December 2009 - Pharo-dev - lists.pharo.org

[Pharo-project] SystemVersion 1.1 Part 2
by Michael Roberts Dec. 14, 2009

Dec. 14, 2009

http://code.google.com/p/pharo/issues/detail?id=1566 I have uploaded to PharoInbox an upgrade for SystemVersion so that the various images can be better aware of their version state. After the upgrade SystemVersion current type --> #PharoCore major --> 1 minor --> 1 suffix --> ALPHA Then the Pharo build script needs SystemVersion current type: #Pharo I didn't want to make it too complicated. It is also backwards compatible with the old API until the deprecation is thought through. The idea of doing this is to make automated image building and testing a little easier. thanks, Mike

2 2

[Pharo-project] Fwd: [squeak-dev] [ANN] CoroutineReadStream (again)
by Mariano Martinez Peck Dec. 14, 2009

Dec. 14, 2009

I didn't understand it too much, but maybe it is worth to check it ? Cheers, Mariano ---------- Forwarded message ---------- From: Stephen Pair <stephen(a)pairhome.net> Date: Mon, Dec 14, 2009 at 2:22 AM Subject: [squeak-dev] [ANN] CoroutineReadStream (again) To: The general-purpose Squeak developers list < squeak-dev(a)lists.squeakfoundation.org> This version fixes a bug in #close that causes the coroutine to not get unwound when the stream is closed. Also, it now uses the stream itself as a marker instead of a special string (a suggestion from Eliot). I've also created unit tests (which is how I caught the problem with close). Below is the class comment: ----- I enable the use of a ReadStream interface in cases where enumeration protocol is available, but no read stream interface is available. For example, I can adapt any object that supports #do: for enumeration: CoroutineReadStream on: #(1 2 3) iterator: #do: The implementation makes use of coroutines to avoid the need to create and schedule a separate Process. Sending #close will abandon the coroutine and unwind all contexts in the coroutine. The coroutine is not invoked until the first request for an object. The iterator is finished when there is an attempt to read past the end (via #next or #peek), or when there is an #atEnd test after the last object was retrieved. A CoroutineReadSteam is not particularly efficient due to its use of thisContext and how stacks are optimized in most Smalltalk implementations. Instance Variables: nextValue <Object | noObjectMarker> Holds the next value to be answered when #next is called (needed to support #peek and #atEnd)...if it holds the noObjectMarker, it inticates that the coroutine needs to be invoked to get the next object. suspendedContext <ContextPart> This is either the context of the iterating stack (when not actively retrieving the next object for a client) or the stack of the user of the stream (when of retrieving the next object) homeContext <MethodContext> Holds the context of #initializeForBlock: called during instantiation and is used to ensure that exception propagation and handling works as one would expect. Copyright (c) 2009 Stephen Pair Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. - Stephen

1 0

[Pharo-project] Can't browse slices in Inbox
by Martin McClure Dec. 14, 2009

Dec. 14, 2009

In a Monticello browser on the PharoInbox, I haven't been able to browse slices. When I press the Browse button I get a browser "Snapshot of SLICE...". In the class category pane the only thing listed is '*Extensions' and that has nothing in it. Is this a bug, or am I going about this wrong? Regards, -Martin

4 6

[Pharo-project] [update 1.1] #11101
by Marcus Denker Dec. 14, 2009

Dec. 14, 2009

11101 ----- Issue 1601: World menu continues to look strange (11099) Thanks: Alain Plantec Issue 1457: CharacterSet removeAll (part two) Thanks: Nicolas Cellier Issue 1065: WriteStream nextPut: has unnecessary coupling to/performance decrease with ByteStrings Thanks: rydier

1 0

[Pharo-project] package documentation viewer
by laurent laffont Dec. 14, 2009

Dec. 14, 2009

Hi, for fun and *for me*, I've made a small Seaside app to quickly browse all documentation in a Pharo image. It's package oriented, all classes and selectors comment on a single page, so I can use search and print functions of my web browser. Someone might have a use of it. It's here: http://www.squeaksource.com/SimpleWebDoc.html Screenshots: http://picasaweb.google.fr/laurent.laffont/Simplewebdoc Laurent

5 10

[Pharo-project] [update 1.1] #11100
by Damien Cassou Dec. 14, 2009

Dec. 14, 2009

11100 ----- Issue 1595: ReadStream upTo: and upToEnd do not honour the readLimit (thank you Nicolas Cellier) -- Damien Cassou http://damiencassou.seasidehosting.st "Lambdas are relegated to relative obscurity until Java makes them popular by not having them." James Iry

1 0

[Pharo-project] Gofer git
by StÃ©phane Ducasse Dec. 14, 2009

Dec. 14, 2009

Hi guys I do not know if you noticed that with the latest version of Gofer you can get a full git behavior Gofer new squeaksource: 'PharoTaskForces'; addPackage: 'RPackage'; pull to pull all the packages pubslihed and fill up your local cache Gofer new squeaksource: 'PharoTaskForces'; addPackage: 'RPackage'; push to push your local saved versions to the repository.... Of course it means that you can build a database and when you run your script the cache is checked first.... so yes you can work in the train :) Tx lukas

5 7

[Pharo-project] [update 1.1] #11098
by Marcus Denker Dec. 14, 2009

Dec. 14, 2009

11098 ----- Issue 1596: DNU when entering a keystroke in the world Issue 1598: Preferences migration part9 * migrate "halo" and "menu" groups * remove ColorTheme class * remove SmalltalkImage>>setPlatformPreferences * make the "about" dialog open at a fixed position (near top left)

2 4

[Pharo-project] [update 1.1] #11099
by Marcus Denker Dec. 14, 2009

Dec. 14, 2009

11099 ----- Issue 1582: Clean Exceptions exception kernel is full of exception that could be put closer to their user.... Issue 1599: After Preferences Migration 9, Menues look strange

1 0

[Pharo-project] FAME: support for multiple inheritance?
by alesch Dec. 14, 2009

Dec. 14, 2009

Hello list: Does FAME supports mutltiple inheritance? I guess the answer is NO, but I thought I would ask to be sure. The scenario is a description language that today is implemented as 4 UML profiles, on top of each other. This profiles are then realized with Eclipse EMF (ecore), and Java. Quite a heavy framework for prototyping, =) I'm evaluating the possibility of replacing this stack with FAME. Thanks in advance! -- View this message in context: http://n2.nabble.com/FAME-support-for-multiple-inheritance-tp4161084p416108… Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

3 3