)]}'
{"oslo_context/context.py":[{"author":{"_account_id":6486,"name":"Brant Knudson","email":"blk@acm.org","username":"blk-u"},"change_message_id":"0507959db6e30c92068c69b37b4649f4a7602d23","unresolved":false,"context_lines":[{"line_number":75,"context_line":""},{"line_number":76,"context_line":"    def to_dict(self):"},{"line_number":77,"context_line":"        user_idt \u003d ("},{"line_number":78,"context_line":"            self.user_idt_format.format("},{"line_number":79,"context_line":"                user\u003dself._safe_encode(self.user or \u0027-\u0027),"},{"line_number":80,"context_line":"                tenant\u003dself._safe_encode(self.tenant or \u0027-\u0027),"},{"line_number":81,"context_line":"                domain\u003dself._safe_encode(self.domain or \u0027-\u0027),"}],"source_content_type":"text/x-python","patch_set":1,"id":"3a29b11f_88622eac","line":78,"updated":"2015-10-20 20:28:28.000000000","message":"does user_idt_format need to be string? why not unicode?","commit_id":"ccc8c780efc1475fb3083fcd95a536f4ae842d25"},{"author":{"_account_id":1297,"name":"Joshua Harlow","email":"jxharlow@godaddy.com","username":"harlowja"},"change_message_id":"f61fe33e514548a0545a0fc2b0154b0a66af7741","unresolved":false,"context_lines":[{"line_number":70,"context_line":"        _request_store.context \u003d self"},{"line_number":71,"context_line":""},{"line_number":72,"context_line":"    def _safe_encode(self, text):"},{"line_number":73,"context_line":"        return text.encode(encoding\u003d\u0027ascii\u0027, errors\u003d\u0027replace\u0027).decode(\u0027ascii\u0027)"},{"line_number":74,"context_line":""},{"line_number":75,"context_line":"    def to_dict(self):"},{"line_number":76,"context_line":"        user_idt \u003d ("}],"source_content_type":"text/x-python","patch_set":4,"id":"ba8a016a_3be66ee8","line":73,"updated":"2015-11-19 18:36:20.000000000","message":"Hmmm, should we prefer https://github.com/openstack/oslo.utils/blob/master/oslo_utils/encodeutils.py or something like it?? That avoids destroying non-ascii characters that people might actually be using (for example in some indian/asian names or whatever)...\n\n    \u003e\u003e\u003e x \u003d u\"John Gāo\"\n    \u003e\u003e\u003e print x\n    John Gāo\n    \u003e\u003e\u003e x.encode(encoding\u003d\u0027ascii\u0027, errors\u003d\u0027replace\u0027)\n    \u0027John G?o\u0027\n    \u003e\u003e\u003e x.encode(encoding\u003d\u0027ascii\u0027, errors\u003d\u0027replace\u0027).decode(\u0027ascii\u0027)\n    u\u0027John G?o\u0027\n\nSeems like we just altered the user name :(","commit_id":"b22872ecf479033449d96089e10477344f121bd8"}]}