gistfile1.txt

#!/usr/bin/env python3

import json
import re
from time import time
from random import randint

import psycopg2
import psycopg2.extras

from unidecode import unidecode

import config

TEST_STRINGS = [
        "massiveattackteardropbartclaessentomfallbootleg",
        "squirrelnutzippersivefoundanewbaby",
        "giantsand1helvakowboysong",
        "ericserrathepantryhideout",
        "bobdylanalabamagetaway", 
        "bobdylanleopardskinpillboxhatincomplete",
        "thereplacementsyouaintgottadancestudiodemo",
        "steviewonderijustcalledtosayiloveyou",
        "thestonerosesmadeofstone808statemix",
        "yolatengotomcourtenayacousticversion",
]

def encode_string(text):
    return unidecode(re.sub(" +", " ", re.sub(r'[^\w ]+', '', text)).strip().lower())

def fuck_string_up(text, num_chars_to_remove):
    
    for i in range(num_chars_to_remove):
        remove = randint(0, len(text))
        text = text[:remove] + text[remove+1:] 
        
    return text

def lookup_test(num_chars_to_remove):
    
    print("Remove %s characters:" % num_chars_to_remove)
    total_time = 0.0
    with psycopg2.connect(config.MBID_MAPPING_DATABASE_URI) as mb_conn:
        with mb_conn.cursor(cursor_factory=psycopg2.extras.DictCursor) as curs:
            for text in TEST_STRINGS:
                query = """SELECT artist_name, recording_name, lookup_nows, similarity(lookup_nows, %s) AS sml
                             FROM mapping.content_resolver
                            WHERE lookup_nows %% %s
                         ORDER BY sml DESC, lookup_nows   
                            LIMIT 1"""
 
                lookup = fuck_string_up(text, num_chars_to_remove)
                t0 = time()     
                curs.execute(query, (lookup, lookup))
                row = curs.fetchone()    
                if row is None:    
                    print("%s not found!!" % lookup)
                    
                if text != row["lookup_nows"]:
                    print("Found wrong entry for:\n   %s (search)\n   %s (result)" % (text, row["lookup_nows"]))
                
                t1 = time()     
                total_time += t1 - t0
                
    print(f"avg time per lookup: {total_time/len(TEST_STRINGS)}\n")
                

if __name__ == "__main__":
    for i in range(5):
        lookup_test(i+1)