Invalid Content within JSON files

[lazaridis_com]

This affects Examples, Tests & Documentation.

Many *.json files within the dojo sources do not contain valid JSON data.

This files should be change to have the names (keys) within " "

{ id: 'CN', name:'China', type:'country' }

becomes:

{ "id": "CN", "name": "China", "type": "country" }

Alternatively, those files could be renamed to e.g. *.jsd (Java Script Data) or something similar, in order to distinguish them clearly from JSON data files.

Related Discussion:

​http://dojo-toolkit.33424.n3.nabble.com/Dojo-usage-of-non-standard-JSON-Format-td2878366.html

[bill]

We'd be happy to take a ​patch to make those files proper JSON format, although you'd first need to file a ​http://dojofoundation.org/about/cla.

[lazaridis_com]

Patching the nearly 500 files is inefficient and critical. A core developer could provide an utility function, which loads the js and then saves it back as json. Or with a regex, to keep formatting.

And then, each developer can convert his files and see if something breaks.

(As for me: I'm aware about the contribution mechanisms, but I have limited time and resources, and the time has not yet come for me to touch the repo, even not indirectly via patches.)

[bill]

OK, I don't think any core developers are interested in doing that. I'm going to close this ticket as wontfix but if someone wants to work on it they can reopen it.

Probably for 2.0 we'll change dojo.xhr() to do strict parsing at which point we'll update the files which need updating.

[bill]

In [27562]:

fix invalid json in dijit, refs #12872 !strict

[lazaridis_com]

Shouldn't this be reopened?