Extensions and Common Helpers für CattleDB
The timeseries classes provide a ordered list of timestamps with a timestamp offset and a float value.
insert Insert is always sorted. O(1) for appending or prepending items. O(log n) for inserts.
access By index access is O(1). The more common access by timestamp is O(log n).
Initialize one of the main classes.
from cdb_ext import TSList, FastTSList, PyTSList
# use fast c++ version
ts = FastTSList("key", "metric")
# use python version
ts = PyTSList("key", "metric")
# use best available
ts = TSList("key", "metric")Add items using Python syntax. Timezone offsets can be given as tuple or using pendulum datetimes.
import pendulum
import random
from cdb_ext import FastTSList
ts = FastTSList("my_device", "temperature")
start = pendulum.now("Europe/Vienna")
for i in range(24):
ts[start.add(hours=i)] = random.random() * 30.0
# access by ts (other timezone)
print(ts[start])
# access by index
print(ts.at_index(2))
# trim to 3 items
ts.trim_index(2, 4)
# remove item
del ts[start.add(hours=3)]
# iterate raw
for t in ts:
print(t)
# iterate iso datetime tuples
for t in ts.iter_iso():
print(t)
# iterate pendulum datetime tuples
for t in ts.iter_datetime():
print(t)To build the extension cmake is needed.
python setup.py develop
pytest tests
# creating windows distribution
python setup.py sdist bdist_wheel
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent