{
/* WARC format files have multiple compressed records. JDK one can't deal with this
* See: http://crawler.archive.org/apidocs/index.html?org/archive/io/arc/ARCWriter.html
* We get around this by using an external zcat process
*/
is = new java.io.BufferedInputStream(new ProcessInputStream("/usr/bin/gzip -dc ", filename));
}
else
is = Files.openFileStream(filename); //throws an IOException, throw upwards
//logger.info("WARCCollection processing "+filename);
//no need to loop again