Finding typos in Django

README.md

Usage

1. Run collectwords.py with file paths as arguments to build a database of words.
2. Run spellcheck.py to mark misspells.
3. Use sqlite3 shell or anything else to output misspelled words to a file.
4. Go through the file eliminating false positives.
5. Search through the codebase and fix typo by typo :-)

collectwords.py

import sys
import re
import sqlite3
c = sqlite3.connect('words.sqlite')

c.execute("""
    CREATE TABLE IF NOT EXISTS words (word TEXT UNIQUE, misspell bool)
""")


def insert_words(words):
    return c.executemany("""
        INSERT OR IGNORE INTO words (word) VALUES (?)
    """, ((w,) for w in words))

for f in sys.argv[1:]:
    print f,
    t = open(f).read()
    words = (mo.group(0) for mo in re.finditer(r"\b\w+\b", t))
    with c:
        cursor = insert_words(words)
    print cursor.rowcount

spellcheck.py

import sqlite3
import enchant

c = sqlite3.connect('words.sqlite')
d = enchant.Dict("en_US")

def check_spelling(word):
    return d.check(word)

c.create_function("check_spelling", 1, check_spelling)

with c:
    cursor = c.execute("""
    UPDATE words
    SET misspell = check_spelling(word)
    WHERE misspell IS NULL
    """)
    print cursor.rowcount