FileNotFoundException caused by aborting the process of downloading Wikipedia dataset

When running the script examples/bin/classify-wikipedia.sh for the first time, it will create a wikixml folder and starts fetching data via curl. If this downloading process is aborted, then in the future when the script is run, it won't extract the .bz2 file (since extracion is guarded by the condition where wikixml doesn't exist) and starts to run Mahout, which will definately end up with throwing up a FileNotFoundException.