sws-spark-dissemination-helper 0.0.166__py3-none-any.whl → 0.0.167__py3-none-any.whl

sws_spark_dissemination_helper/SWSGoldIcebergSparkHelper.py

@@ -296,6 +296,35 @@ class SWSGoldIcebergSparkHelper:
 
         return df
 
+    def write_gold_faostat_unfiltered_data_to_iceberg_and_csv(
+        self, df: DataFrame
+    ) -> DataFrame:
+        """The expected input to this function is the output of the sws disseminated function"""
+        df.writeTo(
+            self.iceberg_tables.GOLD_FAOSTAT_UNFILTERED.iceberg_id
+        ).createOrReplace()
+
+        logging.info(
+            f"Gold FAOSTAT unfiltered table written to {self.iceberg_tables.GOLD_FAOSTAT.iceberg_id}"
+        )
+
+        self.spark.sql(
+            f"ALTER TABLE {self.iceberg_tables.GOLD_FAOSTAT_UNFILTERED.iceberg_id} CREATE OR REPLACE TAG `{self.tag_name}`"
+        )
+
+        logging.info(f"gold FAOSTAT unfiltered tag '{self.tag_name}' created")
+
+        df_1 = df.coalesce(1)
+
+        save_cache_csv(
+            df=df_1,
+            bucket=self.bucket,
+            prefix=self.iceberg_tables.GOLD_FAOSTAT_UNFILTERED.csv_prefix,
+            tag_name=self.tag_name,
+        )
+
+        return df
+
     def write_gold_sws_validated_sws_dissemination_tag(
         self, df: DataFrame, tags: Tags
     ) -> DataFrame:
@@ -589,3 +618,52 @@ class SWSGoldIcebergSparkHelper:
         logging.debug(f"Tag with Added csv Table: {tag}")
 
         return df
+
+    def write_gold_faostat_unfiltered_dissemination_tag(
+        self, df: DataFrame, tags: Tags
+    ) -> DataFrame:
+        # Get or create a new tag
+        tag = get_or_create_tag(tags, self.dataset_id, self.tag_name, self.tag_name)
+        logging.debug(f"Tag: {tag}")
+
+        new_iceberg_table = BaseDisseminatedTagTable(
+            id=f"{self.domain_code.lower()}_gold_faostat_unfiltered_iceberg",
+            name=f"{self.domain_code} gold FAOSTAT unfiltered Iceberg",
+            description="Gold table containing all the tag data in FAOSTAT format",
+            layer=TableLayer.GOLD,
+            private=True,
+            type=TableType.ICEBERG,
+            database=IcebergDatabases.GOLD_DATABASE,
+            table=self.iceberg_tables.GOLD_FAOSTAT_UNFILTERED.table,
+            path=self.iceberg_tables.GOLD_FAOSTAT_UNFILTERED.path,
+            structure={"columns": df.schema.jsonValue()["fields"]},
+        )
+        tag = upsert_disseminated_table(
+            sws_tags=tags,
+            tag=tag,
+            dataset_id=self.dataset_id,
+            tag_name=self.tag_name,
+            table=new_iceberg_table,
+        )
+        logging.debug(f"Tag with Added Iceberg Table: {tag}")
+
+        new_diss_table = BaseDisseminatedTagTable(
+            id=f"{self.domain_code.lower()}_gold_faostat_unfiltered_csv",
+            name=f"{self.domain_code} gold FAOSTAT unfiltered csv",
+            description="Gold table containing the tag data in FAOSTAT format in csv",
+            layer=TableLayer.GOLD,
+            private=True,
+            type=TableType.CSV,
+            path=self.iceberg_tables.GOLD_FAOSTAT_UNFILTERED.csv_path,
+            structure={"columns": df.schema.jsonValue()["fields"]},
+        )
+        tag = upsert_disseminated_table(
+            sws_tags=tags,
+            tag=tag,
+            dataset_id=self.dataset_id,
+            tag_name=self.tag_name,
+            table=new_diss_table,
+        )
+        logging.debug(f"Tag with Added csv Table: {tag}")
+
+        return df

{sws_spark_dissemination_helper-0.0.166.dist-info → sws_spark_dissemination_helper-0.0.167.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sws-spark-dissemination-helper
-Version: 0.0.166
+Version: 0.0.167
 Summary: A Python helper package providing streamlined Spark functions for efficient data dissemination processes
 Project-URL: Repository, https://github.com/un-fao/fao-sws-it-python-spark-dissemination-helper
 Author-email: Daniele Mansillo <danielemansillo@gmail.com>

{sws_spark_dissemination_helper-0.0.166.dist-info → sws_spark_dissemination_helper-0.0.167.dist-info}/RECORD

@@ -1,13 +1,13 @@
 sws_spark_dissemination_helper/SWSBronzeIcebergSparkHelper.py,sha256=ocuau0WtpyRwui0qwdQ_Rxh4nYPOyZoHpGKaWRa6B3Q,28868
 sws_spark_dissemination_helper/SWSDatatablesExportHelper.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sws_spark_dissemination_helper/SWSEasyIcebergSparkHelper.py,sha256=csqKyYglBkJSBvEkEa1_keHarZZAIJHaV0d64gGJy98,26379
-sws_spark_dissemination_helper/SWSGoldIcebergSparkHelper.py,sha256=USEIiPZuPcKxzJdesz-N3PsBSEPpkGk8aavDNUPmb80,22092
+sws_spark_dissemination_helper/SWSGoldIcebergSparkHelper.py,sha256=0dxbVkrhdaASapEffF5PFcgKwAMyJoWBxzgymjZ4JyY,25049
 sws_spark_dissemination_helper/SWSPostgresSparkReader.py,sha256=KpG8gp8Ai9pHDiKhUOTcXWxxmFGeKEE3XKlI_Y-SveU,18453
 sws_spark_dissemination_helper/SWSSilverIcebergSparkHelper.py,sha256=qioLv3SlJEfk0LzTiwfXRtZXVImPOJUeh9k1XwHC-pA,26225
 sws_spark_dissemination_helper/__init__.py,sha256=42TPbk7KxAud_qY3Sr_F4F7VjyofUlxEJkUXAFQsjRo,327
 sws_spark_dissemination_helper/constants.py,sha256=zviO6huxWTWonHv4v2M8zKr7HXCDMBGqjHx-eTfGT2A,13487
 sws_spark_dissemination_helper/utils.py,sha256=G7lQqNRrvqZpgm9WmddD7fWsI8IVn09x1p3cV3458EA,21963
-sws_spark_dissemination_helper-0.0.166.dist-info/METADATA,sha256=tseBASNGGEO46T902JN85SquzCKuVrlILT4W5iPU9Ws,2824
-sws_spark_dissemination_helper-0.0.166.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-sws_spark_dissemination_helper-0.0.166.dist-info/licenses/LICENSE,sha256=zFzeb_j_6pXEHwH8Z0OpIkKFJk7vmhZjdem-K0d4zU4,1073
-sws_spark_dissemination_helper-0.0.166.dist-info/RECORD,,
+sws_spark_dissemination_helper-0.0.167.dist-info/METADATA,sha256=h27GpuoB4elORRQybwqiKoKkF59JOPk6HzkQ4uDsSjo,2824
+sws_spark_dissemination_helper-0.0.167.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+sws_spark_dissemination_helper-0.0.167.dist-info/licenses/LICENSE,sha256=zFzeb_j_6pXEHwH8Z0OpIkKFJk7vmhZjdem-K0d4zU4,1073
+sws_spark_dissemination_helper-0.0.167.dist-info/RECORD,,