I’d like to help but… where is PharoCondenseChanges in the latest image?


On 13.04.2014, at 13:29, Ben Coman <[email protected]> wrote:

> Its cool the new PharoCondenseChanges was integrated 30821, presumably to be 
> used for Pharo 3 release.  I have a problem running it on MS Windows 
> (reported on Case 12729), but that is not critical for the Release since 
> presumably for that it only needs to run properly on one platform. 
> 
> However while investigating my MS Windows problem, it seemed that some 
> sources before and after were not the same.  This could be an consequence of 
> the MS Windows problem, so I was hoping that someone could report the result 
> of the following code on non-Windows platforms (presuming that this provides 
> a valid #condense function check.)
> ---------
> cacheMethodSourcesBlock :=
> [       | classes methods methodSources |
>         classes := (Smalltalk globals select: [  :g | g isClass ]) values.
>         methods := classes flatCollect: [ :c | c methods ].
>         methodSources := Dictionary new.
>         methods do: [ :m | methodSources at: m printString put: m asString ].
>         methodSources
> ].
> 
> methodSourcesBefore := cacheMethodSourcesBlock value.
> 
> PharoCondenseChanges condense.  "btw, this took over an hour for me. Is that 
> reasonable?"
> 
> methodSourcesAfter := cacheMethodSourcesBlock value.
> 
> sameAfter := Dictionary new.
> diffAfter := Dictionary new.
> methodSourcesBefore keys do:
> [       :methodName |  | methodSourceBefore methodSourceAfter |
>         methodSourceBefore := methodSourcesBefore at: methodName.
>         methodSourceAfter := methodSourcesAfter  at: methodName.
>         methodSourceBefore = methodSourceAfter 
>             ifTrue: [ sameAfter at: methodName put: { #beforeAndAfter -> 
> methodSourceBefore } ]
>             ifFalse: [ diffAfter at: methodName put: { #before -> 
> methodSourceBefore . String cr, String cr. #after -> methodSourceAfter }]
> ].
> 
> { #sameAfter -> sameAfter size . #diffAfter -> diffAfter size } inspect
> ---------
> 
> cheer -ben

Reply via email to