weboobcfg: Format capabilities as JSON array

Capabilities was always displayed as a single list of entries
separated by comma. This is not really efficient as the reader has to
parse this string to retrieve atomic informations, even in JSON.

This patch allows to display the capabilities as an array, allowing
standard JSON parsers to immediatly returns a valuable information.

weboob/applications/weboobcfg/weboobcfg.py

@@ -31,6 +31,19 @@ from weboob.tools.application.formatters.iformatter import IFormatter
 __all__ = ['WeboobCfg']
 
 
+class CapabilitiesWrapper(list):
+    """
+    A wrapper class to keep the list nature of capabilities,
+    but provide a comma separated list representation for
+    formaters unable to display a list by themselves.
+
+    Useful for having an array representation in JSON and
+    comma separated list for simple format.
+    """
+    def __repr__(self):
+        return ', '.join(self)
+
+
 class ModuleInfoFormatter(IFormatter):
     def format_dict(self, minfo):
         result = '.------------------------------------------------------------------------------.\n'
@@ -221,7 +234,7 @@ class WeboobCfg(ReplApplication):
         caps = line.split()
         for name, info in sorted(self.weboob.repositories.get_all_modules_info(caps).iteritems()):
             row = OrderedDict([('Name', name),
-                               ('Capabilities', ', '.join(info.capabilities)),
+                               ('Capabilities', CapabilitiesWrapper(info.capabilities)),
                                ('Description', info.description),
                                ])
             self.format(row)