Michael-J-Ward commented on code in PR #666: URL: https://github.com/apache/datafusion-python/pull/666#discussion_r1597650461
########## examples/tpch/q05_local_supplier_volume.py: ########## @@ -0,0 +1,102 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. + +""" +TPC-H Problem Statement Query 5: + +The Local Supplier Volume Query lists for each nation in a region the revenue volume that resulted +from lineitem transactions in which the customer ordering parts and the supplier filling them were +both within that nation. The query is run in order to determine whether to institute local +distribution centers in a given region. The query considers only parts ordered in a given year. The +query displays the nations and revenue volume in descending order by revenue. Revenue volume for all +qualifying lineitems in a particular nation is defined as sum(l_extendedprice * (1 - l_discount)). + +The above problem statement text is copyrighted by the Transaction Processing Performance Council +as part of their TPC Benchmark H Specification revision 2.18.0. +""" + +from datetime import datetime +import pyarrow as pa +from datafusion import SessionContext, col, lit, functions as F + + +DATE_OF_INTEREST = "1994-01-01" +INTERVAL_DAYS = 365 +REGION_OF_INTEREST = "AFRICA" + +date = datetime.strptime(DATE_OF_INTEREST, "%Y-%m-%d").date() + +# Note: this is a hack on setting the values. It should be set differently once +# https://github.com/apache/datafusion-python/issues/665 is resolved. Review Comment: Ditto ########## examples/tpch/README.md: ########## @@ -0,0 +1,57 @@ +<!--- + Licensed to the Apache Software Foundation (ASF) under one + or more contributor license agreements. See the NOTICE file + distributed with this work for additional information + regarding copyright ownership. The ASF licenses this file + to you under the Apache License, Version 2.0 (the + "License"); you may not use this file except in compliance + with the License. You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, + software distributed under the License is distributed on an + "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + KIND, either express or implied. See the License for the + specific language governing permissions and limitations + under the License. +--> + +# DataFusion Python Examples for TPC-H + +These examples reproduce the problems listed in the Transaction Process Council +TPC-H benchmark. The purpose of these examples is to demonstrate how to use +different aspects of Data Fusion and not necessarily geared towards creating the +most performant queries possible. Within each example is a description of the +problem. For users who are familiar with SQL style commands, you can compare the +approaches in these examples with those listed in the specification. + +- https://www.tpc.org/tpch/ + +The examples provided are based on version 2.18.0 of the TPC-H specification. + +## Data Setup + +To run these examples, you must first generate a dataset. The `dbgen` tool +provided by TPC can create datasets of arbitrary scale. For testing it is +typically sufficient to create a 1 gigabyte dataset. For convenience, this +repository has a script which uses docker to create this dataset. From the +`benchmarks/tpch` directory execute the following script. + +```bash +./tpch-gen.sh 1 +``` + +The examples provided use parquet files for the tables generated by `dbgen`. +An python script is provided to convert the text files from `dbgen` into parquet Review Comment: Typo: `A python script` ########## examples/tpch/q04_order_priority_checking.py: ########## @@ -0,0 +1,80 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. + +""" +TPC-H Problem Statement Query 4: + +The Order Priority Checking Query counts the number of orders ordered in a given quarter of a given +year in which at least one lineitem was received by the customer later than its committed date. The +query lists the count of such orders for each order priority sorted in ascending priority order. + +The above problem statement text is copyrighted by the Transaction Processing Performance Council +as part of their TPC Benchmark H Specification revision 2.18.0. +""" + +from datetime import datetime +import pyarrow as pa +from datafusion import SessionContext, col, lit, functions as F + +# Ideally we could put 3 months into the interval. See note below. +INTERVAL_DAYS = 92 +DATE_OF_INTEREST = "1995-04-01" + +# Load the dataframes we need + +ctx = SessionContext() + +df_orders = ctx.read_parquet("data/orders.parquet").select_columns( + "o_orderdate", "o_orderpriority", "o_orderkey" +) +df_lineitem = ctx.read_parquet("data/lineitem.parquet").select_columns( + "l_orderkey", "l_commitdate", "l_receiptdate" +) + +# Create a date object from the string +date = datetime.strptime(DATE_OF_INTEREST, "%Y-%m-%d").date() + +# Note: this is a hack on setting the values. It should be set differently once +# https://github.com/apache/datafusion-python/issues/665 is resolved. Review Comment: Ditto ########## examples/tpch/q01_pricing_summary_report.py: ########## @@ -0,0 +1,90 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. + +""" +TPC-H Problem Statement Query 1: + +The Pricing Summary Report Query provides a summary pricing report for all lineitems shipped as of +a given date. The date is within 60 - 120 days of the greatest ship date contained in the database. +The query lists totals for extended price, discounted extended price, discounted extended price +plus tax, average quantity, average extended price, and average discount. These aggregates are +grouped by RETURNFLAG and LINESTATUS, and listed in ascending order of RETURNFLAG and LINESTATUS. +A count of the number of lineitems in each group is included. + +The above problem statement text is copyrighted by the Transaction Processing Performance Council +as part of their TPC Benchmark H Specification revision 2.18.0. +""" + +import pyarrow as pa +from datafusion import SessionContext, col, lit, functions as F + +ctx = SessionContext() + +df = ctx.read_parquet("data/lineitem.parquet") + +# It may be that the date can be hard coded, based on examples shown. +# This approach will work with any date range in the provided data set. + +greatest_ship_date = df.aggregate( + [], [F.max(col("l_shipdate")).alias("shipdate")] +).collect()[0]["shipdate"][0] + +# From the given problem, this is how close to the last date in the database we +# want to report results for. It should be between 60-120 days before the end. +DAYS_BEFORE_FINAL = 68 + +# Note: this is a hack on setting the values. It should be set differently once +# https://github.com/apache/datafusion-python/issues/665 is resolved. Review Comment: Could you add a TODO in that issue to update this? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For additional commands, e-mail: github-h...@datafusion.apache.org
