> To: discuss-gnustep@gnu.org
> Cc: Nicola Pero <nicola@brainstorm.co.uk>
> Cc: Stan Shebs <shebs@apple.com>
> Subject: How does the GNU runtime encode method signatures?
> Message-ID: <9440FE52-84CA-11D7-B995-000393673036@apple.com>
> Content-Type: text/plain; charset=US-ASCII; format=flowed
> MIME-Version: 1.0 (Apple Message framework v552)
> Content-Transfer-Encoding: 7bit
> Precedence: list
> Message: 4
>
> I've recently cleaned up the NeXT method signature encoding (by the
> compiler)
> quite a bit in Apple's tree, to a point where it is no longer necessary
> to do the awkward handshake with the back-end via
> hack_method_prototype() in
> order to determine the offsets and sizes of parameters. I've also been
> told
> (by Apple's ObjC runtime maintainers) is that the method encodings
> generated
> by the compiler have, historically speaking, been so profoundly broken
> that
> the runtime ignored them and computed its own. :-)
>
> So, what's the situation like on the GNU side of things? Does the GNU
> runtime
> still rely on what the compiler generates? Does the GNU method
> encoding have
> a spec? :-) I'm asking because I'd like to yank out
> hack_method_prototype()
> and generally simplify the encoding machinery in objc-act.c, but am not
> sure
> if it is safe to do so.
>
> Thanks. (And yes, I'd definitely like to put Apple's ObjC fixes,
> including
> method encodings, into the FSF. :-) )
>
> --Zem
> --------------------------------------------------------------
> Ziemowit Laski 1 Infinite Loop, MS 301-2K
> Mac OS X Compiler Group Cupertino, CA USA 95014-2083
> Apple Computer, Inc. +1.408.974.6229 Fax .5477
>
>
>
> ------------------------------
>
> Date: Mon, 12 May 2003 21:51:09 -0700 (PDT)
> From: Gregory Casamento <greg_casamento@yahoo.com>
> To: "Björn" Giesler <giesler@ira.uka.de>, discuss-gnustep@gnu.org
> Subject: Re: Gorm and the GormImage class
> Message-ID: <20030513045109.33936.qmail@web41612.mail.yahoo.com>
> In-Reply-To: <200305111725.16849.giesler@ira.uka.de>
> Content-Type: text/plain; charset=us-ascii
> MIME-Version: 1.0
> Precedence: list
> Message: 5
>
> There was a memory problem in Gorm recently which caused similar issues.
> Could you send me your .gorm file so I can take a look at it?
>
> Also, please update to the latest version of Gorm, if you can.
>
> Thanks, GJC
> --- Björn Giesler <giesler@ira.uka.de> wrote:
> > Hi,
> >
> > I've created a fairly large interface in Gorm (works wonderfully! Thanks,
> > Gregory and all!). Now I've made the mistake of adding an image to it (via
> > the Load Image menu entry), and I can't load the .gorm file anymore -- my
> > app tells me:
> >
> > [NSDataMalloc deserializeDataAt:objObjCType:atCursor:context:] can't find
> > class - GormImage
> > [NSUnarchiver decodeValueOfObjCType:at:] decoded nil class
> >
> > Am I doing something wrong here? If so, what? And if I'm not -- how can I
> > get the image out of my interface again? Just deleting it in Gorm doesn't
> > help, it's still in the objects.gorm file.
> >
> > Thanks very much in advance,
> > --Björn
> > --
> > Dipl.-Inform. Björn Giesler, IAIM, University of Karlsruhe (TH)
> > See http://wwwiaim.ira.uka.de/users/giesler for information.
> >
> > A society that will trade a little freedom for a little order will
> > lose both, and deserves neither. --T. Jefferson
> > Support privacy. Mail me for my PGP/GPG public key.
> >
> >
> > _______________________________________________
> > Discuss-gnustep mailing list
> > Discuss-gnustep@gnu.org
> > http://mail.gnu.org/mailman/listinfo/discuss-gnustep
>
>
> =====
> Gregory John Casamento
> -- bheron on #gnustep, #linuxstep, & #gormtalk ----------------
> Please sign the petition against software patents at:
> http://www.petitiononline.com/pasp01/petition.html
> Petition to make Lighthouse Application Suite Free Software at:
> http://www.petitiononline.com/laafs/petition.html
> --- Main Developer of Gorm (featured in April Linux Journal) ---
>
> __________________________________
> Do you Yahoo!?
> The New Yahoo! Search - Faster. Easier. Bingo.
> http://search.yahoo.com
>
>
> ------------------------------
>
> Date: Tue, 13 May 2003 11:54:40 +0200
> From: Stefan Urbanek <urbanek@host.sk>
> To: discuss-gnustep@gnu.org
> Subject: Using frameworks for scriptign with StepTalk
> Message-ID: <1e29b09dc0b20bb761deaa25ab033f6d@stivi>
> Content-Type: text/plain; charset="us-ascii"; format="flowed"
> MIME-Version: 1.0 (Generated by Pantomime 1.1.0)
> Precedence: list
> Message: 6
>
> Hi,
>
> I have implemented few methods in StepTalk that will make framework scripting
> easier.
>
> In scripts you do:
> Envrionment includeFramework:'FrameworkName'
> and StepTalk will import public classes and objects of that framework, so you
> can use them in scripts. This is useful, for example for using framework
> features without applications or for testing frameworks.
>
> Framework should provide ScriptingInfo.plist with content like this:
> {
> Classes = ( /* array of public classes */ );
> }
> Optionaly you can include a key ScriptingInfoClass specifying name of class
> that will provide framework constants (named objects). All you have to do is
> to implement some (any) class with following method:
> - (NSDictionary *)namedObjectsForScripting
> {
> NSDictionary *dict;
> dict = ... /* some dictionary with objects and their names */;
> return dict;
> }
>
> This ScriptingInfoClass is optional, using framework classes is usually
> sufficient.
>
> I have created development snapshot at usual place:
> http://steptalk.host.sk/Download Code is available on CVS.
>
> Moreover, there are some important fixes in smalltalk source reader.
>
> Enjoy,
>
> Stefan Urbanek
> --
> http://urbanek.host.sk
>
> First they ignore you, then they laugh at you, then they fight you, then you
> win.
> - Mahatma Gandhi
>
>
>
>
> ------------------------------
>
> _______________________________________________
> Discuss-gnustep mailing list
> Discuss-gnustep@gnu.org
> http://mail.gnu.org/mailman/listinfo/discuss-gnustep
>
>
> End of Discuss-gnustep Digest, Vol 6, Issue 24
> **********************************************
| [Prev in Thread] |
Current Thread |
[Next in Thread] |
- Re: Discuss-gnustep Digest, Vol 6, Issue 24,
Dean Bittner <=