mirror of
https://github.com/hanzo-ds/driver.git
synced 2026-10-30 20:32:40 +00:00
zeekay
eb74c996a6
Imports read `datastore_driver`, distribution `datastore-driver`, so nothing has to import another product's name to talk to Hanzo Datastore. Cut from upstream 0.2.9 — the version this stack already ran — rather than from upstream HEAD, so it is a PURE RENAME with no behaviour change. Forking from HEAD would have dragged versions along with the name (connect alone jumped 0.8.17 -> 1.6.0, a major), which is a different and much riskier change wearing a rename's clothes. X-ClickHouse-* headers are untouched: they are wire protocol the server sends and expects, and a different token shape from the module name.
29 lines
803 B
Python
29 lines
803 B
Python
import numpy as np
|
|
import pandas as pd
|
|
from pandas.core.arrays import ExtensionArray
|
|
|
|
|
|
def column_chunks(columns, n):
|
|
for column in columns:
|
|
if not isinstance(
|
|
column, (np.ndarray, pd.DatetimeIndex, ExtensionArray)
|
|
):
|
|
raise TypeError(
|
|
'Unsupported column type: {}. '
|
|
'ndarray/DatetimeIndex/ExtensionArray is expected.'
|
|
.format(type(column))
|
|
)
|
|
|
|
# create chunk generator for every column
|
|
chunked = [
|
|
iter(np.array_split(c, len(c) // n) if len(c) > n else [c])
|
|
for c in columns
|
|
]
|
|
|
|
while True:
|
|
# get next chunk for every column
|
|
item = [next(column, []) for column in chunked]
|
|
if not any(len(x) for x in item):
|
|
break
|
|
yield item
|