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

Re: [Pharo-project] Draft Pharo 1.2.1 Release Announcement
by Sven Van Caekenberghe April 5, 2011

April 5, 2011

On 05 Apr 2011, at 09:19, laurent laffont wrote: > On Tue, Apr 5, 2011 at 9:05 AM, Sven Van Caekenberghe <sven(a)beta9.be> wrote: > > On 05 Apr 2011, at 07:09, laurent laffont wrote: > > > Remarks: > > - group changed by categories: Morphic/Polymorph, Tools, Under the hood, ... > > - Mocketry is not included in Pharo 1.2 > > Thanks for the feedback, I applied your suggestions. > > For more publicity we can have a word on the new success stories: Inceptive ERP, Issys tracking and DrGeo in http://www.pharo-project.org/about/success-stories > > Laurent. > > > More comments ? > > Sven OK, I added a paragraph 'Success Stories', have a look: http://homepage.mac.com/svc/Pharo/pharo-1.2.1-release.html Are the download links correct ?? Sven

1 0

[Pharo-project] IPAddressCountry & Iso3166CountryLongitudeLatitude
by Sven Van Caekenberghe April 4, 2011

April 4, 2011

I want to share a couple of classes. Did you ever want to add this trick to your website where you do something based on the country your visitors are from ? Now you can do so in Smalltalk ! T3EasyUtils getIpAddressCountry atAddress: '81.83.7.35' #BE T3EasyUtils getIso3166CountryLongitudeLatitude at: #BE 4.0(a)50.8333 #('www.beta9.be' 'www.inria.fr' 'nic.st' 'www.world.st' 'www.pharo-project.org') collect: [ :each | T3EasyUtils getIpAddressCountry atAddress: (NetNameResolver addressForName: each) ] #(#BE #FR #SE #US #CH) (T3EasyUtils is holding unique, loaded instances of IPAddressCountry & Iso3166CountryLongitudeLatitude). As this is part of one of our commercial applications, you can even access it over the internet as a JSON REST call: $ curl http://easy.t3-platform.net/rest/geo-ip?address=81.83.7.35 {"latitude":50.8333,"address":"81.83.7.35","country":"BE","longitude":4.0} (Please note that this is not a public service as such, you can try it, but you are not supposed to call this from your own applications.) BTW, T3 Easy is our new entry level track & trace product, an HTML5 web app (using client side HTML+Javascript+CSS) on top of a REST server written in Pharo Smalltalk. It consists of 4 stateless Pharo Cog VM's running behind an Apache load balancer. The server uses Zinc HTTP Components, both for its server part, as well as for clients talking to other systems behind the scenes. Even the Open Street Map tile serving, which is pushing out many megabytes, is currently going through this server. Here is a live demo URL in case your are curious: http://easy.t3-platform.net?t=livedemo (You need a fast web browser like Google Chrome, Safari, Opera 10 or Firefox 4.) The code for IPAddressCountry & Iso3166CountryLongitudeLatitude is on SqueakSource: http://www.squeaksource.com/ADayAtTheBeach.html These are the class comments: --- IPAddressCountry maps IP addresses to their country code. I hold a collection of IPAddressRangeCountry objects that I do a binary search on. I am constructed based on a CSV data file using my class' #readFrom: which expects lines like 3651886848,3651887103,BE or start,stop,code You can download a version of this file from http://homepage.mac.com/svc/data/GeoIPCountry.csv.gz ((IPAddressCountry readFrom: 'GeoIPCountry.csv') atAddress: '81.83.7.35') = #BE Addresses can also be specified as #(81 83 7 35) for #atAddress: or as 1364395811 for #at: You should cache one instance of me. Based on data from http://www.maxmind.com/app/geolitecountry --- Iso3166CountryLongitudeLatitude maps country codes to their average or central longtiude/latitude. My keys are two letter Iso3166 Symbols representing a country, such as #BE My values are Points holding coordinates as longtiude@latitude I am constructed based on a CSV data file using my class' #readFrom: which expects lines like BE,50.8333,4.0000 or code,latitude,longitude You can download a version of this file from http://homepage.mac.com/svc/data/CountryLatLong.csv.gz ((Iso3166CountryLongitudeLatitude readFrom: 'CountryLatLong.csv') at: #BE) = 4.0(a)50.8333 You should cache one instance of me. --- Sven

5 7

[Pharo-project] Announcement real problems - please read and comment.
by StÃ©phane Ducasse April 4, 2011

April 4, 2011

Hi guys We found a problem with ben and RPackage. Here is the scenario and may be you can help us finding the right solutions. Nautilus renames a class the class got renamed -> an announcement is raised - nautilus gets notified, it asks rpackageOrganizer (which also listens to the same event) -> rpackageOrg breaks because the class if was referting to does not exist anymore. Problem 1 - the second announcement was never sent, because the first one broke the second was blocked. >> we should make sure that if an announcement leads to an error, other annoucements on the same emit should pass Problem 2 If rappckageOrganizer would be reached before Nautilus we would not get any problem (beside the problem 1). - we spent one hour playing with alternatives. We thought that if RPackage would points to classes a rename would not break the system. While this is true for class, the same problem will occur for method rename since compiled method are not shared and a new compiled method is compiled. - If RPackage was as deep inside the system as superclass/subclass, the invariant would be directly maintained by the system and we would not have to rely on announcement. But now we should find a solution. >> may be we should be able to say that an announcement has a priority. Typically we would like to say that when RPackageOrganizer register interests it should be served first. we could have a systemLevel priority that could be specified on registration. what do you think? Stef

7 24

[Pharo-project] [COTDC] 41 - ExclusiveWeakMessageSend
by laurent laffont April 4, 2011

April 4, 2011

Today: ExclusiveWeakMessageSend Comment Of The Day Contest - One Day One Comment Rules: #1: Each day a not commented class is elected. Each day the best comment will be integrated with name of the author(s). #2: If you cannot comment it, deprecate it. Results: http://code.google.com/p/pharo/wiki/CommentOfTheDayContest Laurent

1 1

[Pharo-project] On Pharo 1.2
by Torsten Bergmann April 4, 2011

April 4, 2011

Sorry for changing topic ... but with all these [ANN] replies we cant see if there are really new announcements. When 1.2 Hudson was red nobody seemed to care and now that it is green/blue again discussion starts. Good or bad? At least I wished that happened earlier and a red Hudson shows that we have to do our homework. Pharo 1.2 really took time and I think we can expect further releases to take even longer since the more we clean up the harder it would be to keep compatibility with all the existing (old) packages. We all know a core alone would be useless - we have to decide which packages are maintainable by the community and worth to move forward with Pharo-dev. We have two goals that sometimes collide: -> build a small kernel/core -> build a better Smalltalk (feature rich and with good tools) If we work on cleaned up loadable/unloadable and well tested packages it would be less problematic since we can work in both images and load/unload as needed. That should be the main goal for 1.3. and beyond. Hudson is our watchdog. Nonetheless ... with Pharo 1.1 the open source ST world already changed and we have first companies/people using Pharo. Pharo is an accepted Smalltalk version. Since Pharo 1.2 is a much better and cleaner version I hope this will continue/raise and Pharo will find more new friends to help. :) Bye T. -- NEU: FreePhone - kostenlos mobil telefonieren und surfen! Jetzt informieren: http://www.gmx.net/de/go/freephone

1 0

Re: [Pharo-project] [ANN 1.2] 1.2.1 Full Release Image
by Marcus Denker April 4, 2011

April 4, 2011

On Apr 4, 2011, at 6:36 PM, Mariano Martinez Peck wrote: > > > On Mon, Apr 4, 2011 at 5:33 PM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > > On Apr 4, 2011, at 4:58 PM, Mariano Martinez Peck wrote: > > > I started to try to release 1.2. Not only I did the list of changes but also: > > > > 1) I asked if people still wanted the background: http://forum.world.st/Do-you-want-to-add-the-background-picture-to-the-imag… > > results: negative. So, no backgrund. White as it is now. > > > > 2) I asked what people wanted that used to be in 1.1: http://forum.world.st/Do-we-want-this-for-Pharo-1-2-and-futures-td3350721.h… > > guess how many replies? TWO. > > results: I would only nclude the installScript. I guess that Hudson could do that very easy, doesn't it ? marcus ? > > > > https://pharo-ic.lille.inria.fr/hudson/job/Pharo%201.2/lastSuccessfulBuild/… > > > Excellent! and that be included in the zip that hudson creates for the dev images? It used to be in there. But then people where thinking that it is a startup file that is executed when the image starts up... maybe by renaming it this would get clearer? Marcus -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.

2 1

[Pharo-project] fancier tab morph
by Tudor Girba April 4, 2011

April 4, 2011

Hi, In the context of Glamour, I worked on a Morphic tab widget a bit more and here is the status (see the screenshots for some examples): - it computes the pages lazily only when you want to first show the page - it provides the possibility of closing the tab - it offers the possibility of rendering a toolbar for each page displayed to the right of the pages. If you want to play with them, you can: - download the latest moose image: http://hudson.moosetechnology.org/job/moose-latest-dev/lastSuccessfulBuild/… - execute the following to get an example of closable tabs: GLMBasicExamples new accumulator openOn: 42 - execute the following to get an example of tabs with different toolbars attached (to the right): GLMBasicExamples new tabsWithDifferentActions openOn: 42 There are still a couple of issues left: - When the title of a tab is too large, it pushes the bounds of the pane beyond what is visible. - Also, when we have multiple tabs, they again go overboard. - Similar things happen with the toolbar. However, the good news is that Gary is looking into this :) Cheers, Doru -- www.tudorgirba.com "Relationships are of two kinds: those we choose and those that happen. They both matter."

3 5

Re: [Pharo-project] [ANN 1.2] 1.2.1 Full Release Image
by Marcus Denker April 4, 2011

April 4, 2011

On Apr 4, 2011, at 4:58 PM, Mariano Martinez Peck wrote: > I started to try to release 1.2. Not only I did the list of changes but also: > > 1) I asked if people still wanted the background: http://forum.world.st/Do-you-want-to-add-the-background-picture-to-the-imag… > results: negative. So, no backgrund. White as it is now. > > 2) I asked what people wanted that used to be in 1.1: http://forum.world.st/Do-we-want-this-for-Pharo-1-2-and-futures-td3350721.h… > guess how many replies? TWO. > results: I would only nclude the installScript. I guess that Hudson could do that very easy, doesn't it ? marcus ? > https://pharo-ic.lille.inria.fr/hudson/job/Pharo%201.2/lastSuccessfulBuild/… Marcus -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.

2 1

Re: [Pharo-project] [SPAM] Re: [ANN 1.2] 1.2.1 Full Release Image
by Marcus Denker April 4, 2011

April 4, 2011

On Apr 4, 2011, at 4:09 PM, Sven Van Caekenberghe wrote: > > On 04 Apr 2011, at 15:40, StÃ©phane Ducasse wrote: > >> Please help because we have not the steam for the splash organization. > > So we poor users have to do all the work ?? > > Seriously, what is the easiest/fastest way to get to a rough list of the changes between 1.1.1 and 1.2.1 ? > > From that list we could try to make a readable list for release notes and then we could try to make a list of bullet points / highlights. > Mariano started a list: http://code.google.com/p/pharo/wiki/ActionsInPharoOneDotTwo All things done should have a tracker entry: http://code.google.com/p/pharo/issues/list?can=1&q=milestone=1.2 http://code.google.com/p/pharo/issues/list?can=1&q=Milestone%3D1.2-DevImage -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.

5 6

April 4, 2011

Hi guys Since this is interesting for all of us I prefer to publicly discuss it. From: Fernando Olivero <fernando.olivero(a)usi.ch> > Date: March 31, 2011 8:40:26 AM GMT+02:00 > To: alain.plantec(a)free.fr, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > Subject: Regarding SimpleMorphic > Reply-To: fernando.olivero(a)usi.ch > > Hi, i've been working on the necessary steps to ease the adoption of > SimpleMorphic. Excellent! > I found it really nice, and would like to code the new > version of Gaucho on top of it. I post here the conclusions roadmap, > maybe you are starting the same effort, so this is just to collect > feedback or let you know! We want to join forces. I would love to have key widgets: list, tree ported and totally revised so that we do not have 20 of them but just two or four. > I always start from the Morphic in CUIS 3.1, and then go one by one > trying to port the relevant classes, honoring the intent behind Pharo > UI by, for example, adding in the UIManager uses whenever posible, > > Saludos, > Fernando Question: should we look in Cuis3.1 or Pharo1.3 SMx > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > CUIS > System > 1) Review Editor and friends: Separate hierarchy or try to merge with > the existing one in Pharo? Alain's great work on EditorState is > missing in CUIS. Yes I love the selection and highlights of Alain. > 2) import TextModel and hierarchy or maintain a separate hierarchy? > The changes in text model and subclasses seem to be good to have! > 3) (CUIS )DifferenceFinder vs (Pharo)TextDiffBuilder ? ( Seems that > DifferenceFinder has extra funcionality) No idea there. > 4) Transcript and TranscriptMorph. DONE ISSUE 3736 soon in Pharo1.3 > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > CUIS3.1 into Pharo1.3 SimpleMorphic So what is the difference with the one we have. > 1) Make current implementation of SMx load in Pharo1.3. > DONE ISSUE 3878: Implementing enableTextChange > DONE ISSUE 3584: Cleanup removing DisplayText =>to remove > UIManager>>composeForm: > DONE ISSUE 3915: DiffMorph requests thru UIManager: so we don't have > to change senders such as ChangeList and VersionBrowser. Good. > 2) World, ActiveHand, ActiveEvent and ActiveWorld cleanup in Morphic. > To merge Morphic and SimpleMorphic handling of the global state, > centralizing the accessors in UIManager, by removing Globals. ALMOSTE > DONE ISSUE 3903 > 3) general up to date from CUIS 3.1: ALMOST DONE > 4) tools?: Revise the SMxBrowser and family, replace maybe with the > new TextModel>>openAsMorph: , etcâ¦ For the Browser we plan - to simply **Throw** away all the stringHolder and codeHolder classes and subclasses. Benjamin can show you Nautilus beta. - We will remove also ToolBuilder (as soon as we have a new version of the test runner) - We need a model for the debugger (we should check glamour people for that). So we will temporary keep stringHolde but compact it only for the debugger. > 5) Remove SMxTheWorldMenu and replace with MenuPragmaBuilder as in Pharo. Yes > 6) ? > > ++++++++++++++++++++++ > LOAD ORDER (uncomplete) > 3878 > 3584 > 3915 > 3736 > (PENDING) 3903 > ...

7 19