- Pharo-users - lists.pharo.org

Re: [Pharo-users] Connecting to a C++ dll

by Esteban Lorenzano

btw, I don't know what are you trying to do, but some c++ libraries provides an "ffi capable" extra library, to handle this kind of issues. On Nov 28, 2012, at 1:31 PM, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: > no, it is not possible, since NativeBoost is "a kind of FFI" after all... and the problem is not with ffi libraries but with c++ (lack of) standards > > On Nov 28, 2012, at 1:25 PM, shanukalra <shanukalra25(a)gmail.com> wrote: > >> Is it possible if I use NativeBoost? And if possible can you please tell me >> exactly how to call an external function from a dll and display the ouput in >> pharo. >> >> Thanks >> Shanu >> >> >> >> -- >> View this message in context: http://forum.world.st/Connecting-to-a-C-dll-tp4656451p4657114.html >> Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com. >> >

Nov. 28, 2012

Re: [Pharo-users] Connecting to a C++ dll

by Esteban Lorenzano

no, it is not possible, since NativeBoost is "a kind of FFI" after all... and the problem is not with ffi libraries but with c++ (lack of) standards On Nov 28, 2012, at 1:25 PM, shanukalra <shanukalra25(a)gmail.com> wrote: > Is it possible if I use NativeBoost? And if possible can you please tell me > exactly how to call an external function from a dll and display the ouput in > pharo. > > Thanks > Shanu > > > > -- > View this message in context: http://forum.world.st/Connecting-to-a-C-dll-tp4656451p4657114.html > Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com. >

Nov. 28, 2012

Re: [Pharo-users] Connecting to a C++ dll

by shanukalra

Is it possible if I use NativeBoost? And if possible can you please tell me exactly how to call an external function from a dll and display the ouput in pharo. Thanks Shanu -- View this message in context: http://forum.world.st/Connecting-to-a-C-dll-tp4656451p4657114.html Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com.

Nov. 28, 2012

Re: [Pharo-users] VAST-style Compile-time constants

by Mariano Martinez Peck

On Tue, Nov 27, 2012 at 9:04 AM, Mariano Martinez Peck < marianopeck(a)gmail.com> wrote: > > > On Tue, Nov 27, 2012 at 12:54 AM, Bob Hartwig <bobjectssmalltalk(a)gmail.com > > wrote: > >> Hi all, >> >> I'm investigating the feasibility of porting a large VAST application to >> Pharo, so I may have some questions in the coming days. Here's my first >> question: >> >> VAST has the concept of compile-time constants, using the ##() syntax. >> Whatever is in the parentheses is evaluated at compile time, and the >> result is stored, so the evaluation does not need to happen again when the >> method is sent later. E.g., when this method is sent: >> >> printTheNumberToTheTranscript >> >> Transcript nextPutAll: ##( 10 factorial asString ) . >> >> no calculation or string conversion is necessary, because 10 factorial >> asString was evaluated at compile time and the answer was stored. Please >> note that this is different from array literals, which is how Pharo seems >> to treat this syntax. >> >> As far as I can tell, Pharo does not have this feature (certainly not >> with that syntax), but I would love to be proved wrong. Am I? >> > > Yes, it has it also. The array is stored as a literal in the compiled > method.....I think the sintax was with only one #, like: #(10 factorial > asString) > > can you try? > > uffff I should not answer emails before waking up....what I say is only for some kind of objects, not expressions. So, no, you don't have it out of the box. > Does anyone else have words of wisdom regarding porting a VAST app to >> Pharo? Any other language features that are missing? I'm not too concerned >> about Pragmas, and this app has no GUI, so I'm not concerned about UI >> framework differences. >> >> Thanks, >> Bob Hartwig >> >> > > > -- > Mariano > http://marianopeck.wordpress.com > > -- Mariano http://marianopeck.wordpress.com

Nov. 27, 2012

Re: [Pharo-users] VAST-style Compile-time constants

by StÃ©phane Ducasse

Hi bob Why could be done is to extend the compiler (marcus did that in the past) to have compile time evaluation. Now how many times such feature is used in your application? Stef > printTheNumberToTheTranscript > > Transcript nextPutAll: ##( 10 factorial asString ) . > > no calculation or string conversion is necessary, because 10 factorial asString was evaluated at compile time and the answer was stored. Please note that this is different from array literals, which is how Pharo seems to treat this syntax. > > As far as I can tell, Pharo does not have this feature (certainly not with that syntax), but I would love to be proved wrong. Am I? No it does not. > Does anyone else have words of wisdom regarding porting a VAST app to Pharo? Any other language features that are missing? Not really Did you check the float precision? > I'm not too concerned about Pragmas, and this app has no GUI, so I'm not concerned about UI framework differences. Do not hesitate to let us know the problem you encounter. > > Thanks, > Bob Hartwig >

Nov. 27, 2012

Re: [Pharo-users] VAST-style Compile-time constants

by Mariano Martinez Peck

On Tue, Nov 27, 2012 at 12:54 AM, Bob Hartwig <bobjectssmalltalk(a)gmail.com>wrote: > Hi all, > > I'm investigating the feasibility of porting a large VAST application to > Pharo, so I may have some questions in the coming days. Here's my first > question: > > VAST has the concept of compile-time constants, using the ##() syntax. > Whatever is in the parentheses is evaluated at compile time, and the > result is stored, so the evaluation does not need to happen again when the > method is sent later. E.g., when this method is sent: > > printTheNumberToTheTranscript > > Transcript nextPutAll: ##( 10 factorial asString ) . > > no calculation or string conversion is necessary, because 10 factorial > asString was evaluated at compile time and the answer was stored. Please > note that this is different from array literals, which is how Pharo seems > to treat this syntax. > > As far as I can tell, Pharo does not have this feature (certainly not with > that syntax), but I would love to be proved wrong. Am I? Does anyone else > have words of wisdom regarding porting a VAST app to Pharo? Any other > language features that are missing? > Not that I know. The first step is to fileout or somehow get the code from ENVY. Then start importing in Pharo. It would help if the app has tests in which case I recommend start porting the tests. > I'm not too concerned about Pragmas, and this app has no GUI, so I'm not > concerned about UI framework differences. > > Pragmas are used in Pharo for more things than just UI. Anyway, Pragmas do also exist in VAST. Finally, I was going to say that the most complicated part of the port would be the UI, but if there is none, then easier. Do you have a database? which frameworks/libraries do you use from VAST? > Thanks, > Bob Hartwig > > -- Mariano http://marianopeck.wordpress.com

Nov. 27, 2012

Re: [Pharo-users] VAST-style Compile-time constants

by Mariano Martinez Peck

On Tue, Nov 27, 2012 at 12:54 AM, Bob Hartwig <bobjectssmalltalk(a)gmail.com>wrote: > Hi all, > > I'm investigating the feasibility of porting a large VAST application to > Pharo, so I may have some questions in the coming days. Here's my first > question: > > VAST has the concept of compile-time constants, using the ##() syntax. > Whatever is in the parentheses is evaluated at compile time, and the > result is stored, so the evaluation does not need to happen again when the > method is sent later. E.g., when this method is sent: > > printTheNumberToTheTranscript > > Transcript nextPutAll: ##( 10 factorial asString ) . > > no calculation or string conversion is necessary, because 10 factorial > asString was evaluated at compile time and the answer was stored. Please > note that this is different from array literals, which is how Pharo seems > to treat this syntax. > > As far as I can tell, Pharo does not have this feature (certainly not with > that syntax), but I would love to be proved wrong. Am I? > Yes, it has it also. The array is stored as a literal in the compiled method.....I think the sintax was with only one #, like: #(10 factorial asString) can you try? > Does anyone else have words of wisdom regarding porting a VAST app to > Pharo? Any other language features that are missing? I'm not too concerned > about Pragmas, and this app has no GUI, so I'm not concerned about UI > framework differences. > > Thanks, > Bob Hartwig > > -- Mariano http://marianopeck.wordpress.com

Nov. 27, 2012

VAST-style Compile-time constants

by Bob Hartwig

Hi all, I'm investigating the feasibility of porting a large VAST application to Pharo, so I may have some questions in the coming days. Here's my first question: VAST has the concept of compile-time constants, using the ##() syntax. Whatever is in the parentheses is evaluated at compile time, and the result is stored, so the evaluation does not need to happen again when the method is sent later. E.g., when this method is sent: printTheNumberToTheTranscript Transcript nextPutAll: ##( 10 factorial asString ) . no calculation or string conversion is necessary, because 10 factorial asString was evaluated at compile time and the answer was stored. Please note that this is different from array literals, which is how Pharo seems to treat this syntax. As far as I can tell, Pharo does not have this feature (certainly not with that syntax), but I would love to be proved wrong. Am I? Does anyone else have words of wisdom regarding porting a VAST app to Pharo? Any other language features that are missing? I'm not too concerned about Pragmas, and this app has no GUI, so I'm not concerned about UI framework differences. Thanks, Bob Hartwig

Nov. 26, 2012

Re: [Pharo-users] CollectionExtensions missing in Pharo 2.0

by Esteban Lorenzano

and you need to be fast, because as soon as we declare beta (next week), nothing else will be added :) On Nov 23, 2012, at 7:03 PM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > usman > > we would love to get some of the nice collection extensions made in Moose. > Now somebody has to push it to us. > open a ticket and submit code and you will get it. > Stef > > On Nov 22, 2012, at 4:37 PM, Usman Bhatti wrote: > >> Hello all, >> >> I remarked that the package CollectionExtensions is missing in Pharo 2.0. We are using a useful method provided by the package on strings that provides character intervals for line numbers in a string. Here is the code from 1.4: >> >> intervalOfLine: aLineNumber >> >> "Answer an interval spanning between the first and the last character of the line from aLineNumber" >> >> | lineIndex | >> lineIndex := 0. >> self lineIndicesDo: [ :start :endWithoutDelimiters :end | >> lineIndex := lineIndex + 1. >> lineIndex = aLineNumber ifTrue: [ ^ start to: end ] ]. >> ^ 0 to: 0 >> >> Now, this method is missing in Pharo 2.0. Is there a particular reason for omitting the package? Is there a replacement for the pckage? >> >> tx, >> Usman > >

Nov. 23, 2012

Re: [Pharo-users] CollectionExtensions missing in Pharo 2.0

by StÃ©phane Ducasse

usman we would love to get some of the nice collection extensions made in Moose. Now somebody has to push it to us. open a ticket and submit code and you will get it. Stef On Nov 22, 2012, at 4:37 PM, Usman Bhatti wrote: > Hello all, > > I remarked that the package CollectionExtensions is missing in Pharo 2.0. We are using a useful method provided by the package on strings that provides character intervals for line numbers in a string. Here is the code from 1.4: > > intervalOfLine: aLineNumber > > "Answer an interval spanning between the first and the last character of the line from aLineNumber" > > | lineIndex | > lineIndex := 0. > self lineIndicesDo: [ :start :endWithoutDelimiters :end | > lineIndex := lineIndex + 1. > lineIndex = aLineNumber ifTrue: [ ^ start to: end ] ]. > ^ 0 to: 0 > > Now, this method is missing in Pharo 2.0. Is there a particular reason for omitting the package? Is there a replacement for the pckage? > > tx, > Usman

Nov. 23, 2012