Emergency evaluator broken?
Locked
 
|
I've not run into the emergency evaluator, but I wonder: Transcript
class >> #emergencyEvaluator calls Transcript >> #newInFrame:, which calls #initInFrame: which checks for a class called Paragraph. (Trunk doesn't have such a class, because - AFAIK at least - it's an ST80 class). If there's no such class, _it raises an #inform:_. In an emergency, that's probably not what you want! So: has anyone used the emergency evaluator of late? Have you used it in an image without ST80? Certainly the example usage in Transcript >> #newInFrame: is broken: (Transcripter newInFrame: (0@0 extent: 100@200)) nextPutAll: 'Hello there'; endEntry; cr; print: 355.0/113; endEntry; readEvalPrint. frank |
Re: Emergency evaluator broken?
|
|
Move Transcripter to ST80. Then check for presence of class named #Transcripter in Object>>#primitiveError: to determine whether to present and accept that option.
On Thu, Dec 12, 2013 at 1:33 PM, Frank Shearar <[hidden email]> wrote: I've not run into the emergency evaluator, but I wonder: Transcript |
Re: Emergency evaluator broken?
|
|
In reply to this post by Frank Shearar-3
Good catch. It's been a while since I tried unloading/reloading MVC, but
I'm fairly sure that I never tried to do anything that would invoke it while I had MVC unloaded from the image. So I cannot say if it works in Morphic-only or not. Assuming that it is a problem, we should see how Juan handles it in Cuis (which has no MVC, and presumably does have a working emergency evaluator). I can definitely confirm that the emergency evaluator works in MVC, because I fell into it about a zillion times in the course of getting the debugger to work in MVC again ;-) Dave > I've not run into the emergency evaluator, but I wonder: Transcript > class >> #emergencyEvaluator calls Transcript >> #newInFrame:, which > calls #initInFrame: which checks for a class called Paragraph. (Trunk > doesn't have such a class, because - AFAIK at least - it's an ST80 > class). If there's no such class, _it raises an #inform:_. In an > emergency, that's probably not what you want! > > So: has anyone used the emergency evaluator of late? Have you used it > in an image without ST80? > > Certainly the example usage in Transcript >> #newInFrame: is broken: > > (Transcripter newInFrame: (0@0 extent: 100@200)) > nextPutAll: 'Hello there'; endEntry; > cr; print: 355.0/113; endEntry; > readEvalPrint. > > frank > |
Re: Emergency evaluator broken?
|
|
In reply to this post by Chris Muller-3
On 12 December 2013 19:57, Chris Muller <[hidden email]> wrote:
> Move Transcripter to ST80. > > Then check for presence of class named #Transcripter in > Object>>#primitiveError: to determine whether to present and accept that > option. ... and also get an emergency evaluator working in Morphic. frank > On Thu, Dec 12, 2013 at 1:33 PM, Frank Shearar <[hidden email]> > wrote: >> >> I've not run into the emergency evaluator, but I wonder: Transcript >> class >> #emergencyEvaluator calls Transcript >> #newInFrame:, which >> calls #initInFrame: which checks for a class called Paragraph. (Trunk >> doesn't have such a class, because - AFAIK at least - it's an ST80 >> class). If there's no such class, _it raises an #inform:_. In an >> emergency, that's probably not what you want! >> >> So: has anyone used the emergency evaluator of late? Have you used it >> in an image without ST80? >> >> Certainly the example usage in Transcript >> #newInFrame: is broken: >> >> (Transcripter newInFrame: (0@0 extent: 100@200)) >> nextPutAll: 'Hello there'; endEntry; >> cr; print: 355.0/113; endEntry; >> readEvalPrint. >> >> frank >> > > > > |
Re: Emergency evaluator broken?
|
|
On 12 December 2013 20:17, Frank Shearar <[hidden email]> wrote:
> On 12 December 2013 19:57, Chris Muller <[hidden email]> wrote: >> Move Transcripter to ST80. >> >> Then check for presence of class named #Transcripter in >> Object>>#primitiveError: to determine whether to present and accept that >> option. > > ... and also get an emergency evaluator working in Morphic. I'll do the class moving, but I'm not tackling Morphic. frank > frank > >> On Thu, Dec 12, 2013 at 1:33 PM, Frank Shearar <[hidden email]> >> wrote: >>> >>> I've not run into the emergency evaluator, but I wonder: Transcript >>> class >> #emergencyEvaluator calls Transcript >> #newInFrame:, which >>> calls #initInFrame: which checks for a class called Paragraph. (Trunk >>> doesn't have such a class, because - AFAIK at least - it's an ST80 >>> class). If there's no such class, _it raises an #inform:_. In an >>> emergency, that's probably not what you want! >>> >>> So: has anyone used the emergency evaluator of late? Have you used it >>> in an image without ST80? >>> >>> Certainly the example usage in Transcript >> #newInFrame: is broken: >>> >>> (Transcripter newInFrame: (0@0 extent: 100@200)) >>> nextPutAll: 'Hello there'; endEntry; >>> cr; print: 355.0/113; endEntry; >>> readEvalPrint. >>> >>> frank >>> >> >> >> >> |
Re: Emergency evaluator broken?
|
|
On 12 December 2013 20:18, Frank Shearar <[hidden email]> wrote:
> On 12 December 2013 20:17, Frank Shearar <[hidden email]> wrote: >> On 12 December 2013 19:57, Chris Muller <[hidden email]> wrote: >>> Move Transcripter to ST80. >>> >>> Then check for presence of class named #Transcripter in >>> Object>>#primitiveError: to determine whether to present and accept that >>> option. >> >> ... and also get an emergency evaluator working in Morphic. > > I'll do the class moving, but I'm not tackling Morphic. Or not, because moving the class is the wrong thing to do. Instead, the _right_ thing to do is to make a MorphicTranscripter and an St80Transcripter. The latter is just Transcripter with a new name, and the new Transcripter is an AppRegistry subclass, presumably in System-Applications. Morphic and ST80 then register their preferred Transcripter-protocol-fulfiller. > frank > >> frank >> >>> On Thu, Dec 12, 2013 at 1:33 PM, Frank Shearar <[hidden email]> >>> wrote: >>>> >>>> I've not run into the emergency evaluator, but I wonder: Transcript >>>> class >> #emergencyEvaluator calls Transcript >> #newInFrame:, which >>>> calls #initInFrame: which checks for a class called Paragraph. (Trunk >>>> doesn't have such a class, because - AFAIK at least - it's an ST80 >>>> class). If there's no such class, _it raises an #inform:_. In an >>>> emergency, that's probably not what you want! >>>> >>>> So: has anyone used the emergency evaluator of late? Have you used it >>>> in an image without ST80? >>>> >>>> Certainly the example usage in Transcript >> #newInFrame: is broken: >>>> >>>> (Transcripter newInFrame: (0@0 extent: 100@200)) >>>> nextPutAll: 'Hello there'; endEntry; >>>> cr; print: 355.0/113; endEntry; >>>> readEvalPrint. >>>> >>>> frank >>>> >>> >>> >>> >>> |
Re: Emergency evaluator broken?
|
|
On Thu, Dec 12, 2013 at 08:25:40PM +0000, Frank Shearar wrote:
> On 12 December 2013 20:18, Frank Shearar <[hidden email]> wrote: > > On 12 December 2013 20:17, Frank Shearar <[hidden email]> wrote: > >> On 12 December 2013 19:57, Chris Muller <[hidden email]> wrote: > >>> Move Transcripter to ST80. > >>> > >>> Then check for presence of class named #Transcripter in > >>> Object>>#primitiveError: to determine whether to present and accept that > >>> option. > >> > >> ... and also get an emergency evaluator working in Morphic. > > > > I'll do the class moving, but I'm not tackling Morphic. > > Or not, because moving the class is the wrong thing to do. Instead, > the _right_ thing to do is to make a MorphicTranscripter and an > St80Transcripter. The latter is just Transcripter with a new name, and > the new Transcripter is an AppRegistry subclass, presumably in > System-Applications. Morphic and ST80 then register their preferred > Transcripter-protocol-fulfiller. Eeek! Holy overengineering batman. Hopefully Collections-dtl.548 will resolve the problem. Note to self: Touching a method that carries the initials 'di' is a fast path to public embarrassment. Not that I did not already know this... Dave |
«
Return to Squeak - Dev
|
1 view|%1 views
| Free forum by Nabble | Edit this page |