From 9cb7c902c4850bca450ce512a4df4d5ecf6965b0 Mon Sep 17 00:00:00 2001
From: Pierre Chanial <pierre.chanial@ego-gw.it>
Date: Wed, 4 Aug 2021 04:23:06 +0200
Subject: [PATCH] Create table from SQL query involving tables in the same
project.
---
README.md | 27 +++++++++++++++++++++++++++
esap_client/models/datasets.py | 16 ++++++++++++++++
2 files changed, 43 insertions(+)
diff --git a/README.md b/README.md
index 6746c2f..7b4a948 100644
--- a/README.md
+++ b/README.md
@@ -74,6 +74,33 @@ External CSV files can be imported into a table:
table = dataset.create_table_from('https://www.data.gouv.fr/fr/datasets/r/c0f59f00-3ab2-4f31-8a05-d317b43e9055', sep=';')
df = table.aspandas()
```
+
+## Creation of a table from a query
+The tables from different datasets can be combined in a query.
+
+```python
+import pandas as pd
+
+dataset1 = project.create_dataset('dataset1')
+df_fruits = pd.DataFrame({'x': 6 * ['fruit'], 'y': list('🍓🥝🍇🍐🍏🍍')})
+fruits = dataset1.create_table_from(df_fruits, 'fruits')
+fruits.aspandas()
+
+dataset2 = project.create_dataset('dataset2')
+df_vegetables = pd.DataFrame({'x': 5 * ['vegetable'], 'y': list('🥑🌽🥒🍆🥦')})
+vegetables = dataset2.create_table_from(df_vegetables, 'vegetables')
+vegetables.aspandas()
+
+dataset3 = project.create_dataset('dataset3')
+query = """
+SELECT * FROM dataset1.fruits
+UNION
+SELECT * FROM dataset2.vegetables
+ORDER BY x, y
+"""
+food = dataset3.create_table_as(query, 'food')
+food.aspandas()
+```
## Creation of a table from an ESAP ESFRI query
We can store the result of an ESAP-API query into a table that belongs to the dataset that we have just created.
diff --git a/esap_client/models/datasets.py b/esap_client/models/datasets.py
index bda40c1..bb50e45 100644
--- a/esap_client/models/datasets.py
+++ b/esap_client/models/datasets.py
@@ -71,6 +71,22 @@ def create_table_from(
raise_for_status(response)
return Table.deserialize(response.json())
+ def create_table_as(
+ self, query: str, name: Optional[str] = None, description: str = ''
+ ) -> Table:
+ """Creates a table from a SQL query."""
+ if not isinstance(query, str):
+ raise TypeError('The query is not a string.')
+
+ payload = {
+ 'name': name,
+ 'description': description,
+ 'query': query,
+ }
+ response = self.session.post('/tables', payload)
+ raise_for_status(response)
+ return Table.deserialize(response.json())
+
def create_table_from_esap_gateway_query(
self, query: dict, name: str, description: str = ''
) -> Table:
--
GitLab