zpaqlist

List the contents of zpaq files in more concise way
A patched zpaq 6.60 (http://mattmahoney.net/dc/zpaq.html)

Provided as-is, with no warranty whatsoever, by Franco Corbelli, franco@francocorbelli.com

More concise and easier to parse output: reducing the time necessary for subsequent use (e.g. GUIs) on Windows.

Why?

When the size of the output is large (can become hundreds of MB) the savings both in writing (on magnetic disks), reading and parsing can be considerable

For small archives (KB) there is obviously no difference compared to zpaq's time

What?

A classic method for an extracting GUI for zpaq is to redirect the output of the command
zpaq l (list) to a temporary file, read , parse and then process, but it takes time, even minutes.

The output of zpaqlist is composed by
- version # with !. In this example, 1266

!1266

• version list, with fixed-length version number, datetime (std format)

|      1 2019-05-12 15:42:22
|      2 2019-05-13 09:22:49
|      3 2019-05-14 17:18:06
|      4 2019-05-16 14:17:25
|      5 2019-05-16 15:30:17
|      6 2019-05-16 23:30:17
(...)

• total row number (with +), in this example 38+ millions to be parsed

+38915424

• A sorted list of version and file name (for a time machine-like use)
• and, by default, 'deduplicated' in file names (future: file size too)
• version_number
• datetime (or D for deleted)
• size (with dots)
• filename or ? (if not changed from previous)

Example (two record)

The file f:/zarc/ihsv/pakka/30_3/zpaqfranz.exe is 3.089.462 bytes long,
and was found in the 946 version, @ 02/10/2020 14:25:34 (European-style date format)
In the version 959 the file result deleted (not present)

-946
02/10/2020 14:25:34
3.089.462
f:/zarc/ihsv/pakka/30_3/zpaqfranz.exe
-959
D
0
?

Optional switches

-pakka for more verbose
-distinct do not 'deduplicate'
-all show all versions
-key password
-until version
-out logfile.txt

How to compile (Windows)

Nothing hard

g++ -O3  zpaqlist.cpp -o zpaqlist -static