tika-extract.py

#!/usr/bin/python
# Runs in Pyhton 3
# Processes MSWord files with Tika. Zero exception or error control, use at own risk.
# Assumes lowercase .doc and .docx extensions.
# Output is written to new files with corresponding file exts in the same dir.

import os
import sys
import subprocess
import glob

# Usage: python convertdoc.py [dir] [meta/txt/html/xml]

curDir = "."
tika = "/home/pepe/code/tika-1.7/tika-app-1.7.jar"
mode = "--metadata"	# Default conversion mode. Use "txt", "html"
					# and "xml" on the command line for the others

def newName(name, mode):
	newName = name[:-3]
	if mode == "--metadata":
		newName += "meta"
	elif mode == "--text":
		newName += "txt"
	elif mode == "--html":
		newName += "html"
	elif mode == "--xml":
		newName += "xml"
	else:
		newName += "meta"
	
	return newName
				
if len(sys.argv) >= 2:
	curDir = sys.argv[1]

if len(sys.argv) == 3:
	if sys.argv[2] == "txt":
		mode = "--text"
	elif sys.argv[2] == "html":
		mode = "--html"
	elif sys.argv[2] == "xml":
		mode = "--xml"
	else:
		sys.argv[2] == "--metadata"

# filelist = glob.glob(os.path.join(curDir, "*.DOC"))
filelist = glob.glob(os.path.join(curDir, "*.do*"))
		
for filename in filelist:
	print("Processing " + filename + "...")
	newFile = newName(filename, mode)
	f = open(newFile, "w")
	subprocess.call(["java", "-jar", tika, mode, filename], stdout=f, timeout=None)
	f.close()