[ANN]
Locked
12
12
 
|
- Issue 1144: Many occurrences of "ReadStream on: aCollection" which
can be replaced with "aCollection readStream" _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
the problem that not all classes define #readStream..
as far as i remember, i had some issues with it, and forced to use "ReadStream on:" instead 2009/9/3 Stéphane Ducasse <[hidden email]>: > - Issue 1144: Many occurrences of "ReadStream on: aCollection" which > can be replaced with "aCollection readStream" > > > > > _______________________________________________ > Pharo-project mailing list > [hidden email] > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > -- Best regards, Igor Stasenko AKA sig. _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
Sig,
One of the great things about Smalltalk is that we are seldom forced to do anything, and certainly not in this case. It is easy to define #readStream where appropriate, and IMHO Pharo should do so. Not only does it give cleaner code, it will help us insert Nile when the time comes, and we can simply add the missing methods. I submit that the problem is instead all of the external code that will still explicitly reference ReadStream. That is not a reason to not make the change, and I am hoping that a little leadership by example will cause the right thing to happen across the board. More challenging still are the inconsistencies in exhaustion behavior. Referring to my proposal on the wiki, I stand ready to help if we want to enforce EndOfStreamError on any "unauthorized" exhaustion, but I have my doubts that we will do so; Cincom considered and declined the proposal. I can also not afford to wait for the decision to be made and imlemented. For now, I am gradually moving my code toward #nextOne, #nextMany:, etc., which again can be done simply by adding new code. If Pharo adopts EndOfStreamError, I can easily change from my new protocol to the more generic messages. FWIW, one of the first things I did was define #readStream on various collections and collection classes. Dolphin (sorry to keep beating the drum, but it sets a wonderful example!) has had #readStream "everywhere" for a long time if not from its earliest days. Bill -----Original Message----- From: [hidden email] [mailto:[hidden email]] On Behalf Of Igor Stasenko Sent: Thursday, September 03, 2009 10:42 AM To: [hidden email] Subject: Re: [Pharo-project] [ANN] the problem that not all classes define #readStream.. as far as i remember, i had some issues with it, and forced to use "ReadStream on:" instead 2009/9/3 Stéphane Ducasse <[hidden email]>: > - Issue 1144: Many occurrences of "ReadStream on: aCollection" which > can be replaced with "aCollection readStream" > > > > > _______________________________________________ > Pharo-project mailing list > [hidden email] > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > -- Best regards, Igor Stasenko AKA sig. _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
In reply to this post by Stéphane Ducasse
On Fri, Sep 4, 2009 at 3:21 AM, Stéphane Ducasse <[hidden email]> wrote: Gulik.- Issue 1144: Many occurrences of "ReadStream on: aCollection" which For what it's worth, I prefer the style of "ReadStream on: xxx". I prefer to choose the stream class rather than let the collection choose it for me. -- http://gulik.pbwiki.com/ _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
Then #readStreamXYZ would do the job. Depending
on how you write the methods, both you and the collection have a say, or you can
also have complete control.
From: [hidden email] [mailto:[hidden email]] On Behalf Of Michael van der Gulik Sent: Thursday, September 03, 2009 4:33 PM To: [hidden email] Subject: Re: [Pharo-project] [ANN] On Fri, Sep 4, 2009 at 3:21 AM, Stéphane Ducasse <[hidden email]>
wrote: Gulik.- Issue 1144: Many occurrences of "ReadStream on: aCollection" which For what it's worth, I prefer the style of "ReadStream on: xxx". I prefer to choose the stream class rather than let the collection choose it for me. -- http://gulik.pbwiki.com/ _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
In reply to this post by Michael van der Gulik-2
El vie, 04-09-2009 a las 09:33 +1200, Michael van der Gulik escribió:
> > > On Fri, Sep 4, 2009 at 3:21 AM, Stéphane Ducasse > <[hidden email]> wrote: > - Issue 1144: Many occurrences of "ReadStream on: > aCollection" which > can be replaced with "aCollection readStream" > > > For what it's worth, I prefer the style of "ReadStream on: xxx". I > prefer to choose the stream class rather than let the collection > choose it for me. But don't you always have the option to call ReadStream on: object, even though object class has a readStream method? > > > Gulik. > > -- > http://gulik.pbwiki.com/ > _______________________________________________ > Pharo-project mailing list > [hidden email] > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project Miguel Cobá http://miguel.leugim.com.mx _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
Sure.
-----Original Message----- From: [hidden email] [mailto:[hidden email]] On Behalf Of Miguel Enrique Cobá Martinez Sent: Thursday, September 03, 2009 4:56 PM To: [hidden email] Subject: Re: [Pharo-project] [ANN] El vie, 04-09-2009 a las 09:33 +1200, Michael van der Gulik escribió: > > > On Fri, Sep 4, 2009 at 3:21 AM, Stéphane Ducasse > <[hidden email]> wrote: > - Issue 1144: Many occurrences of "ReadStream on: > aCollection" which > can be replaced with "aCollection readStream" > > > For what it's worth, I prefer the style of "ReadStream on: xxx". I > prefer to choose the stream class rather than let the collection > choose it for me. But don't you always have the option to call ReadStream on: object, even though object class has a readStream method? > > > Gulik. > > -- > http://gulik.pbwiki.com/ > _______________________________________________ > Pharo-project mailing list > [hidden email] > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project Miguel Cobá http://miguel.leugim.com.mx _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
In reply to this post by Miguel Cobá
2009/9/4 Miguel Enrique Cobá Martinez <[hidden email]> El vie, 04-09-2009 a las 09:33 +1200, Michael van der Gulik escribió: In Pharo, sure. However, in SecureSqueak, there won't be >>readStream methods. I'll spare the details because I hate multi-page emails. Ask me if you're interested. Gulik. -- http://gulik.pbwiki.com/ _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
Ok, I'm interested :) Is there a
security argument? Off the top, it does not sound like something I would
want to use. If you can make a case that #readStream is dangerous, I'm all
ears. If it is simply that it closes off modifications to the base
classes, it is probably too restictive for my tastes. Locking it down
**after** I add the missing methods and deploy on a server, that's another
matter. But then, there would be #readStream
methods...
Bill
From: [hidden email] [mailto:[hidden email]] On Behalf Of Michael van der Gulik Sent: Thursday, September 03, 2009 5:54 PM To: [hidden email] Subject: Re: [Pharo-project] [ANN] 2009/9/4 Miguel Enrique Cobá Martinez <[hidden email]> El vie, 04-09-2009 a las 09:33 +1200, Michael van der Gulik escribió: In Pharo, sure. However, in SecureSqueak, there won't be >>readStream methods. I'll spare the details because I hate multi-page emails. Ask me if you're interested. Gulik. -- http://gulik.pbwiki.com/ _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
In reply to this post by Schwab,Wilhelm K
2009/9/3 Schwab,Wilhelm K <[hidden email]>:
> Sig, > > One of the great things about Smalltalk is that we are seldom forced to do anything, and certainly not in this case. It is easy to define #readStream where appropriate, and IMHO Pharo should do so. Not only does it give cleaner code, it will help us insert Nile when the time comes, and we can simply add the missing methods. > I understanding this, but at that moment i weren't sure if this is a missing method, or it is missing by intent. Also i had to make sure that my code works also for 3.10. So, for me it was an easiest thing to leave things as they are. > I submit that the problem is instead all of the external code that will still explicitly reference ReadStream. That is not a reason to not make the change, and I am hoping that a little leadership by example will cause the right thing to happen across the board. > Right: i'm trying to not use globals if its possible, except well known classes like Arrays, collections etc. > More challenging still are the inconsistencies in exhaustion behavior. Referring to my proposal on the wiki, I stand ready to help if we want to enforce EndOfStreamError on any "unauthorized" exhaustion, but I have my doubts that we will do so; Cincom considered and declined the proposal. I can also not afford to wait for the decision to be made and imlemented. For now, I am gradually moving my code toward #nextOne, #nextMany:, etc., which again can be done simply by adding new code. If Pharo adopts EndOfStreamError, I can easily change from my new protocol to the more generic messages. i missed this part of discussion, so i can't comment the above. > > FWIW, one of the first things I did was define #readStream on various collections and collection classes. Dolphin (sorry to keep beating the drum, but it sets a wonderful example!) has had #readStream "everywhere" for a long time if not from its earliest days. > The arguments like 'it is done like that somewhere' is a bad way to convince me (and i assume many others). I think you are already given enough arguments before and they already good enough to convince everyone (i hope), except me, because you don't need to convince me on that :) > Bill -- Best regards, Igor Stasenko AKA sig. _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
Sig,
I doubt you missed anything; I dragged EndOfStreamError into it because IMHO, that's where the real problems wait for us. If I had to make code work on 3.10, I would fill in missing pieces (in this case #readStream) there rather than hold back my Pharo code; in fact, that's more or less what I am doing with Pharo too. Bill -----Original Message----- From: [hidden email] [mailto:[hidden email]] On Behalf Of Igor Stasenko Sent: Thursday, September 03, 2009 6:42 PM To: [hidden email] Subject: Re: [Pharo-project] [ANN] 2009/9/3 Schwab,Wilhelm K <[hidden email]>: > Sig, > > One of the great things about Smalltalk is that we are seldom forced to do anything, and certainly not in this case. It is easy to define #readStream where appropriate, and IMHO Pharo should do so. Not only does it give cleaner code, it will help us insert Nile when the time comes, and we can simply add the missing methods. > I understanding this, but at that moment i weren't sure if this is a missing method, or it is missing by intent. Also i had to make sure that my code works also for 3.10. So, for me it was an easiest thing to leave things as they are. > I submit that the problem is instead all of the external code that will still explicitly reference ReadStream. That is not a reason to not make the change, and I am hoping that a little leadership by example will cause the right thing to happen across the board. > Right: i'm trying to not use globals if its possible, except well known classes like Arrays, collections etc. > More challenging still are the inconsistencies in exhaustion behavior. Referring to my proposal on the wiki, I stand ready to help if we want to enforce EndOfStreamError on any "unauthorized" exhaustion, but I have my doubts that we will do so; Cincom considered and declined the proposal. I can also not afford to wait for the decision to be made and imlemented. For now, I am gradually moving my code toward #nextOne, #nextMany:, etc., which again can be done simply by adding new code. If Pharo adopts EndOfStreamError, I can easily change from my new protocol to the more generic messages. i missed this part of discussion, so i can't comment the above. > > FWIW, one of the first things I did was define #readStream on various collections and collection classes. Dolphin (sorry to keep beating the drum, but it sets a wonderful example!) has had #readStream "everywhere" for a long time if not from its earliest days. > The arguments like 'it is done like that somewhere' is a bad way to convince me (and i assume many others). I think you are already given enough arguments before and they already good enough to convince everyone (i hope), except me, because you don't need to convince me on that :) > Bill -- Best regards, Igor Stasenko AKA sig. _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
In reply to this post by Schwab,Wilhelm K
Argh. Please reply below other people's emails.
No, it's not really to do with security. It's to do with this package system I invented. Basically, I want to reduce package dependencies. Putting >>readStream in kernel classes would make a package dependency between the Collection package and the Stream package. If the Collections package was loaded into SecureSqueak, the package architecture would be forced to load the Stream package as well even if it isn't used. If your own package depended on both Collection and a different Streams package, anybody who loaded your code would need to load two Streams packages. There's no reason that >>readStream couldn't be included in a Collections package, provided that it's not in the Collections package provided with the SecureSqueak Kernel. I want the Kernel package to have as few dependencies as possible. Packages are read-only in SecureSqueak and the Kernel package contains special objects, meaning that Kernel and its dependencies can not be upgraded without distributing a new image. Collections would be a non-upgradable dependency of Kernel, so I'd be trying hard to not also make Streams also a non-upgradable dependency. More info: http://gulik.pbwiki.com/Namespaces http://gulik.pbwiki.com/Packages Gulik. -- http://gulik.pbwiki.com/ _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
2009/9/4 Michael van der Gulik <[hidden email]>
Argh. Please reply below other people's emails. Michael, This is one of those cases that really makes me miss ENVY, and its Class extentions. Under that system you would avoid the problem completely, by extending the Collection class in the Stream package and adding the methods that depend on Streams in that package. I guess everyone knows what ENVY Class extensions are, but just in case someone doesn't, they are basically collections of methods to be added to a Class already in the image (i.e. not subclasses), when the extension is loaded. -- Cheers, Peter _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
2009/9/4 Peter Hugosson-Miller <[hidden email]>
Assuming you're referring to me. There're lots of Michaels around.
Unless there's more magic here than I understand, Monticello also does this. My package system can't do that. I can't work out a way to modify existing classes that won't cause complete chaos. SecureSqueak is designed to run completely untrusted code loaded from the Internet; allowing loaded code to override methods in other packages allows for too much bad stuff to happen. If you can think of a way to do that securely, I'm all ears. Gulik. -- http://gulik.pbwiki.com/ _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
2009/9/4 Michael van der Gulik <[hidden email]>
Yes, my answer was meant for you.
I'm *really* new to PharO and Squeak, Monticello and lots of other things, so I don't feel qualified to say one way or the other.
It isn't a question of overriding specifically, but of adding methods to classes (which may then of course override methods in their superclasses). But since I don't know enough about SecureSqueak either, I don't really understand what your packaging system is intending to do either. But now I'm afraid I've really dragged the discussion off-topic, for which I apologise :-)
-- Cheers, Peter _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
In reply to this post by Michael van der Gulik-2
2009/9/3 Michael van der Gulik <[hidden email]>:
> Argh. Please reply below other people's emails. > > >> Bill >> >> ________________________________ >> From: [hidden email] >> [mailto:[hidden email]] On Behalf Of Michael >> van der Gulik >> Sent: Thursday, September 03, 2009 5:54 PM >> To: [hidden email] >> Subject: Re: [Pharo-project] [ANN] >> >> >> >> 2009/9/4 Miguel Enrique Cobá Martinez <[hidden email]> >>> >>> El vie, 04-09-2009 a las 09:33 +1200, Michael van der Gulik escribió: >>> > >>> > >>> > On Fri, Sep 4, 2009 at 3:21 AM, Stéphane Ducasse >>> > <[hidden email]> wrote: >>> > - Issue 1144: Many occurrences of "ReadStream on: >>> > aCollection" which >>> > can be replaced with "aCollection readStream" >>> > >>> > >>> > For what it's worth, I prefer the style of "ReadStream on: xxx". I >>> > prefer to choose the stream class rather than let the collection >>> > choose it for me. >>> >>> But don't you always have the option to call ReadStream on: object, >>> even though object class has a readStream method? >> >> >> In Pharo, sure. >> >> However, in SecureSqueak, there won't be >>readStream methods. I'll spare >> the details because I hate multi-page emails. Ask me if you're interested. > > >> >> 2009/9/4 Schwab,Wilhelm K <[hidden email]> >> Ok, I'm interested :) Is there a security argument? Off the top, it does >> not sound like something I would want to use. If you can make a case that >> #readStream is dangerous, I'm all ears. If it is simply that it closes off >> modifications to the base classes, it is probably too restictive for my >> tastes. Locking it down **after** I add the missing methods and deploy on >> a server, that's another matter. But then, there would be #readStream >> methods... > > > Excuse the funny formatting. GMail sometimes doesn't cooperate. Also, this > is getting a bit off topic on the Pharo list. > > No, it's not really to do with security. It's to do with this package system > I invented. > > Basically, I want to reduce package dependencies. Putting >>readStream in > kernel classes would make a package dependency between the Collection > package and the Stream package. If the Collections package was loaded into > SecureSqueak, the package architecture would be forced to load the Stream > package as well even if it isn't used. If your own package depended on both > Collection and a different Streams package, anybody who loaded your code > would need to load two Streams packages. > > There's no reason that >>readStream couldn't be included in a Collections > package, provided that it's not in the Collections package provided with the > SecureSqueak Kernel. I want the Kernel package to have as few dependencies > as possible. Packages are read-only in SecureSqueak and the Kernel package > contains special objects, meaning that Kernel and its dependencies can not > be upgraded without distributing a new image. Collections would be a > non-upgradable dependency of Kernel, so I'd be trying hard to not also make > Streams also a non-upgradable dependency. > Interesting dilemma. Right thing, that the package who responsible for implementing #readStream should be the one, who defines any streams of sort. So, then collection could delegate to specific class the responsibility for returning a stream. Following the same logic, an Object class should not contain things like #isCollection, if we imagine things w/o collections, and then load a Collections package, it should provide this method by extending existing behavior. But the problem is, that you need to extend an existing behavior, and once you'll allow that, you are not in secure shell anymore. I think there is solution. By example: The Collection class , in Collection class could define a class var, like ReadStreamDelegate and then a simple method which allows to assign object to it. And then, #readStream could be implemented as: readStream ^ ReadStreamDelegate on: self just don't say that classvars is read-only as well :) You still have to make something non-readonly. So, if classvar is readonly, then i could use one of a slots of an object held in classvar to keep value there :) > More info: > http://gulik.pbwiki.com/Namespaces > http://gulik.pbwiki.com/Packages > > Gulik. > > -- > http://gulik.pbwiki.com/ > > _______________________________________________ > Pharo-project mailing list > [hidden email] > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > -- Best regards, Igor Stasenko AKA sig. _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
In reply to this post by Michael van der Gulik-2
I'd love to - how do you get LookOut! 2003 to comment
text? It amazes me that people who write software for a living don't grok
the need for that ability, but I've been saddled with this piece of junk, and
the piece of junk before it, that appear to be unable to do
it.
From: [hidden email] [mailto:[hidden email]] On Behalf Of Michael van der Gulik Sent: Thursday, September 03, 2009 6:51 PM To: [hidden email] Subject: Re: [Pharo-project] [ANN]
No, it's not really to do with security. It's to do with this package system I invented. Basically, I want to reduce package dependencies. Putting >>readStream in kernel classes would make a package dependency between the Collection package and the Stream package. If the Collections package was loaded into SecureSqueak, the package architecture would be forced to load the Stream package as well even if it isn't used. If your own package depended on both Collection and a different Streams package, anybody who loaded your code would need to load two Streams packages. There's no reason that >>readStream couldn't be included in a Collections package, provided that it's not in the Collections package provided with the SecureSqueak Kernel. I want the Kernel package to have as few dependencies as possible. Packages are read-only in SecureSqueak and the Kernel package contains special objects, meaning that Kernel and its dependencies can not be upgraded without distributing a new image. Collections would be a non-upgradable dependency of Kernel, so I'd be trying hard to not also make Streams also a non-upgradable dependency. More info: http://gulik.pbwiki.com/Namespaces http://gulik.pbwiki.com/Packages Gulik. -- http://gulik.pbwiki.com/ _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
In reply to this post by Peter Hugosson-Miller
Ok.
Please note that your package system and limits it imposes might affect "market
share".
Bill
From: [hidden email] [mailto:[hidden email]] On Behalf Of Peter Hugosson-Miller Sent: Thursday, September 03, 2009 7:00 PM To: [hidden email] Subject: Re: [Pharo-project] [ANN] Argh. Please reply below other people's emails. Michael, This is one of those cases that really makes me miss ENVY, and its Class extentions. Under that system you would avoid the problem completely, by extending the Collection class in the Stream package and adding the methods that depend on Streams in that package. I guess everyone knows what ENVY Class extensions are, but just in case someone doesn't, they are basically collections of methods to be added to a Class already in the image (i.e. not subclasses), when the extension is loaded. -- Cheers, Peter _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
In reply to this post by Peter Hugosson-Miller
Peter you should have a look at Squeak by example and soon pharo by
example in short if you had a category with *MyPackage in a class of another package then your package MyPackage will contain the extensions. You can see the extension by clicking on the Monticello browser and clicking browse. In fact this is a good practice to check the extensions of a package before publishing it. I can show you during the break :) Stef _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
|
|
I also prefer ReadStream on: since it is explicit about the stream you
want rather than expecting the collection to know and have the protocol. I don't really see the need to rewrite the whole image. Cheers mike On Friday, September 4, 2009, Stéphane Ducasse <[hidden email]> wrote: > Peter you should have a look at Squeak by example and soon pharo by > example > > in short if you had a category with *MyPackage in a class of another > package then > your package MyPackage will contain the extensions. > > You can see the extension by clicking on the Monticello browser and > clicking browse. > > In fact this is a good practice to check the extensions of a package > before publishing it. > > I can show you during the break :) > > Stef > > _______________________________________________ > Pharo-project mailing list > [hidden email] > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
«
Return to Pharo Smalltalk Developers
|
1 view|%1 views
| Free forum by Nabble | Edit this page |