Re: [web2py] Possible bug in JSONRPC in v2.3.2

2012-12-22 Thread Niphlod
both of you, jump on the bandwagon here . https://groups.google.com/d/topic/web2py/5kCYP-wiELo/discussion On Saturday, December 22, 2012 7:04:18 AM UTC+1, rochacbruno wrote: On Sat, Dec 22, 2012 at 3:57 AM, Mike D mike@gmail.com javascript:wrote: s = methods[method](**params)

Re: [web2py] Possible bug in JSONRPC in v2.3.2

2012-12-22 Thread Massimo Di Pierro
Turns out we do not have a choice. params is a variable specified by jsonrpc. It can be an array or an object. The change in trunk was wrong as we need to check whether it is an array or an object (serialized as dict). and act accordingly. It should now be fixed in trunk. Massimo On Saturday,

Re: [web2py] Possible bug in JSONRPC in v2.3.2

2012-12-22 Thread Mike D
Thanks guys. On Saturday, December 22, 2012 6:59:10 AM UTC-8, Massimo Di Pierro wrote: Turns out we do not have a choice. params is a variable specified by jsonrpc. It can be an array or an object. The change in trunk was wrong as we need to check whether it is an array or an object

[web2py] Possible bug in JSONRPC in v2.3.2

2012-12-21 Thread Mike D
Hi, I recently updated to v2.3.2 and noticed that my JSONRPC calls were failing with a message of argument after ** must be a mapping, not str. After investigating some changes in the versions I noticed that in tools.py, in the serve_jsonrpc() function, the following line changed from: s =

Re: [web2py] Possible bug in JSONRPC in v2.3.2

2012-12-21 Thread Bruno Rocha
On Sat, Dec 22, 2012 at 3:57 AM, Mike D mike.dic...@gmail.com wrote: s = methods[method](**params) I guess it should be s = methods[method](*args, **params) to be more flexible --