Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

dv job altered to add optional src-prq read-1 #9285

Merged
merged 5 commits into from
Jan 10, 2025
+110 −38
Merged

dv job altered to add optional src-prq read-1 #9285

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
4923d2b
dv job altered to add optional src-prq read-1
madhu-k-sr2 Jan 9, 2025
2dc4b4d
dv job altered to add optional src-prq read-2
madhu-k-sr2 Jan 9, 2025
5b22c25
Merge branch 'main' of https://github.com/ministryofjustice/modernisa…
madhu-k-sr2 Jan 9, 2025
e99ab88
dv job altered to add optional src-prq read-3
madhu-k-sr2 Jan 9, 2025
80de5ca
dv job altered to add optional src-prq read-4
madhu-k-sr2 Jan 10, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
53 changes: 27 additions & 26 deletions terraform/environments/electronic-monitoring-data/dms_data_validation_glue_job_v2.tf
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -564,7 +564,7 @@ resource "aws_glue_job" "etl_rds_tbl_rows_hashvalue_to_s3_prq_yyyy_mm" {
execution_property {
max_concurrent_runs = 12
}

default_arguments = {
"--script_bucket_name" = module.s3-glue-job-script-bucket.bucket.id
"--rds_db_host_ep" = split(":", aws_db_instance.database_2022.endpoint)[0]
Expand Down Expand Up @@ -815,31 +815,32 @@ resource "aws_glue_job" "dms_dv_on_rows_hashvalue_partitionby_yyyy_mm" {
worker_type = "G.2X"
number_of_workers = 4
default_arguments = {
"--script_bucket_name" = module.s3-glue-job-script-bucket.bucket.id
"--rds_db_host_ep" = split(":", aws_db_instance.database_2022.endpoint)[0]
"--rds_db_pwd" = aws_db_instance.database_2022.password
"--rds_database_folder" = ""
"--rds_db_schema_folder" = "dbo"
"--rds_table_orignal_name" = ""
"--table_pkey_column" = ""
"--date_partition_column_name" = ""
"--rds_hashed_rows_prq_parent_dir" = "rds_tables_rows_hashed"
"--dms_prq_output_bucket" = module.s3-dms-target-store-bucket.bucket.id
"--dms_prq_table_folder" = ""
"--rds_only_where_clause" = ""
"--prq_df_where_clause" = ""
"--skip_columns_for_hashing" = ""
"--rds_hashed_rows_prq_bucket" = module.s3-dms-data-validation-bucket.bucket.id
"--glue_catalog_dv_bucket" = module.s3-dms-data-validation-bucket.bucket.id
"--glue_catalog_db_name" = aws_glue_catalog_database.dms_dv_glue_catalog_db.name
"--glue_catalog_tbl_name" = "glue_df_output"
"--extra-py-files" = "s3://${module.s3-glue-job-script-bucket.bucket.id}/${aws_s3_object.aws_s3_object_pyzipfile_to_s3folder.id}"
"--continuous-log-logGroup" = "/aws-glue/jobs/${aws_cloudwatch_log_group.dms_dv_on_rows_hashvalue_partitionby_yyyy_mm.name}"
"--enable-continuous-cloudwatch-log" = "true"
"--enable-continuous-log-filter" = "true"
"--enable-metrics" = "true"
"--enable-auto-scaling" = "true"
"--conf" = <<EOF
"--script_bucket_name" = module.s3-glue-job-script-bucket.bucket.id
"--rds_db_host_ep" = split(":", aws_db_instance.database_2022.endpoint)[0]
"--rds_db_pwd" = aws_db_instance.database_2022.password
"--rds_database_folder" = ""
"--rds_db_schema_folder" = "dbo"
"--rds_table_orignal_name" = ""
"--table_pkey_column" = ""
"--date_partition_column_name" = ""
"--rds_hashed_rows_prq_parent_dir" = "rds_tables_rows_hashed"
"--dms_prq_output_bucket" = module.s3-dms-target-store-bucket.bucket.id
"--dms_prq_table_folder" = ""
"--rds_only_where_clause" = ""
"--prq_df_where_clause" = ""
"--skip_columns_for_hashing" = ""
"--read_rds_tbl_agg_stats_from_parquet" = "false"
"--rds_hashed_rows_prq_bucket" = module.s3-dms-data-validation-bucket.bucket.id
"--glue_catalog_dv_bucket" = module.s3-dms-data-validation-bucket.bucket.id
"--glue_catalog_db_name" = aws_glue_catalog_database.dms_dv_glue_catalog_db.name
"--glue_catalog_tbl_name" = "glue_df_output"
"--extra-py-files" = "s3://${module.s3-glue-job-script-bucket.bucket.id}/${aws_s3_object.aws_s3_object_pyzipfile_to_s3folder.id}"
"--continuous-log-logGroup" = "/aws-glue/jobs/${aws_cloudwatch_log_group.dms_dv_on_rows_hashvalue_partitionby_yyyy_mm.name}"
"--enable-continuous-cloudwatch-log" = "true"
"--enable-continuous-log-filter" = "true"
"--enable-metrics" = "true"
"--enable-auto-scaling" = "true"
"--conf" = <<EOF
spark.sql.legacy.parquet.datetimeRebaseModeInRead=CORRECTED
--conf spark.sql.sources.partitionOverwriteMode=dynamic
--conf spark.sql.parquet.aggregatePushdown=true
Expand Down
57 changes: 45 additions & 12 deletions ...ments/electronic-monitoring-data/glue-job/dms_dv_on_rows_hashvalue_partitionby_yyyy_mm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -63,7 +63,8 @@
OPTIONAL_INPUTS = [
"rds_only_where_clause",
"prq_df_where_clause",
"skip_columns_for_hashing"
"skip_columns_for_hashing",
"read_rds_tbl_agg_stats_from_parquet"
]

AVAILABLE_ARGS_LIST = CustomPysparkMethods.resolve_args(DEFAULT_INPUTS_LIST+OPTIONAL_INPUTS)
Expand Down Expand Up @@ -216,12 +217,47 @@ def write_parquet_to_s3(df_dv_output: DataFrame, database, db_sch_tbl_name):
LOGGER.info(f"""TABLE_PKEY_COLUMN = {TABLE_PKEY_COLUMN}""")
LOGGER.info(f"""DATE_PARTITION_COLUMN_NAME = {DATE_PARTITION_COLUMN_NAME}""")

group_by_cols_list = ['year', 'month']
prq_df_where_clause = args.get("prq_df_where_clause", None)

# EVALUATE RDS-DATAFRAME ROW-COUNT
rds_table_row_stats_df_agg = rds_jdbc_conn_obj.get_min_max_count_groupby_yyyy_mm(
rds_table_orignal_name,
DATE_PARTITION_COLUMN_NAME,
TABLE_PKEY_COLUMN,
args.get("rds_only_where_clause", None))
read_rds_tbl_agg_stats_from_parquet = args.get("read_rds_tbl_agg_stats_from_parquet", None)

if read_rds_tbl_agg_stats_from_parquet == 'true':
rds_table_row_stats_df_agg = CustomPysparkMethods.get_s3_parquet_df_v2(
f"""s3://{rds_hashed_rows_bucket_parent_dir}/rds_table_row_stats_df_agg""",
CustomPysparkMethods.get_year_month_min_max_count_schema(TABLE_PKEY_COLUMN)
)

if prq_df_where_clause is not None:
rds_table_row_stats_df_agg = rds_table_row_stats_df_agg.where(f"{prq_df_where_clause}")
# -----------------------------------------------------------------------------------------
else:
rds_table_row_stats_df_agg = rds_jdbc_conn_obj.get_min_max_count_groupby_yyyy_mm(
rds_table_orignal_name,
DATE_PARTITION_COLUMN_NAME,
TABLE_PKEY_COLUMN,
args.get("rds_only_where_clause", None))

if S3Methods.check_s3_folder_path_if_exists(RDS_HASHED_ROWS_PRQ_BUCKET,
f"{rds_hashed_rows_bucket_parent_dir}/rds_table_row_stats_df_agg"):
prq_rds_table_row_stats_df_agg = CustomPysparkMethods.get_s3_parquet_df_v2(
f"""s3://{rds_hashed_rows_bucket_parent_dir}/rds_table_row_stats_df_agg""",
rds_table_row_stats_df_agg.schema
)
prq_rds_table_row_stats_df_agg_updated = CustomPysparkMethods.update_df1_with_df2(prq_rds_table_row_stats_df_agg,
rds_table_row_stats_df_agg,
group_by_cols_list,
[e.name
for e in rds_table_row_stats_df_agg.schema.fields
if e.name not in group_by_cols_list
]
)
prq_rds_table_row_stats_df_agg_updated.write.mode("overwrite").parquet(f"""s3://{rds_hashed_rows_bucket_parent_dir}/rds_table_row_stats_df_agg""")
else:
rds_table_row_stats_df_agg.write.mode("overwrite").parquet(f"""s3://{rds_hashed_rows_bucket_parent_dir}/rds_table_row_stats_df_agg""")
# --------------------------------------------------------------------
# --------------------------------------------------------------------
# +----+-----+-----------------+-----------------+-------------------+
# |year|month|min_GPSPositionID|max_GPSPositionID|count_GPSPositionID|
# +----+-----+-----------------+-----------------+-------------------+
Expand Down Expand Up @@ -265,21 +301,18 @@ def write_parquet_to_s3(df_dv_output: DataFrame, database, db_sch_tbl_name):
)


group_by_cols_list = ['year', 'month']
prq_df_where_clause = args.get("prq_df_where_clause", None)


if skipped_struct_fields_list:
rds_hashed_rows_prq_df = CustomPysparkMethods.get_s3_parquet_df_v2(
rds_hashed_rows_fulls3path,
CustomPysparkMethods.get_pyspark_hashed_table_schema(
TABLE_PKEY_COLUMN, skipped_struct_fields_list)
TABLE_PKEY_COLUMN,
skipped_struct_fields_list)
)
else:
rds_hashed_rows_prq_df = CustomPysparkMethods.get_s3_parquet_df_v2(
rds_hashed_rows_fulls3path,
CustomPysparkMethods.get_pyspark_hashed_table_schema(
TABLE_PKEY_COLUMN)
TABLE_PKEY_COLUMN)
)

if prq_df_where_clause is not None:
Expand Down
38 changes: 38 additions & 0 deletions ...ironments/electronic-monitoring-data/glue-job/reusable_module/glue_data_validation_lib.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -838,3 +838,41 @@ def get_pyspark_hashed_table_schema(in_pkey_column, sf_list=None):
schema = schema.add(T.StructField("RowHash", T.StringType(), False))

return schema

@staticmethod
def get_year_month_min_max_count_schema(in_pkey_column_str):

agg_schema = T.StructType([
T.StructField("year", T.IntegerType(), False),
T.StructField("month", T.IntegerType(), False),
T.StructField(f"min_{in_pkey_column_str}", T.LongType(), False),
T.StructField(f"max_{in_pkey_column_str}", T.LongType(), False),
T.StructField(f"count_{in_pkey_column_str}", T.LongType(), False)]
)
return agg_schema

@staticmethod
def update_df1_with_df2(df1: DataFrame, df2: DataFrame,
join_columns_list,
all_remaining_columns_list):
# Step 1: Find unmatched rows from df1 / df_parquet
df_unmatched_rows = df1.join(df2, join_columns_list, "left_anti")

# Step 2: Update matched rows between df2 / df_JDBC AND df1 / df_parquet
update_select_columns = [df2[c].alias(c) for c in all_remaining_columns_list]
df_updated_rows = df1.join(df2, join_columns_list, "inner") \
.select(
df1['year'],
df1['month'],
*update_select_columns
)

# Step 3: Include new rows from df2 / df_JDBC not in df1 / df_parquet
df_new_rows = df2.join(df1, join_columns_list, "left_anti")

# Step 4: Combine all type of rows in dataframes
final_df = df_unmatched_rows.union(df_updated_rows).union(df_new_rows)

final_df = final_df.orderBy(join_columns_list)

return final_df
Loading