Re: [Zorba-coders] [Merge] lp:~zorba-coders/zorba/fix-1046415 into lp:zorba

2012-09-13 Thread Ghislain Fourny
Review: Approve -- https://code.launchpad.net/~zorba-coders/zorba/fix-1046415/+merge/123283 Your team Zorba Coders is subscribed to branch lp:zorba. -- Mailing list: https://launchpad.net/~zorba-coders Post to : zorba-coders@lists.launchpad.net Unsubscribe :

Re: [Zorba-coders] [Merge] lp:~zorba-coders/zorba/fix-1046415 into lp:zorba

2012-09-13 Thread Ghislain Fourny
Hi Cezar, In general, an object constructor expression is not always valid JSON, as it is more general. For example, { 1 : 1 } is valid JSONiq although it is not valid JSON. A key expression, can be anything, and will be atomized and cast to xs:string. Also, { null : 1 } (which is what is

Re: [Zorba-coders] [Merge] lp:~zorba-coders/zorba/fix-1046415 into lp:zorba

2012-09-13 Thread Cezar Andrei
I understand, it make sense in this way but it's very dangerous, in this way I don't think it helps the users on contrary can leave bad code falsely working. Cezar On Thu, Sep 13, 2012 at 6:38 AM, Ghislain Fourny ghislain.fou...@28msec.com wrote: Hi Cezar, In general, an object constructor

Re: [Zorba-coders] [Merge] lp:~zorba-coders/zorba/fix-1046415 into lp:zorba

2012-09-13 Thread Zorba Build Bot
The attempt to merge lp:~zorba-coders/zorba/fix-1046415 into lp:zorba failed. Below is the output from the failed tests. CMake Error at /home/ceej/zo/testing/zorbatest/tester/TarmacLander.cmake:274 (message): Validation queue job fix-1046415-2012-09-13T14-49-49.139Z is finished. The final

Re: [Zorba-coders] [Merge] lp:~zorba-coders/zorba/fix-1046415 into lp:zorba

2012-09-13 Thread Zorba Build Bot
Voting does not meet specified criteria. Required: Approve 1, Disapprove 1, Needs Fixing 1, Pending 1. Got: 2 Approve, 1 Pending. -- https://code.launchpad.net/~zorba-coders/zorba/fix-1046415/+merge/123283 Your team Zorba Coders is subscribed to branch lp:zorba. -- Mailing list:

Re: [Zorba-coders] [Merge] lp:~zorba-coders/zorba/fix-1046415 into lp:zorba

2012-09-13 Thread Chris Hillery
Review: Approve -- https://code.launchpad.net/~zorba-coders/zorba/fix-1046415/+merge/123283 Your team Zorba Coders is subscribed to branch lp:zorba. -- Mailing list: https://launchpad.net/~zorba-coders Post to : zorba-coders@lists.launchpad.net Unsubscribe :

Re: [Zorba-coders] [Merge] lp:~zorba-coders/zorba/fix-1046415 into lp:zorba

2012-09-13 Thread Zorba Build Bot
The attempt to merge lp:~zorba-coders/zorba/fix-1046415 into lp:zorba failed. Below is the output from the failed tests. CMake Error at /home/ceej/zo/testing/zorbatest/tester/TarmacLander.cmake:274 (message): Validation queue job fix-1046415-2012-09-13T20-14-47.795Z is finished. The final

Re: [Zorba-coders] [Merge] lp:~zorba-coders/zorba/fix-1046415 into lp:zorba

2012-09-13 Thread Zorba Build Bot
The attempt to merge lp:~zorba-coders/zorba/fix-1046415 into lp:zorba failed. Below is the output from the failed tests. CMake Error at /home/ceej/zo/testing/zorbatest/tester/TarmacLander.cmake:274 (message): Validation queue job fix-1046415-2012-09-13T20-54-55.036Z is finished. The final

Re: [Zorba-coders] [Merge] lp:~zorba-coders/zorba/fix-1046415 into lp:zorba

2012-09-12 Thread Ghislain Fourny
Hi Matthias, - I think the empty string is a valid JSON name. - The current semantics of { null : 1 }(null) in the spec is indeed to return 1. I think I am fine with this. - I uploaded more tests (somehow they did not get uploaded yesterday). --

Re: [Zorba-coders] [Merge] lp:~zorba-coders/zorba/fix-1046415 into lp:zorba

2012-09-12 Thread Matthias Brantner
Review: Approve -- https://code.launchpad.net/~zorba-coders/zorba/fix-1046415/+merge/123283 Your team Zorba Coders is subscribed to branch lp:zorba. -- Mailing list: https://launchpad.net/~zorba-coders Post to : zorba-coders@lists.launchpad.net Unsubscribe :

Re: [Zorba-coders] [Merge] lp:~zorba-coders/zorba/fix-1046415 into lp:zorba

2012-09-11 Thread Matthias Brantner
Review: Needs Information - Is the empty string a valid name? Should we raise an error if the name of a pair is an empty string? Currently, the following query works fine: { : 1}() - What should happen if the key expression evaluates to null. Currently, the following query returns 1; { null :

Re: [Zorba-coders] [Merge] lp:~zorba-coders/zorba/fix-1046415 into lp:zorba

2012-09-11 Thread Chris Hillery
As far as I can make out, an empty string is a valid key for an object. jsonlint.com at least agrees. null is not a valid key, and we should raise an error if such a case arises. Also, the fact that using the *string* null as an accessor worked in that case frightens me. --

Re: [Zorba-coders] [Merge] lp:~zorba-coders/zorba/fix-1046415 into lp:zorba

2012-09-11 Thread Matthias Brantner
As far as I can make out, an empty string is a valid key for an object. jsonlint.com at least agrees. Fair enough. null is not a valid key, and we should raise an error if such a case arises. Also, the fact that using the *string* null as an accessor worked in that case frightens me.

Re: [Zorba-coders] [Merge] lp:~zorba-coders/zorba/fix-1046415 into lp:zorba

2012-09-07 Thread Matthias Brantner
Review: Needs Fixing It works great for the usual cases, e.g. if the name is an integer. However, the following query raises a strange error: { { foo : bar} : foo } [jerr:JNTY0004]: object items do not have typed value. This might be hard to understand by users. --

Re: [Zorba-coders] [Merge] lp:~zorba-coders/zorba/fix-1046415 into lp:zorba

2012-09-07 Thread Zorba Build Bot
Voting does not meet specified criteria. Required: Approve 1, Disapprove 1, Needs Fixing 1, Pending 1. Got: 1 Needs Fixing, 1 Pending. -- https://code.launchpad.net/~zorba-coders/zorba/fix-1046415/+merge/123283 Your team Zorba Coders is subscribed to branch lp:zorba. -- Mailing list: