From rbb at techgame.net Sat Jul 7 02:06:27 2007 From: rbb at techgame.net (Brian Brown) Date: Fri, 6 Jul 2007 18:06:27 -0600 Subject: Displaying graphics question Message-ID: <6D5C4BB7-2E70-4330-84D9-ABD12EEB3426@techgame.net> Is there a Magritte description that know how to display graphics? I don't see anything that looks promising in the Magritte-Model- Description category, but I thought I had read something about it at one time. - Brian From renggli at iam.unibe.ch Sat Jul 7 11:45:28 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Sat, 7 Jul 2007 11:45:28 +0200 Subject: Displaying graphics question In-Reply-To: <6D5C4BB7-2E70-4330-84D9-ABD12EEB3426@techgame.net> References: <6D5C4BB7-2E70-4330-84D9-ABD12EEB3426@techgame.net> Message-ID: <824F4352-7D3E-4AF2-AE9A-CE72AE408C84@iam.unibe.ch> > Is there a Magritte description that know how to display graphics? I > don't see anything that looks promising in the Magritte-Model- > Description category, but I thought I had read something about it at > one time. MAFileDescription does that if the element is displayed in read-only mode. The editor in MAFileUploadComponent doens't do that per default anymore, because if it displayed huge images within forms wasn't that user friendly. However you can change that by subclassing MAFileUploadComponent. Lukas -- Lukas Renggli http://www.lukas-renggli.ch From renggli at iam.unibe.ch Thu Jul 12 18:15:26 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Thu, 12 Jul 2007 18:15:26 +0200 Subject: Differences w/ exception handling in Magritte..? In-Reply-To: <45074.157.127.124.14.1184255325.squirrel@www.ca-flower.com> References: <45074.157.127.124.14.1184255325.squirrel@www.ca-flower.com> Message-ID: <07649D84-71B7-4947-92D1-00C53C190269@iam.unibe.ch> Where exactly do you get a recursion? It looks like that this is exactly the point in time where I added code to avoid recursion when dealing with recursive descriptions. I guess the method causing trouble is MAValidatorVisitor>>#visitDescription:? Lukas On 12.07.2007, at 17:48, Rick Flower wrote: > > Lukas, > > I dropped a note to Thomas Koschate, who did the first ports of > Magritte > over to VW, asking about moving the current version of Magritte > over to VW > in the hopes of having something to use w/ Seaside 2.7 or 2.8a > (instead of > 2.6).. Anyway, he indicated that he had tried porting over > Magritte-all-lr.224 last month but ran into an issue w/ infinite > recursion > as part of some perceived exception handling changes.. Anyway, do > you know > off the top of your head if anything changed between versions 192/193 > (from Feb 16th, 2007) and 224 in how exception handling is dealt > with in > Magritte? I've not had a chance to look into it yet after his latest > comments early this morning (I don't have VW/Squeak at work). Anyway, > just thought I'd ask.. Thanks! > > -- Rick > > > -- Lukas Renggli http://www.lukas-renggli.ch From rickf at ca-flower.com Thu Jul 12 18:29:59 2007 From: rickf at ca-flower.com (Rick Flower) Date: Thu, 12 Jul 2007 09:29:59 -0700 (PDT) Subject: Differences w/ exception handling in Magritte..? In-Reply-To: <07649D84-71B7-4947-92D1-00C53C190269@iam.unibe.ch> References: <45074.157.127.124.14.1184255325.squirrel@www.ca-flower.com> <07649D84-71B7-4947-92D1-00C53C190269@iam.unibe.ch> Message-ID: <49591.157.127.124.14.1184257799.squirrel@www.ca-flower.com> On Thu, July 12, 2007 9:15 am, Lukas Renggli wrote: > Where exactly do you get a recursion? > > It looks like that this is exactly the point in time where I added > code to avoid recursion when dealing with recursive descriptions. I > guess the method causing trouble is > MAValidatorVisitor>>#visitDescription:? I'm not sure exactly -- I'll have to load it to find out -- I'll get back to you with some more details.. Thanks for the heads-up though -- at least it's a place to start looking. From rickf at ca-flower.com Thu Jul 12 18:35:13 2007 From: rickf at ca-flower.com (Rick Flower) Date: Thu, 12 Jul 2007 09:35:13 -0700 (PDT) Subject: What version of Seaside for current version of Magritte? Message-ID: <11431.157.127.124.14.1184258113.squirrel@www.ca-flower.com> Lukas -- Just to make sure I'm not trying to plug together apples & oranges, what version of Seaside does the latest version of Magritte work with? Ultimately, I'm trying to avoid the Seaside 2.6 tree and move into either 2.7 or perhaps even 2.8 and wanted to check.. I didn't see anything indicating this on the web page, but may have missed it perhaps. Thx! From koschate at omasko.com Thu Jul 12 18:35:24 2007 From: koschate at omasko.com (Thomas Koschate) Date: Thu, 12 Jul 2007 12:35:24 -0400 Subject: Differences w/ exception handling in Magritte..? In-Reply-To: <07649D84-71B7-4947-92D1-00C53C190269@iam.unibe.ch> References: <45074.157.127.124.14.1184255325.squirrel@www.ca-flower.com> <07649D84-71B7-4947-92D1-00C53C190269@iam.unibe.ch> Message-ID: <4cb10a4f0707120935v37942648pd61e98f240dced08@mail.gmail.com> On 7/12/07, Lukas Renggli wrote: > > Where exactly do you get a recursion? > > It looks like that this is exactly the point in time where I added > code to avoid recursion when dealing with recursive descriptions. I > guess the method causing trouble is > MAValidatorVisitor>>#visitDescription:? > That's exactly the spot. The #resignalAs: is causing VW some heartburn. -- ============================================================ Thomas Koschate -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070712/03bdcbae/attachment.html From renggli at iam.unibe.ch Thu Jul 12 18:44:56 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Thu, 12 Jul 2007 18:44:56 +0200 Subject: Differences w/ exception handling in Magritte..? In-Reply-To: <4cb10a4f0707120935v37942648pd61e98f240dced08@mail.gmail.com> References: <45074.157.127.124.14.1184255325.squirrel@www.ca-flower.com> <07649D84-71B7-4947-92D1-00C53C190269@iam.unibe.ch> <4cb10a4f0707120935v37942648pd61e98f240dced08@mail.gmail.com> Message-ID: > That's exactly the spot. The #resignalAs: is causing VW some > heartburn. I don't exactly remember why I copy the exception there. That seems rather strange and presumably causes VW to end up in an infinite recursion. Please try with the following code that works -- as far as I can see right now -- well in Squeak: visitDescription: aDescription "Validate the current object using aDescription within an exception handler to avoid running further tests that might cause error-cascades." aDescription isVisible ifFalse: [ ^ self ]. [ self validate: self object using: aDescription ] on: MAValidationError do: [ :err | err beResumable; pass ] Lukas -- Lukas Renggli http://www.lukas-renggli.ch From cbeler at enit.fr Thu Jul 12 18:59:17 2007 From: cbeler at enit.fr (=?ISO-8859-1?Q?C=E9drick_B=E9ler?=) Date: Thu, 12 Jul 2007 18:59:17 +0200 Subject: #invalidDispatcher error - causes an improper store into indexable object Message-ID: <46965DE5.1090604@enit.fr> Hi, in my image, PRPierMain>>invalidDispatcher: causes an improper store into indexable object WAResponse internalError: 'bla bla' executing the last line causes the error...in SeasideSupportPlatform>>walkbackStringFor: is it because error are not meant to be a string (anymore) ? ... don't know how to fix... I use the last seaside and pier version. C?drick From renggli at iam.unibe.ch Thu Jul 12 19:23:46 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Thu, 12 Jul 2007 19:23:46 +0200 Subject: Differences w/ exception handling in Magritte..? In-Reply-To: <4cb10a4f0707120935v37942648pd61e98f240dced08@mail.gmail.com> References: <45074.157.127.124.14.1184255325.squirrel@www.ca-flower.com> <07649D84-71B7-4947-92D1-00C53C190269@iam.unibe.ch> <4cb10a4f0707120935v37942648pd61e98f240dced08@mail.gmail.com> Message-ID: Name: Magritte-All-lr.226 Author: lr Time: 12 July 2007, 7:22:28 pm UUID: 72307372-3019-412b-94df-ab0524468fca Ancestors: Magritte-All-lr.225 Dependencies: Magritte-Model-lr.271, Magritte-Tests-lr.115, Magritte- Seaside-lr.232, Magritte-Morph-lr.38 - trying to improve the way resumeable errors are thrown, so that it works on VW as well - added tests to exactly check the way resumeable errors are handled On 12.07.2007, at 18:35, Thomas Koschate wrote: > > On 7/12/07, Lukas Renggli wrote: Where > exactly do you get a recursion? > > It looks like that this is exactly the point in time where I added > code to avoid recursion when dealing with recursive descriptions. I > guess the method causing trouble is > MAValidatorVisitor>>#visitDescription:? > > That's exactly the spot. The #resignalAs: is causing VW some > heartburn. > > -- > ============================================================ > Thomas Koschate > > _______________________________________________ > SmallWiki, Magritte, Pier and Related Tools ... > https://www.iam.unibe.ch/mailman/listinfo/smallwiki -- Lukas Renggli http://www.lukas-renggli.ch From renggli at iam.unibe.ch Thu Jul 12 19:24:40 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Thu, 12 Jul 2007 19:24:40 +0200 Subject: What version of Seaside for current version of Magritte? In-Reply-To: <11431.157.127.124.14.1184258113.squirrel@www.ca-flower.com> References: <11431.157.127.124.14.1184258113.squirrel@www.ca-flower.com> Message-ID: <4C133B65-A167-4ED3-898A-8C4A34D630B5@iam.unibe.ch> > Just to make sure I'm not trying to plug together apples & oranges, > what > version of Seaside does the latest version of Magritte work with? > Ultimately, I'm trying to avoid the Seaside 2.6 tree and move into > either > 2.7 or perhaps even 2.8 and wanted to check.. I didn't see anything > indicating this on the web page, but may have missed it perhaps. I am using it with 2.8. It should also work with 2.7 as far as I know, but I didn't try. Lukas -- Lukas Renggli http://www.lukas-renggli.ch From renggli at iam.unibe.ch Thu Jul 12 19:49:05 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Thu, 12 Jul 2007 19:49:05 +0200 Subject: #invalidDispatcher error - causes an improper store into indexable object In-Reply-To: <46965DE5.1090604@enit.fr> References: <46965DE5.1090604@enit.fr> Message-ID: <580A7984-3CD7-4BA3-9211-C8AE90317039@iam.unibe.ch> Thanks for the bug-report. Try this version: Name: Pier-All-lr.216 Author: lr Time: 12 July 2007, 7:48:34 pm UUID: 786c04c0-059e-40c7-9761-ef69974dc9d8 Ancestors: Pier-All-lr.215 Dependencies: Pier-Model-lr.154, Pier-Tests-lr.72, Pier-Seaside-lr. 180, Pier-OmniBrowser-lr.25, Pier-Security-lr.81, Pier-Blog-lr.56 - setup propre configuration hierarchy with decent default values - fixed bug in #invalidDispatcher: > in my image, PRPierMain>>invalidDispatcher: causes an improper store > into indexable object > WAResponse internalError: 'bla bla' > executing the last line causes the error...in > SeasideSupportPlatform>>walkbackStringFor: > is it because error are not meant to be a string (anymore) ? ... don't > know how to fix... > > I use the last seaside and pier version. > > C?drick > > _______________________________________________ > SmallWiki, Magritte, Pier and Related Tools ... > https://www.iam.unibe.ch/mailman/listinfo/smallwiki > -- Lukas Renggli http://www.lukas-renggli.ch From koschate at omasko.com Thu Jul 12 20:42:34 2007 From: koschate at omasko.com (Thomas Koschate) Date: Thu, 12 Jul 2007 14:42:34 -0400 Subject: Differences w/ exception handling in Magritte..? In-Reply-To: References: <45074.157.127.124.14.1184255325.squirrel@www.ca-flower.com> <07649D84-71B7-4947-92D1-00C53C190269@iam.unibe.ch> <4cb10a4f0707120935v37942648pd61e98f240dced08@mail.gmail.com> Message-ID: <4cb10a4f0707121142h16aa982ase97f288e170463d8@mail.gmail.com> On 7/12/07, Lukas Renggli wrote: > > Please try with the following code that works -- as far as I can see > right now -- well in Squeak: > > visitDescription: aDescription That solves the problem nicely for that one. However, we run into the same issue with MAStringReader>>#visitElementDescription:. Morever, there's actually a bug with this method, since #resignalAs: is expecting an instance of an exception, and you're passing in a class. While I remember, a request to help with future ports: Please avoid the use of {} to define arrays, as VisualWorks doesn't know how to deal with this. -- ============================================================ Thomas Koschate -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070712/04bb607e/attachment.html From renggli at iam.unibe.ch Thu Jul 12 21:10:00 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Thu, 12 Jul 2007 21:10:00 +0200 Subject: Differences w/ exception handling in Magritte..? In-Reply-To: <4cb10a4f0707121142h16aa982ase97f288e170463d8@mail.gmail.com> References: <45074.157.127.124.14.1184255325.squirrel@www.ca-flower.com> <07649D84-71B7-4947-92D1-00C53C190269@iam.unibe.ch> <4cb10a4f0707120935v37942648pd61e98f240dced08@mail.gmail.com> <4cb10a4f0707121142h16aa982ase97f288e170463d8@mail.gmail.com> Message-ID: <8C95C649-37BB-4C12-9E8D-990DB960A272@iam.unibe.ch> > That solves the problem nicely for that one. However, we run into > the same issue with MAStringReader>>#visitElementDescription:. > Morever, there's actually a bug with this method, since > #resignalAs: is expecting an instance of an exception, and you're > passing in a class. In Squeak a class and an instance works. I fixed that to create a new instance instead. > While I remember, a request to help with future ports: Please > avoid the use of {} to define arrays, as VisualWorks doesn't know > how to deal with this. I found one occurrence in a test. Are there others? The problems are fixed: Name: Magritte-All-lr.227 Author: lr Time: 12 July 2007, 9:08:32 pm UUID: eadc01b1-8663-4f91-8acc-c67a30f133d4 Ancestors: Magritte-All-lr.226 Dependencies: Magritte-Model-lr.272, Magritte-Tests-lr.116, Magritte- Seaside-lr.232, Magritte-Morph-lr.38 - for platform compatibility: - create an instance for #resignalAl: - removed single use of {} in test -- Lukas Renggli http://www.lukas-renggli.ch From koschate at omasko.com Thu Jul 12 21:23:51 2007 From: koschate at omasko.com (Thomas Koschate) Date: Thu, 12 Jul 2007 15:23:51 -0400 Subject: Differences w/ exception handling in Magritte..? In-Reply-To: <8C95C649-37BB-4C12-9E8D-990DB960A272@iam.unibe.ch> References: <45074.157.127.124.14.1184255325.squirrel@www.ca-flower.com> <07649D84-71B7-4947-92D1-00C53C190269@iam.unibe.ch> <4cb10a4f0707120935v37942648pd61e98f240dced08@mail.gmail.com> <4cb10a4f0707121142h16aa982ase97f288e170463d8@mail.gmail.com> <8C95C649-37BB-4C12-9E8D-990DB960A272@iam.unibe.ch> Message-ID: <4cb10a4f0707121223h530f6d38lff04e6ab1987c666@mail.gmail.com> On 7/12/07, Lukas Renggli wrote: In Squeak a class and an instance works. I fixed that to create a new > instance instead. I'm sorry, I may not have made myself understood. The infinite recursion error occurs as a result of the #resignalAs:, even if an instance is correctly supplied. I found one occurrence in a test. Are there others? > No, it was only the one occurrence. Thanks for the quick turnaround! -- ============================================================ Thomas Koschate +1 416.716.5949 koschate at omasko.com FAX: +1 905.451.9268 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070712/caa806f3/attachment.html From renggli at iam.unibe.ch Thu Jul 12 21:51:14 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Thu, 12 Jul 2007 21:51:14 +0200 Subject: Differences w/ exception handling in Magritte..? In-Reply-To: <4cb10a4f0707121223h530f6d38lff04e6ab1987c666@mail.gmail.com> References: <45074.157.127.124.14.1184255325.squirrel@www.ca-flower.com> <07649D84-71B7-4947-92D1-00C53C190269@iam.unibe.ch> <4cb10a4f0707120935v37942648pd61e98f240dced08@mail.gmail.com> <4cb10a4f0707121142h16aa982ase97f288e170463d8@mail.gmail.com> <8C95C649-37BB-4C12-9E8D-990DB960A272@iam.unibe.ch> <4cb10a4f0707121223h530f6d38lff04e6ab1987c666@mail.gmail.com> Message-ID: <393EC678-B7A5-4F38-8733-DB9916974E5C@iam.unibe.ch> Ok, no more senders of #resignalAl: in Name: Magritte-All-lr.228 Author: lr Time: 12 July 2007, 9:50:42 pm UUID: d7bdc877-4341-4986-8a7d-810edd165050 Ancestors: Magritte-All-lr.227 Dependencies: Magritte-Model-lr.273, Magritte-Tests-lr.116, Magritte- Seaside-lr.232, Magritte-Morph-lr.38 - avoid another infinite look with #resignalAs: On 12.07.2007, at 21:23, Thomas Koschate wrote: > > > On 7/12/07, Lukas Renggli wrote: > > In Squeak a class and an instance works. I fixed that to create a new > instance instead. > > I'm sorry, I may not have made myself understood. The infinite > recursion error occurs as a result of the #resignalAs:, even if an > instance is correctly supplied. > > I found one occurrence in a test. Are there others? > > No, it was only the one occurrence. > > Thanks for the quick turnaround! > -- > ============================================================ > Thomas Koschate +1 416.716.5949 > koschate at omasko.com FAX: +1 905.451.9268 -- Lukas Renggli http://www.lukas-renggli.ch From koschate at omasko.com Thu Jul 12 22:29:20 2007 From: koschate at omasko.com (Thomas Koschate) Date: Thu, 12 Jul 2007 16:29:20 -0400 Subject: Differences w/ exception handling in Magritte..? In-Reply-To: <393EC678-B7A5-4F38-8733-DB9916974E5C@iam.unibe.ch> References: <45074.157.127.124.14.1184255325.squirrel@www.ca-flower.com> <07649D84-71B7-4947-92D1-00C53C190269@iam.unibe.ch> <4cb10a4f0707120935v37942648pd61e98f240dced08@mail.gmail.com> <4cb10a4f0707121142h16aa982ase97f288e170463d8@mail.gmail.com> <8C95C649-37BB-4C12-9E8D-990DB960A272@iam.unibe.ch> <4cb10a4f0707121223h530f6d38lff04e6ab1987c666@mail.gmail.com> <393EC678-B7A5-4F38-8733-DB9916974E5C@iam.unibe.ch> Message-ID: <4cb10a4f0707121329w65016495r1c0e7ed2817f4028@mail.gmail.com> On 7/12/07, Lukas Renggli wrote: > > Ok, no more senders of #resignalAl: in > We're getting close now. I just need to work through a few more Squeak/VW dialect differences and we'll be up to date again. -- ============================================================ Thomas Koschate +1 416.716.5949 koschate at omasko.com FAX: +1 905.451.9268 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070712/18f42fef/attachment.html From stephan at stack.nl Fri Jul 13 15:02:27 2007 From: stephan at stack.nl (Stephan Eggermont) Date: Fri, 13 Jul 2007 15:02:27 +0200 Subject: Color editing magritte In-Reply-To: <580A7984-3CD7-4BA3-9211-C8AE90317039@iam.unibe.ch> References: <46965DE5.1090604@enit.fr> <580A7984-3CD7-4BA3-9211-C8AE90317039@iam.unibe.ch> Message-ID: <20070713130227.GB29509@stack.nl> The magritte version I'm using has a color chooser that doesn't allow editing: MAColorChooser renderContentOn: html html image " callback: [ :point | self answer: (self palette colorAt: point) ];" form: self palette. Is that the latest version, or is there one that allows changing colors? Stephan From rickf at ca-flower.com Fri Jul 13 20:07:03 2007 From: rickf at ca-flower.com (Rick Flower) Date: Fri, 13 Jul 2007 11:07:03 -0700 Subject: Updating Magritte Tutorial to not elicit complaints about deprecated functions being used... Message-ID: <4697BF47.7060009@ca-flower.com> Lukas et-al, I figured I'd try to update the Magritte Tutorial on the VW side (Magritte is now running well on VW care of yesterday's efforts -- Thanks!) (and I'll see if I can move the changes over to Squeak if there's interest) to no longer issue complaints about using deprecated methods and was wondering if someone could point me in the right direction for one of the updates as a sample.. Below is one of the complaining methods: MADescription>>auto: aSelector label: aString priority: aNumber self deprecated: 'Obsolete, use instance side configuration methods.'. ^ self accessor: (MAAutoSelectorAccessor selector: aSelector) label: aString priority: aNumber Magritte.MAAddressModel>>descriptionCanton ^ (MASingleOptionDescription auto: 'canton' label: 'Canton' priority: 40) options: #( 'Zurich' 'Bern' 'Luzern' 'Uri' 'Schwyz' 'Unterwalden' 'Glarus' 'Zug' 'Freiburg' 'Solothurn' 'Basel' 'Schaffhausen' 'Appenzell' 'St. Gallen' 'Graubunden' 'Aargau' 'Thurgau' 'Ticino' 'Vaud' 'Valais' 'Neuchatel' 'Geneve' 'Jura' ); reference: MAStringDescription new; beSorted; yourself. I tried converting the above #descriptionCanton to the following but it didn't work -- perhaps I'm misinterpretting the object layout.. Any suggestions? descriptionCanton " OLD CODE WAS: ^ (MASingleOptionDescription auto: 'canton' label: 'Canton' priority: 40)" ^(self accessor: (MAAutoSelectorAccessor selector: 'canton' label: 'Canton' priority: 40)) options: #( 'Zurich' 'Bern' 'Luzern' 'Uri' 'Schwyz' 'Unterwalden' 'Glarus' 'Zug' 'Freiburg' 'Solothurn' 'Basel' 'Schaffhausen' 'Appenzell' 'St. Gallen' 'Graubunden' 'Aargau' 'Thurgau' 'Ticino' 'Vaud' 'Valais' 'Neuchatel' 'Geneve' 'Jura' ); reference: MAStringDescription new; beSorted; yourself. From renggli at iam.unibe.ch Sat Jul 14 09:02:13 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Sat, 14 Jul 2007 09:02:13 +0200 Subject: Color editing magritte In-Reply-To: <20070713130227.GB29509@stack.nl> References: <46965DE5.1090604@enit.fr> <580A7984-3CD7-4BA3-9211-C8AE90317039@iam.unibe.ch> <20070713130227.GB29509@stack.nl> Message-ID: <6D904CB2-03E7-4DB1-9955-798B3D55BD04@iam.unibe.ch> Please try: Name: Magritte-All-lr.229 Author: lr Time: 14 July 2007, 9:01:44 am UUID: 4fb99b08-0a6c-4156-a83f-b529b1e64256 Ancestors: Magritte-All-lr.228 Dependencies: Magritte-Model-lr.273, Magritte-Tests-lr.116, Magritte- Seaside-lr.233, Magritte-Morph-lr.38 - made the color chooser work again On 13.07.2007, at 15:02, Stephan Eggermont wrote: > The magritte version I'm using has a color chooser that > doesn't allow editing: > > MAColorChooser > renderContentOn: html > html image > " callback: [ :point | self answer: (self palette colorAt: point) ];" > form: self palette. > > Is that the latest version, or is there one that allows changing > colors? > > Stephan > > _______________________________________________ > SmallWiki, Magritte, Pier and Related Tools ... > https://www.iam.unibe.ch/mailman/listinfo/smallwiki -- Lukas Renggli http://www.lukas-renggli.ch -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 3774 bytes Desc: not available Url : http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070714/d6233aa9/smime.bin From renggli at iam.unibe.ch Sat Jul 14 09:23:29 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Sat, 14 Jul 2007 09:23:29 +0200 Subject: Updating Magritte Tutorial to not elicit complaints about deprecated functions being used... In-Reply-To: <4697BF47.7060009@ca-flower.com> References: <4697BF47.7060009@ca-flower.com> Message-ID: > I figured I'd try to update the Magritte Tutorial on the VW side > (Magritte is now running well on VW care of yesterday's efforts -- > Thanks!) Great news! > (and I'll see if I can move the changes over to Squeak if > there's interest) to no longer issue complaints about using deprecated > methods and was wondering if someone could point me in the right > direction for one of the updates as a sample.. Where do you get those warnings? > Below is one of the complaining methods: > > MADescription>>auto: aSelector label: aString priority: aNumber > self deprecated: 'Obsolete, use instance side configuration > methods.'. > ^ self accessor: (MAAutoSelectorAccessor selector: aSelector) > label: > aString priority: aNumber Basically I expect all those ugly long constructor methods to be split up in simple cascaded. I found that the code becomes much easier to read like this. Please try the fixed version: Name: Magritte-Tutorial-lr.5 Author: lr Time: 14 July 2007, 9:22:09 am UUID: 57bc0745-71f8-469d-aa5e-7e59d0ad8015 Ancestors: Magritte-Tutorial-lr.4 - fixed deprecated code -- Lukas Renggli http://www.lukas-renggli.ch -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 3774 bytes Desc: not available Url : http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070714/7b953ea9/smime.bin From rickf at ca-flower.com Sun Jul 15 00:24:08 2007 From: rickf at ca-flower.com (Rick Flower) Date: Sat, 14 Jul 2007 15:24:08 -0700 Subject: Updating Magritte Tutorial to not elicit complaints about deprecated functions being used... In-Reply-To: References: <4697BF47.7060009@ca-flower.com> Message-ID: <46994D08.1070809@ca-flower.com> Lukas Renggli wrote: >> (and I'll see if I can move the changes over to Squeak if >> there's interest) to no longer issue complaints about using deprecated >> methods and was wondering if someone could point me in the right >> direction for one of the updates as a sample.. > > Where do you get those warnings? > I see then running the tutorial code on VW.. I see lots and lots of those messages.. > > Basically I expect all those ugly long constructor methods to be split > up in simple cascaded. I found that the code becomes much easier to > read like this. Please try the fixed version: > Cool.. I'll check it out.. Thanks! From rickf at ca-flower.com Sun Jul 15 00:29:12 2007 From: rickf at ca-flower.com (Rick Flower) Date: Sat, 14 Jul 2007 15:29:12 -0700 Subject: Wrapping parts of model objects in divs for display purposes? Message-ID: <46994E38.2010707@ca-flower.com> Hi all.. Just curious as a Magritte newbie.. If I've got a domain model object that has 10 fields (name, address,etc) and want some of them wrapped in div's when the form is displayed, what's the best way to achieve that? Ideally, I use the divs to apply css and might have for a user record a box drawn (in css) around the person's address info and perhaps a different box drawn around other attributes such as account info, etc. Anyway, I figure I should probably be subclassing something to do this but thought I'd ask first before I start going down that path.. Thanks! -- Rick From koschate at omasko.com Sun Jul 15 03:22:49 2007 From: koschate at omasko.com (Thomas Koschate) Date: Sat, 14 Jul 2007 21:22:49 -0400 Subject: Wrapping parts of model objects in divs for display purposes? In-Reply-To: <46994E38.2010707@ca-flower.com> References: <46994E38.2010707@ca-flower.com> Message-ID: <4cb10a4f0707141822r6a5de242u72b717cf6495af4@mail.gmail.com> On 7/14/07, Rick Flower wrote: > Ideally, I use the divs to apply css and might have for a user record > a box drawn (in css) around the person's address info and perhaps a > different box drawn around other attributes such as account info, etc. > Anyway, I figure I should probably be subclassing something to do this > but thought I'd ask first before I start going down that path.. My inclination would be to create separate classes for the groups of attributes that you wish to display differently. You'll be glad of it later for reasons beyond your immediate display needs. And, really, it's just another code smell telling you that there's objects there wanting to escape. -- ============================================================ Thomas Koschate From rickf at ca-flower.com Sun Jul 15 08:12:04 2007 From: rickf at ca-flower.com (Rick Flower) Date: Sat, 14 Jul 2007 23:12:04 -0700 Subject: Wrapping parts of model objects in divs for display purposes? In-Reply-To: <4cb10a4f0707141822r6a5de242u72b717cf6495af4@mail.gmail.com> References: <46994E38.2010707@ca-flower.com> <4cb10a4f0707141822r6a5de242u72b717cf6495af4@mail.gmail.com> Message-ID: <4699BAB4.1060508@ca-flower.com> Thomas Koschate wrote: > On 7/14/07, Rick Flower wrote: > >> Ideally, I use the divs to apply css and might have for a user record >> a box drawn (in css) around the person's address info and perhaps a >> different box drawn around other attributes such as account info, etc. >> Anyway, I figure I should probably be subclassing something to do this >> but thought I'd ask first before I start going down that path.. > > My inclination would be to create separate classes for the groups of > attributes that you wish to display differently. You'll be glad of it > later for reasons beyond your immediate display needs. And, really, > it's just another code smell telling you that there's objects there > wanting to escape. Thanks Thomas! One last question (for now) -- If I use two classes to separate the groups for a single class model, how do I tell Magritte that I want the first 4 attributes as part of Class #1 and the second group of 6 attributes as part of Class #2? Thanks! From stephan at stack.nl Sun Jul 15 09:44:37 2007 From: stephan at stack.nl (Stephan Eggermont) Date: Sun, 15 Jul 2007 09:44:37 +0200 Subject: Color editing magritte In-Reply-To: <6D904CB2-03E7-4DB1-9955-798B3D55BD04@iam.unibe.ch> References: <46965DE5.1090604@enit.fr> <580A7984-3CD7-4BA3-9211-C8AE90317039@iam.unibe.ch> <20070713130227.GB29509@stack.nl> <6D904CB2-03E7-4DB1-9955-798B3D55BD04@iam.unibe.ch> Message-ID: <20070715074437.GA19562@stack.nl> Thanx. I see I need a newer seaside too. Stephan From stephan at stack.nl Sun Jul 15 09:49:56 2007 From: stephan at stack.nl (Stephan Eggermont) Date: Sun, 15 Jul 2007 09:49:56 +0200 Subject: Wrapping parts of model objects in divs for display purposes? In-Reply-To: <4699BAB4.1060508@ca-flower.com> References: <46994E38.2010707@ca-flower.com> <4cb10a4f0707141822r6a5de242u72b717cf6495af4@mail.gmail.com> <4699BAB4.1060508@ca-flower.com> Message-ID: <20070715074956.GB19562@stack.nl> On Sat, Jul 14, 2007 at 11:12:04PM -0700, Rick Flower wrote: > Thanks Thomas! One last question (for now) -- If I use two classes > to separate the groups for a single class model, how do I tell Magritte > that I want the first 4 attributes as part of Class #1 and the second > group of 6 attributes as part of Class #2? Thanks! Take a look at: http://www.iam.unibe.ch/pipermail/smallwiki/2007-March/002663.html for a somewhat similar problem. Stephan From rickf at ca-flower.com Sun Jul 15 18:51:45 2007 From: rickf at ca-flower.com (Rick Flower) Date: Sun, 15 Jul 2007 09:51:45 -0700 Subject: Wrapping parts of model objects in divs for display purposes? In-Reply-To: <20070715074956.GB19562@stack.nl> References: <46994E38.2010707@ca-flower.com> <4cb10a4f0707141822r6a5de242u72b717cf6495af4@mail.gmail.com> <4699BAB4.1060508@ca-flower.com> <20070715074956.GB19562@stack.nl> Message-ID: <469A50A1.9090008@ca-flower.com> Stephan Eggermont wrote: > On Sat, Jul 14, 2007 at 11:12:04PM -0700, Rick Flower wrote: >> Thanks Thomas! One last question (for now) -- If I use two classes >> to separate the groups for a single class model, how do I tell Magritte >> that I want the first 4 attributes as part of Class #1 and the second >> group of 6 attributes as part of Class #2? Thanks! > > Take a look at: > http://www.iam.unibe.ch/pipermail/smallwiki/2007-March/002663.html > for a somewhat similar problem. Thanks Stephan.. I'll take a look at it! From koschate at omasko.com Mon Jul 16 17:50:46 2007 From: koschate at omasko.com (Thomas Koschate) Date: Mon, 16 Jul 2007 11:50:46 -0400 Subject: Updating Magritte Tutorial to not elicit complaints about deprecated functions being used... In-Reply-To: <46994D08.1070809@ca-flower.com> References: <4697BF47.7060009@ca-flower.com> <46994D08.1070809@ca-flower.com> Message-ID: <4cb10a4f0707160850y5495810r5dd3ab9a60e5030@mail.gmail.com> On 7/14/07, Rick Flower wrote: > > Basically I expect all those ugly long constructor methods to be split > > up in simple cascaded. I found that the code becomes much easier to > > read like this. Please try the fixed version: > > > Cool.. I'll check it out.. > Thanks! I've ported Magritte lr.230 and Magritte-Tutorial lr.5 over to VW. They're in the public StORE. Let me know of any issues. -- ============================================================ Thomas Koschate -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070716/5f8fc749/attachment.html From Markus.Gaelli at suisa.ch Thu Jul 19 10:06:34 2007 From: Markus.Gaelli at suisa.ch (Markus Gaelli) Date: Thu, 19 Jul 2007 10:06:34 +0200 Subject: Tree View for Magritte: Anyone? Message-ID: Hi, I have a tree datastructure to display (currently the nodes are strings and read only), so I thought I better ask the list, if someone already is using trees together with Magritte. I couldn't find anything via google, so could someone help me here? Thanks a lot, Markus From philippe.marschall at gmail.com Thu Jul 19 18:09:09 2007 From: philippe.marschall at gmail.com (Philippe Marschall) Date: Thu, 19 Jul 2007 18:09:09 +0200 Subject: Tree View for Magritte: Anyone? In-Reply-To: References: Message-ID: <66666f210707190909j1b6f920at425965aa71552c91@mail.gmail.com> 2007/7/19, Markus Gaelli : > Hi, > > I have a tree datastructure to display (currently the nodes are strings > and read only), so I thought I better ask the list, if someone already is > using trees together with Magritte. > I couldn't find anything via google, so could someone help me here? IIRC the structure description of Pier renders trees. Philippe > Thanks a lot, > > Markus > > _______________________________________________ > SmallWiki, Magritte, Pier and Related Tools ... > https://www.iam.unibe.ch/mailman/listinfo/smallwiki > From rbb at techgame.net Fri Jul 20 00:49:10 2007 From: rbb at techgame.net (Brian Brown) Date: Thu, 19 Jul 2007 16:49:10 -0600 Subject: Problem Inspecting Magritte components Message-ID: <2D02CCDE-E9B6-4E0C-B8BD-165C8C0FE831@techgame.net> I'm doing the following: renderContentOn: html html render: ((self parent person asComponent) readonly: true; yourself ). html form with: [html submitButton callback: [self editClient]; value: 'edit'] To view a model object... everything seems to work fine. When I toggle halos and try to see source or inspect one of the Magritte components, it fails, because they aren't being returned by #children anywhere. Is there a way I can cache the component, or do this a different way? Thanks, Brian From rbb at techgame.net Fri Jul 20 00:58:06 2007 From: rbb at techgame.net (Brian Brown) Date: Thu, 19 Jul 2007 16:58:06 -0600 Subject: Displaying graphics question In-Reply-To: <824F4352-7D3E-4AF2-AE9A-CE72AE408C84@iam.unibe.ch> References: <6D5C4BB7-2E70-4330-84D9-ABD12EEB3426@techgame.net> <824F4352-7D3E-4AF2-AE9A-CE72AE408C84@iam.unibe.ch> Message-ID: <8E1FA81D-EB01-46F7-82E6-A29ED8B51DB7@techgame.net> On Jul 7, 2007, at 3:45 AM, Lukas Renggli wrote: >> Is there a Magritte description that know how to display graphics? I >> don't see anything that looks promising in the Magritte-Model- >> Description category, but I thought I had read something about it at >> one time. > > MAFileDescription does that if the element is displayed in read-only > mode. The editor in MAFileUploadComponent doens't do that per default > anymore, because if it displayed huge images within forms wasn't that > user friendly. However you can change that by subclassing > MAFileUploadComponent. > I subclassed MAFileUploadComponent and MAFileDescription and am now using my components. I have looked at the #upload: method and #renderViewerOn: but don't see anything related to displaying it readonly. I also have looked at MAFileModel, and I'm clueless once again :-) Any ideas? Brian > Lukas > > -- > Lukas Renggli > http://www.lukas-renggli.ch > > > _______________________________________________ > SmallWiki, Magritte, Pier and Related Tools ... > https://www.iam.unibe.ch/mailman/listinfo/smallwiki From ramon.leon at allresnet.com Fri Jul 20 02:08:29 2007 From: ramon.leon at allresnet.com (Ramon Leon) Date: Thu, 19 Jul 2007 17:08:29 -0700 Subject: Problem Inspecting Magritte components In-Reply-To: <2D02CCDE-E9B6-4E0C-B8BD-165C8C0FE831@techgame.net> References: <2D02CCDE-E9B6-4E0C-B8BD-165C8C0FE831@techgame.net> Message-ID: <00bd01c7ca62$1a972fb0$f000a8c0@hq.allresnet.com> > > I'm doing the following: > > renderContentOn: html > html render: ((self parent person asComponent) > readonly: true; > yourself ). > html form with: [html submitButton callback: [self editClient]; > value: 'edit'] > > > To view a model object... everything seems to work fine. When > I toggle halos and try to see source or inspect one of the > Magritte components, it fails, because they aren't being > returned by #children anywhere. Is there a way I can cache > the component, or do this a different way? > > Thanks, > > Brian Common mistake, you don't use asComponent on the fly like that. You need to stick the result of asComponent into an instance variable, return it from children, and then use html render: on the instance variable. Seaside components are stateful, instance variables stick around between page requests. Ramon Leon http://onsmalltalk.com From stephan at stack.nl Fri Jul 20 16:46:12 2007 From: stephan at stack.nl (stephan@stack.nl) Date: Fri, 20 Jul 2007 16:46:12 +0200 Subject: Displaying graphics question In-Reply-To: <6D5C4BB7-2E70-4330-84D9-ABD12EEB3426@techgame.net> References: <6D5C4BB7-2E70-4330-84D9-ABD12EEB3426@techgame.net> Message-ID: <20070720164612.t943s2ptogssg84c@webmail0.stack.nl> Quoting Brian Brown : > Is there a Magritte description that know how to display graphics? I > don't see anything that looks promising in the Magritte-Model- > Description category, but I thought I had read something about it at > one time. A snippet of code is attached. The interesting (but not refactored) part is the form method of TestDrawing Stephan -------------- next part -------------- A non-text attachment was scrubbed... Name: Magritte-Drawing.st Type: application/octet-stream Size: 7273 bytes Desc: not available Url : http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070720/dcd77da3/Magritte-Drawing-0001.obj From rbb at techgame.net Fri Jul 20 17:18:12 2007 From: rbb at techgame.net (Brian Brown) Date: Fri, 20 Jul 2007 09:18:12 -0600 Subject: Problem Inspecting Magritte components In-Reply-To: <00bd01c7ca62$1a972fb0$f000a8c0@hq.allresnet.com> References: <2D02CCDE-E9B6-4E0C-B8BD-165C8C0FE831@techgame.net> <00bd01c7ca62$1a972fb0$f000a8c0@hq.allresnet.com> Message-ID: <51B58724-906B-4803-B675-120A6C0BCA52@techgame.net> On Jul 19, 2007, at 6:08 PM, Ramon Leon wrote: >> >> I'm doing the following: >> >> renderContentOn: html >> html render: ((self parent person asComponent) >> readonly: true; >> yourself ). >> html form with: [html submitButton callback: [self editClient]; >> value: 'edit'] >> >> >> To view a model object... everything seems to work fine. When >> I toggle halos and try to see source or inspect one of the >> Magritte components, it fails, because they aren't being >> returned by #children anywhere. Is there a way I can cache >> the component, or do this a different way? >> >> Thanks, >> >> Brian > > Common mistake, you don't use asComponent on the fly like that. > You need to > stick the result of asComponent into an instance variable, return > it from > children, and then use html render: on the instance variable. Seaside > components are stateful, instance variables stick around between page > requests. > Well, sure - I've known this for years, but with all the Magritte examples I've seen in the image, caching wasn't being used, mostly just #call: and #answer:, so I guess I became confused :) thanks, Brian > Ramon Leon > http://onsmalltalk.com > > > _______________________________________________ > SmallWiki, Magritte, Pier and Related Tools ... > https://www.iam.unibe.ch/mailman/listinfo/smallwiki From stephan at stack.nl Sun Jul 22 10:15:11 2007 From: stephan at stack.nl (Stephan Eggermont) Date: Sun, 22 Jul 2007 10:15:11 +0200 Subject: Working combination In-Reply-To: <20070715074437.GA19562@stack.nl> References: <46965DE5.1090604@enit.fr> <580A7984-3CD7-4BA3-9211-C8AE90317039@iam.unibe.ch> <20070713130227.GB29509@stack.nl> <6D904CB2-03E7-4DB1-9955-798B3D55BD04@iam.unibe.ch> <20070715074437.GA19562@stack.nl> Message-ID: <20070722081511.GB52151@stack.nl> On Sun, Jul 15, 2007 at 09:44:37AM +0200, Stephan Eggermont wrote: > Thanx. I see I need a newer seaside too. Hmm, anyone know how to create a working combination of this magritte, a recent seaside 2.8 and Damiens dev-image? I don't like: Error: MessageNotUnderstood: UndefinedObject>>pixelValueForDepth Stephan From azreal1977 at hotmail.com Sun Jul 22 12:53:42 2007 From: azreal1977 at hotmail.com (J J) Date: Sun, 22 Jul 2007 10:53:42 +0000 Subject: Differences w/ exception handling in Magritte..? Message-ID: Date: Thu, 12 Jul 2007 14:42:34 -0400From: koschate at omasko.comTo: renggli at iam.unibe.chSubject: Re: Differences w/ exception handling in Magritte..?CC: renggli at gmail.com; smallwiki at iam.unibe.chWhile I remember, a request to help with future ports: Please avoid the use of {} to define arrays, as VisualWorks doesn't know how to deal with this.-- ============================================================ Thomas KoschateHrm, I use this a bit in my recurrence rules code. How do you get the same functionality in VW? _________________________________________________________________ Missed the show?? Watch videos of the Live Earth Concert on MSN. http://liveearth.msn.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070722/83980a4c/attachment.html From philippe.marschall at gmail.com Sun Jul 22 13:16:32 2007 From: philippe.marschall at gmail.com (Philippe Marschall) Date: Sun, 22 Jul 2007 13:16:32 +0200 Subject: Differences w/ exception handling in Magritte..? In-Reply-To: References: Message-ID: <66666f210707220416v232646fdj133b1ca974be6d58@mail.gmail.com> 2007/7/22, J J : > > > > ________________________________ > Date: Thu, 12 Jul 2007 14:42:34 -0400 > From: koschate at omasko.com > To: renggli at iam.unibe.ch > Subject: Re: Differences w/ exception handling in Magritte..? > CC: renggli at gmail.com; smallwiki at iam.unibe.ch > > > While I remember, a request to help with future ports: Please avoid the use > of {} to define arrays, as VisualWorks doesn't know how to deal with this. > -- > ============================================================ > Thomas Koschate > > > > Hrm, I use this a bit in my recurrence rules code. How do you get the same > functionality in VW? Array with: with:... Philippe From damien.cassou at gmail.com Sun Jul 22 15:56:47 2007 From: damien.cassou at gmail.com (Damien Cassou) Date: Sun, 22 Jul 2007 15:56:47 +0200 Subject: Maintainer needed for Magritte/Pier/Seaside on the Universe Message-ID: <6ac749c10707220656v37a5f94cpba229f5351c27497@mail.gmail.com> Hi, I've currently no time to keep track of magritte, pier and seaside on the Universe. It's a very simple task, but it requires to be informed about what version of pier blog works with which version of RSRSS, magritte and seaside... The universe is interesting to help people install whatever they want easily. Does anyone wants to help me? -- Damien Cassou From azreal1977 at hotmail.com Sun Jul 22 22:31:01 2007 From: azreal1977 at hotmail.com (J J) Date: Sun, 22 Jul 2007 20:31:01 +0000 Subject: Differences w/ exception handling in Magritte..? Message-ID: I was afraid you would say that. :( Is it limited to 4 with:'s like Squeak? In any case, VW does allow one to add behaviors to # easier last I heard. Maybe I can just make a #{} as part of my package. :)> Date: Sun, 22 Jul 2007 13:16:32 +0200> From: philippe.marschall at gmail.com> To: smallwiki at iam.unibe.ch> Subject: Re: Differences w/ exception handling in Magritte..?> > 2007/7/22, J J :> >> >> >> > ________________________________> > Date: Thu, 12 Jul 2007 14:42:34 -0400> > From: koschate at omasko.com> > To: renggli at iam.unibe.ch> > Subject: Re: Differences w/ exception handling in Magritte..?> > CC: renggli at gmail.com; smallwiki at iam.unibe.ch> >> >> > While I remember, a request to help with future ports: Please avoid the use> > of {} to define arrays, as VisualWorks doesn't know how to deal with this.> > --> > ============================================================> > Thomas Koschate> >> >> >> > Hrm, I use this a bit in my recurrence rules code. How do you get the same> > functionality in VW?> > Array with: with:...> > Philippe> > _______________________________________________> SmallWiki, Magritte, Pier and Related Tools ...> https://www.iam.unibe.ch/mailman/listinfo/smallwiki _________________________________________________________________ See what you?re getting into?before you go there. http://newlivehotmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070722/ec9b5da4/attachment.html From philippe.marschall at gmail.com Sun Jul 22 22:42:14 2007 From: philippe.marschall at gmail.com (Philippe Marschall) Date: Sun, 22 Jul 2007 22:42:14 +0200 Subject: Differences w/ exception handling in Magritte..? In-Reply-To: References: Message-ID: <66666f210707221342o779a036dmb631687cba98169f@mail.gmail.com> 2007/7/22, J J : > > I was afraid you would say that. :( Is it limited to 4 with:'s like Squeak? Then use streams. See senders of #braceStream: > In any case, VW does allow one to add behaviors to # easier last I heard. > Maybe I can just make a #{} as part of my package. :) No. Just don't use curly brace arrays. Philippe > ________________________________ > > Date: Sun, 22 Jul 2007 13:16:32 +0200 > > From: philippe.marschall at gmail.com > > To: smallwiki at iam.unibe.ch > > Subject: Re: Differences w/ exception handling in Magritte..? > > > > 2007/7/22, J J : > > > > > > > > > > > > ________________________________ > > > Date: Thu, 12 Jul 2007 14:42:34 -0400 > > > From: koschate at omasko.com > > > To: renggli at iam.unibe.ch > > > Subject: Re: Differences w/ exception handling in Magritte..? > > > CC: renggli at gmail.com; smallwiki at iam.unibe.ch > > > > > > > > > While I remember, a request to help with future ports: Please avoid the > use > > > of {} to define arrays, as VisualWorks doesn't know how to deal with > this. > > > -- > > > > ============================================================ > > > Thomas Koschate > > > > > > > > > > > > Hrm, I use this a bit in my recurrence rules code. How do you get the > same > > > functionality in VW? > > > > Array with: with:... > > > > Philippe > > > > _______________________________________________ > > SmallWiki, Magritte, Pier and Related Tools ... > > https://www.iam.unibe.ch/mailman/listinfo/smallwiki > > ________________________________ > See what you're getting into?before you go there. Check it out! > > _______________________________________________ > SmallWiki, Magritte, Pier and Related Tools ... > https://www.iam.unibe.ch/mailman/listinfo/smallwiki > From azreal1977 at hotmail.com Sun Jul 22 22:45:25 2007 From: azreal1977 at hotmail.com (J J) Date: Sun, 22 Jul 2007 20:45:25 +0000 Subject: Differences w/ exception handling in Magritte..? Message-ID: > Date: Sun, 22 Jul 2007 22:42:14 +0200> From: philippe.marschall at gmail.com> To: smallwiki at iam.unibe.ch> Subject: Re: Differences w/ exception handling in Magritte..?>> Then use streams. See senders of #braceStream:I suppose that's doable. I know how to do it, but it's not as concise as the curly-array syntax.> No. Just don't use curly brace arrays.> > PhilippeIt's not the braces I care about, it's the concise way of constructing an array from values that are evaluated at runtime instead of compile time. _________________________________________________________________ Missed the show?? Watch videos of the Live Earth Concert on MSN. http://liveearth.msn.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070722/6529d4ba/attachment-0001.html From norbert at hartl.name Thu Jul 26 11:14:35 2007 From: norbert at hartl.name (Norbert Hartl) Date: Thu, 26 Jul 2007 11:14:35 +0200 Subject: OrderedCollection>>copyWithAll: Message-ID: <1185441275.19183.111.camel@localhost> Hi, OrderedCollection>>copyWithAll: does not work. (OrderedCollection with: 1 with: 2) copyWithAll: {3 . 4 . 5 } throws an error. It uses replaceFrom:to:with: which tries to operate on non-existing indexes. copyWithAll: is just used at two places MADistribution>>addPostscript: and MADistribution>>addPreamble: These make use of Collection>>copyWithAll: which works for String objects. OrderedCollection>>copyWithAll: is never used. So it isn't that important and might be better removed regards, Norbert From tallman at inbox.ru Thu Jul 26 12:56:12 2007 From: tallman at inbox.ru (tallman@inbox.ru) Date: Thu, 26 Jul 2007 14:56:12 +0400 Subject: Strange accessors/instvars bug Message-ID: <46A87DCC.1060207@inbox.ru> I locate some strange bug when I working for Seaside on Sails. How to look at it: 1) load http://richart-consalt.ru/Sails.bugged.zip, unpack it (all needed including VM inside) 2) run it 3) in left top corner you'll see workspace with strings SailsTest test2. blg:=BlogPost new. blg title:'1111'. blg title blg comments 4) say "do it" to strings blg:=BlogPost new. blg title:'1111'. 5) and then say "print it" to strings blg title blg comments It prints '1111' both for comments and title accessors. But BlogPost is quite simple object: Object child with 3 instvars: title, post, comments. comments:=OrderedCollection new in initialize. Accessors (automatically made) for all 3 instvars. Nothing more. So "blg comments" must return emty ordered collection, not a title string... Some more comments: a) Sometimes both title and comments accessors returns empty ordered collection b) If you inspect blg, you will see strange instvars values sometimes and sometimes - normal values (but accessos still return buggy values) c) Bug disappear as fast as I locate it and put breakpoints to accessor to catch it. d) Bug can interact not only with workspace, as far as it disappeare my Seaside example become functional (it failed before, becouse #add: was applied to title string returned by comments accessor). ============== Assargadon tallman at inbox.ru http://headache.h1.ru From damien.cassou at gmail.com Thu Jul 26 15:18:56 2007 From: damien.cassou at gmail.com (Damien Cassou) Date: Thu, 26 Jul 2007 15:18:56 +0200 Subject: Strange accessors/instvars bug In-Reply-To: <46A87DCC.1060207@inbox.ru> References: <46A87DCC.1060207@inbox.ru> Message-ID: <6ac749c10707260618p4a79be84hfb87abf1ccb50ec0@mail.gmail.com> I haven't tried your image, but I guess some of your methods must be recompiled. 2007/7/26, tallman at inbox.ru : > I locate some strange bug when I working for Seaside on Sails. > How to look at it: > 1) load http://richart-consalt.ru/Sails.bugged.zip, unpack it (all > needed including VM inside) > 2) run it > 3) in left top corner you'll see workspace with strings > SailsTest test2. > > blg:=BlogPost new. > blg title:'1111'. > blg title > blg comments > 4) say "do it" to strings > blg:=BlogPost new. > blg title:'1111'. > 5) and then say "print it" to strings > blg title > blg comments > > It prints '1111' both for comments and title accessors. > But BlogPost is quite simple object: Object child with 3 instvars: > title, post, comments. comments:=OrderedCollection new in initialize. > Accessors (automatically made) for all 3 instvars. Nothing more. > So "blg comments" must return emty ordered collection, not a title string... > > Some more comments: > a) Sometimes both title and comments accessors returns empty ordered > collection > b) If you inspect blg, you will see strange instvars values sometimes > and sometimes - normal values (but accessos still return buggy values) > c) Bug disappear as fast as I locate it and put breakpoints to accessor > to catch it. > d) Bug can interact not only with workspace, as far as it disappeare my > Seaside example become functional (it failed before, becouse #add: was > applied to title string returned by comments accessor). > > ============== > Assargadon > tallman at inbox.ru > http://headache.h1.ru > > _______________________________________________ > SmallWiki, Magritte, Pier and Related Tools ... > https://www.iam.unibe.ch/mailman/listinfo/smallwiki > -- Damien Cassou From mathk.sue at gmail.com Thu Jul 26 15:46:21 2007 From: mathk.sue at gmail.com (Mathieu Suen) Date: Thu, 26 Jul 2007 15:46:21 +0200 Subject: Strange accessors/instvars bug In-Reply-To: <46A87DCC.1060207@inbox.ru> References: <46A87DCC.1060207@inbox.ru> Message-ID: <60FE519E-076E-4580-80E5-A2B6DC1A1F98@gmail.com> It seems to be a bug from the refactoring which create the accessor. To recompile the class simply execute: BlogPost compileAll Mth On Jul 26, 2007, at 12:56 PM, tallman at inbox.ru wrote: > I locate some strange bug when I working for Seaside on Sails. > How to look at it: > 1) load http://richart-consalt.ru/Sails.bugged.zip, unpack it (all > needed including VM inside) > 2) run it > 3) in left top corner you'll see workspace with strings > SailsTest test2. > > blg:=BlogPost new. > blg title:'1111'. > blg title > blg comments > 4) say "do it" to strings > blg:=BlogPost new. > blg title:'1111'. > 5) and then say "print it" to strings > blg title > blg comments > > It prints '1111' both for comments and title accessors. > But BlogPost is quite simple object: Object child with 3 instvars: > title, post, comments. comments:=OrderedCollection new in initialize. > Accessors (automatically made) for all 3 instvars. Nothing more. > So "blg comments" must return emty ordered collection, not a title > string... > > Some more comments: > a) Sometimes both title and comments accessors returns empty ordered > collection > b) If you inspect blg, you will see strange instvars values sometimes > and sometimes - normal values (but accessos still return buggy values) > c) Bug disappear as fast as I locate it and put breakpoints to > accessor > to catch it. > d) Bug can interact not only with workspace, as far as it > disappeare my > Seaside example become functional (it failed before, becouse #add: was > applied to title string returned by comments accessor). > > ============== > Assargadon > tallman at inbox.ru > http://headache.h1.ru > > _______________________________________________ > SmallWiki, Magritte, Pier and Related Tools ... > https://www.iam.unibe.ch/mailman/listinfo/smallwiki From rbb at techgame.net Thu Jul 26 22:02:40 2007 From: rbb at techgame.net (Brian Brown) Date: Thu, 26 Jul 2007 14:02:40 -0600 Subject: "Creating Custom Forms" video question Message-ID: <14DCC2D3-49F4-4116-8F7E-639E002660A0@techgame.net> At some point I thought I saw a video demonstrating how to create the birthday forms used in "Creating Custom Forms" video using the web interface. The "Creating Custom Forms" video shows how to copy one and modify it, but not how they got there in the first place. Is that still around some where? Thanks, Brian From rbb at techgame.net Thu Jul 26 22:43:06 2007 From: rbb at techgame.net (Brian Brown) Date: Thu, 26 Jul 2007 14:43:06 -0600 Subject: Pier pointers Message-ID: <3434AB74-DB97-4391-B6C3-EE9A5BE7DA8F@techgame.net> I have an existing seaside application that I would like to port into the Pier framework, and need some pointers :) 1. I store some things on session; what is the Pier equivalent? 2. Do I have to move my renderContentOn: methods to renderWidgetOn:? 3. Should I subclass MAComponent or PRViewComponent or something entirely different? Thanks! Brian (I've looked on the pier Wiki and read the FAQ and have got a minimal grasp of Magritte (and am using it in a non Pier context), but I haven't found any developer getting started type documentation) From keith_hodges at yahoo.co.uk Fri Jul 27 01:42:25 2007 From: keith_hodges at yahoo.co.uk (Keith Hodges) Date: Fri, 27 Jul 2007 00:42:25 +0100 Subject: Serving Pier from a directory Message-ID: <46A93161.50909@yahoo.co.uk> Hello, is it possible to serve pier from a seaside entry point contained in a 'directory' sublevel? thanks in advance Keith From keith_hodges at yahoo.co.uk Fri Jul 27 03:18:40 2007 From: keith_hodges at yahoo.co.uk (Keith Hodges) Date: Fri, 27 Jul 2007 02:18:40 +0100 Subject: Serving Pier from a directory In-Reply-To: <46A93161.50909@yahoo.co.uk> References: <46A93161.50909@yahoo.co.uk> Message-ID: <46A947F0.9090701@yahoo.co.uk> Keith Hodges wrote: > Hello, > > is it possible to serve pier from a seaside entry point contained in a > 'directory' sublevel? > > thanks in advance > > Keith > I made an attempt at fixing this in the Pier-ServeAllKernels package available in http://mc.lukas-renggli.ch/pieraddons I dont understand the convoluted code in PRPierMain-structureFrom: Why go to all the trouble of breaking up the path into pieces and then putting it back together again? cheers Keith From keith_hodges at yahoo.co.uk Fri Jul 27 04:30:25 2007 From: keith_hodges at yahoo.co.uk (Keith Hodges) Date: Fri, 27 Jul 2007 03:30:25 +0100 Subject: Help needed to run widgets in other contexts Message-ID: <46A958C1.6010708@yahoo.co.uk> Help! Hi All, A while ago Lukas gave me the following code for opening a pier widget so that it would be able to access its context even though we are not even in a pier application! self - is my control panel which runs in the WADispatcherEditor's session. widget is PUUsers or PUGroups ControlPanel-callWidget: widget titled: title PRCurrentContext use: self during: [ WARenderLoop new call: (widget addDecoration: (WAWindowDecoration new title: title; yourself); yourself) ] This works to the extent that the widget displays correctly, however the buttons do not work due to PRCurrentContext value being nil. I am using the latest everything in the development universes. Does anyone have any ideas? thanks in advance Keith From damien.cassou at gmail.com Fri Jul 27 08:05:40 2007 From: damien.cassou at gmail.com (Damien Cassou) Date: Fri, 27 Jul 2007 08:05:40 +0200 Subject: Pier pointers In-Reply-To: <3434AB74-DB97-4391-B6C3-EE9A5BE7DA8F@techgame.net> References: <3434AB74-DB97-4391-B6C3-EE9A5BE7DA8F@techgame.net> Message-ID: <6ac749c10707262305m198c1320we7d213469a65c66e@mail.gmail.com> Hi Brian, I can't answer your question. Before, I need to know why you want to use Pier for. How will you benefit from Pier? Then, what do you want to keep in your application? You have to know that Seaside component can be integrated into Pier without any change. 2007/7/26, Brian Brown : > I have an existing seaside application that I would like to port into > the Pier framework, and need some pointers :) > > > 1. I store some things on session; what is the Pier equivalent? > > 2. Do I have to move my renderContentOn: methods to renderWidgetOn:? > > 3. Should I subclass MAComponent or PRViewComponent or something > entirely different? > > > Thanks! > > Brian > (I've looked on the pier Wiki and read the FAQ and have got a minimal > grasp of Magritte (and am using it in a non Pier context), but I > haven't found any developer getting started type documentation) > > _______________________________________________ > SmallWiki, Magritte, Pier and Related Tools ... > https://www.iam.unibe.ch/mailman/listinfo/smallwiki > -- Damien Cassou From tallman at inbox.ru Fri Jul 27 09:13:29 2007 From: tallman at inbox.ru (tallman@inbox.ru) Date: Fri, 27 Jul 2007 11:13:29 +0400 Subject: Strange accessors/instvars bug (Mathieu Suen) Message-ID: <46A99B19.6040300@inbox.ru> Thanks, but I force this class to work properly before sending this message. "My class don't work" is not a trouble now, trouble is "something wrong with some of squeaks details". I try recompiling, it works. So, as far as I understand, it will be good to locate the action generates this bug. From akuhn at gmx.ch Sat Jul 28 10:21:51 2007 From: akuhn at gmx.ch (Adrian Kuhn) Date: Sat, 28 Jul 2007 10:21:51 +0200 Subject: Maybe Smallwiki RSS broken? Message-ID: <7EDD9C57-E3FC-44A7-B5F8-2F52B0994F99@gmx.ch> Dear all, Smallwiki's RSS feed does, at least for Vienna, not report all changes. Since the name of an RSS feed entry acts as unique key, changes on a page are only reported once and then never ever again. Is this a Vienna only problem, or general? cheers, AA From nishis at zephyr.dti.ne.jp Sat Jul 28 17:20:47 2007 From: nishis at zephyr.dti.ne.jp (NISHIHARA Satoshi) Date: Sun, 29 Jul 2007 00:20:47 +0900 Subject: PRViewRenderer>>visitFile: Message-ID: <46AB5ECF.4040108@zephyr.dti.ne.jp> At Pier 1.0.9.201, should it be *file*? PRViewRenderer>>visitFile: anObject | file url | file := anObject file ifNil: [ ^ self ]. file isText ifTrue: [ ^ html preformatted: anObject contents ]. "^^^^^^^^" " file?" url := file urlOn: html. file isImage ifTrue: [ ^ html image url: url; altText: anObject title ]. html anchor url: url; with: anObject title -- -------------------------------------------- ???? (NISHIHARA Satoshi) URL: http://www.zephyr.dti.ne.jp/~nishis/ -------------------------------------------- From norbert at hartl.name Tue Jul 31 12:45:56 2007 From: norbert at hartl.name (Norbert Hartl) Date: Tue, 31 Jul 2007 12:45:56 +0200 Subject: form element ids and names and testing Message-ID: <1185878756.23841.13.camel@localhost> Hi, I'm a user of seaside testing. SeasideTesting provides two methods for accessing form elements. It's by index and by element id. Requesting by index is too error prone as the change of the order of form elements breaks the test. Requesting by id is rendered impossible by using Magritte. The ajaxId: stuff includes the hash of an object and is not predictable when used inside a test. I changed ajaxId: in MAComponent to use the following: ajaxId: aSymbol | accessor | ^ String streamContents: [ :stream| accessor := self description accessor. [ accessor respondsTo: #next ] whileTrue: [ stream nextPutAll: accessor accessor readSelector. stream nextPut: $.. accessor := accessor next ]. stream nextPutAll: accessor readSelector. ]. This gives you the name of the selector as id and for nested descriptions you get nestedObjectSelector.selector. Therefor the ids per object are unique and are easily used in SeasideTesting. But my colleague reminded me that this only works if I have a single form on the page. He said the id has to be unique across the whole page. So my approach is bit critical. Is there anybody who uses SeasideTesting for testing magritte components? And if yes how do you do that? thanks, Norbert From dave at thedesignexperience.org Tue Jul 31 13:52:03 2007 From: dave at thedesignexperience.org (Dave Bauer) Date: Tue, 31 Jul 2007 07:52:03 -0400 Subject: form element ids and names and testing In-Reply-To: <1185878756.23841.13.camel@localhost> References: <1185878756.23841.13.camel@localhost> Message-ID: <6c4e07720707310452q3d7b034cjdd974935533663f3@mail.gmail.com> > This gives you the name of the selector as id and for nested > descriptions you get nestedObjectSelector.selector. Therefor > the ids per object are unique and are easily used in SeasideTesting. > But my colleague reminded me that this only works if I have a > single form on the page. He said the id has to be unique across > the whole page. So my approach is bit critical. > Hi I usually just lurk here, but I know the general solution to this problem. This is where using Xpath would be handy or another way to define how to find the element by first specifying the id of the form, then the id of the element within the form. Selenium IDE uses this and its pretty useful. Dave From renggli at iam.unibe.ch Tue Jul 31 21:34:16 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Tue, 31 Jul 2007 21:34:16 +0200 Subject: OrderedCollection>>copyWithAll: In-Reply-To: <1185441275.19183.111.camel@localhost> References: <1185441275.19183.111.camel@localhost> Message-ID: <5DFFD9A0-38EF-49ED-BA0D-A1AD1DBC5123@iam.unibe.ch> I added another test and fixed this issue in Magritte-All-lr.231. I moved the method from SequenceableCollection to ArrayedCollection as it is used in Pier and some other projects. Thanks for reporting. Lukas On 26.07.2007, at 11:14, Norbert Hartl wrote: > Hi, > > OrderedCollection>>copyWithAll: does not work. > > (OrderedCollection with: 1 with: 2) copyWithAll: {3 . 4 . 5 } > > throws an error. It uses replaceFrom:to:with: which tries > to operate on non-existing indexes. copyWithAll: is just > used at two places > > MADistribution>>addPostscript: > and > MADistribution>>addPreamble: > > These make use of Collection>>copyWithAll: which works for > String objects. OrderedCollection>>copyWithAll: is never > used. > > So it isn't that important and might be better removed > > regards, > > Norbert > > > _______________________________________________ > SmallWiki, Magritte, Pier and Related Tools ... > https://www.iam.unibe.ch/mailman/listinfo/smallwiki -- Lukas Renggli http://www.lukas-renggli.ch -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 3774 bytes Desc: not available Url : http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070731/d0a66913/smime.bin From renggli at iam.unibe.ch Tue Jul 31 21:39:12 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Tue, 31 Jul 2007 21:39:12 +0200 Subject: "Creating Custom Forms" video question In-Reply-To: <14DCC2D3-49F4-4116-8F7E-639E002660A0@techgame.net> References: <14DCC2D3-49F4-4116-8F7E-639E002660A0@techgame.net> Message-ID: <0C497FA0-6877-40E3-A41B-6C987721AD43@iam.unibe.ch> > At some point I thought I saw a video demonstrating how to create > the birthday forms used in "Creating Custom Forms" video using the > web interface. The "Creating Custom Forms" video shows how to copy > one and modify it, but not how they got there in the first place. > Is that still around some where? I don't think that I created a video about that. As you say I just show how to modify an existing one. Creating a new one is not much different. Of course any contributions with new videos are welcome. Cheers, Lukas -- Lukas Renggli http://www.lukas-renggli.ch -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 3774 bytes Desc: not available Url : http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070731/895c68fe/smime-0001.bin From renggli at iam.unibe.ch Tue Jul 31 21:46:57 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Tue, 31 Jul 2007 21:46:57 +0200 Subject: Pier pointers In-Reply-To: <3434AB74-DB97-4391-B6C3-EE9A5BE7DA8F@techgame.net> References: <3434AB74-DB97-4391-B6C3-EE9A5BE7DA8F@techgame.net> Message-ID: <8F87828B-CDB1-45AB-9EF0-9A1572A8E399@iam.unibe.ch> > I have an existing seaside application that I would like to port > into the Pier framework, and need some pointers :) I am sure you saw the video 'Embedding Seaside Application' at http:// www.lukas-renggli.ch/smalltalk/pier/videos/sushi.mov > 1. I store some things on session; what is the Pier equivalent? The problem when using sessions is that they are global and introduce a tight coupling between your application/ components and the session. If you want to combine different components that expect different sessions you are screwed up. Of course Pier (or Seaside) could fake different Sessions per component (if necessary) but it doesn't at the moment. Exactly for the reasons given above, Pier doesn't use the Seaside session. Just tell your Pier application configuration to use your own Session class and proceed as shown in the Video. You application will be the only one to bother about the session so you shouldn't run into troubles. > 2. Do I have to move my renderContentOn: methods to renderWidgetOn:? No, in the latest versions #renderWidgetOn: is not used anymore. Pier uses #renderContentOn: like any other component. > 3. Should I subclass MAComponent or PRViewComponent or something > entirely different? Not necessarily, unless you want to use some of the convenience methods provided by Pier. A normal subclass of WAComponent does just fine in most cases. Hope this helps, Lukas -- Lukas Renggli http://www.lukas-renggli.ch -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 3774 bytes Desc: not available Url : http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070731/4acb83d1/smime.bin From renggli at iam.unibe.ch Tue Jul 31 21:49:36 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Tue, 31 Jul 2007 21:49:36 +0200 Subject: Serving Pier from a directory In-Reply-To: <46A947F0.9090701@yahoo.co.uk> References: <46A93161.50909@yahoo.co.uk> <46A947F0.9090701@yahoo.co.uk> Message-ID: >> is it possible to serve pier from a seaside entry point contained >> in a 'directory' sublevel? I don't know, I've never tried. > I dont understand the convoluted code in PRPierMain-structureFrom: > Why go to all the trouble of breaking up the path into pieces and then > putting it back together again? This makes it possible to serve Pier trough Apache at any URL. It also avoids some problems with older versions of Seaside where the parent node in the dispatcher was not available. I guess this is not necessary with Seaside 2.8 anymore and could be simplified. I simply didn't had the time to look at it yet. Cheers, Lukas -- Lukas Renggli http://www.lukas-renggli.ch -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 3774 bytes Desc: not available Url : http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070731/0ed5350a/smime.bin From renggli at iam.unibe.ch Tue Jul 31 21:56:28 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Tue, 31 Jul 2007 21:56:28 +0200 Subject: PRViewRenderer>>visitFile: In-Reply-To: <46AB5ECF.4040108@zephyr.dti.ne.jp> References: <46AB5ECF.4040108@zephyr.dti.ne.jp> Message-ID: <79119EFF-0BE9-4864-A02A-3D3D52FE8207@iam.unibe.ch> Thanks for reporting. This is fixed in Pier-All-lr.219. Lukas On 28.07.2007, at 17:20, NISHIHARA Satoshi wrote: > At Pier 1.0.9.201, should it be *file*? > > > PRViewRenderer>>visitFile: anObject > | file url | > file := anObject file ifNil: [ ^ self ]. > file isText > ifTrue: [ ^ html preformatted: anObject contents ]. > "^^^^^^^^" > > " file?" > url := file urlOn: html. > file isImage > ifTrue: [ ^ html image url: url; altText: anObject title ]. > html anchor url: url; with: anObject title > > -- > -------------------------------------------- > ???? (NISHIHARA Satoshi) > URL: http://www.zephyr.dti.ne.jp/~nishis/ > -------------------------------------------- > > > _______________________________________________ > SmallWiki, Magritte, Pier and Related Tools ... > https://www.iam.unibe.ch/mailman/listinfo/smallwiki -- Lukas Renggli http://www.lukas-renggli.ch -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 3774 bytes Desc: not available Url : http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070731/2ddf798c/smime.bin From renggli at iam.unibe.ch Tue Jul 31 22:01:00 2007 From: renggli at iam.unibe.ch (Lukas Renggli) Date: Tue, 31 Jul 2007 22:01:00 +0200 Subject: form element ids and names and testing In-Reply-To: <6c4e07720707310452q3d7b034cjdd974935533663f3@mail.gmail.com> References: <1185878756.23841.13.camel@localhost> <6c4e07720707310452q3d7b034cjdd974935533663f3@mail.gmail.com> Message-ID: <51CCF313-ADD9-4787-AAAD-147FE2DFFC3F@iam.unibe.ch> Sorry, I can't help you directly with your problem. Two ideas: 1. Modify Magritte so that you can specify the id manually. Again this could potentially lead to duplicated IDs on a single page, if you use the same description multiple times. 2. Use a different testing framework. Albatross is well known to work with Magritte (and Pier) and has been used for several scenarios. Have a look at: http://www.iam.unibe.ch/~scg/Archive/Projects/ Brue06a.pdf (Chapter 2). It uses a fuzzy JavaScript search (similar to your Xpath idea) to identify the form-element using its label. Cheers, Lukas -- Lukas Renggli http://www.lukas-renggli.ch -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 3774 bytes Desc: not available Url : http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20070731/68cd3bb0/smime-0001.bin