September 2016 - Pharo-users

Re: [Pharo-users] Pharo not running with macOS Sierra

by MartinW

vivus-ignis wrote > which image were you using? The "unofficial" VM from the link presents a > dialog to open a file and once I select an image, it just disappears and > nothing happens after that. The cog_macos32x86_pharo.cog.spur_201609201401.tar.gz works for me, the cog_macos64x64_pharo.cog.spur_201609201401.tar.gz does not. -- View this message in context: http://forum.world.st/Pharo-not-running-with-macOS-Sierra-tp4916468p4916700… Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com.

Sept. 23, 2016

Re: [Pharo-users] Using a Unix filedescriptor in a FileStream?

by Ronie Salgado

Hi Holger, Sorry for not noticing this thread before. Some months ago, I did some experiments myself using inotify and epoll, with the UFFI. I was creating a file system monitoring API similar in design to OSWindow. I just did some minor changes for making this usable along with making a configure script. I left all of the code and some scripts for loading the FileSystemMonitor API at: http://smalltalkhub.com/#!/~ronsaldo/OSCalls . For testing the monitoring API, try the following in a playground: monitor := FileSystemMonitor for: '.' do: [ :event | Transcript show: event; cr. ]. If you do this, open a Transcript and you start browsing some methods in Nautilus, you will start receiving some events where Pharo is accessing the .changes file. For the destroying the monitor and stop receiving the events, you can do the following: monitor destroy. Currently the events given by this API are only holding a copy of the raw inotify event in rawEvent and a copy of the raw path in rawPath. Further work is required for translating the events into a more platform independent interface, and for supporting OS X and Windows with this API. However, this is already something that could be useful to you. Best regards, Ronie 2016-09-20 15:26 GMT-03:00 Holger Freyther <holger(a)freyther.de>: > Hi, > > I explored to use the Linux inotify API using UFFI. The > inotify_init/inotify_init1 routine will me a Unix filedescriptor and I > would like to > > a.) monitor it for being readable > b.) read from it with a stream > > > For a.) I have found the AioEventHandler and think I will be able to call > >>#descriptor: directly and then can use it (still figuring out the API, > probably just wait for the >>#changed: call). > > For b.) I thought I could use AttachableFileStream but that required a > "fileId" but not a filedescriptor. Is there a way I can read from my fd > using the standard stream API (otherwise I can try to use UFFI for read) > > > regards > holger >

Sept. 23, 2016

Re: [Pharo-users] Using a Unix filedescriptor in a FileStream?

by Holger Freyther

> On 22 Sep 2016, at 23:33, Mariano Martinez Peck <marianopeck(a)gmail.com> wrote: > > Hi Holger, Hey! > > I just run all OSSubprocess tests and they all worked! (tested in Pharo 5.0). > > I guess I will commit this on the dev branch and hopefully when this is integrated into the VM I can merge that for my next OSSubprocess release. cool and thank you for trying it so quickly. I thought it make sense to mark non files the same as isStdioStream but that triggers a funny case: do { clearerr(file); if (fread(dst, 1, 1, file) == 1) { bytesRead += 1; if (dst[bytesRead-1] == '\n' || dst[bytesRead-1] == '\r') break; } } while (bytesRead <= 0 && ferror(file) && errno == EINTR); which means 0 or 1 char is read with >>#primRead:into:startingAt:count: and in case of inotify the event is lost (partially read and the rest discarded). I will have to start a discussion why primRead should be line buffered at all. Anyway. My good news is that: "an INotify" | arr fileID | self init: 8r00004000. self prim_add_watch: '/tmp' flags: 16r00000100. arr := ByteArray new: 4096. fileID := StandardFileStream new primFdOpen: fd writable: false. StandardFileStream new primRead: fileID into: arr startingAt: 1 count: 4096. arr has read a file notification event.. :)

Sept. 22, 2016

Re: [Pharo-users] Using a Unix filedescriptor in a FileStream?

by Mariano Martinez Peck

Hi Holger, OK, I made a new VM with your PR after I reviewed it. Good news? It works!!! If you want to check OSSubprocess, I am using now the primCreatePipe from OSProcess because that would answer me directly the SQFiles of the pipe. Originally (before ending up doing that), I was trying to make the pipes myself via FFI ( pipe() ) but I came to the problem we discussed earlier (remember the #name:attachTo:writable:). So now I can go back to my original solution. The code is in # *makePipeWithReadBlocking*: and now it looks like this: | pipePointer returnValue fileDescriptors pipe fileIDsArray fileDescriptor1 fileDescriptor2 | pipePointer := ExternalAddress allocate: 8. [ returnValue := self primitivePipe: pipePointer. (returnValue = -1) ifTrue: [ self perror: 'pipe()' ]. fileIDsArray := Array new: 2. fileDescriptor1 := pipePointer nbUInt32AtOffset: 0. fileDescriptor2 := pipePointer nbUInt32AtOffset: 4. fileIDsArray at: 1 put: *(self primitiveFileOpenUseFileDescriptor: fileDescriptor1 writeFlag: false)*. fileIDsArray at: 2 put: *(self primitiveFileOpenUseFileDescriptor: fileDescriptor2 writeFlag: true).* pipe := OSSPipe newWith: fileIDsArray readBlocking: aBoolean. ] ensure:[ pipePointer free. ]. ^ pipe I just run all OSSubprocess tests and they all worked! (tested in Pharo 5.0). I guess I will commit this on the dev branch and hopefully when this is integrated into the VM I can merge that for my next OSSubprocess release. Thank you very much Holger On Thu, Sep 22, 2016 at 12:35 PM, Mariano Martinez Peck < marianopeck(a)gmail.com> wrote: > > > On Thu, Sep 22, 2016 at 12:12 PM, Holger Freyther <holger(a)freyther.de> > wrote: > >> >> > On 21 Sep 2016, at 15:09, Mariano Martinez Peck <marianopeck(a)gmail.com> >> wrote: >> > >> > >> > >> > Exactly. I have been wanting this a couple of times while doing >> OSSubprocess. >> >> >> https://github.com/pharo-project/pharo-vm/pull/108. Would be nice if you >> could review it and give it a try. It adds two primitive (one to work on fd >> one to work on FILE). >> >> I probably also want to do: >> >> sqFile->isStdioStream = isatty(fileno(file)) >> >> >> > Yes, exactly. I remember now. And as I said, I also wanted to be able >> to work at fd or FILE* level and I failed. >> >> Could you give the above a try and then I try to get it into the >> Opensmalltalk VM. >> > > Hi Holger, > > I just reviewed the PR and it looks really good. Please, allow me some > time to get updated to the VM compiling instructions and I will give it a > try. Probably, I will be testing it by doing my original (unused now) > #name:attachToCFile:writable: use the new primitive and running the > OSSubprocess tests. Does this make sense to you? > > Thanks a LOT for going deep and fix it! > > > > > -- > Mariano > http://marianopeck.wordpress.com > -- Mariano http://marianopeck.wordpress.com On Thu, Sep 22, 2016 at 12:35 PM, Mariano Martinez Peck < marianopeck(a)gmail.com> wrote: > > > On Thu, Sep 22, 2016 at 12:12 PM, Holger Freyther <holger(a)freyther.de> > wrote: > >> >> > On 21 Sep 2016, at 15:09, Mariano Martinez Peck <marianopeck(a)gmail.com> >> wrote: >> > >> > >> > >> > Exactly. I have been wanting this a couple of times while doing >> OSSubprocess. >> >> >> https://github.com/pharo-project/pharo-vm/pull/108. Would be nice if you >> could review it and give it a try. It adds two primitive (one to work on fd >> one to work on FILE). >> >> I probably also want to do: >> >> sqFile->isStdioStream = isatty(fileno(file)) >> >> >> > Yes, exactly. I remember now. And as I said, I also wanted to be able >> to work at fd or FILE* level and I failed. >> >> Could you give the above a try and then I try to get it into the >> Opensmalltalk VM. >> > > Hi Holger, > > I just reviewed the PR and it looks really good. Please, allow me some > time to get updated to the VM compiling instructions and I will give it a > try. Probably, I will be testing it by doing my original (unused now) > #name:attachToCFile:writable: use the new primitive and running the > OSSubprocess tests. Does this make sense to you? > > Thanks a LOT for going deep and fix it! > > > > > -- > Mariano > http://marianopeck.wordpress.com > -- Mariano http://marianopeck.wordpress.com

Sept. 22, 2016

strange cache behaviour for DSAdaptiveModel magritte description

by Yanni Chiu

Does any one have any insight into the caching of magritte descriptions for DSAdaptiveModel? What I see is that there seem to be multiple copies of the description elements. A problem appears when a MACachedMemento seems to have a different description instance (a different hash value anyway), but the rest of the values seem to be the same as the other copy.

Sept. 22, 2016

Re: [Pharo-users] Spec: Class pop list code completion

by Johan Fabry

Hi Brad, AFAIK there is nothing prebuilt, you will have to roll your own. -- Does this mail seem too brief? Sorry for that, I donât mean to be rude! Please see http://emailcharter.org . Johan Fabry - http://pleiad.cl/~jfabry PLEIAD and RyCh labs - Computer Science Department (DCC) - University of Chile > On Sep 22, 2016, at 16:04, Brad Selfridge <bsselfridge(a)gmail.com> wrote: > > Is there a way in Spec to pop a list widget that contains class names when > entering a partial class name in a test field, (like GTPlayground does)? > > p.s. I do not want to go down the Glamour route. > > > > ----- > Brad Selfridge > -- > View this message in context: http://forum.world.st/Spec-Class-pop-list-code-completion-tp4916672.html > Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com. > >

Sept. 22, 2016

Spec: Class pop list code completion

by Brad Selfridge

Is there a way in Spec to pop a list widget that contains class names when entering a partial class name in a test field, (like GTPlayground does)? p.s. I do not want to go down the Glamour route. ----- Brad Selfridge -- View this message in context: http://forum.world.st/Spec-Class-pop-list-code-completion-tp4916672.html Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com.

Sept. 22, 2016

Re: [Pharo-users] Sharing Seaside URLs between users (security) + what is actual way for session expiration page

by Mariano Martinez Peck

OK, "TooMuchPressureExpcetion signal". I will do it. There is another open-source-task related to OSSubprocess that I must address first. Then will come to this one. Probably I can also put the code in github. Will let you know when ready. Cheers, On Thu, Sep 22, 2016 at 2:53 PM, stepharo <stepharo(a)free.fr> wrote: > Mariano > > > It would be cool. > > Why don't you paste it on your blog? > > I hope to get back to the seaside book one of these days and I would like > to add such tips and tricks > > > Stef > > Le 21/9/16 Ã 18:00, Mariano Martinez Peck a Ã©crit : > > > > On Wed, Sep 21, 2016 at 11:50 AM, Johan Brichau <johan(a)inceptive.be> > wrote: > >> >> > On 21 Sep 2016, at 12:31, Petr Fischer <petr.fischer(a)me.com> wrote: >> > >> > Hello, two questions about Seaside sessions: >> > >> > 1) URL sharing between different users - what if "boss" shares URL from >> his browser and send it to another regular user - of course, easy way, >> whole URL with session (_s=xxxx) - when another/regular user opens that >> link -> whole "boss" session opens in regular user's browser, with all >> "boss" permissions, UI state etc etc - very bad, is there any solution for >> this? Rewrite every (!) URL with updateURL: is not solution :( >> >> If this is a concern, you can use a cookie for session tracking, but that >> means you cannot have multiple Seaside sessions running in the same browser >> at the same time. >> >> There are probably other ways, but I think the solution is not to rely on >> a session key for authentication. >> Hereâs a strategy: >> Keep the Seaside session key in the url for session tracking but use an >> authorization cookie for authorization. Put that cookie when the user logs >> in and check its presence when requests come in for a session. >> I think that using a filter for that is a good choice. >> >> Whenever another user copy/pastes the url, he cannot âhijackâ the session >> because he lacks the correct authentication cookie. >> >> > That's exactly what I did in my case. And the way to implement that was > with a custom session tracker that dealt with the cookie plus a filter for > the checking and kickout. > > I can share this if someone wants it (I think I already shared it before) > > > >> > 2) What is the actual way for "session expiration/login page"? There is >> few tutorials and books on the inet - but info about session expiration is >> obsolete :( Methods from tutorials not exists in Seaside 3.2.0. >> > Some trick with WAApplication subclass is actual? >> >> Iâm not sure what the question is. Do you want to redirect users to a >> page whenever the session is expired? >> >> cheers >> Johan >> > > > > -- > Mariano > http://marianopeck.wordpress.com > > > -- Mariano http://marianopeck.wordpress.com

Sept. 22, 2016

Re: [Pharo-users] Sharing Seaside URLs between users (security) + what is actual way for session expiration page

by stepharo

Mariano It would be cool. Why don't you paste it on your blog? I hope to get back to the seaside book one of these days and I would like to add such tips and tricks Stef Le 21/9/16 Ã 18:00, Mariano Martinez Peck a Ã©crit : > > > On Wed, Sep 21, 2016 at 11:50 AM, Johan Brichau <johan(a)inceptive.be > <mailto:johan@inceptive.be>> wrote: > > > > On 21 Sep 2016, at 12:31, Petr Fischer <petr.fischer(a)me.com > <mailto:petr.fischer@me.com>> wrote: > > > > Hello, two questions about Seaside sessions: > > > > 1) URL sharing between different users - what if "boss" shares > URL from his browser and send it to another regular user - of > course, easy way, whole URL with session (_s=xxxx) - when > another/regular user opens that link -> whole "boss" session opens > in regular user's browser, with all "boss" permissions, UI state > etc etc - very bad, is there any solution for this? Rewrite every > (!) URL with updateURL: is not solution :( > > If this is a concern, you can use a cookie for session tracking, > but that means you cannot have multiple Seaside sessions running > in the same browser at the same time. > > There are probably other ways, but I think the solution is not to > rely on a session key for authentication. > Hereâs a strategy: > Keep the Seaside session key in the url for session tracking but > use an authorization cookie for authorization. Put that cookie > when the user logs in and check its presence when requests come in > for a session. > I think that using a filter for that is a good choice. > > Whenever another user copy/pastes the url, he cannot âhijackâ the > session because he lacks the correct authentication cookie. > > > That's exactly what I did in my case. And the way to implement that > was with a custom session tracker that dealt with the cookie plus a > filter for the checking and kickout. > > I can share this if someone wants it (I think I already shared it before) > > > 2) What is the actual way for "session expiration/login page"? > There is few tutorials and books on the inet - but info about > session expiration is obsolete :( Methods from tutorials not > exists in Seaside 3.2.0. > > Some trick with WAApplication subclass is actual? > > Iâm not sure what the question is. Do you want to redirect users > to a page whenever the session is expired? > > cheers > Johan > > > > > -- > Mariano > http://marianopeck.wordpress.com

Sept. 22, 2016

Re: [Pharo-users] Pharo 3.0 Issue with instance variables

by stepharo

Hi Pablo This is great to know that Cordoba is teaching OOP in Pharo. Let us know how we can help you. Why don't you move to Pharo 50 because we fixed a lot of issues in 3 years. And the mooc http://mooc.pharo.org (which has been evaluated to be worth 180 000 Euros - yes three salaries for some months and professional videos processing and translation) is based on Pharo 50. So you get 30 live videos for free. Not counting all the lectures. Stef Le 22/9/16 Ã 16:58, Pablo Frias a Ã©crit : > Hi all, > > We are teaching Smalltalk at the Universtity of Cordoba, Argentina > with Pharo, but we are having the following issue. > Sometimes y in different workstations, we notice that when > instantiating an object and setting values using the accesors methods, > the values are added in the object, but in different instance variables. > > For example: > > |a| > > a := MyPersonObject new. > a age: 35. > a name: 'Person name'. > > But when using the object, we notice that 35 has been saved to name > and the string to age. > > Is that a known issue? Unfortunately we haven't tested Pharo 5.0 yet, > so I cannot tell you if it is already fixed. > > Our current image is: Pharo 3.0 on Windows. > > Thanks in advance, > Pablo

Sept. 22, 2016