Cool. I just merged it and I can confirm that it solves my issue.
> On 05 Jan 2016, at 16:19, Marcus Denker <[email protected]> wrote: > > yes, > > this is fixed by > > https://pharo.fogbugz.com/f/cases/17276/Semantic-Analysis-needs-refactoring > > I am now back from holidays and will review/intergrate this soon. > > >> On 05 Jan 2016, at 16:15, Sven Van Caekenberghe <[email protected]> wrote: >> >> I am seeing the following very annoying error in 50511 spur. >> >> Open a Playground, >> type [ (both open en close are inserted, with the cursor in-between), >> type : (to start a block arg), >> boom: >> >> RBParseErrorNode(Object)>>doesNotUnderstand: #isInstance >> OCASTClosureAnalyzer>>changeVariableClass: >> OCASTClosureAnalyzer>>visitArgumentNode: >> [ :each | self visitArgumentNode: each ] in >> OCASTClosureAnalyzer(RBProgramNodeVisitor)>>visitArgumentNodes: >> OrderedCollection>>do: >> OCASTClosureAnalyzer(RBProgramNodeVisitor)>>visitArgumentNodes: >> OCASTClosureAnalyzer>>visitBlockNode: >> RBBlockNode>>acceptVisitor: >> OCASTClosureAnalyzer(RBProgramNodeVisitor)>>visitNode: >> [ :each | self visitNode: each ] in >> OCASTClosureAnalyzer(RBProgramNodeVisitor)>>visitSequenceNode: >> OrderedCollection>>do: >> OCASTClosureAnalyzer(RBProgramNodeVisitor)>>visitSequenceNode: >> RBSequenceNode>>acceptVisitor: >> OCASTClosureAnalyzer(RBProgramNodeVisitor)>>visitNode: >> OCASTClosureAnalyzer>>visitMethodNode: >> RBMethodNode>>acceptVisitor: >> OCASTClosureAnalyzer(RBProgramNodeVisitor)>>visitNode: >> OCASTSemanticAnalyzer>>analyze: >> RBMethodNode>>doSemanticAnalysis >> RBBlockNode(RBProgramNode)>>doSemanticAnalysis >> RubSHTextStylerST80(SHRBTextStyler)>>privateStyle: >> [ self privateStyle: text. >> view ifNotNil: [ view stylerStyledInBackground: text ] ] in [ >> backgroundProcess := [ self privateStyle: text. >> view ifNotNil: [ view stylerStyledInBackground: text ] ] >> forkAt: Processor userBackgroundPriority ] in >> RubSHTextStylerST80(SHTextStyler)>>styleInBackgroundProcess: >> [ self value. >> Processor terminateActive ] in BlockClosure>>newProcess >> >> Has this already been reported ? >> >> It makes the image almost unusable... >> >> Sven >> >> > >
