collective/mediawiki_dump_tools
16
0
Fork 0
Code Issues Pull Requests Activity

bugfix. column name text_chars

Browse Source
This commit is contained in:
Nathan TeBlunthuis 2022-03-08 20:17:20 -08:00
parent 1aea601a30
commit 3e645b5e58
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
wikiq
View File

@ -254,7 +254,7 @@ class RevDataBase():
pa.field("title",pa.string()), pa.field("title",pa.string()),
pa.field("namespace",pa.int32()), pa.field("namespace",pa.int32()),
pa.field("deleted",pa.bool_()), pa.field("deleted",pa.bool_()),
pa.field("test_chars",pa.int32()), pa.field("text_chars",pa.int32()),
pa.field("revert",pa.bool_()), pa.field("revert",pa.bool_()),
pa.field("reverteds",pa.list_(pa.int64())), pa.field("reverteds",pa.list_(pa.int64())),
pa.field("sha1",pa.string()), pa.field("sha1",pa.string()),
@ -661,7 +661,7 @@ class WikiqParser():
outtable = rows_to_table(self.parquet_buffer, self.schema) outtable = rows_to_table(self.parquet_buffer, self.schema)
if self.pq_writer is None: if self.pq_writer is None:
self.pq_writer = pq.ParquetWriter(self.output_file, schema, flavor='spark') self.pq_writer = pq.ParquetWriter(self.output_file, self.schema, flavor='spark')
self.pq_writer.write_table(outtable) self.pq_writer.write_table(outtable)
self.parquet_buffer = [] self.parquet_buffer = []