Roteiro2Arc
===========================================================
Roteiro2Arc is a tool that converts to Internet Archive ARC
files the local archive of the Portuguese Web present in the 
CD-ROM featured with the book "Novo Roteiro Prático da Internet"
by José Magalhães.

Roteiro2Arc is a tool created by the SAW group at FCCN
<sawfccn@google.com> to be used by the Portuguese Web
Archive for internal projects (http://www.arquivo.pt/).

This software is distributed as is and is licensed with
GNU LGPL licence.


-----------------------------------------------------------
Running Roteiro2Arc
-----------------------------------------------------------
Roteiro2Arc conversion is a two step process:
1) Search the files for an the embedded original URL and
write to a database the information that relates the local
path of the file to the found URL.

2) Reads the files and export them ARC files.

This second steps is further divided in sub-steps.
	2.1) Read the files that we know the original URL(it is in
	the DB), convert the local paths in the file to URLs and
	write it to an ARC file. 
	During this step, every file that had no known
	original URL receives a generated URL (we use the
	URL of the file that called it and append the local
	path of the unknown file) and a reference to this
	file and its generated URL is stored in a special
	recovery DB.

	2.2) The recovery DB is read and every file referenced
	is exported to ARC files using the generated URL.
	This way, it is possible that webpages can maintain
	the reference to images, stylesheets and other
	resources.

The easiest way to run Roteiro2Arc is to use the "run" Ant task.
This task handles both steps of running the conversion process:

> ant run -Dsource=<source_dir> -Ddestination=<destination_dir> 
	\ -Ddb=<db_dir> -Drecovery-db=<recovery-db_dir>

The other way is to run each step separately:

1) URL extraction
> ant run-url-extraction -Dsource=<source_dir> -Ddb=<db_dir>	(with Ant)
> java ExtractURL <source_dir> <db_dir>				(calling the class) 

2) Writing the ARC files
> ant run-arc-conversion -Dsource=<source_dir> -Ddestination=<destination_dir>
	\ -Ddb=<db_dir> -Drecovery-db=<recovery-db_dir>		(with Ant)
> java ArcConverter <source_dir> <destination_dir> <db_dir> 	
	\ <recovery-db_dir>					(calling the class)

source_dir: the dir with the local files.
destination_dir: the destination for the converted ARC files.
db_dir: the path where the Berkeley DB with the detected
	paths/URLs will be stored.
recovery-db_dir: the path where the Berkeley DB with the recovery
	information of the files without the original URL. This
	DB will hold path/generated-url informations.