Issue 146 in metacello: unrecognized SystemVersion results in assuming system is on #gemstone...

Status: Accepted
Owner: [hidden email]
Labels: Type-Defect Priority-Medium Product-Core Version-1.0-beta.30

New issue 146 by [hidden email]: unrecognized SystemVersion results in  
assuming system is on #gemstone...
http://code.google.com/p/metacello/issues/detail?id=146

see  
http://forum.world.st/Seaside-3-0-Metacello-Configuration-broken-for-Squeak-td3720943.html 
for more details.

Here's the method in MetacelloPlatform:

   defaultPlatformAttributes
        | versionString |
        Smalltalk at: #SystemVersion ifPresent: [:cl |
                versionString := cl current version.
                (versionString beginsWith: 'Pharo')
                        ifTrue: [ ^ #(#squeakCommon #pharo ) ].
                (versionString beginsWith: 'Squeak')
                        ifTrue: [^ #(#squeakCommon #squeak )]].
        ^ #(#gemstone )

I'm pretty sure that the logic is "required" for the system to bootstrap in  
GemStone, but the fact that SystemVersion should preclude answering  
#gemstone ... guessing wrong ends up with surprising results, so should  
throw an error if the versionString doesn't begin with 'Pharo'  
or 'Squeak'...should have a solid test for GemStone as well so that an  
error can be thrown as well ...

Comment #1 on issue 146 by [hidden email]: unrecognized SystemVersion  
results in assuming system is on #gemstone...
http://code.google.com/p/metacello/issues/detail?id=146

Note that I've been thinking that the base system should implement a  
#defaultPlatformAttributes method, so I wouldn't have to poke around in  
various places to guess what the attributes should be ... same thing goes  
for the version-specific attributes (like #gemstone3.x or squeak4.2 or  
#pharo1.3) since I'd rather not have to ship a new version Metacello to do  
the additional tests ... don't know if I can get across the board  
adoption...

Comment #2 on issue 146 by [hidden email]: unrecognized  
SystemVersion results in assuming system is on #gemstone...
http://code.google.com/p/metacello/issues/detail?id=146

Tell us what would make your life easier. For me I would love to have a  
simple class that avoid all the duplication :)

Great ... I'll think on it a bit and let you know ...

Dale

----- Original Message -----
| From: [hidden email]
| To: [hidden email]
| Sent: Monday, August 8, 2011 12:19:26 AM
| Subject: [Metacello] Re: Issue 146 in metacello: unrecognized SystemVersion results in assuming system is on
| #gemstone...
|
|
| Comment #2 on issue 146 by [hidden email]: unrecognized
| SystemVersion results in assuming system is on #gemstone...
| http://code.google.com/p/metacello/issues/detail?id=146
|
| Tell us what would make your life easier. For me I would love to have
| a
| simple class that avoid all the duplication :)
|
|

Comment #3 on issue 146 by [hidden email]: unrecognized SystemVersion  
results in assuming system is on #gemstone...
http://code.google.com/p/metacello/issues/detail?id=146

http://code.google.com/p/pharo/issues/detail?id=4613 is open for collecting  
input ... I can make a similar addition for GemStone ....

Updates:
        Status: Fixed
        Labels: Fixed-1.0-beta.30

Comment #4 on issue 146 by [hidden email]: unrecognized SystemVersion  
results in assuming system is on #gemstone...
http://code.google.com/p/metacello/issues/detail?id=146

took care of both problems: extra checks for match before throwing an error  
and use SmalltalkImage>>metacelloPlatformAttributes for attributes, if  
present

Metacello 1.0-beta.30 (dkh.627)

Comment #5 on issue 146 by [hidden email]: unrecognized  
SystemVersion results in assuming system is on #gemstone...
http://code.google.com/p/metacello/issues/detail?id=146

Issue 142 has been merged into this issue.