zulu

Reads stdin and writes to stdout while replacing epoch times in text with human-readable strings, defaults to UTC/Zulu time. If you have a file or other stream of text data that contains epoch times:

{ "text": "I am some JSON", "ts": 1574736728 }

You can pipe the stdout from a process like cat, kafakcat, jq, etc. to zulu to reformat these timestamps with minimal effort:

{ "text": "I am some JSON", "ts": "2019-11-26T02:52:08Z" }

What values are converted?

• 9-10 digit values are treated as seconds since epoch
• 12-13 digit values are treated as milliseconds since epoch
• 15-16 digit values are treated as microseconds since epoch

This program makes some assumptions about the range of years to limit false matches by the regex, effectively restricting replacements to the years 1973-2128. For example 5574739532 would convert to Sunday, August 28, 2146 10:45:32 AM. It starts with a 5 which means it would be later than 2128 and probably not an intentional timestamp - so it is not converted. It could be considered "greedy" in what it tries to convert, with these restrictions in place. UPCs are a good example of a numeric string that can be unintentionally converted - it is handy for triaging things and poking around, but not recommended to use zulu for data conversion because of this.

Installation

• You can download a binary from the Releases section for supported platforms
• If you'd like to build from source:
  • have Rust 2018 installed (developed with 1.38, although slightly older versions should work)
  • clone this repository on your local filesystem
  • run cargo build --release and copy the target/release/zulu binary to somewhere on your path, ie. /usr/local/bin/

Usage

Use zulu on the command line as part of a series of pipes, ie:

cat my_file | grep 'something to limit' | zulu > new_file

Optional parameters:

• -f <format string> Specify output format, see Chrono's strftime module for more details
• -l Use local timezone for output
• -s Stringify output by wrapping with double quotes, useful for working with JSON data

TODO

• Benchmarks
• Field aware, allow operations only on specific columns of data as output by something like awk
• brew installation support