Reputation: 7210
I'm working with an AWS SQS queue where python is putting strings into the queue using boto. Most of the strings are unicode, but some are in pythons basic string type.
I'm trying to read these messages with GO and running into JSON decode errors:
JSON Decode Error: 2 invalid character 'e' looking for beginning of value
All these decode errors happen when GO encounters a messages that is not a unicode string but pythons basic string format.
Is there a way of converting python strings into unicode strings from GO? ascii -> unicode?
EDIT:
Here's an example of a working and non working json string. The only way I'm able to get the non working one is through python, GO gives the decode error above
Working:
u'{"queue_time": "1374523279747", "object_id": "5efc90c0912fef247f028f1758082a299c018e8a2c6499289f3426217b14f0ae", "source_bucket": "ap1-cache"}',
Doesn't work:
'{"queue_time": "1374523279.75026", "source_bucket": "eu1-cache", "object_id": "42e8b7b01ed5919a23d6d91fdc0b540944a5f45e4c387fa1c2bc20e1bf81bdeb"}',
Upvotes: 2
Views: 2539
Reputation: 940
To produce proper JSON from Python, you can use the json
library:
>>> d = {"queue_time": "1374523279747", "object_id": "...", "source_bucket": "ap1-cache"}
>>> d
{'queue_time': '1374523279747', 'source_bucket': 'ap1-cache', 'object_id': '...'}
>>> import json
>>> print json.dumps(d)
{"queue_time": "1374523279747", "source_bucket": "ap1-cache", "object_id": "..."}
Note that the JSON output looks almost entirely similar for this simple example, but that's not necessarily true for more complicated things. E.g. tuples and unicode strings have different representations:
>>> t = [u"Hello", ("World", "!"), u"\xa0"]
>>> t
[u'Hello', ('World', '!'), u'\xa0']
>>> print json.dumps(t)
["Hello", ["World", "!"], "\u00a0"]
Upvotes: 2