- Pharo-dev - lists.pharo.org

Re: [Pharo-project] [ENH] Syntax extension for continuations

by Levente Uzonyi

On Mon, 1 Nov 2010, Igor Stasenko wrote: > Hello, > > Out of curiosity, i tried to look how it would be hard to change > parser to support extended continuation syntax. > And it was quite simple, i had to change only single method! > > A new syntax using double semicolon ';;' to indicate a continuation. > > So, code like: > > self foo ;; bar ;; zork > > is equivalent to: > > self foo bar zork > > But when its going to binary, or keyword messages, it allows to write > more clean code, > because it not requires using parenthesis. > > For instance: > > (1/10) asFloat > > with new syntax can be written as: > > 1/10 ;; asFloat > > More complex example: > > ((self foo: bar) + 10 ) baz > > can be written as > > self foo: bar ;; + 10 ;; baz > > > Note, that it is same number of characters to type.. but is much more > clean, and less time expensive to code, > because when you coding, you usually type first message: > > self foo: bar ` > > and then you realising that next message is binary and hence you need > to go back to beginning of line and put open paren there: > > `( self foo: bar > > then again, go to the end of message, and continue typing: > > ( self foo: bar ` ) + 10 > > ` - is cursor position. > > so, it is much more keystrokes & navigation than just typing two semicolons :) > > P.S. don't burn me for my herecy, please :) This idea comes up every few years. This post sums up well the last attempt: http://blog.3plus4.org/2007/08/30/message-chains/ . I don't think it's worth changing the syntax for this. IMHO 1-2 pairs of parentheses usually improve readability. More parenthesis is rarely needed. Also mixing ;; with ; makes the code harder to understand. Try this: self foo ; bar ;; baz ; foo ;; bar ; baz. Note that the term 'continuation' means a totally different thing: http://en.wikipedia.org/wiki/Continuation . Levente > > -- > Best regards, > Igor Stasenko AKA sig. >

Nov. 1, 2010

Re: [Pharo-project] [ENH] Syntax extension for continuations

by Igor Stasenko

Here is more examples (found by searching .sources file with 3 or more '(((' ). ContextPart>>handleSignal: (((self tempAt: 1) handles: exception) and: [self tempAt: 3]) ifFalse: [ ^ self nextHandlerContext handleSignal: exception]. self tempAt: 1 ;; handles: exception ;; and: [self tempAt: 3] ;; ifFalse: [ ^ self nextHandlerContext handleSignal: exception]. Bezier3Segment. before: bezier2SegmentCount: pixelError "Compute the number of quadratic bezier segments needed to approximate this cubic with no more than a specified error" | a | a := (start x negated @ start y negated) + (3 * via1) - (3 * via2) + (end). ^ (((a r / (20.0 * pixelError)) raisedTo: 0.333333) ceiling) max: 1. after: bezier2SegmentCount: pixelError "Compute the number of quadratic bezier segments needed to approximate this cubic with no more than a specified error" | a | a := start x negated @ start y negated ;; + (3 * via1) - (3 * via2) + end. ^ a r / (20.0 * pixelError) raisedTo: 0.333333 ;; ceiling max: 1. ZipArchive before: lastIndexOfPKSignature: aSignature in: data "Answer the last index in data where aSignature (4 bytes long) occurs, or 0 if not found" | a b c d | a := aSignature first. b := aSignature second. c := aSignature third. d := aSignature fourth. (data size - 3) to: 1 by: -1 do: [ :i | (((data at: i) = a) and: [ ((data at: i + 1) = b) and: [ ((data at: i + 2) = c) and: [ ((data at: i + 3) = d) ]]]) ifTrue: [ ^i ] ]. ^0 after: lastIndexOfPKSignature: aSignature in: data "Answer the last index in data where aSignature (4 bytes long) occurs, or 0 if not found" | a b c d | a := aSignature first. b := aSignature second. c := aSignature third. d := aSignature fourth. data size - 3 to: 1 by: -1 do: [ :i | data at: i ;; = a and: [ data at: i + 1 ;; = b and: [ data at: i + 2 ;; = c and: [ data at: i + 3 ;; = d ]]] ifTrue: [ ^i ] ]. ^0 Character before: asUppercase "If the receiver is lowercase, answer its matching uppercase Character." "A tentative implementation. Eventually this should consult the Unicode table." | v | v := self charCode. (((8r141 <= v and: [v <= 8r172]) or: [16rE0 <= v and: [v <= 16rF6]]) or: [16rF8 <= v and: [v <= 16rFE]]) ifTrue: [^ Character value: value - 8r40] ifFalse: [^ self] after: asUppercase "If the receiver is lowercase, answer its matching uppercase Character." "A tentative implementation. Eventually this should consult the Unicode table." | v | v := self charCode. 8r141 <= v and: [v <= 8r172] ;; or: [16rE0 <= v and: [v <= 16rF6]] ;; or: [16rF8 <= v and: [v <= 16rFE]] ;; ifTrue: [^ Character value: value - 8r40] ifFalse: [^ self] -- Best regards, Igor Stasenko AKA sig.

Nov. 1, 2010

[Pharo-project] [ENH] Syntax extension for continuations

by Igor Stasenko

Hello, Out of curiosity, i tried to look how it would be hard to change parser to support extended continuation syntax. And it was quite simple, i had to change only single method! A new syntax using double semicolon ';;' to indicate a continuation. So, code like: self foo ;; bar ;; zork is equivalent to: self foo bar zork But when its going to binary, or keyword messages, it allows to write more clean code, because it not requires using parenthesis. For instance: (1/10) asFloat with new syntax can be written as: 1/10 ;; asFloat More complex example: ((self foo: bar) + 10 ) baz can be written as self foo: bar ;; + 10 ;; baz Note, that it is same number of characters to type.. but is much more clean, and less time expensive to code, because when you coding, you usually type first message: self foo: bar ` and then you realising that next message is binary and hence you need to go back to beginning of line and put open paren there: `( self foo: bar then again, go to the end of message, and continue typing: ( self foo: bar ` ) + 10 ` - is cursor position. so, it is much more keystrokes & navigation than just typing two semicolons :) P.S. don't burn me for my herecy, please :) -- Best regards, Igor Stasenko AKA sig.

Nov. 1, 2010

Re: [Pharo-project] Failing tests in 1.2

by Nicolas Cellier

Please, please, fix http://code.google.com/p/pharo/issues/detail?id=3200 before release. Cheers Nicolas 2010/10/30 Marcus Denker <marcus.denker(a)inria.fr>: > ... bug Issues for all of them > > Â Â Â Â http://code.google.com/p/pharo/issues/list?can=2&q=type=FailingTest > > > -- > Marcus Denker Â -- http://www.marcusdenker.de > INRIA Lille -- Nord Europe. Team RMoD. > > >

Oct. 31, 2010

Re: [Pharo-project] InputEventFetcher starving all other processes

by StÃ©phane Ducasse

what is strange is that pharo does not change the vm code. One which OS are you? > Hi: > > On 31 Oct 2010, at 19:15, Stefan Marr wrote: > >> But the symptoms are that InputEventFetcher>>primGetNextEvent: delivers 500-1000x more events than EventSensor>>primGetNextEvent: in MVC or Squeak 4.1 images deliver. > Turns out that the array which is supposed to be filled in by '>>primGetNextEvent: array' does not get filled in correctly. > For some reason the old values stay in the array when there isn't any new event. may be a semaphore or something like that prevents some actions to happen. > > However, the very same code works for the non-Pharo images. > > Was there any change to the code in or related to this primitive (<primitive: 94>) lately? > > Thanks > Stefan > > > -- > Stefan Marr > Software Languages Lab > Vrije Universiteit Brussel > Pleinlaan 2 / B-1050 Brussels / Belgium > http://soft.vub.ac.be/~smarr > Phone: +32 2 629 2974 > Fax: +32 2 629 3525 > >

Oct. 31, 2010

Re: [Pharo-project] About literalDictionary

by StÃ©phane Ducasse

ok thanks! On Oct 31, 2010, at 9:23 PM, Levente Uzonyi wrote: > On Sun, 31 Oct 2010, stephane ducasse wrote: > >> Hi levente >> >> I saw that literalDictionary was removed from squeak compiler. >> Can you tell us the reason? > > - one class less in the system > - one class less in the HashedCollection hierarchy, so it will be easier to migrate HashedCollections to the new object format > - we have a nice example for PluggableDictionary > > > Levente > >> Tx >> >> Stef >> >> >

Oct. 31, 2010

Re: [Pharo-project] Fwd: about OPAX

by StÃ©phane Ducasse

Thanks!!! On Oct 31, 2010, at 8:06 PM, Tudor Girba wrote: > Done. > > Doru > > On 31 Oct 2010, at 12:17, StÃ©phane Ducasse wrote: > >> Hi doru >> >> could you take 15 min and add comments to the OPAX classes? >> in XMLSupport. There is not a line of comments, so I know that this is your private code that got there but still. >> >> Stef >> >> >> >> > > -- > www.tudorgirba.com > > "Don't give to get. Just give." > > > > > >

Oct. 31, 2010

Re: [Pharo-project] InputEventFetcher starving all other processes

by Stefan Marr

Hi: On 31 Oct 2010, at 19:15, Stefan Marr wrote: > But the symptoms are that InputEventFetcher>>primGetNextEvent: delivers 500-1000x more events than EventSensor>>primGetNextEvent: in MVC or Squeak 4.1 images deliver. Turns out that the array which is supposed to be filled in by '>>primGetNextEvent: array' does not get filled in correctly. For some reason the old values stay in the array when there isn't any new event. However, the very same code works for the non-Pharo images. Was there any change to the code in or related to this primitive (<primitive: 94>) lately? Thanks Stefan -- Stefan Marr Software Languages Lab Vrije Universiteit Brussel Pleinlaan 2 / B-1050 Brussels / Belgium http://soft.vub.ac.be/~smarr Phone: +32 2 629 2974 Fax: +32 2 629 3525

Oct. 31, 2010

Re: [Pharo-project] About literalDictionary

by Levente Uzonyi

On Sun, 31 Oct 2010, stephane ducasse wrote: > Hi levente > > I saw that literalDictionary was removed from squeak compiler. > Can you tell us the reason? - one class less in the system - one class less in the HashedCollection hierarchy, so it will be easier to migrate HashedCollections to the new object format - we have a nice example for PluggableDictionary Levente > Tx > > Stef > >

Oct. 31, 2010

Re: [Pharo-project] how to load magritte in seaside 3.0?

by Esteban Lorenzano

MAReport from magritte 1 does not work well with Seaside 3 (it needs a small fix), all other things works ok. cheers, Esteban On 2010-10-31 06:40:31 -0300, Lukas Renggli <renggli(a)gmail.com> said: > Magritte 2 uses Grease as a portability layer and works well on Seaside 3. > > Magritte 1 has its own portability layer, works well on Seaside 2.8, > but also 3 (I think). > > You should use Magritte 2 on Seaside 3. > > Lukas > > > > On 31 October 2010 10:36, StÃ©phane Ducasse > <stephane.ducasse(a)inria.fr> wr > ote: >> thanks lukas >> >> what is magritte2 vs magritte? >> Because I 'm trying to rebuild an old image of mine that use an old versi > on of magritte >> >> Stef >> >> On Oct 31, 2010, at 10:11 AM, Lukas Renggli wrote: >> >>> Until somebody figures out why the configuration does not work you >>> could use the following script from >>> http://github.com/renggli/builder/blob/master/scripts/magritte2.st. Of >>> course this assumes that Seaside 3 is already loaded. >>> >>> Gofer new >>> Â Â Â renggli: 'magritte2'; >>> Â Â Â package: 'Magritte-Model'; >>> Â Â Â package: 'Magritte-Pharo-Model'; >>> Â Â Â package: 'Magritte-Seaside'; >>> Â Â Â package: 'Magritte-Pharo-Seaside'; >>> Â Â Â package: 'Magritte-Morph'; >>> Â Â Â package: 'Magritte-Tests-Model'; >>> Â Â Â package: 'Magritte-Tests-Pharo-Model'; >>> Â Â Â load. >>> >>> Lukas >>> >>> On 31 October 2010 09:38, stephane ducasse >>> <stephane.ducasse(a)free.fr> wr > ote: >>>> Hi >>>> >>>> I want to load magritte with Seaside 3.0 and I followed >>>> >>>> http://book.pharo-project.org/book/PharoTools/Magritte/ >>>> >>>> Magritte2 For Pharo With Seaside 3 >>>> >>>> Gofer it >>>> Â Â Â squeaksource: 'MetacelloRepository'; >>>> Â Â Â package: 'ConfigurationOfMagritte'; >>>> Â Â Â load. >>>> (ConfigurationOfMagritte2 project version: '2.0.5') load >>>> >>>> but I does not load. >>>> >>>> Stef >>>> >>>> >>>> >>>> >>> >>> >>> >>> -- >>> Lukas Renggli >>> www.lukas-renggli.ch

Oct. 31, 2010