formatting and use default compression

Author: joocer

mabel/data/readers/internals/cursor.py

@@ -3,7 +3,7 @@
 
 Cursor is made of three parts:
 - map      : a bit array representing all of the blobs in the set - unread blobs
-             are 0s and read blobs are 1s. This allows for blobs to be read in 
+             are 0s and read blobs are 1s. This allows for blobs to be read in
              an arbitrary order - although currently only implemented linearly.
 - partition: the active parition (blob) that is being read
 - location : the record in the active partition (blob), so we can resume reading

mabel/data/readers/internals/threaded_wrapper.py

@@ -1,6 +1,4 @@
-"""
-
-"""
+""" """
 
 import logging
 import threading

mabel/data/validator/__init__.py

@@ -11,7 +11,7 @@
 
 
 def schema_loader(
-    definition: Union[str, List[Dict[str, Any]], dict, RelationSchema, bool]
+    definition: Union[str, List[Dict[str, Any]], dict, RelationSchema, bool],
 ) -> Union[RelationSchema, bool]:
     if definition is None:
         raise ValueError(

mabel/data/writers/internals/blob_writer.py

@@ -165,7 +165,7 @@ def commit(self):
                         pytable = self._normalize_arrow_schema(pytable, self.schema)
 
                     tempfile = io.BytesIO()
-                    pyarrow.parquet.write_table(pytable, where=tempfile, compression="zstd")
+                    pyarrow.parquet.write_table(pytable, where=tempfile)
 
                     tempfile.seek(0)
                     write_buffer = tempfile.read()

mabel/utils/dates.py

@@ -41,7 +41,7 @@ def parse_delta(delta: str) -> datetime.timedelta:
 
 
 def parse_iso(
-    value: Union[str, int, float, datetime.datetime, datetime.date]
+    value: Union[str, int, float, datetime.datetime, datetime.date],
 ) -> Optional[datetime.datetime]:
     """
     Parses an ISO date string into a datetime object, with an emphasis on speed.

mabel/version.py

@@ -1,6 +1,6 @@
 # Store the version here so:
 # 1) we don't load dependencies by storing it in __init__.py
 # 2) we can import it in setup.py for the same reason
-__version__ = "0.6.24"
+__version__ = "0.6.25"
 
 # nodoc - don't add to the documentation wiki

tests/performance/index_performance.py

@@ -1,11 +1,11 @@
 """
 Results (seconds to search for a username in 65,500 rows):
 
- indexed | row exists | time  
+ indexed | row exists | time
 -------------------------------
  yes     |    yes     |   0.094   <- about 3.5x faster when is match
  yes     |    no      |   0.006   <- over 50x faster when no match
- no      |    yes     |   0.357 
+ no      |    yes     |   0.357
  no      |    no      |   0.332
 -------------------------------
 """

tests/performance/indexing.py

@@ -1,6 +1,4 @@
-"""
-
-"""
+""" """
 
 import sys
 import os

tests/performance/json_performance.py

@@ -1,13 +1,13 @@
 """
-JSON parsing and serialization performance tests so a decision on 
+JSON parsing and serialization performance tests so a decision on
 which library(s) to use can be made - previously the selection was
 inconsistent.
 
 Results (seconds to process 10m rows):
 
- library | parsing | serialize  
+ library | parsing | serialize
 -------------------------------
- json    |    36.6 |      1.74 
+ json    |    36.6 |      1.74
  ujson   |    16.5 |      0.86
  orjson  |    10.4 |      0.66   <- lower is better
  simd    |     2.8 |       N/A   <- lower is better

tests/test_utils_common.py

@@ -1,6 +1,4 @@
-"""
-
-"""
+""" """
 
 import os
 import sys