- Pharo-dev - lists.pharo.org

[Pharo-project] selectionForDoitAsStream smells
by Nicolas Cellier Feb. 27, 2012

Feb. 27, 2012

Compare this which is like old squeak selectionAsStream SmalltalkEditor>>selectionForDoitAsStream ^ ReadWriteStream on: self string from: self startIndex to: self stopIndex - 1 and this which has been changed conforming to Cuis TextEditor>>selectionAsStream "Answer a ReadStream on the text in the paragraph that is currently selected." ^ReadStream on: (self string copyFrom: self startIndex to: self stopIndex - 1) (or self selection asString readStream) The two differ. The former stream position starts at editor startIndex - 1, while the later stream position starts at 0. The stream position is used by the Compiler/Parser for notifying the SmalltalkEditor (or TextMorph) requestor. Now look at SmalltalkEditor>>notify: aString at: anInteger in: aStream "The compilation of text failed. The syntax error is noted as the argument, aString. Insert it in the text at starting character position anInteger." | pos | pos := self selectionInterval notEmpty ifTrue: [self startIndex + anInteger - 1 ] ifFalse: [anInteger]. self insertAndSelect: aString at: (pos max: 1) You see that it adds (startIndex - 1) to the error message insertion position which is a Cuis change. (The old ParagraphEditor did not). When #notify:at:in: is used for evaluating #selectionAsStream, all is OK, the (startIndex - 1) is added only once. But when it is used with #selectionForDoitAsStream, the offset is counted twice (because already in the stream location passed in anInteger parameter). I guess the hack #selectionForDoitAsStream that undo the Cuis clean-up has been added because some other part of Parser/Encoder/Compiler is messing with the requestor selection (like highlighting some part of the source for user interaction in case of ambiguity). Tthis is not a valid workaround. In Squeak, I did revert the Cuis clean-up just to keep some simple compatibility with ParagraphEditor, but Pharo does not care about that, so it should apply full simplification. I suggest removing #selectionForDoitAsStream, and replace senders with #selectionAsStream. Then if there are some other feature broken we'll fix'em. OK? Nicolas

2 4

Re: [Pharo-project] [Pharo-users] Nautilus version 4.0.4
by Mariano Martinez Peck Feb. 26, 2012

Feb. 26, 2012

> Ben, don't forget to cc pharo-users ;) > > > I do not get it :S > > When you do an ANN, do not only send it to pharo-project(a)lists.gforge.inria.fr but also to pharo-users(a)lists.gforge.inria.fr > > > Thanks again, > > Ben > > > Cheers > > > >> >> Gofer it >> url: 'http://ss3.gemstone.com/ss/Nautilus'; >> package: 'ConfigurationOfNautilus'; >> load. >> >> (Smalltalk at: #ConfigurationOfNautilus) loadRelease. >> >> >> >>> Thx, >>> >>> Sven >>> >>> >>> >> >> >> -- >> Mariano >> http://marianopeck.wordpress.com >> >> > > > -- > Mariano > http://marianopeck.wordpress.com > > > -- Mariano http://marianopeck.wordpress.com

1 0

[Pharo-project] [update 1.4] #14362
by Marcus Denker Feb. 26, 2012

Feb. 26, 2012

14362 ----- Issue 5402: AbstractTool fix http://code.google.com/p/pharo/issues/detail?id=5402 Issue 5403: remove some more references of #blockCopy: http://code.google.com/p/pharo/issues/detail?id=5403 -- Marcus Denker -- http://marcusdenker.de

1 0

[Pharo-project] can you test 14360 because it behaves strangely
by stephane ducasse Feb. 26, 2012

Feb. 26, 2012

The debugger opens really slowly. I cannot access ss3 repositoriesâ¦. So there is definitively something strange. Stf

3 2

Re: [Pharo-project] can you test 14360 because it behaves strangely
by Marcus Denker Feb. 26, 2012

Feb. 26, 2012

On Feb 26, 2012, at 5:48 PM, stephane ducasse wrote: > The debugger opens really slowly. > I cannot access ss3 repositoriesâ¦. > So there is definitively something strange. > No problem here with the latest from Jenkins: https://ci.lille.inria.fr/pharo/job/Pharo%201.4/ -- Marcus Denker -- http://marcusdenker.de

1 0

[Pharo-project] [update 1.4] #14361
by Marcus Denker Feb. 26, 2012

Feb. 26, 2012

14361 ----- Issue 5400: Remove reference to #blockCopy: from SpecialObjectsArray http://code.google.com/p/pharo/issues/detail?id=5400 Issue 5399: remove two empty Multiligual packages http://code.google.com/p/pharo/issues/detail?id=5399 Issue 5193: Nuke BorderedBlahBlahMorph http://code.google.com/p/pharo/issues/detail?id=5193 -- Marcus Denker -- http://marcusdenker.de

1 0

Re: [Pharo-project] How to get refactorings with nautilus
by Marcus Denker Feb. 26, 2012

Feb. 26, 2012

On Feb 26, 2012, at 3:39 PM, StÃ©phane Ducasse wrote: > I took Nautilus with refactoring on jenkins but I do not find the refactorings. > In the menu? But you need to take care that there is right know no way to get the menu other then right-click. Command-Click which is normally the same is not the same in this case... Marcus -- Marcus Denker -- http://marcusdenker.de

4 6

Feb. 26, 2012

14360 ----- - Issue 5392: Add some announcement to Metacello. Thanks Benjamin van Ryseghem. http://code.google.com/p/pharo/issues/detail?id=5392 - Issue 5398: Clean up #isBlueBookCompiled.Thanks Marcus Denker. http://code.google.com/p/pharo/issues/detail?id=5398 - Issue 5397: Cleanup Multilingual Fonts more... Thanks Marcus Denker. http://code.google.com/p/pharo/issues/detail?id=5397 - Issue 5391: Clean Font Donwloading. Thanks Marcus Denker. http://code.google.com/p/pharo/issues/detail?id=5391 - Issue 5388: Clean BlockNode old style code. Thanks Marcus Denker. http://code.google.com/p/pharo/issues/detail?id=5388 Stef

1 0

Feb. 26, 2012

I took Nautilus with refactoring on jenkins but I do not find the refactorings. Stef

1 0

Re: [Pharo-project] [update 1.4] #14355
by Marcus Denker Feb. 26, 2012

Feb. 26, 2012

On Feb 25, 2012, at 5:20 PM, StÃ©phane Ducasse wrote: > Excellent! > Pavel > We should add Pharo Kernel as dependent of the head so that we get immediate feedback like that. > because this is really great. > It already is configured to be build after every update: https://ci.lille.inria.fr/pharo/job/Pharo%20Kernel%201.4/ tests, too: https://ci.lille.inria.fr/pharo/job/Pharo%20Kernel%20Tests/ Now we need to take action when it breaks... -- Marcus Denker -- http://marcusdenker.de

4 6