Quantcast

Amusing (kinda) bug in SqueakMap release editor

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Amusing (kinda) bug in SqueakMap release editor

timrowledge
If one submits a new (or edited) release in the SqueakMap release editor tool that happens to have the word ‘error’ in the notes field then the save ‘fails’ even if it succeeds.

Why? Because the test used in SMReleaseBrowser>>savePackageRelease: carefu;ly checks the response content string for ‘error’ being included. The response appears to include the entire entry sent to the SM server and thus… success is failure. I think it needs more knowledge of what web servers do than I have to solve this. Presumably there is some tighter parsing needed?

tim
--
tim Rowledge; [hidden email]; http://www.rowledge.org/tim
Strange OpCodes: IML: Invoke Murphy's Laws



Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Amusing (kinda) bug in SqueakMap release editor

Chris Muller-3
Hi Tim,

Yes, maybe some tighter parsing there, but the SMServer needs to be
rewritten in any case.  We can't even only _run_ it on a modern cog
image because some of the methods are longer than Cog can accept.  It
uses Göran's HttpView2 framework, which is is a neat approach to a web
server, but ultimately not what we want for the rich SM domain model.

Squeak really needs some more alternatives besides Seaside and Aida
for hosting domain models in the web browser..

On Thu, Feb 9, 2017 at 3:42 PM, tim Rowledge <[hidden email]> wrote:

> If one submits a new (or edited) release in the SqueakMap release editor tool that happens to have the word ‘error’ in the notes field then the save ‘fails’ even if it succeeds.
>
> Why? Because the test used in SMReleaseBrowser>>savePackageRelease: carefu;ly checks the response content string for ‘error’ being included. The response appears to include the entire entry sent to the SM server and thus… success is failure. I think it needs more knowledge of what web servers do than I have to solve this. Presumably there is some tighter parsing needed?
>
> tim
> --
> tim Rowledge; [hidden email]; http://www.rowledge.org/tim
> Strange OpCodes: IML: Invoke Murphy's Laws
>
>
>

Loading...