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

Re: [Pharo-project] Issue 3740 in pharo: Provide Settings to directly open the full Debugger View

by pharo＠googlecode.com

Comment #3 on issue 3740 by camillob...(a)gmail.com: Provide Settings to directly open the full Debugger View http://code.google.com/p/pharo/issues/detail?id=3740 Yep you are right, I forgot to add the changes to the Debugger class. The attached fileout contains the changed method. Attachments: Debugger class-openOncontextlabelcontentsfullView.st 646 bytes

Feb. 23, 2011

Re: [Pharo-project] Accessing OS windows within smalltalk VM

by Mariano Martinez Peck

but what about using OSProcess? On Tue, Feb 22, 2011 at 11:30 PM, Igor Stasenko <siguctua(a)gmail.com> wrote: > On 22 February 2011 19:59, Nathan Tuttle <nathan.tuttle(a)gmail.com> wrote: > > Is it possible to access operating system windows (screen captures) in > > pharo. Since this would be OS specific does anyone know of a primitive > > plugin that would allow this? > > > > there is no such primitive. > > > -- > Best regards, > Igor Stasenko AKA sig. > >

Feb. 23, 2011

[Pharo-project] XML browser

by laurent laffont

As I haven't found something to browse XML the way I want, I've written a little tool: http://magaloma.blogspot.com/2011/02/xml-browser-with-pharo.html If there's already a good tool please tell me. Laurent Laffont - @lolgzs <http://twitter.com/#!/lolgzs> Pharo Smalltalk Screencasts: http://www.pharocasts.com/ Blog: http://magaloma.blogspot.com/

Feb. 23, 2011

Re: [Pharo-project] Fwd: [ANN] pharo focused sprint - bern, feb 26

by Tudor Girba

Great. Doru On 22 Feb 2011, at 19:47, Camillo Bruni wrote: > I'll be there. > > From what I have done this week so far, I would like to work on extending the IDE and reduce the amount of clicks / keystrokes to browse things... > > m(o_o)m > camillo > > On 2011-02-21, at 19:42, Tudor Girba wrote: > >> Hi, >> >> I would like to remind you that this Saturday, February 26, we organize a Sprint at SCG, University of Berne, Switzerland (room 107): >> http://scg.unibe.ch/contact/maps >> >> Further details can be found below. Please let us know by mail if you plan to attend and what you would wish to work on (see below). >> >> Cheers, >> Doru >> >> >> >> Begin forwarded message: >> >>> From: Tudor Girba <tudor.girba(a)gmail.com> >>> Date: 1 February 2011 16:38:30 CET >>> To: "Pharo-project(a)lists.gforge.inria.fr Development" <pharo-project(a)lists.gforge.inria.fr> >>> Cc: Adrian Lienhard <alienhard(a)netstyle.ch> >>> Subject: [ANN] pharo focused sprint - bern, feb 26 >>> >>> Hi, >>> >>> We would like to organize a Pharo Sprint on Saturday, February 26 at SCG, University of Berne, Switzerland (room 107): >>> http://scg.unibe.ch/contact/maps >>> >>> Unlike other sprints, this would not be about fixing things from the tracker, but about building something new and reviewing the situation around one central topic. For this one, the main topic will be Morphic and the IDE. Why? Because it is time to rethink our day-to-day tools and bring them closer to this century (ok, maybe decade :)). It is Ok if you are not a specialist. It is more important to want to participate and learn. The rest will come in time. >>> >>> ***** Please let us know by mail if you plan to attend and what you would wish to work on (see below). >>> >>> Proposed tasks (other tasks are possible in the same area): >>> - Enhancements of Morphic widgets: >>> --- TabGroupMorph with lazy pages, closable tabs, overflow handling and toolbars (a start exists in Glamour) >>> --- High level support for collapsable panes (based on the input of Gary) >>> --- SearchMorph with multiple groups of items (like in Spotlight) >>> --- MorphTreeMorph integrated with GeneralScrollPane to allow for space filling >>> --- BreadcrumbMorph >>> --- Hyperlinks in TextMorphs >>> - Test and enhance the current IDE >>> - Evaluate TextMorphs: NewTextMorph, PluggableTextMorph, SMxPluggableTextMorph etc >>> - Prototype a new ToolSet solution >>> - Enhancements of Glamour and of the Glamorous Toolkit >>> --- Debugger >>> --- Coder >>> --- Multipage Workspace >>> --- CodeBubbles >>> - Develop WeakAnnouncement (start from the implementation of Lukas) >>> - Evaluate the various keybinding implementations >>> - Evaluate ToolBuilder >>> - Evaluate the path to adopt SimpleMorphic >>> >>> A second working topic is getting Monticello faster, but this will only be tackled if there are enough people for the first one. >>> >>> Cheers, >>> Adrian and Doru >>> >>> >>> >>> -- >>> www.tudorgirba.com >>> >>> "Be rather willing to give than demanding to get." >>> >>> >>> >> >> -- >> www.tudorgirba.com >> >> "It's not how it is, it is how we see it." >> >> > > -- www.tudorgirba.com "Value is always contextual."

Feb. 23, 2011

[Pharo-project] Issue 3747 in pharo: ' 'Morph' asMorph openInworld' creates a Morph that can't be deleted using meta-click menu

by pharo＠googlecode.com

Status: Accepted Owner: smagooe...(a)gmail.com New issue 3747 by smagooe...(a)gmail.com: ' 'Morph' asMorph openInworld' creates a Morph that can't be deleted using meta-click menu http://code.google.com/p/pharo/issues/detail?id=3747 This is a small issue but it makes for a messy workspace. I couldn't find any way to delete the morph that was created this way. Pharo image: Pharo-1.1.1-dev10.09.1.image Pharo core version: Pharo-1.1.1-- Latest update: #11414 Virtual machine used: Squeak-4.4.7.2357-linux_i386 Class browser used if applicable: OBSystemBrowserAdaptor Operating System: Ubuntu 10.04 Lucid Lynx Steps to reproduce: 1. Type 'Morph' asMorph openInWorld in a workspace. 2. Try to use meta-click to select the StringMorph that is created by above step, and delete it. On my machine this is shift-ctrl-left mouse button.

Feb. 23, 2011

Re: [Pharo-project] Pattern for reifying XML doc?

by StÃ©phane Ducasse

On Feb 22, 2011, at 3:41 PM, Guillermo Polito wrote: > From the Java world, I know Digester http://commons.apache.org/digester/xmlrules.html, letting you map an xml to objects with a set of rules... > I don't know if that's what you were expecting as an answer :). > > BTW, great example! :P Magic is cool :) Henrik has crazy games too :) Stef

Feb. 22, 2011

Re: [Pharo-project] Pattern for reifying XML doc?

by StÃ©phane Ducasse

Thanks johan What I like with having just xml with my format is that I can hack the files without breaking the objects it represents. I can evolve my objects without losing my filed backup in xml. On Feb 22, 2011, at 5:07 PM, Johan Brichau wrote: > SIXX works very good to write/read objects, but it means you have to give up on your own xml schema. > > And it's almost to easy to use: > > sws := SixxWriteStream newFileNamed: aFileName. > sws nextPut: (self objectsToExport). > sws close. > > srs := SixxReadStream readOnlyFileNamed: (aFileName). > objects := OrderedCollection new. > [srs atEnd] whileFalse: [objects add: srs next]. > srs close. > > On 21 Feb 2011, at 10:11, stephane ducasse wrote: > >> Hi guys >> >> I often encounter the following situation: I have an xmldoc and I want to load it and get the corresponding objects. >> Here is an example >> >> <group> >> <groupName>Test2Cards</groupName> >> <cards><card> >> <cardName>All Is Dust</cardName> >> <manaCost>7</manaCost> >> <convertedManaCost>7</convertedManaCost> >> <type>Tribal Sorcery â Eldrazi</type> >> <cardText>Each player sacrifices all colored permanents he or she controls.</cardText> >> <rarity>mythic</rarity> >> <cardNumber>1</cardNumber> >> <artist>Jason Felix</artist> >> <kind>STD</kind> >> <expansionId>ELDRAZI</expansionId> >> </card> >> <card> >> <cardName>Kird Ape</cardName> >> <manaCost>G</manaCost> >> <convertedManaCost>1</convertedManaCost> >> <type>Tribal Sorcery â Eldrazi</type> >> <cardText>get +1/+1 when you control</cardText> >> <rarity>common</rarity> >> <cardNumber>33</cardNumber> >> <artist>Zork</artist> >> <kind>STD</kind> >> <expansionId>WORLDWAKE</expansionId> >> </card></cards> >> </group> >> >> I have card and group. But I could have several kind of nodes. So I was wondering if there are patterns to read in >> and map that to objects. In the future I would like to have magritte like type of first class instance variables that >> perform i/o in xml or other formats because this is always the same. >> >> stef >> >> >> > >

Feb. 22, 2011

Re: [Pharo-project] Pattern for reifying XML doc?

by StÃ©phane Ducasse

Yes I was thinking about using magritte too because I did that too in the past. I will look at what you did. I think that in the future we should really get - first class instance variables - internal to class or external mapper from to different formats like XML On Feb 22, 2011, at 5:21 PM, Norbert Hartl wrote: > Stef, > > I'm using Magritte for all these cases. We developed a XML binding for magritte (http://source.lukas-renggli.ch/magritte2addons/) so it is easy as > > User class>>descriptionEmailAddress > ^ MAStringDescription new > accessor: #emailAddress; > priority: 100; > label: 'Email Address'; > beRequired; > beXmlAttribute; > xmlAttributeName: 'email'; > yourself > > but you need magritte and it doesn't sound you like to have a lot of dependencies. For read support I hacked something with Pastell for a quick and flexible read > > Example class>>xmodel > ^ Dictionary new; > at: #name put: [:node| (node / #Name ) contentString ]; > at: #childObject put: [:node| ExampleReferencedClass node: (node / #ChildNode) first ]; > yourself > > Example>>at: aSymbol > ^ (self class xmodel at: aSymbol) value: self node > > This is more flexible than easy. For your case you can map just the values. The biggest problem will probably be the type conversion. You could have setters like cardNumberString: that convert the type. But if I imagine you right now I imagine you faceplaming :) > > Norbert > > On 21.02.2011, at 10:11, stephane ducasse wrote: > >> Hi guys >> >> I often encounter the following situation: I have an xmldoc and I want to load it and get the corresponding objects. >> Here is an example >> >> <group> >> <groupName>Test2Cards</groupName> >> <cards><card> >> <cardName>All Is Dust</cardName> >> <manaCost>7</manaCost> >> <convertedManaCost>7</convertedManaCost> >> <type>Tribal Sorcery â Eldrazi</type> >> <cardText>Each player sacrifices all colored permanents he or she controls.</cardText> >> <rarity>mythic</rarity> >> <cardNumber>1</cardNumber> >> <artist>Jason Felix</artist> >> <kind>STD</kind> >> <expansionId>ELDRAZI</expansionId> >> </card> >> <card> >> <cardName>Kird Ape</cardName> >> <manaCost>G</manaCost> >> <convertedManaCost>1</convertedManaCost> >> <type>Tribal Sorcery â Eldrazi</type> >> <cardText>get +1/+1 when you control</cardText> >> <rarity>common</rarity> >> <cardNumber>33</cardNumber> >> <artist>Zork</artist> >> <kind>STD</kind> >> <expansionId>WORLDWAKE</expansionId> >> </card></cards> >> </group> >> >> I have card and group. But I could have several kind of nodes. So I was wondering if there are patterns to read in >> and map that to objects. In the future I would like to have magritte like type of first class instance variables that >> perform i/o in xml or other formats because this is always the same. >> >> stef >> >> >> > >

Feb. 22, 2011

Re: [Pharo-project] FileSystem. Possible FSReference rename

by StÃ©phane Ducasse

On Feb 21, 2011, at 2:54 PM, Denis Kudriashov wrote: > Hello, > > I hear many people does not like name FSReference in FileSystem package. Me too. Its not clear to me. But it is main object for end user. > Maybe we can rename it to FSLocation? FSReference = FSSystem + FSPath So may be it could be renamed FSLocation. > > I frequently name variables that point to some file/dir like imagesLocation, settingsLocation and others. Therefore it is natural to me have FSLocation instead FSReference. > What do you think? >

Feb. 22, 2011

Re: [Pharo-project] [squeak-dev] Re: Cog VM -- Thanks and Performance / Optimization Questions

by StÃ©phane Ducasse

On Feb 21, 2011, at 7:04 PM, Eliot Miranda wrote: > > > On Mon, Feb 21, 2011 at 6:05 AM, Juan Vuletich <juan(a)jvuletich.org> wrote: > StÃ©phane Ducasse wrote: > Thanks juan > We will look at it :) > Having a fast and safe transcripter is good. > > Good. At least there one person in our communities is interested on this. > > I am disappointed, as the number of people who actually commented on the relevance of said TranscriptStream problems, or on how my new implementations succeeds or fails to fix them is zero. > > Not that I really believed I would "instantly become a hero"... But I can't understand how nobody else had trouble with the Transcript in the past, to the point of taking a look at the code, trying it, and commenting on it. > > Igor and I agree. We want TranscriptStream to remain a subclass of WriteStream, by all means thread-safe (I think the one in the Teleplace images is thread-safe), and including a log file is fine. But the view must be separate. sorry but TranscriptStream was not thread safe else I would not have written threadSafeTranscript. Stef

Feb. 22, 2011