예제 #1
0
    def _get_schema_metadata_for_datasource(
        self, datasource_fields: List[dict]
    ) -> Optional[SchemaMetadata]:
        fields = []
        schema_metadata = None
        for field in datasource_fields:
            # check datasource - custom sql relations from a field being referenced
            self._track_custom_sql_ids(field)

            nativeDataType = field.get("dataType", "UNKNOWN")
            TypeClass = FIELD_TYPE_MAPPING.get(nativeDataType, NullTypeClass)

            schema_field = SchemaField(
                fieldPath=field["name"],
                type=SchemaFieldDataType(type=TypeClass()),
                description=make_description_from_params(
                    field.get("description", ""), field.get("formula")
                ),
                nativeDataType=nativeDataType,
                globalTags=get_tags_from_params(
                    [
                        field.get("role", ""),
                        field.get("__typename", ""),
                        field.get("aggregation", ""),
                    ]
                )
                if self.config.ingest_tags
                else None,
            )
            fields.append(schema_field)

        if fields:
            schema_metadata = SchemaMetadata(
                schemaName="test",
                platform=f"urn:li:dataPlatform:{self.platform}",
                version=0,
                fields=fields,
                hash="",
                platformSchema=OtherSchema(rawSchema=""),
            )

        return schema_metadata
예제 #2
0
    def emit_sheets_as_charts(self,
                              workbook: Dict) -> Iterable[MetadataWorkUnit]:
        sheet_upstream_datasources = self.get_sheetwise_upstream_datasources(
            workbook)
        for sheet in workbook.get("sheets", []):
            chart_snapshot = ChartSnapshot(
                urn=builder.make_chart_urn(self.platform, sheet.get("id")),
                aspects=[],
            )

            creator = workbook.get("owner", {}).get("username", "")
            created_at = sheet.get("createdAt", datetime.now())
            updated_at = sheet.get("updatedAt", datetime.now())
            last_modified = self.get_last_modified(creator, created_at,
                                                   updated_at)

            if sheet.get("path"):
                site_part = f"/site/{self.config.site}" if self.config.site else ""
                sheet_external_url = (
                    f"{self.config.connect_uri}/#{site_part}/views/{sheet.get('path')}"
                )
            elif sheet.get("containedInDashboards"):
                # sheet contained in dashboard
                site_part = f"/t/{self.config.site}" if self.config.site else ""
                dashboard_path = sheet.get("containedInDashboards")[0].get(
                    "path", "")
                sheet_external_url = f"{self.config.connect_uri}{site_part}/authoring/{dashboard_path}/{sheet.get('name', '')}"
            else:
                # hidden or viz-in-tooltip sheet
                sheet_external_url = None
            fields = {}
            for field in sheet.get("datasourceFields", ""):
                description = make_description_from_params(
                    get_field_value_in_sheet(field, "description"),
                    get_field_value_in_sheet(field, "formula"),
                )
                fields[get_field_value_in_sheet(field, "name")] = description

            # datasource urn
            datasource_urn = []
            data_sources = sheet_upstream_datasources.get(
                sheet.get("id"), set())

            for ds_id in data_sources:
                if ds_id is None or not ds_id:
                    continue
                ds_urn = builder.make_dataset_urn(self.platform, ds_id,
                                                  self.config.env)
                datasource_urn.append(ds_urn)
                if ds_id not in self.datasource_ids_being_used:
                    self.datasource_ids_being_used.append(ds_id)

            # Chart Info
            chart_info = ChartInfoClass(
                description="",
                title=sheet.get("name", ""),
                lastModified=last_modified,
                externalUrl=sheet_external_url,
                inputs=sorted(datasource_urn),
                customProperties=fields,
            )
            chart_snapshot.aspects.append(chart_info)

            # Browse path
            browse_path = BrowsePathsClass(paths=[
                f"/{self.platform}/{workbook.get('projectName', '').replace('/', REPLACE_SLASH_CHAR)}"
                f"/{workbook.get('name', '')}"
                f"/{sheet.get('name', '').replace('/', REPLACE_SLASH_CHAR)}"
            ])
            chart_snapshot.aspects.append(browse_path)

            # Ownership
            owner = self._get_ownership(creator)
            if owner is not None:
                chart_snapshot.aspects.append(owner)

            #  Tags
            tag_list = sheet.get("tags", [])
            if tag_list and self.config.ingest_tags:
                tag_list_str = [
                    t.get("name", "").upper() for t in tag_list
                    if t is not None
                ]
                chart_snapshot.aspects.append(
                    builder.make_global_tag_aspect_with_tag_list(tag_list_str))

            yield self.get_metadata_change_event(chart_snapshot)

            yield from add_entity_to_container(self.gen_workbook_key(workbook),
                                               "chart", chart_snapshot.urn)