Samoylov Pavel AT-18

HW.py

@@ -0,0 +1,53 @@
+import sqlite3
+import pandas as pd
+import re
+
+def strip_tags(value):
+    return re.sub(r'<[^>]*>', '', str(value))
+
+con = sqlite3.connect('works.sqlite')
+cursor = con.cursor()
+cursor.execute('PRAGMA foreign_keys = true')
+con.commit()
+
+cursor.execute('DROP TABLE IF EXISTS works')
+cursor.execute('CREATE TABLE works ('
+               'ID INTEGER PRIMARY KEY AUTOINCREMENT,'
+               'salary INTEGER,'
+               'educationType TEXT,'
+               'jobTitle TEXT,'
+               'qualification TEXT,'
+               'gender TEXT,'
+               'dateModify TEXT,'
+               'skills TEXT,'
+               'otherInfo TEXT)')
+
+df = pd.read_csv("works.csv")
+
+df['skills'] = df['skills'].apply(strip_tags)
+df['otherInfo'] = df['otherInfo'].apply(strip_tags)
+df.to_sql("works", con, if_exists='append', index=False)
+con.commit()
+
+cursor.execute('DROP TABLE IF EXISTS genders')
+cursor.execute('CREATE TABLE genders(genderName TEXT PRIMARY KEY )')
+cursor.execute('INSERT INTO genders SELECT DISTINCT gender FROM works WHERE gender IS NOT NULL')
+cursor.execute('DROP TABLE IF EXISTS educations')
+cursor.execute('CREATE TABLE educations(educationType TEXT PRIMARY KEY )')
+cursor.execute('INSERT INTO educations SELECT DISTINCT educationType FROM works WHERE works.educationType IS NOT NULL')
+con.commit()
+
+cursor.execute('CREATE TABLE new_works ('
+               'ID INTEGER PRIMARY KEY AUTOINCREMENT,'
+               'salary INTEGER,'
+               'educationType TEXT REFERENCES educations(educationType) ON DELETE CASCADE ON UPDATE CASCADE,'
+               'jobTitle TEXT,'
+               'qualification TEXT,'
+               'gender TEXT REFERENCES genders(genderName) ON DELETE CASCADE ON UPDATE CASCADE,'
+               'dateModify TEXT,'
+               'skills TEXT,'
+               'otherInfo TEXT)')
+cursor.execute('INSERT INTO new_works SELECT * FROM works')
+cursor.execute('DROP TABLE works')
+cursor.execute('ALTER TABLE new_works RENAME TO works')
+con.commit()

class.py

@@ -0,0 +1,62 @@
+import sqlite3
+import pandas as pd
+import matplotlib.pyplot as plt
+import numpy as np
+
+#1
+con = sqlite3.connect('works.sqlite')
+cursor = con.cursor()
+cursor.execute('drop table if exists works')
+cursor.execute('create table works ('
+               'ID INTEGER PRIMARY KEY AUTOINCREMENT,'
+               'salary INTEGER,'
+               'educationType TEXT,'
+               'jobTitle TEXT,'
+               'qualification TEXT,'
+               'gender TEXT,'
+               'dateModify TEXT,'
+               'skills TEXT,'
+               'otherInfo TEXT)')
+con.commit()
+
+df = pd.read_csv("works.csv")
+df.to_sql("works", con, if_exists='append', index=False)
+con.commit()
+
+#2
+cursor.execute('create index salary_index on works (salary)')
+con.commit()
+
+#3
+cursor.execute('SELECT count(*) FROM works')
+print(cursor.fetchall()[0][0])
+
+#4
+cursor.execute("SELECT gender, count(*) FROM works GROUP BY works.gender")
+print(cursor.fetchall())
+
+#5
+cursor.execute("SELECT count(*) FROM works WHERE works.skills IS NOT NULL")
+print(cursor.fetchall()[0][0])
+
+#6
+cursor.execute("SELECT * FROM works WHERE works.skills IS NOT NULL")
+print(cursor.fetchall())
+
+#7
+cursor.execute("SELECT salary FROM works WHERE skills LIKE '%Python%'")
+print(cursor.fetchall())
+
+#8
+cursor.execute("SELECT salary FROM works WHERE gender = 'Мужской'")
+m_salary = [i[0] for i in cursor.fetchall()]
+cursor.execute("SELECT salary FROM works WHERE gender = 'Женский'")
+w_salary = [i[0] for i in cursor.fetchall()]
+
+m_quantile = np.quantile(m_salary, np.linspace(0.1, 1, 10))
+w_quantile = np.quantile(w_salary, np.linspace(0.1, 1, 10))
+plt.hist(m_quantile, 100, color='blue')
+plt.show()
+plt.hist(w_quantile, 100, color='red')
+plt.show()
+con.commit()