Checking for duplicate values in database with Python

Question

I'm working on a Python application where I read and extract data from an HTML file. The data is stored in a list, and the number of items in a list is, on average, 50,000+. The items from the list are added to a database. I cannot allow duplicate datetime values in one of the columns, and so, I have to check if a specific value exists in the database already, and if it does, keep incrementing it until there's no duplicates.

Here's what my code looks like (I'm using SQLAlchemy):

import re from bs4 import BeautifulSoup from datetime import datetime, timedelta from models import db, Notes def import_file(filename, user): file = open(filename, 'r') html = file.read() file.close() soup = BeautifulSoup(html, "lxml") for link in soup.find_all('a'): validurl = re.compile( r'^(?:[a-z0-9\.\-]*)://' r'(?:(?:[A-Z0-9](?:[A-Z0-9-]{0,61}[A-Z0-9])?\.)+(?:[A-Z]{2,6}\.?|[A-Z0-9-]{2,}(?<!-)\.?)|' r'\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}|' r'\[?[A-F0-9]*:[A-F0-9:]+\]?)' r'(?::\d+)?' r'(?:/?|[/?]\S+)$', re.IGNORECASE) url = validurl.match(link.get('href')) if url: url = link.get('href') if len(url) <= 2000: if link.get('add_date'): date = datetime.utcfromtimestamp(int(link.get('add_date'))) while Notes.query.filter_by(user=user, added_on=date).first(): date += timedelta(0, 1) title = link.string import_notes = Notes() import_notes.main_url = url import_notes.title = title import_notes.user = user import_notes.added_on = date db.session.add(import_notes) db.session.commit() 

However, this code runs very slowly. I calculated how long it was taking per insert, and it came out at exactly 13 seconds per insert. I ran a profiler, and the filter_by query was taking the majority of the time. I figured it was because the script had to run a query so many times per item, and so I changed my code to use a static list instead:

import re from bs4 import BeautifulSoup from datetime import datetime, timedelta from models import db, Notes def import_file(filename, user): file = open(filename, 'r') html = file.read() file.close() soup = BeautifulSoup(html, "lxml") for link in soup.find_all('a'): validurl = re.compile( r'^(?:[a-z0-9\.\-]*)://' r'(?:(?:[A-Z0-9](?:[A-Z0-9-]{0,61}[A-Z0-9])?\.)+(?:[A-Z]{2,6}\.?|[A-Z0-9-]{2,}(?<!-)\.?)|' r'\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}|' r'\[?[A-F0-9]*:[A-F0-9:]+\]?)' r'(?::\d+)?' r'(?:/?|[/?]\S+)$', re.IGNORECASE) url = validurl.match(link.get('href')) if url: url = link.get('href') if len(url) <= 2000: if link.get('add_date'): date = datetime.utcfromtimestamp(int(link.get('add_date'))) while Notes.query.filter_by(user=user, added_on=date).first(): date += timedelta(0, 1) title = link.string import_notes = Notes() import_notes.main_url = url import_notes.title = title import_notes.user = user import_notes.added_on = date db.session.add(import_notes) db.session.commit() 

I thought this might make the code run faster, but there was little to no change in runtime. For comparison, if I remove the duplicate check, I get ~300 inserts per second.

I cannot let duplicate datetime values into the database. What can I do to make my code faster?

Answer 1

In python checking membership of an item in a set is extremely fast.

I suggest you use sets.

Answer 2

In addition to @Caridorc's answer about using sets instead of lists, a few more things beg for improvement.

Working with files

Use with ... as ... when working with files, to make sure that you cannot forget closing them. So instead of this:

file = open(filename, 'r') html = file.read() file.close() 

Write like this:

with open(filename) as fh: html = fh.read() 

(I also omitted the mode parameter of open, as 'r' is the default.

Move non-changing calculation out of a loop

In this loop, validurl is the same in every iteration:

for link in soup.find_all('a'): validurl = re.compile( r'^(?:[a-z0-9\.\-]*)://' r'(?:(?:[A-Z0-9](?:[A-Z0-9-]{0,61}[A-Z0-9])?\.)+(?:[A-Z]{2,6}\.?|[A-Z0-9-]{2,}(?<!-)\.?)|' r'\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}|' r'\[?[A-F0-9]*:[A-F0-9:]+\]?)' r'(?::\d+)?' r'(?:/?|[/?]\S+)$', re.IGNORECASE) url = validurl.match(link.get('href')) 

It should be done just once before the loop begings:

validurl = re.compile( r'^(?:[a-z0-9\.\-]*)://' r'(?:(?:[A-Z0-9](?:[A-Z0-9-]{0,61}[A-Z0-9])?\.)+(?:[A-Z]{2,6}\.?|[A-Z0-9-]{2,}(?<!-)\.?)|' r'\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}|' r'\[?[A-F0-9]*:[A-F0-9:]+\]?)' r'(?::\d+)?' r'(?:/?|[/?]\S+)$', re.IGNORECASE) for link in soup.find_all('a'): url = validurl.match(link.get('href')) 

Indentation

Indentation in Python is very important, and so it's important to do it consistently. For example this looks suspicious:

if url: url = link.get('href') if len(url) <= 2000: if link.get('add_date'): 

In most of the program you correctly indented with 4 spaces, but after if url there you indented by 8 spaces. This is confusing, and it should be corrected.