jnget 0.2.1 review

jnget is a command-line tool for retrieving binaries from Usenet. jnget project allows multi-thread downloading and partial or ful

License: The Apache License 2.0 File size: 0K Developer: Santa

0

jnget is a command-line tool for retrieving binaries from Usenet.

jnget project allows multi-thread downloading and partial or full header updating and caching.

Requirements:
Java 1.5.0

Installation:

- Unzip the release file.
- Set the INSTALLDIR variable into the batch file you use (jnget.bat for windows, jnget.sh for Linuxes)
- Fill at least the mandatory fields of the jnget.properties file.
- Try to always start a new download into a fresh new directory.

Usage:

-g < newsgroup > newsgroup to grab from
-G < newsgroup > same as -g but no cache update
-p < #PAR blocks > download #PAR blocks (-p all for all PAR blocks)
-r < regex > regular expression of articles to get
-R < regex > same has -r but case sensitive
-l < #articles > only grab the last #articles
-t < #threads > number of threads to connect to news server
-help print this message

For example:

jnget -g alt.binaries.linux -r "linux_kernel_v[123]$" -p 80
This will:
- update the local header cache of the alt.binaries.linux newsgroup
- download all articles which subject matching the "linux_kernel_v[123]$" regular expression
- download 80 par blocks associated with the same regular expression

jnget -G alt.binaries.linux -r "linux_kernel_v[123]$" -p 80
Will do the same as above except it won't update the header cache file

jnget -g alt.binaries.linux -r "linux_kernel_v[123]$" -p 80 -l 100000

This time the header cache is only updated with the last 100000 article headers (too old headers cached are purged but the good one are kept, even if older than the 100000th).

(Note for Linux users: It can be usefull to wrap that into a nohup call in order for the download to last after user disconnection:

nohup jnget -g alt.binaries.linux -r "linux_kernel_v[123]$" -p 80 -l 100000 &
And one can then follow the progress with 'tail -f nohup.out')

If you have no idea of what a regular expression is, just use -r followed with a word that is in the articles you want to get and avoid characters like "[]*$^". Or learn regular expressions (this can be usefull):

http://en.wikipedia.org/wiki/Regular_expression

For information, you can read cached header files in your cache.dir directory, those are just zipped text file.

For example: 'zcat alt.binaries.linux.gz | grep kernel' launched into the cache.dir directory will return all articles containing the word kernel in its subject.

I would appreciate it if you make modifications if you would send them to me for possible inclusion in the main source.

What's New in This Release:
Reconnection on broken pipe and socket timeout
cache.dir is not mandatory when using NZB file