Merge pull request #12 from chrishardoc/develop

Develop

.gitignore

@@ -2,4 +2,6 @@
 venv/
 
 *.csv
+provisoire.py
+test_db.py
 

answers/beverages_and_food.sql

@@ -0,0 +1,2 @@
+SELECT * FROM beverages
+CROSS JOIN coucou

app.py

@@ -1,83 +1,69 @@
 # pylint: disable = missing-module-docstring
-
 import io
-
+import ast
 import duckdb
 import pandas as pd
 import streamlit as st
 
-CSV = """
-beverage,price
-orange juice,2.5
-Expresso,2
-Tea,3
-"""
-
-beverages = pd.read_csv(io.StringIO(CSV))
+#from init_db import beverages
 
-CSV2 = """
-food_item,food_price
-cookie juice,2.5
-chocolatine,2
-muffin,3
-"""
+con = duckdb.connect(database="data/exercices_sql_tables.duckdb", read_only=False)
 
-food_items = pd.read_csv(io.StringIO(CSV2))
-
-ANSWER_STR = """
-SELECT * FROM beverages
-CROSS JOIN food_items
-"""
-
-solution_df = duckdb.sql(ANSWER_STR).df()
+# solution_df = duckdb.sql(ANSWER_STR).df()
 
 with st.sidebar:
-    option = st.selectbox(
+    theme = st.selectbox(
         "What would you like to review?",
-        ("Joins", "GroupBy", "Windows Functions"),
+        ("cross_joins", "GroupBy", "windows_functions"),
         index=None,
         placeholder="Select a theme...",
     )
-    st.write("You selected:", option)
+    st.write("You selected:", theme)
+
+    exercice = con.execute(f"SELECT * FROM memory_state WHERE theme = '{theme}'").df()
+    st.write(exercice)
 
 
 st.header("enter your code")
 query = st.text_area(label="votre code SQL ici", key="user_input")
 if query:
-    result = duckdb.sql(query).df()
+    result = con.execute(query).df()
     st.dataframe(result)
-
-    # if len(result.columns) != len(     #comparaison du nombre de colonnes avec celui attendu
-    #    solution_df.columns
-    # ):      # replace with try result = result(solution_df.columns)
-    #    st.write("Some columns are missing depuis if")
-
-    try:
-        result = result[solution_df.columns]
-        st.dataframe(result.compare(solution_df))
-        # comparaison deux dataframes, avec la méthode compare de pandas, mais cela ne fonctionne
-        # que sur des dataframes ayant le même nombre de colonnes et dans le même ordre
-    except KeyError as e:
-        st.write("Some columns are missing depuis except keyerror")
-
-    n_line_difference = (
-        result.shape[0] - solution_df.shape[0]
-    )  # comparaison du nombre de lignes avec celui attendu
-    if n_line_difference != 0:
-        st.write(
-            f"result has a {n_line_difference} lines difference with the solution_df"
-        )
-
-
-tab2, tab3 = st.tabs(["Tables", "Solutions"])
+#
+#     # if len(result.columns) != len(     #comparaison du nombre de colonnes avec celui attendu
+#     #    solution_df.columns
+#     # ):      # replace with try result = result(solution_df.columns)
+#     #    st.write("Some columns are missing depuis if")
+#
+#     try:
+#         result = result[solution_df.columns]
+#         st.dataframe(result.compare(solution_df))
+#         # comparaison deux dataframes, avec la méthode compare de pandas, mais cela ne fonctionne
+#         # que sur des dataframes ayant le même nombre de colonnes et dans le même ordre
+#     except KeyError as e:
+#         st.write("Some columns are missing depuis except keyerror")
+#
+#     n_line_difference = (
+#         result.shape[0] - solution_df.shape[0]
+#     )  # comparaison du nombre de lignes avec celui attendu
+#     if n_line_difference != 0:
+#         st.write(
+#             f"result has a {n_line_difference} lines difference with the solution_df"
+#         )
+#
+#
+tab2, tab3 = st.tabs(["Tables", "Solution"])
 
 with tab2:
-    st.write("table : beverages")
-    st.dataframe(beverages)
-    st.write("table : food_items")
-    st.dataframe(food_items)
-    st.write("expected")
-    st.dataframe(solution_df)
+    exercice_tables = ast.literal_eval(exercice.loc[0,"tables"])
+    for table in exercice_tables:
+        st.write(f"table: {table}")
+        df_table=con.execute(f"SELECT * FROM {table}").df()
+        st.dataframe(df_table)
 
 with tab3:
-    st.write(ANSWER_STR)
+    exercice_name = exercice.loc[0, "exercice_name"]
+    with open(f"answers/{exercice_name}.sql", "r") as f:
+        answer = f.read()
+    st.write(answer)
+

init_db.py

@@ -0,0 +1,48 @@
+# pylint: disable = missing-module-docstring
+
+import io
+
+import duckdb
+import pandas as pd
+
+con = duckdb.connect(database = "data/exercices_sql_tables.duckdb", read_only=False)
+
+
+# --------------------------------------------------------------
+# EXERCISES LIST
+# --------------------------------------------------------------
+
+data = {
+    "theme": ["cross_joins","windows_functions"],
+    "exercice_name": ["beverages_and_food","simple_window"],
+    "tables": [["beverages", "food_items"],"simple_window"],
+    "Last_review": ["1970-01-01", "1970-01-01"],
+}
+memory_state_df = pd.DataFrame(data)
+con.execute("CREATE TABLE IF NOT EXISTS memory_state AS SELECT * FROM memory_state_df")
+
+
+# --------------------------------------------------------------
+# CROSS JOIN EXERCISES
+# --------------------------------------------------------------
+
+CSV = """
+beverage,price
+orange juice,2.5
+Expresso,2
+Tea,3
+"""
+
+beverages = pd.read_csv(io.StringIO(CSV))
+con.execute("CREATE TABLE IF NOT EXISTS beverages AS SELECT * FROM beverages")
+
+CSV2 = """
+food_item,food_price
+cookie juice,2.5
+chocolatine,2
+muffin,3
+"""
+
+food_items = pd.read_csv(io.StringIO(CSV2))
+con.execute("CREATE TABLE IF NOT EXISTS food_items AS SELECT * FROM food_items")
+