December 2012 - Pharo-users

Re: [Pharo-users] Running remote command via SSH causes Pharo to hang

by Ben Coman

Frank Church wrote: > On 28 December 2012 23:44, GÃ¶ran Krampe <goran(a)krampe.se> wrote: > > >> On 12/28/2012 04:02 PM, Frank Church wrote: >> >> >>> I using the PipeableOSProcess to run ssh commands which return output >>> and Pharo hangs when I try it. Commands executing locally work okay, but >>> the ssh ones generally hang. >>> >>> >> I do this alot in my current project and I have a small little framework >> around this because IIRC the "problem" is that your call to #output will >> block until #atEnd. So what you want to do if you do not want to block is >> to fork off a background process that reads without blocking. This works >> really well for me, so I run multiple of these in parallell and each is >> given its own background Process that reads and copies into a WriteStream >> on a String. So if I ask my "wrapper" for the output it will ask this >> stream for its contents instead. >> >> regards, GÃ¶ran >> >> >> > I jumped straight in and tried to hack without learning some Smalltalk > basics but after spending a day with Pharo By Example I am beginning to > make some progress. > > Is the Transcript in Pharo 1.4 different from the one used in the PBE > image? It displays ThreadedTranscript when I check its properties and it > doesn't seem to hang like the PBE image. > > Some example code would really help here. > > Thanks > > > > > Transcript was replaced at some point - I "think" it was after PBE publication. I link to some related posts below. Regarding sample code... what are you trying to do? The way I commonly use the Transcript is... Transcript crShow: something asString, '---', somethingElse asString. asString might not be the most efficient way, but its simple. Without 'asString', using the comma to concatenate strings often does does not work since non-string classes generally don't understand comma. http://code.google.com/p/pharo/issues/detail?id=3948 http://www.mail-archive.com/pharo-project@lists.gforge.inria.fr/msg51417.ht… http://code.google.com/p/pharo/issues/list?can=1&q=summary%3Atranscript&sor…

Dec. 29, 2012

Re: [Pharo-users] Running remote command via SSH causes Pharo to hang

by Sean P. DeNigris

Frank Church wrote > I using the PipeableOSProcess... but the > ssh ones generally hang. I don't know if this has anything to do with your problem, but I often run into hanging output with OSP and usually send... I think it's #setNonBlockingOutput... to the PipeableOSProcess before reading the output. Then IIRC #output should immediately return whatever's currently available, and you can keep looping to get more. HTH, Sean -- View this message in context: http://forum.world.st/Running-remote-command-via-SSH-causes-Pharo-to-hang-t… Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com.

Dec. 29, 2012

What is the difference between GlorpDBX and SqueakDBX?

by Frank Church

What is the difference between GlorpDBX and SqueakDBX/DBXTalk and where does the original Glorp fit in? -- Frank Church ======================= http://devblog.brahmancreations.com

Dec. 29, 2012

Re: [Pharo-users] Running remote command via SSH causes Pharo to hang

by Frank Church

On 28 December 2012 23:44, GÃ¶ran Krampe <goran(a)krampe.se> wrote: > On 12/28/2012 04:02 PM, Frank Church wrote: > >> >> I using the PipeableOSProcess to run ssh commands which return output >> and Pharo hangs when I try it. Commands executing locally work okay, but >> the ssh ones generally hang. >> > > I do this alot in my current project and I have a small little framework > around this because IIRC the "problem" is that your call to #output will > block until #atEnd. So what you want to do if you do not want to block is > to fork off a background process that reads without blocking. This works > really well for me, so I run multiple of these in parallell and each is > given its own background Process that reads and copies into a WriteStream > on a String. So if I ask my "wrapper" for the output it will ask this > stream for its contents instead. > > regards, GÃ¶ran > > I jumped straight in and tried to hack without learning some Smalltalk basics but after spending a day with Pharo By Example I am beginning to make some progress. Is the Transcript in Pharo 1.4 different from the one used in the PBE image? It displays ThreadedTranscript when I check its properties and it doesn't seem to hang like the PBE image. Some example code would really help here. Thanks -- Frank Church ======================= http://devblog.brahmancreations.com

Dec. 29, 2012

Re: [Pharo-users] Running remote command via SSH causes Pharo to hang

by GÃ¶ran Krampe

On 12/28/2012 04:02 PM, Frank Church wrote: > > I using the PipeableOSProcess to run ssh commands which return output > and Pharo hangs when I try it. Commands executing locally work okay, but > the ssh ones generally hang. I do this alot in my current project and I have a small little framework around this because IIRC the "problem" is that your call to #output will block until #atEnd. So what you want to do if you do not want to block is to fork off a background process that reads without blocking. This works really well for me, so I run multiple of these in parallell and each is given its own background Process that reads and copies into a WriteStream on a String. So if I ask my "wrapper" for the output it will ask this stream for its contents instead. regards, GÃ¶ran

Dec. 28, 2012

Running remote command via SSH causes Pharo to hang

by Frank Church

I using the PipeableOSProcess to run ssh commands which return output and Pharo hangs when I try it. Commands executing locally work okay, but the ssh ones generally hang. Transcript show: (PipeableOSProcess command: '~/deploy_scripts/linkproxy.sh -f 1591') output eg The linkproxy.sh backgrounds an SSH command that enables both remote and local forwarding via the "command &" option. Is there some other option I can use to stop the script hanging? The other part is when I use execute a command remotely via SSH. eg - ssh user@localhost 'execute remote command'. This option also results in a hang. Anyway I will make some further checks and report back. Transcript show: (PipeableOSProcess command: '~/deploy_scripts/linkproxy.sh -f 1591') output -- Frank Church ======================= http://devblog.brahmancreations.com

Dec. 28, 2012

Re: [Pharo-users] Two questions about Smalltalk language design

by S Krish

1) Why do ST methods return "self" if nothing is explicitly returned? Blue book talks of: *extent: newExtent* corner <- origin + newExtent *Bytecode:* Rectangle extent: 0 push the value of the receiver's first instance variable (origin) onto the stack 16 push the argument (newExtent) onto the stack 176 send a binary message with the selector + 97 pop the top object off of the stack and store it in the receiver's second instance variable (corner) *120 return the receiver as the value of the message (extent:)* For blocks: "Since a block's expressions may not have been evaluated when it returns from fork, the value of fork must be independent of the value of the block's expressions. A block returns itself as the value of fork." also: http://www.mirandabanda.org/bluebook/bluebook_chapter28.html#ReturnBytecode… Semantically in a pure OOPS land sending a message to an object returning a nil / void may seem not so object oriented either. Message modifies or actions something on the object, the default return of the object seems sensible option to return nil Facilitates I believe the easy operation of chaining method sends without each time having explicitly to return the object in each of those methods. viz: Aircraft airforceOne flyToDestination haltForRefuelling returnToStation. or its other forms with bracketed keyword messages. Though using cascades one can do similar stuff but will read less english like. 2) In Collections, why does "add:" return the object being added, and not "self"? Blue Book: "Note that the add: and remove: messages answer the argument rather than the collection itself so that computed arguments can be accessed." I have always pondered why this is essential and returning the collection would be fine..or return true / false rather than argument / exception, infact for Set new add: 2; add: 2 . will return 2.. though the second one does not add anything to the Set. On Thu, Dec 27, 2012 at 6:02 AM, Sebastian Nozzi <sebnozzi(a)gmail.com> wrote: > Hello Smalltalkers, > > I was interacting the other day with an user in a Scala mailing list, > and he had some questions about Smalltalk. From his FP (functional > programming) point of view there were some things that looked strange. > I tried to reply to the best of my ability, but still I don't really > know the background for these design decisions. So I come to you, real > experts, hoping for some answers. > > The questions were: > > 1) Why do ST methods return "self" if nothing is explicitly returned? > (he would have expected something like "Unit" in Scala, or "Void" - > which I know makes little sense in a world like ST's, so I didn't > include this in the Stackoverflow question) > > > http://stackoverflow.com/questions/14047887/why-do-methods-return-self-by-d… > > 2) In Collections, why does "add:" return the object being added, and > not "self"? > > > http://stackoverflow.com/questions/14047940/why-does-add-return-the-object-… > > ---- > > The original interaction in the Scala mailing list (nice language, by > the way) was here: > > https://groups.google.com/d/topic/scala-user/gPEJNAEZO-8/discussion > > > Thanks a lot in advance! > > > Sebastian > >

Dec. 28, 2012

[Pharo-Users] FFI question. How read external byte* to image Form?

by Denis Kudriashov

Hello. My application works extensively with png images. So Image based PNGReadWriter speed not enough for me. I have C library dll which has function "byte* readPNG(filename)". It returns buffer (pointer address) with same structure as Form Bitmap has. I use FFI to call this function. And I have very naive converter of returned buffer to Form: buffer := self primReadImageBytesFrom: imagePath. buffer getHandle asInteger == 0 ifTrue: [self error: 'bad image: ', imagePath ]. form := Form extent: width @ height depth: 32. x:= 0. y:=0. [y <= height ] whileTrue: [ x :=0. [x <= width] whileTrue: [ offset := x * 4 + (y * width * 4) + 1 + 8. value := buffer getHandle integerAt: offset size: 4 signed: false. form pixelValueAt: x@y put: value. x := x + 1. ]. y := y + 1. ]. ^form It's work and it is improve that buffer format similar to Form bits format. But such algorithm very slow. So my question is: How efficiently convert external byte array from FFI call to internal ByteArray (Bitmap, Form). I know about BitBlt class and blt := (BitBlt current toForm: f) sourceForm: source. blt combinationRule: Form over. "store" blt sourceX: 0; sourceY: 0; height: height; width: width. blt destX: 0; destY: 0. blt copyBits. And it works very fast. I try create source Form by source := Form extent: width@height depth: 32. source bits: buffer getHandle But BitBlt raise error for such form. #copyBits primitive failed and method execution continue with attempt to unhibernate given forms. Any ideas?

Dec. 27, 2012

Two questions about Smalltalk language design

by Sebastian Nozzi

Hello Smalltalkers, I was interacting the other day with an user in a Scala mailing list, and he had some questions about Smalltalk. From his FP (functional programming) point of view there were some things that looked strange. I tried to reply to the best of my ability, but still I don't really know the background for these design decisions. So I come to you, real experts, hoping for some answers. The questions were: 1) Why do ST methods return "self" if nothing is explicitly returned? (he would have expected something like "Unit" in Scala, or "Void" - which I know makes little sense in a world like ST's, so I didn't include this in the Stackoverflow question) http://stackoverflow.com/questions/14047887/why-do-methods-return-self-by-d… 2) In Collections, why does "add:" return the object being added, and not "self"? http://stackoverflow.com/questions/14047940/why-does-add-return-the-object-… ---- The original interaction in the Scala mailing list (nice language, by the way) was here: https://groups.google.com/d/topic/scala-user/gPEJNAEZO-8/discussion Thanks a lot in advance! Sebastian

Dec. 27, 2012

Re: [Pharo-users] How to try out XTreams

by Sean P. DeNigris

Camillo Bruni-3 wrote > I would love to tackle that in early 3.0 :) Let me know when you're ready and I'll dig up the code... -- View this message in context: http://forum.world.st/How-to-try-out-XTreams-tp4657923p4660883.html Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com.

Dec. 25, 2012