March 2012 - Pharo-dev - lists.pharo.org

Re: [Pharo-project] Unicode depending on Zinc

by StÃ©phane Ducasse

I would prefer to have a method that contains this data. Because any Smalltalk at:â¦. will lead to dead code when ring is not loaded. Stef On Mar 16, 2012, at 12:18 AM, Guillermo Polito wrote: > That's why I generated a method from the file, and then put the link in the comment... > > On Thu, Mar 15, 2012 at 8:17 PM, Guillermo Polito <guillermopolito(a)gmail.com> wrote: > But what if I want to have an image without Zinc? Ok, I will not use that method, but I'll get an Undeclared entry and therefore a dirty image... :/ > > > On Thu, Mar 15, 2012 at 12:37 PM, Sven Van Caekenberghe <sven(a)beta9.be> wrote: > Hi Guillermo, > > On 15 Mar 2012, at 16:25, Guillermo Polito wrote: > > > Of course with a nice comment saying something like: > > > > "Initialize the mappings from unicode lower to upper case and vice versa. sourceCodes and destinyCodes contains the unicode values for the mappings, and they match one by one. > > The source of this mapping can be found at: http://www.unicode.org/Public/UNIDATA/CaseFolding.txt' " > > Have a look at what I did with http://code.google.com/p/pharo/issues/detail?id=4187 > > The spec tables are constants, like in your latest example, but I added a method that can generate these from an internet accessible resource (using Zn as HTTP client). That way, there is some kind of dependency, but not really at run time. I think it is cool to literally use these public spec documents. > > Sven > > >

March 16, 2012

Re: [Pharo-project] Does anyone have tried/succeeded to use PetitParser for parsing dates?

by StÃ©phane Ducasse

> how did you handle things like > 1/1/11 (or didn't you?) > or worse > 1/1/1 -> 01.01.2001? This is not related to parsing but to the interpretation of the data :) There is no magic no? BTW this is the birthday of one of my sons :) > > how have you treated dates like > 1/12/2012 -> 12.01.2012 or 1.12.2012? > > or even more "problematic" > 12/12/2012 Again why a computer looking at only one sample would be prescient? > and the like there are so many way to parse the dates.... > > Regards > Friedrich > > -- > Q-Software Solutions GmbH; Sitz: Bruchsal; Registergericht: Mannheim > Registriernummer: HRB232138; Geschaeftsfuehrer: Friedrich Dominicus >

March 16, 2012

Re: [Pharo-project] [Moose-dev] Moose Windows VM problem

by StÃ©phane Ducasse

> Hi, > > It would be great to get a job for running the Moose on Windows. Yes usman was talking about that with igor last week. Now just running. Stef

March 16, 2012

Re: [Pharo-project] Unicode depending on Zinc

by Sven Van Caekenberghe

If you want to avoid hardcoding a class reference to Zn, you could do a simple download like this: 'http://zn.stfx.eu/zn/numbers.txt' asZnUrl retrieveContents It avoids an entry to Undeclared or the ugly test, and it since it is implemented as retrieveContents "Download and return the resource that I refer to. This will typically return a String or a ByteArray (see ZnClient>>#get:). If something goes wrong, an exception will be signalled." " 'http://zn.stfx.eu/zn/numbers.txt' asZnUrl retrieveContents. " ^ ZnClient new beOneShot; enforceHttpSuccess: true; get: self it is easy to use for simple scripting. Sven On 16 Mar 2012, at 00:24, Igor Stasenko wrote: > you can always make dependency less hardcoded: > > Smalltalk at: #ZnClient ifAbsent: [ self error: 'sorry dude' ] > ifPresent: [:client | > ... download stuff , whatever ] > > > On 16 March 2012 00:18, Guillermo Polito <guillermopolito(a)gmail.com> wrote: >> That's why I generated a method from the file, and then put the link in the >> comment... >> >> >> On Thu, Mar 15, 2012 at 8:17 PM, Guillermo Polito >> <guillermopolito(a)gmail.com> wrote: >>> >>> But what if I want to have an image without Zinc? Ok, I will not use that >>> method, but I'll get an Undeclared entry and therefore a dirty image... :/ >>> >>> >>> On Thu, Mar 15, 2012 at 12:37 PM, Sven Van Caekenberghe <sven(a)beta9.be> >>> wrote: >>>> >>>> Hi Guillermo, >>>> >>>> On 15 Mar 2012, at 16:25, Guillermo Polito wrote: >>>> >>>>> Of course with a nice comment saying something like: >>>>> >>>>> "Initialize the mappings from unicode lower to upper case and vice >>>>> versa. sourceCodes and destinyCodes contains the unicode values for the >>>>> mappings, and they match one by one. >>>>> The source of this mapping can be found at: >>>>> http://www.unicode.org/Public/UNIDATA/CaseFolding.txt' " >>>> >>>> Have a look at what I did with >>>> http://code.google.com/p/pharo/issues/detail?id=4187 >>>> >>>> The spec tables are constants, like in your latest example, but I added a >>>> method that can generate these from an internet accessible resource (using >>>> Zn as HTTP client). That way, there is some kind of dependency, but not >>>> really at run time. I think it is cool to literally use these public spec >>>> documents. >>>> >>>> Sven >>>> >>> >> > > >

March 16, 2012

Re: [Pharo-project] Does anyone have tried/succeeded to use PetitParser for parsing dates?

by Friedrich Dominicus

Tudor Girba <tudor(a)tudorgirba.com> writes: > Hi, > > I used it for various forms of dates. What is the question? E.g handling of optional leading zeroes how did you handle things like 1/1/11 (or didn't you?) or worse 1/1/1 -> 01.01.2001? how have you treated dates like 1/12/2012 -> 12.01.2012 or 1.12.2012? or even more "problematic" 12/12/2012 and the like there are so many way to parse the dates.... Regards Friedrich -- Q-Software Solutions GmbH; Sitz: Bruchsal; Registergericht: Mannheim Registriernummer: HRB232138; Geschaeftsfuehrer: Friedrich Dominicus

March 16, 2012

Re: [Pharo-project] Grouping of Tools?

by Friedrich Dominicus

StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> writes: > On Mar 13, 2012, at 11:25 AM, Friedrich Dominicus wrote: > >> I wonder if one can get Pharo to group the tools e.g the Workspace on >> one tab, Browsers on another etc. >> >> Any suggestions? > > did you check the TWM project? No, I did not know anything about it. Thanks for your information. Regards Friedrich -- Q-Software Solutions GmbH; Sitz: Bruchsal; Registergericht: Mannheim Registriernummer: HRB232138; Geschaeftsfuehrer: Friedrich Dominicus

March 16, 2012

[Pharo-project] [update 1.4] #14394

by Marcus Denker

14394 ----- Issue 5466: Lots of failing test: test seems to create undeletable file on linux (trying a workaroung for now...) http://code.google.com/p/pharo/issues/detail?id=5466 Issue 5486: Fix in Abstract Tool http://code.google.com/p/pharo/issues/detail?id=5486 -- Marcus Denker -- http://marcusdenker.de

March 16, 2012

Re: [Pharo-project] TestCase>>#assert:equals: optimization

by Pavel Krivanek

On Fri, Mar 16, 2012 at 8:28 AM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > > On Mar 15, 2012, at 11:20 PM, Nicolas Cellier wrote: > >> Being able to automate is very good, however... >> <rant> >> > > I am 100% with you... my "perfect system" looks *completely* different > to what we have... Can you, please, describe the basic points? :-) -- Pavel > -- > Marcus Denker -- http://marcusdenker.de > >

March 16, 2012

Re: [Pharo-project] Unicode depending on Zinc

by Pavel Krivanek

On Fri, Mar 16, 2012 at 8:20 AM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > > On Mar 16, 2012, at 12:24 AM, Igor Stasenko wrote: > >> you can always make dependency less hardcoded: >> >> Smalltalk at: #ZnClient ifAbsent: [ self error: 'sorry dude' ] >> ifPresent: [:client | >> Â ... download stuff , whatever ] > > > Why don't we make that the default? That is, compile all globals > to something like "self environment lookUp: #ClassName" ? > In general it will lead to unimplemented calls so it is always better to solve it with polymorphism. However we need some standard way how to work with that. - the expression "self environment" is not good because the message #environment is defined only in classes and for nil, not in Object. - we should rise a standard exception with default action taken from UIManager. - there should be an optional variant messages that take a package name as an argument so the error handling will be able to tell user what package should be installed. -- Pavel > Â Â Â Â Marcus > > > -- > Marcus Denker -- http://marcusdenker.de > >

March 16, 2012

Re: [Pharo-project] TestCase>>#assert:equals: optimization

by Marcus Denker

On Mar 15, 2012, at 11:20 PM, Nicolas Cellier wrote: > Being able to automate is very good, however... > <rant> > I am 100% with you... my "perfect system" looks *completely* different to what we have... There are two options: 1) stop fixing bugs and integrating until we have the perfect system for handling it. 2) Accept that it might be better to integrate what people are doing anyway even without yet having the perfect source-code-management/bug tracker/integration. Doing 2 while *of course* alligning things to improve the integration system eventually... but I do not see how we can build the system that I dream about if we don't move already now. And even if it is just for generating small successes that allow us to build on top. It's hard to aquire resources out of nothing, but if you have something to show (as bad as it is) it magically *is* possible! Marcus -- Marcus Denker -- http://marcusdenker.de

March 16, 2012