Download Google Docs without a browser

The script looks something like this:

#!/bin/bash
token=$(curl -s www.google.com/accounts/ClientLogin -d Email=user.name@gmail.com -d Passwd=qwerty -d accountType=GOOGLE -d service=writely -d Gdata-version=3.0 |cut -d "=" -f 2)
set $token
curl --silent --header "Gdata-Version: 3.0" --header "Authorization: GoogleLogin auth=$3" "http://docs.google.com/feeds/user.name@gmail.com/private/full" | tidy -xml -indent -utf8 -quiet > /tmp/gdocs.xml
file=(doc1 doc2 doc3 doc4)
resourceId =(document:0ASpnxqo7zPlfZGQ3Z2Zwc183NWZteDNiM2Y ...) # 3 resourceId
i=0
while (( i < 4 )) # match num_files
do
wget --header "Gdata-Version: 3.0" --header "Authorization: GoogleLogin auth=$3" "https://docs.google.com/feeds/download/documents/Export?docID=${resourceId[$i]}&exportFormat=txt" -O /tmp/${file[$i]}.txt

sleep 2
i=$i+1
done

And now, as is customary, exposure.

• The first line is the authentication token. Next, we present it for downloading an RSS feed, which we save to a file. Next, take resourceId from it and download the docks. There should be a reservation that resourceId is taken "manually" from the saved RSS file. You can, of course, automate this, but if there is not a lot of documents, it is not necessary to inflate the script.
• Link for downloading docks:
  GET /feeds/download/documents/Export?docID=resource_id&exportFormat=format
• service = writely. The list of Google service code names is taken from here .
• resourceId key element. This is the dock id. We take it from the RSS feed downloaded to the /tmp/gdocs.xml file. Relevant tags look like and are in the same entry where the name is docks.
  <gd:resourceId>
document:0ASpnxqo7zPlfZGQ3Z2Zwc183Zzc3NnZkY2g</gd:resourceId>

• exportFormat = txt. You can choose other formats to download. The list is here .

Written in haste, so any comments and criticism are welcome.