coalesce.py

#!/usr/bin/env python
"""
Coalesce a single file containing multiple JS arrays (undelimited) into
one giant JS array.

Ex. "[{...}]\n[{..}]" becomes "[{...},{...}]

This assumes each JS array contains a list of JS object literals.

Usage:

> coalesce.py broken.json > fixed.json

"""
import os
import sys
import re

# Source: http://stackoverflow.com/questions/3862010/is-there-a-generator-version-of-string-split-in-python
def itersplit(s, sep=None):
    exp = re.compile(r'\s+' if sep is None else re.escape(sep))
    pos = 0
    while True:
        m = exp.search(s, pos)
        if not m:
            if pos < len(s) or sep is not None:
                yield s[pos:]
            break
        if pos < m.start() or sep is not None:
            yield s[pos:m.start()]
        pos = m.end()

contents = open(sys.argv[1], 'r').read()

i = 0
sys.stdout.write('[')
for line in itersplit(contents, '\n'):
    # Strip leading and trailing characters.
    line = line.strip(',[]')

    # Filter out empty rows.
    if len(line) == 0:
        continue

    # Filter out rows that do not contain whole JSON objects.
    if line[0] != '{':
        continue
    if line[len(line) - 1] != '}':
        continue

    if i > 0:
        sys.stdout.write('\n,')

    sys.stdout.write(line)
    i += 1
sys.stdout.write(']')