February 2011 - Pharo-dev - lists.pharo.org

[Pharo-project] Issue 3682 in pharo: multiple undo/redo is broken in 1.2 because of Ocompletion.
by pharo＠googlecode.com Feb. 10, 2011

Feb. 10, 2011

Status: Accepted Owner: aplan...(a)gmail.com Labels: Milestone-1.2 New issue 3682 by aplan...(a)gmail.com: multiple undo/redo is broken in 1.2 because of Ocompletion. http://code.google.com/p/pharo/issues/detail?id=3682 I've found the fix. It has two parts, one part for the core and the second is for ocompletion.

1 9

[Pharo-project] Issue 3675 in pharo: ProcessBrowser>>cleanup
by pharo＠googlecode.com Feb. 10, 2011

Feb. 10, 2011

Status: Accepted Owner: stephane...(a)gmail.com CC: laurent....(a)gmail.com Labels: Milestone-1.3 Type-Squeak Difficulty-Easy New issue 3675 by stephane...(a)gmail.com: ProcessBrowser>>cleanup http://code.google.com/p/pharo/issues/detail?id=3675 Name: Tools-mtf.290 Author: mtf Time: 12 January 2011, 7:36:30.491 pm UUID: 4f8986b8-f97f-416f-989f-2b0b686d9c73 Ancestors: Tools-cmm.289 add a cleanUp method to ProcessBrowser as it can sometimes hold on to terminated processes =============== Diff against Tools-cmm.289 =============== Item was added: + ----- Method: ProcessBrowser class>>cleanUp (in category 'class initialization') ----- + cleanUp + "Remove terminated processes from my suspended list" + + self suspendedProcesses keys do: [:ea | + ea isTerminated ifTrue: [self suspendedProcesses removeKey: ea]]!

1 2

[Pharo-project] Issue 3659 in pharo: Task bar appearance does not make sense
by pharo＠googlecode.com Feb. 10, 2011

Feb. 10, 2011

Status: Accepted Owner: damien.p...(a)gmail.com Labels: Type-RequestForEnhancement Importance-Low New issue 3659 by damien.p...(a)gmail.com: Task bar appearance does not make sense http://code.google.com/p/pharo/issues/detail?id=3659 Pharo 1.2rc2 Pharo core version: 12330 The buttons in the task bar are: - frontmost window: gray background + black text - background windows: gray background + gray text - hidden windows: blue gradient + gray text To me these are bad affordances, because gray text signifies "inactive or hidden" and color background means "active" or "current" (see the instance/class side buttons in the browser. So I'd change the appearances to reflect that: - window visibility = text color (black = visible, gray = hidden) - gradient color = blue for the frontmost window\

1 14

[Pharo-project] Issue 3661 in pharo: fixed the calculation of base in PNGReadWriter>>copyPixelsGray: (floor vs ceil)
by pharo＠googlecode.com Feb. 10, 2011

Feb. 10, 2011

Status: FixedWaitingToBePharoed Owner: stephane...(a)gmail.com Labels: Milestone-1.3 Type-Squeak New issue 3661 by stephane...(a)gmail.com: fixed the calculation of base in PNGReadWriter>>copyPixelsGray: (floor vs ceil) http://code.google.com/p/pharo/issues/detail?id=3661 Name: Graphics-ul.177 Author: ul Time: 5 February 2011, 4:53:19.27 am UUID: dfac9cf3-85a4-3c4b-9256-1c8214f36fe6 Ancestors: Graphics-nice.176 - fixed the calculation of base in PNGReadWriter>>copyPixelsGray: (floor vs ceil) =============== Diff against Graphics-nice.176 =============== Item was changed: ----- Method: PNGReadWriter>>copyPixelsGray: (in category 'pixel copies') ----- copyPixelsGray: y "Handle non-interlaced grayscale color mode (colorType = 0)" | base bits bytesLeft word | bitsPerChannel = 16 ifTrue: [ "Warning: This is extremely slow. Besides we are downsampling to 8 bits!!" | blitter | blitter := BitBlt current bitPokerToForm: form. 0 to: width - 1 do: [ :x | blitter pixelAt: x @ y put: 255 - (thisScanline at: x * 2 + 1) ]. ^self ]. "Just copy the bits" "This Smalltalk version might be easier to understand than the others below." + base := y * (form width * bitsPerChannel + 31 // 32) + 1. - base := y * form width * bitsPerChannel // 32 + 1. bits := form bits. 0 to: thisScanline size // 4 - 1 do: [ :i | | ii | ii := i * 4. "This somewhat weird mixture of (#* and #+) with (#bitShift: and #bitOr:) is to make use of faster arithmetic bytecodes, but not of slow largeintegers." word := (((thisScanline at: ii + 1) * 256 + (thisScanline at: ii + 2) * 256 + (thisScanline at: ii + 3)) bitShift: 8) bitOr: (thisScanline at: ii + 4). bits at: base + i put: word ]. (bytesLeft := thisScanline size bitAnd: 3) = 0 ifFalse: [ word := 0. thisScanline size - bytesLeft + 1 to: thisScanline size do: [ :ii | word := word * 256 + (thisScanline at: ii) ]. word := word bitShift: 8 * (4 - bytesLeft). bits at: base + (thisScanline size // 4) put: word ]. "This interesting technique (By Andreas Raab) is faster for very large images, but might be slower for small ones" "^self copyPixelsGrayWeirdBitBltHack: y ". "It uses the following method: PNGReadWriter >> copyPixelsGrayWeirdBitBltHack: y ""Handle non-interlaced black and white color mode (colorType = 0) By Andreas Raab"" | source dest cmap | source := Form extent: 1 @ (thisScanline size // 4) depth: 32 bits: thisScanline. dest := Form extent: 1 @ (form bits size) depth: 32 bits: form bits. cmap := Smalltalk isLittleEndian ifTrue:[ColorMap shifts: #(-24 -8 8 24) masks: #(16rFF000000 16r00FF0000 16r0000FF00 16r000000FF)]. (BitBlt toForm: dest) sourceForm: source; destX: 0 destY: (y * form width*bitsPerChannel//32) width: 1 height: (form width+31*bitsPerChannel//32); colorMap: cmap; combinationRule: 3; copyBits." "This interesting technique (By Yoshiki Ohshima) is faster for very large images, but might be slower for small ones" "form bits copyFromByteArray2: thisScanline to: y * (form width* bitsPerChannel // 32)". "It uses the following method: BitMap >> copyFromByteArray2: byteArray to: i ""This method should work with either byte orderings"" | myHack byteHack | myHack := Form new hackBits: self. byteHack := Form new hackBits: byteArray. Smalltalk isLittleEndian ifTrue: [byteHack swapEndianness]. byteHack displayOn: myHack at: 0@i"!

1 3

[Pharo-project] Issue 3676 in pharo: Remove #example* class messages from HTTPClient
by pharo＠googlecode.com Feb. 10, 2011

Feb. 10, 2011

Status: New Owner: ---- New issue 3676 by patchi...(a)gmail.com: Remove #example* class messages from HTTPClient http://code.google.com/p/pharo/issues/detail?id=3676 Pharo image: Pharo-core Pharo core version: all of them Since we're documenting HTTPClient it would be great to remove class side messages #examplePostArgs and #examplePostMultipart

1 2

[Pharo-project] Issue 3609 in pharo: Integrate RPackage
by pharo＠googlecode.com Feb. 10, 2011

Feb. 10, 2011

Status: Accepted Owner: cy.delau...(a)gmail.com New issue 3609 by cy.delau...(a)gmail.com: Integrate RPackage http://code.google.com/p/pharo/issues/detail?id=3609 RPackage aims to replace PackageInfo with a faster and cleaner implementation. What RPackage is already able to do (or at least until you find something wrong;)) is: - synchronize and initialize with the current PackageOrganizer in the system (RPackageOrhanizer class>> initializeDefault) - update itself when changes are made in the system (new category, new class, new method, ... ) (see all tests in RPackageMCSynchronisationTest). To point I am working on is to modify monticello to make it work using RPackage. I have currently A version that does not seem to crash all (:) that means that it save and load from a repository without error, and viewing changes seems to match with what you would expect). Now what could be cool is that the monticello tests green. To have all this running in your image, that's what should evaluate: Gofer new squeaksource: 'PharoTaskForces'; package: 'ConfigurationOfRPackage'; load. (Smalltalk globals at: #ConfigurationOfRPackage) loadDefault. "to be sure that the organizer is initialized before loading anything else:" (Smalltalk globals at: #RPackageOrganizer) initializeDefault. Gofer new squeaksource: 'DelaunayTmpStuffs'; version: 'Monticello-cyrilledelaunay.505'; merge. Gofer new squeaksource: 'DelaunayTmpStuffs'; version: 'MonticelloGUI-cyrilledelaunay.88'; merge.

1 10

[Pharo-project] Issue 3688 in pharo: Network primitives missing in the image
by pharo＠googlecode.com Feb. 10, 2011

Feb. 10, 2011

Status: Accepted Owner: jannik.l...(a)gmail.com Labels: Difficulty-Easy New issue 3688 by jannik.l...(a)gmail.com: Network primitives missing in the image http://code.google.com/p/pharo/issues/detail?id=3688 There are some primitives missing in Pharo, there are in eToys. Maybe we can include them primGetAddressInfoFamily <primitive: 'primitiveResolverGetAddressInfoFamily' module: 'SocketPlugin'> self primitiveFailed === primGetAddressInfoHost: hostName service: servName flags: flags family: family type: type protocol: protocol <primitive: 'primitiveResolverGetAddressInfo' module: 'SocketPlugin'> self primitiveFailed === primGetAddressInfoNext <primitive: 'primitiveResolverGetAddressInfoNext' module: 'SocketPlugin'> self primitiveFailed === primGetAddressInfoProtocol <primitive: 'primitiveResolverGetAddressInfoProtocol' module: 'SocketPlugin'> self primitiveFailed === primGetAddressInfoResult: socketAddress <primitive: 'primitiveResolverGetAddressInfoResult' module: 'SocketPlugin'> self primitiveFailed === primGetAddressInfoSize <primitive: 'primitiveResolverGetAddressInfoSize' module: 'SocketPlugin'> self primitiveFailed === primGetAddressInfoType <primitive: 'primitiveResolverGetAddressInfoType' module: 'SocketPlugin'> self primitiveFailed === primGetNameInfo: socketAddress flags: flags <primitive: 'primitiveResolverGetNameInfo' module: 'SocketPlugin'> flags == 0 ifTrue: [^self primGetNameInfo: socketAddress flags: SocketAddressInformation numericFlag]. self primitiveFailed === primGetNameInfoHostResult: aString <primitive: 'primitiveResolverGetNameInfoHostResult' module: 'SocketPlugin'> self primitiveFailed === primGetNameInfoHostSize <primitive: 'primitiveResolverGetNameInfoHostSize' module: 'SocketPlugin'> self primitiveFailed === primGetNameInfoServiceResult: aString <primitive: 'primitiveResolverGetNameInfoServiceResult' module: 'SocketPlugin'> self primitiveFailed === primGetNameInfoServiceSize <primitive: 'primitiveResolverGetNameInfoServiceSize' module: 'SocketPlugin'> self primitiveFailed === primHostNameResult: aString <primitive: 'primitiveResolverHostNameResult' module: 'SocketPlugin'> self primitiveFailed === primHostNameSize <primitive: 'primitiveResolverHostNameSize' module: 'SocketPlugin'> self primitiveFailed

1 4

Feb. 10, 2011

13045 ----- - Issue 3688: Network primitives missing in the image. Thanks Jannik Laval Stef and the Anneciens sprinting :)

1 0

[Pharo-project] Issue 3690 in pharo: Monticello Browser: improve submit slice dialog
by pharo＠googlecode.com Feb. 10, 2011

Feb. 10, 2011

Status: Accepted Owner: laurent....(a)gmail.com CC: aplan...(a)gmail.com New issue 3690 by laurent....(a)gmail.com: Monticello Browser: improve submit slice dialog http://code.google.com/p/pharo/issues/detail?id=3690 We are currently working on version 1.2, so please verify your issue with the latest 1.2 Unstable version before reporting it - we improved a lot the system. If you fill an issue for the first time, please read "How to report bugs" at http://www.pharo-project.org/community/issue-tracking Pharo image: Pharo-core Pharo core version: Pharo1.2rc2 Latest update: #12332 In Monticello Browser, improve GUI for submitting a fix: - +Slice -> rename to 'Contribute a fix' - in Slice GUI: - put a hint like 'Copy paste issue number and summary from http://code.google.com/p/pharo/issues' - change label 'Included packages' to 'Choose packages to include' - in packages list => checkbox From PharoConf Annecy 2011

1 1

Feb. 10, 2011

In SMxCanvas>>copyClipRect: aRectangle ^ self copyOrigin: origin clipRect: (aRectangle translateBy: origin) SMxFormCanvas>>copyOrigin: aPoint clipRect: aRectangle "Return a copy of this canvas with the given origin. The clipping rectangle of this canvas is the intersection of the given rectangle and the receiver's current clipping rectangle. This allows the clipping rectangles of nested clipping morphs to be composed." ^ self copy setOrigin: aPoint clipRect: (clipRect intersect: aRectangle) Canvas>>copyClipRect: newClipRect ^ ClippingCanvas canvas: self clipRect: newClipRect FormCanvas>>copyOrigin: aPoint clipRect: aRectangle "Return a copy of this canvas with the given origin. The clipping rectangle of this canvas is the intersection of the given rectangle and the receiver's current clipping rectangle. This allows the clipping rectangles of nested clipping morphs to be composed." ^ self copy setOrigin: aPoint clipRect: (clipRect intersect: aRectangle) Do you think that we can remove ClippingCanvas. There are too many canvases in Pharo. Stef

2 1