influxdb/models
Edd Robinson 6b350edf59 Ensure correct number of tags parsed
This commit fixes a parsing bug that was causing extra tags to be
generated when the incoming point contained escaped commas.

As an optimisation, the slice of tags associated with a point was
being pre-allocated using the number of commas in the series key as a hint to
the appropriate size. The hinting did not consider literal comma values in
the key though, and so it was possible for extra (empty) tag key and
value pairs to be part of the tags structure associated with a parsed
point.
2018-03-14 13:28:55 +00:00
..
consistency.go Update godoc for package models 2016-12-30 18:02:52 -08:00
inline_fnv.go Update godoc for package models 2016-12-30 18:02:52 -08:00
inline_fnv_test.go Implement and use zero-alloc FNV64a. 2016-09-26 11:43:27 -07:00
inline_strconv_parse.go Taught tsm1 storage engine how to read and write uint64 values 2017-07-24 09:03:22 -07:00
inline_strconv_parse_test.go models: Add FieldIterator type 2016-10-03 16:30:21 -06:00
points.go Ensure correct number of tags parsed 2018-03-14 13:28:55 +00:00
points_internal_test.go Add test for marshalling a point without fields 2016-12-28 12:09:30 -08:00
points_test.go Ensure correct number of tags parsed 2018-03-14 13:28:55 +00:00
rows.go Update godoc for package models 2016-12-30 18:02:52 -08:00
statistic.go Update godoc for package models 2016-12-30 18:02:52 -08:00
statistic_test.go Ensure we don't mutate provided statistics tags 2016-08-24 11:40:13 +01:00
time.go Support negative timestamps in the query engine 2016-08-25 12:52:41 -05:00
uint_support.go Add uint support into the write protocol 2017-09-19 10:44:26 -05:00