January 2013 - Pharo-dev - lists.pharo.org

Re: [Pharo-project] [Update] Zinc-SSO now supports Dropbox !

by Sven Van Caekenberghe

On 27 Jan 2013, at 15:55, Denis Kudriashov <dionisiydk(a)gmail.com> wrote: > Cool! > Do you already have FileSystem backend for this? Denis, I had not yet thought of that. It *is* a cool idea though. An ideal student project I would say. > 2013/1/27 Sven Van Caekenberghe <sven(a)stfx.eu> > Hi, > > I am happy to report that Zinc-SSO now supports the Dropbox API. Which means you can write (web) applications that access (read/write) files in a user's Dropbox account. > > Jan extended Zinc-SSO-OAuth1-Core to support both version 1 and version 1a of OAuth1 which I then used for a demo. > > Again, the actual setup is non-trivial, and a public demo is not easy either, so there is only this screenshot: > > <Screen Shot 2013-01-27 at 13.15.36.png> > > This is what you see after allowing the Zinc-SSO app access to a special 'sandbox' folder called Apps/Zinc-SSO (full 'root' Dropbox access is possible as well). Here you can see there is one file, readme.txt in the folder. > > If anyone is interested, I can write another email explaining how you can run this demo for yourself on your own account. > > Sven > > -- > Sven Van Caekenberghe > http://stfx.eu > Smalltalk is the Red Pill > >

Jan. 27, 2013

Re: [Pharo-project] [Update] Zinc-SSO now supports Dropbox !

by Sven Van Caekenberghe

On 27 Jan 2013, at 13:32, Sven Van Caekenberghe <sven(a)stfx.eu> wrote: > If anyone is interested, I can write another email explaining how you can run this demo for yourself on your own account. How to run the Dropbox code yourself: 1. Get yourself a free Dropbox account at http://www.dropbox.com 2. Install some client software on one or more of your computers (Mac, Windows, Linux) or devices (iOS [iPhone, iPad, iPod], Android, BlackBerry, Kindle Fire) and move some files around to get a feel for the service. 3. When you give other applications access to (part of) your Dropbox files, you can check that at https://www.dropbox.com/account#applications 4. Define your own app that will try to access Dropbox files. Go to https://www.dropbox.com/developers (Developer home) and start defining a new app by clicking the 'Create an app' button at https://www.dropbox.com/developers/apps - start with the access type 'App folder'. 5. The two important elements you need are the 'App key' and 'App secret', make sure to copy them correctly. 6. Get the very latest code in your image (we develop in Pharo 2.0, but 1.4 or even 1.3 should work - SSL has to be fully operational). Gofer it url: 'http://mc.stfx.eu/ZincHTTPComponents'; package: 'ConfigurationOfZincHTTPComponents'; load. ConfigurationOfZincHTTPComponents project latestVersion load: 'SSO'. 7. Start a Zn server with an appropriate delegate, configured for your app (ZnServer startDefaultOn: 8100) logToTranscript; debugMode: true; delegate: (ZnSSOServerDelegate new dropboxData: (ZnOAuth1ConsumerData newForDropbox consumer: 'vqz-app-key-zoca'; consumerSecret: '13u-app-secret-8l006'; yourself); yourself). (The above is a demo setup limited to Dropbox, the other demos are not configured and won't work). 8. Go to http:://localhost:8100/sso-dropbox and click the link to log in - you will be transferred to Dropbox to log in (if you aren't already) and asked to authorise your demo app, after which Dropbox will redirect you back to your app. 9. Once back in your app, sso-dropbox-callback (#ssoDropboxCallback:) will access your account info and list the app root folder, which will then be shown. 10. For the rest of the API, (currently not yet implemented), see https://www.dropbox.com/developers/reference/api - you don't have to worry about the whole authentication thanks to Zinc-SSO. We would love for others to try this (as well as the other Zinc-SSO options, Google, Facebook, Twitter) out, in their own projects, and to provide actual feedback and generally help improve the code and documentation. As always, May the Source be with you! Sven PS: The demo app is not (yet) a production app: only you and up to 5 explicitly listed users can use it. You have to apply for production status later on. Also, the whole ZnSSOServerDelegate code is a demo, for an actual project, it would be best to extract the necessary parts out of it, reusing the underlying Zinc-SSO-OAuth1-Core framework. -- Sven Van Caekenberghe http://stfx.eu Smalltalk is the Red Pill

Jan. 27, 2013

Re: [Pharo-project] testInstanceSideMethodsWithNilKeyInLastLiteral is still red

by StÃ©phane Ducasse

Ok excellent I forgot to run the script! Thanks I will do that in the next update. Stef On Jan 27, 2013, at 2:24 PM, wam wrote: > No, it's ok after running the script. > > > On 27 janv. 2013, at 18:17, wam wrote: > >> Did your run the script from issue 7169? It is this one: >> SystemNavigation new instanceSideMethodsWithNilKeyInLastLiteral do: [ :thisShouldNotBeAStringButAMethod | >> | method classAndSelector | >> classAndSelector := ' ' split: thisShouldNotBeAStringButAMethod. >> method := (Smalltalk at: classAndSelector first asSymbol) methodNamed: classAndSelector second asSymbol. >> method recompile. >> ]. >> >> But if I remember it introduces a conflict with Fuel in MetalevelTests. >> I'm checking. >> >> >> On 27 janv. 2013, at 17:19, StÃ©phane Ducasse wrote: >> >>> Camille I integrated your second version of the issue but I still get >>> >>> >>> testInstanceSideMethodsWithNilKeyInLastLiteral >>> >>> self assert: SystemNavigation new instanceSideMethodsWithNilKeyInLastLiteral isEmpty >>> >> >

Jan. 27, 2013

[Pharo-project] TextEmphasis Questions

by Sean P. DeNigris

1. TextEmphasis class>>narrow ^ TextKern kern: -1 All the other instance creation methods return instances of TextEmphasis, but this one returns aTextKern. Why? I want to implement composable TextEmphases, but this is a thorn in my side. Also, it doesn't smell right because TextEmphases all have an emphasis code, but TextKern doesn't use it. So it can't adeuqately be passed as an argument where aTextEmphasis is expected: e.g. given "MyClass>>#myMethod: aTextEmphasis", "aMyClass myMethod: (TextEmphasis narrow)" will not work as I would expect because the emphasisCode will be 0 (i.e. normal, no special emphasis). 2. TextEmphasis>>writeScanOn: strm emphasisCode = 1 ifTrue: [strm nextPut: $b]. emphasisCode = 2 ifTrue: [strm nextPut: $i]. emphasisCode = 0 ifTrue: [strm nextPut: $n]. emphasisCode = 16 ifTrue: [strm nextPut: $=]. emphasisCode = 4 ifTrue: [strm nextPut: $u]. This seems to be used by RunArray... What if there is a compound emphasis code? e.g. 3 (1 | 2) for bold and italic? Cheers, Sean -- View this message in context: http://forum.world.st/TextEmphasis-Questions-tp4665742.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

Jan. 27, 2013

Re: [Pharo-project] The monkey is back in town

by Sean P. DeNigris

Camillo Bruni-3 wrote > The build service got accidentally deleted, so we will have to readd that > thing first :) Do you want me to open an issue? -- View this message in context: http://forum.world.st/The-monkey-is-back-in-town-tp4658091p4665739.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

Jan. 27, 2013

Re: [Pharo-project] testInstanceSideMethodsWithNilKeyInLastLiteral is still red

by wam

No, it's ok after running the script. On 27 janv. 2013, at 18:17, wam wrote: > Did your run the script from issue 7169? It is this one: > SystemNavigation new instanceSideMethodsWithNilKeyInLastLiteral do: [ :thisShouldNotBeAStringButAMethod | > | method classAndSelector | > classAndSelector := ' ' split: thisShouldNotBeAStringButAMethod. > method := (Smalltalk at: classAndSelector first asSymbol) methodNamed: classAndSelector second asSymbol. > method recompile. > ]. > But if I remember it introduces a conflict with Fuel in MetalevelTests. > I'm checking. > > > On 27 janv. 2013, at 17:19, StÃ©phane Ducasse wrote: > >> Camille I integrated your second version of the issue but I still get >> >> >> testInstanceSideMethodsWithNilKeyInLastLiteral >> >> self assert: SystemNavigation new instanceSideMethodsWithNilKeyInLastLiteral isEmpty >> >

Jan. 27, 2013

Re: [Pharo-project] testInstanceSideMethodsWithNilKeyInLastLiteral is still red

by wam

Did your run the script from issue 7169? It is this one: SystemNavigation new instanceSideMethodsWithNilKeyInLastLiteral do: [ :thisShouldNotBeAStringButAMethod | | method classAndSelector | classAndSelector := ' ' split: thisShouldNotBeAStringButAMethod. method := (Smalltalk at: classAndSelector first asSymbol) methodNamed: classAndSelector second asSymbol. method recompile. ]. But if I remember it introduces a conflict with Fuel in MetalevelTests. I'm checking. On 27 janv. 2013, at 17:19, StÃ©phane Ducasse wrote: > Camille I integrated your second version of the issue but I still get > > > testInstanceSideMethodsWithNilKeyInLastLiteral > > self assert: SystemNavigation new instanceSideMethodsWithNilKeyInLastLiteral isEmpty >

Jan. 27, 2013

Re: [Pharo-project] testInstanceSideMethodsWithNilKeyInLastLiteral is still red

by Marcus Denker

On Jan 27, 2013, at 5:20 PM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > Camille I integrated your second version of the issue but I still get > > > testInstanceSideMethodsWithNilKeyInLastLiteral > > self assert: SystemNavigation new instanceSideMethodsWithNilKeyInLastLiteral isEmpty > I think we need to recompile the offending methods once. Marcus

Jan. 27, 2013

Re: [Pharo-project] The monkey is back in town

by Bruni camillo

The build service got accidentally deleted, so we will have to readd that thing first :) On 27 Jan 2013 18:12, "Sean P. DeNigris" <sean(a)clipperadams.com> wrote: > Camillo Bruni-3 wrote > > I had a sudden urge to program something so I revived the monkey (partly) > > ... > > the only thing missing are the couple of changes I proposed for the main > > image: > > Those three issues have all been integrated. Can we unleash the monkey > now?! > > > > -- > View this message in context: > http://forum.world.st/The-monkey-is-back-in-town-tp4658091p4665734.html > Sent from the Pharo Smalltalk mailing list archive at Nabble.com. > >

Jan. 27, 2013

Re: [Pharo-project] The monkey is back in town

by Sean P. DeNigris

Camillo Bruni-3 wrote > I had a sudden urge to program something so I revived the monkey (partly) > ... > the only thing missing are the couple of changes I proposed for the main > image: Those three issues have all been integrated. Can we unleash the monkey now?! -- View this message in context: http://forum.world.st/The-monkey-is-back-in-town-tp4658091p4665734.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

Jan. 27, 2013