BigQuery：加载日期列数据类型为 long 转换为时间戳的 avro 文件

Question

我无法让 BigQuery 从 avro 文件正确加载时间戳。

avro 文件的日期列存储时间很长，逻辑类型为 timestamp-micros。根据文档，BigQuery 应将其存储为时间戳数据类型。我也尝试过 timestamp-millis 作为逻辑类型。

数据是这样存储在avro中的：

{'id': '<masked>', '<masked>': '<masked>', 'tm': 1553990400000, '<masked>': <masked>, '<masked>': <masked>, 'created': 1597056958864}

字段tm和created是多头，2019-03-31T00:00:00Z 和 2020-08-10T11:50:58.986816592Z。

avro 的架构是

{"type":"record","name":"SomeMessage","namespace":"com.df",
"fields":
[{"name":"id","type":"string"},
{"name":"<masked>","type":"string"},
{"name":"tm","type":"long","logicalType":"timestamp-micros"},
{"name":"<masked>","type":"int"},
{"name":"<masked>","type":"float"},
{"name":"created","type":"long","logicalType":"timestamp-micros"}]}";

当通过 bq load 导入到 BigQuery 时，记录最终如下所示：

<masked>   <masked>   tm                            <masked>   <masked>  created
________________________________________________________________________________________________________
<masked> | <masked> | 1970-01-18 23:39:50.400 UTC | <masked> | <masked> | 1970-01-19 11:37:36.958864 UTC
________________________________________________________________________________________________________

使用的导入命令是：

bq load --source_format=AVRO --use_avro_logical_types some_dataset.some_table "gs://some-bucket/some.avro"

BigQuery 中的时间戳与 avro 中提供的实际值相去甚远。

有人知道如何正确执行此操作吗？

Answer 1

我发现 avro 模式实际上是错误的。时间戳字段应该是这样的：

{"name":"created","type":{"type":"long", "logicalType":"timestamp-millis"}}

BigQuery：加载日期列数据类型为 long 转换为时间戳的 avro 文件

BigQuery: Load avro-files with date column data type as long converted to timestamp

google-bigquery

avro

bq