- Pharo-users - lists.pharo.org

Re: [Pharo-users] modifying the compiler adding an alias for :=

by Thierry Goubier

Hi Erik, You need to make sure you are using the right compiler. In a Pharo3 image, you'll find two compilers: Opal and the old one. In the old one, you need to change: Scanner>>xColon In RB/Opal, what you can try with your changes is to inspect: RBParser parseExpression: 'a :> 5' And, if you get a RBAssignmentNode, then it is working. Thierry 2014-12-10 13:32 GMT+01:00 Erik Itter <erik.itter(a)gmail.com>: > Thanks, tried > > scanSpecialCharacter > | character | > currentCharacter = $: > ifTrue: [ > self step. > ^ currentCharacter = $= > ifTrue: [ > self step. > RBAssignmentToken start: tokenStart] > ifFalse: [ RBSpecialCharacterToken value: $: start: > tokenStart ]]. > > currentCharacter = $: > ifTrue: [ > self step. > ^ currentCharacter = $> > ifTrue: [ > self step. > RBAssignmentToken start: tokenStart] > ifFalse: [ RBSpecialCharacterToken value: $: start: > tokenStart ]]. > > currentCharacter = $_ ifTrue: [ > self step. > ^ RBShortAssignmentToken start: tokenStart ]. > > character := currentCharacter. > self step. > ^ RBSpecialCharacterToken value: character start: tokenStart > > but while it does not seem to shred my image to pieces it also seems to > have no effect at all. I guess the UI components still check differently > and neither workplace nor SystemBrowser allow me to write code using the :> > operator > > Am 10.12.2014 um 13:11 schrieb Thierry Goubier: > > Hi, > > If you are using the OpalCompiler (Pharo4, sure, Pharo3, I'm not so > sure), have a look at: > > RBScanner>>scanSpecialCharacter > > Thierry > > 2014-12-10 12:57 GMT+01:00 Erik Itter <erik.itter(a)gmail.com>: > >> Hi, >> >> I need to modify the compiler adding ":>" as an alias for ":=" for >> assignment. I have been reading all over the Compiler package but do not >> find where a tokenizer parses for the assignement operator or where it is >> defined. >> >> Any hint where to start or do it is appreciated. >> >> best regards >> Erik >> >> > >

Dec. 10, 2014

Re: [Pharo-users] modifying the compiler adding an alias for :=

by Erik Itter

Thanks, tried scanSpecialCharacter | character | currentCharacter = $: ifTrue: [ self step. ^ currentCharacter = $= ifTrue: [ self step. RBAssignmentToken start: tokenStart] ifFalse: [ RBSpecialCharacterToken value: $: start: tokenStart ]]. currentCharacter = $: ifTrue: [ self step. ^ currentCharacter = $> ifTrue: [ self step. RBAssignmentToken start: tokenStart] ifFalse: [ RBSpecialCharacterToken value: $: start: tokenStart ]]. currentCharacter = $_ ifTrue: [ self step. ^ RBShortAssignmentToken start: tokenStart ]. character := currentCharacter. self step. ^ RBSpecialCharacterToken value: character start: tokenStart but while it does not seem to shred my image to pieces it also seems to have no effect at all. I guess the UI components still check differently and neither workplace nor SystemBrowser allow me to write code using the :> operator Am 10.12.2014 um 13:11 schrieb Thierry Goubier: > Hi, > > If you are using the OpalCompiler (Pharo4, sure, Pharo3, I'm not so > sure), have a look at: > > RBScanner>>scanSpecialCharacter > > Thierry > > 2014-12-10 12:57 GMT+01:00 Erik Itter <erik.itter(a)gmail.com > <mailto:erik.itter@gmail.com>>: > > Hi, > > I need to modify the compiler adding ":>" as an alias for ":=" for > assignment. I have been reading all over the Compiler package but > do not find where a tokenizer parses for the assignement operator > or where it is defined. > > Any hint where to start or do it is appreciated. > > best regards > Erik > >

Dec. 10, 2014

Re: [Pharo-users] modifying the compiler adding an alias for :=

by Thierry Goubier

Hi, If you are using the OpalCompiler (Pharo4, sure, Pharo3, I'm not so sure), have a look at: RBScanner>>scanSpecialCharacter Thierry 2014-12-10 12:57 GMT+01:00 Erik Itter <erik.itter(a)gmail.com>: > Hi, > > I need to modify the compiler adding ":>" as an alias for ":=" for > assignment. I have been reading all over the Compiler package but do not > find where a tokenizer parses for the assignement operator or where it is > defined. > > Any hint where to start or do it is appreciated. > > best regards > Erik > >

Dec. 10, 2014

modifying the compiler adding an alias for :=

by Erik Itter

Hi, I need to modify the compiler adding ":>" as an alias for ":=" for assignment. I have been reading all over the Compiler package but do not find where a tokenizer parses for the assignement operator or where it is defined. Any hint where to start or do it is appreciated. best regards Erik

Dec. 10, 2014

Re: [Pharo-users] Left mouse button not working in Pharo 3.0

by kilon alios

just upload your image file so we can take a look, one of the great things about pharo ;) you can also upload the whole pharo folder On Wed, Dec 10, 2014 at 10:04 AM, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: > can you share a screenshot of your processes? (World Menu/Tools/Process > Browsers) > > Esteban > > > > On 10 Dec 2014, at 02:58, David Boeren <boerend(a)gmail.com> wrote: > > Didn't seem to have any effect. The broken windows are still there, and > the file size is still 99.5MB. > > On Tue, Dec 9, 2014 at 8:49 PM, Mariano Martinez Peck < > marianopeck(a)gmail.com> wrote: > >> The size could have been the Monticello cache for all the stuff you >> loaded? >> In the big image try ImageCleaner new cleanUpForRelease, then save it and >> compare size. >> >> CHeers, >> >> On Tue, Dec 9, 2014 at 9:22 PM, stepharo <stepharo(a)free.fr> wrote: >> >>> This is indeed quite strange. >>> >>> Le 9/12/14 14:31, dboeren a Ã©crit : >>> >>> OK, some progress. I made a copy of Pharo to experiment with and found >>>> World->Windows->DeleteAllWindowsDiscardingEdits. >>>> >>>> This destroyed the three rogue windows and at least to a first glance >>>> the >>>> system is working correctly again. >>>> >>>> But, I also still have the original copy which is still malfunctioning. >>>> >>>> Ready for more weird? The broken copy of Pharo is 99.5MB. The working >>>> copy >>>> is only 70.2MB. >>>> >>>> Here is the full list of what I did to create the working copy: >>>> 1. Copy/paste the Pharo3.0 file in OSX Finder, rename it to "Pharo 3.0 >>>> test" >>>> 2. Open Pharo 3.0 Test >>>> 3. World->Windows->DeleteAllWindowsDiscardingEdits >>>> 4. Recreate my 3 windows (Workspace, Transcript, Browser) and resize >>>> them >>>> to approximately the same layout I had before >>>> 5. Quit and save the new image >>>> >>>> So whatever was wrong with those three windows was taking up 29.3MB? >>>> That's >>>> really bizarre... >>>> >>>> If someone knows how to debug this sort of thing I can upload my >>>> corrupted >>>> copy to you and perhaps that would be the best way to shed some light, >>>> but >>>> at least I seem to have a working environment again so I can proceed on >>>> the >>>> next step of my program :) >>>> >>>> >>>> >>>> >>>> -- >>>> View this message in context: http://forum.world.st/Left- >>>> mouse-button-not-working-in-Pharo-3-0-tp4795107p4795126.html >>>> Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com. >>>> >>>> >>>> >>> >>> >> >> >> -- >> Mariano >> http://marianopeck.wordpress.com >> > > >

Dec. 10, 2014

Re: [Pharo-users] Working with zipped files

by Markus Fritsche

On 2014-12-10 01:23, Ð®ÑÐ¸Ð¹ ÐÐ¸ÑÐ¾Ð½ÐµÐ½ÐºÐ¾ wrote: > General idea was to use ZipArchive instead of MCZ-related code. Sorry, I wasn't clear. MCZ was just my example. The real question was, if the ZipFileSystem API can be used or not... I will go with ZipArchive and friends then.

Dec. 10, 2014

Re: [Pharo-users] Working with zipped files

by Markus Fritsche

On 2014-12-09 20:47, HernÃ¡n Morales Durand wrote: > | zipArchive fileRef | > zipArchive := ZipArchive new. > fileRef := 'myFile.zip' asFileReference. > [ zipArchive > readFrom: fileRef fullName; > extractAllTo: FileSystem workingDirectory ] > ensure: [ zipArchive close ]. Thank you!

Dec. 10, 2014

Re: [Pharo-users] Left mouse button not working in Pharo 3.0

by Esteban Lorenzano

can you share a screenshot of your processes? (World Menu/Tools/Process Browsers) Esteban > On 10 Dec 2014, at 02:58, David Boeren <boerend(a)gmail.com> wrote: > > Didn't seem to have any effect. The broken windows are still there, and the file size is still 99.5MB. > > On Tue, Dec 9, 2014 at 8:49 PM, Mariano Martinez Peck <marianopeck(a)gmail.com <mailto:marianopeck@gmail.com>> wrote: > The size could have been the Monticello cache for all the stuff you loaded? > In the big image try ImageCleaner new cleanUpForRelease, then save it and compare size. > > CHeers, > > On Tue, Dec 9, 2014 at 9:22 PM, stepharo <stepharo(a)free.fr <mailto:stepharo@free.fr>> wrote: > This is indeed quite strange. > > Le 9/12/14 14:31, dboeren a Ã©crit : > > OK, some progress. I made a copy of Pharo to experiment with and found > World->Windows->DeleteAllWindowsDiscardingEdits. > > This destroyed the three rogue windows and at least to a first glance the > system is working correctly again. > > But, I also still have the original copy which is still malfunctioning. > > Ready for more weird? The broken copy of Pharo is 99.5MB. The working copy > is only 70.2MB. > > Here is the full list of what I did to create the working copy: > 1. Copy/paste the Pharo3.0 file in OSX Finder, rename it to "Pharo 3.0 > test" > 2. Open Pharo 3.0 Test > 3. World->Windows->DeleteAllWindowsDiscardingEdits > 4. Recreate my 3 windows (Workspace, Transcript, Browser) and resize them > to approximately the same layout I had before > 5. Quit and save the new image > > So whatever was wrong with those three windows was taking up 29.3MB? That's > really bizarre... > > If someone knows how to debug this sort of thing I can upload my corrupted > copy to you and perhaps that would be the best way to shed some light, but > at least I seem to have a working environment again so I can proceed on the > next step of my program :) > > > > > -- > View this message in context: http://forum.world.st/Left-mouse-button-not-working-in-Pharo-3-0-tp4795107p… <http://forum.world.st/Left-mouse-button-not-working-in-Pharo-3-0-tp4795107p…> > Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com. > > > > > > > > -- > Mariano > http://marianopeck.wordpress.com <http://marianopeck.wordpress.com/> >

Dec. 10, 2014

Re: [Pharo-users] [ Article ] LampSort, a non-recursive QuickSort implementation

by Sven Van Caekenberghe

Because the array sorts in place, and it seems good practice to copy literal constants (it does not matter much in a workspace, but it does lead to problems in actual code where the constants are stored in methods). I could have added the sort: / sorted: distinction but I went for as little methods as possible. > On 10 Dec 2014, at 03:05, stepharo <stepharo(a)free.fr> wrote: > > Sven > > why do you copy the array passed to sort? > > Stef > Le 8/12/14 10:36, Sven Van Caekenberghe a Ã©crit : >> Hi, >> >> Here is another article I just published >> >> LampSort, a non-recursive QuickSort implementation >> >> The divide and conquer partitioning is at the heart of QuickSort >> >> https://medium.com/@svenvc/lampsort-a-non-recursive-quicksort-implementatio… >> >> Pharo makes it easy to implement this non-recursive version of QuickSort - and beautiful as well. >> >> Sven >> >> >> > >

Dec. 10, 2014

Re: [Pharo-users] [ Article ] LampSort, a non-recursive QuickSort implementation

by stepharo

Sven why do you copy the array passed to sort? Stef Le 8/12/14 10:36, Sven Van Caekenberghe a Ã©crit : > Hi, > > Here is another article I just published > > LampSort, a non-recursive QuickSort implementation > > The divide and conquer partitioning is at the heart of QuickSort > > https://medium.com/@svenvc/lampsort-a-non-recursive-quicksort-implementatio… > > Pharo makes it easy to implement this non-recursive version of QuickSort - and beautiful as well. > > Sven > > >

Dec. 10, 2014