|
From: | Antonio Diaz Diaz |
Subject: | Re: [Lzip-bug] Feature Request/Question - Compression Level Information |
Date: | Wed, 25 Oct 2017 17:24:03 +0200 |
User-agent: | Mozilla/5.0 (X11; U; Linux i586; en-US; rv:1.9.1.19) Gecko/20110420 SeaMonkey/2.0.14 |
Timothy Beryl Grahek wrote:
That is extremely helpful, thank you so much.
You are welcome. :-)
What is the reason that you chose to indicate to some extent compression levels by dictionary size rather than by the level numbers themselves? Is it because it's more informative to use dictionary sizes exclusively?
The lzip format is supposed to be as simple as possible, and the dictionary size is the only information needed to decompress the file. The other parameter (match length limit) is not needed.
The compression level does not offer accurate information because it is an arbitrary mapping to tuples {dictionary size, match length limit}. This mapping was different in old versions of lzip.
Moreover, any combination of '--dictionary-size' and '--match-length' can be specified in the command line, which will usually not correspond to any concrete compression level.
Best regards, Antonio.
[Prev in Thread] | Current Thread | [Next in Thread] |