sort the given jsonl file by the given key, writing the output to STDOUT.

sort-jsonl-by-key.py

"""
sort the given jsonl document (distinct json documents separated by newline)
by the given key, writing the output to STDOUT.

example:

    python sort-jsonl-by-key.py log.jsonl "timestamp"

this does require reading the entire document into memory, first.
a future revision could maybe use a mmap to avoid keeping things in memory.
"""
import re
import sys
import json

with open(sys.argv[1], "rb") as f:
    buf = f.read().decode("utf-8")

key = sys.argv[2]

lines = []
for match in re.finditer(r"^(.*)$", buf, re.M):
    if not match:
        continue

    line = buf[match.start():match.end()]
    if not line:
        continue

    linedoc = json.loads(line)
    linekey = linedoc[key]

    lines.append((linekey, match.start(), match.end()))

lines.sort()

for _, start, end in lines:
    print(buf[start:end])