hr4e_gen_patient_files.py

'''
Created on May 17, 2011

    Simple python script to generate a file name for an HR4E patient file.  
    
    The input to this script is an instance of an hr4e patient data xml document that 
    was generated from hr4e_patient_template.xml.
    
    The patient's given and family name and the document ID are extracted from the file and 
    concatenated to produce a file name with a .xml suffix that will be stored in the local directory.

@author: torkroth
@modified: superrawr
'''

import libxml2
import libxslt
import os
import sys
import StringIO, sys
import subprocess
import argparse
import cPickle as pickle
 #
 # Parse the command line
 #
 # -i <input file>: patient data file to write out
 #    The default is hr4e_patient_template.xml in current directory
 #

parser = argparse.ArgumentParser()
parser.add_argument('-i', metavar='in-file', type=str, default="hr4e_patient.xml")
parser.add_argument('-g', metavar='xpath_given_name', type=str, default="//gcda:greenCCD/gcda:header/gcda:personalInformation/gcda:patientInformation/gcda:personInformation/gcda:personName/gcda:given")
parser.add_argument('-f', metavar='xpath_family_name', type=str, default="//gcda:greenCCD/gcda:header/gcda:personalInformation/gcda:patientInformation/gcda:personInformation/gcda:personName/gcda:family")
parser.add_argument('-d', metavar='xpath_docID', type=str, default="//gcda:greenCCD/gcda:header/gcda:documentID/@root")
parser.add_argument('-n', metavar='namespace', type=str, default="AlschulerAssociates::GreenCDA")
parser.add_argument('-green', metavar='green-prefix', type=str, default="green_")
parser.add_argument('-cda', metavar='cda-prefix', type=str, default="cda_")
parser.add_argument('-hxslt', metavar='hr4e-xslt-file', type=str, default="templates/hr4e_patient_to_ccd.xslt")
parser.add_argument('-gxslt', metavar='green-xslt-file', type=str, default="templates/green_ccd.xslt")

try:
    results = parser.parse_args()
    input = results.i
    given_xpath = results.g
    family_xpath = results.f
    docID_xpath = results.d
    namespace_name = results.n
    green_filename_prefix = results.green
    cda_filename_prefix = results.cda
    hr4e_xslt = results.hxslt
    green_xslt = results.gxslt    
#    print 'Input file:', input
#    print 'XPath to given name', given_xpath
#    print 'XPath to family name', family_xpath
#    print 'XPath to docID', docID_xpath
#    print 'Namespace', namespace_name
#    print 'Green filename prefix ', green_filename_prefix
#    print 'CDA filename prefix ', cda_filename_prefix
#    print 'XSLT file to transform HR4E format to green CCD formta:', hr4e_xslt
#   print 'XSLT file to transform green CCD to CDA:', green_xslt
except IOError, msg:
    parser.error(str(msg))
    

#Let's get the patient pickle dict
#patient = pickle.load(file('tmp/patient.pkl','r+b'))
#Do some csv magic...
#print patient



#
# Parse the file
#
patient_doc = libxml2.parseFile(input)


#
# Use XPath to get given name, family name and document ID
#
# For this to work, we need to set the namespace and then prefix each element name with it
# in the XPath expression. Yuck!
#
ctxt = patient_doc.xpathNewContext()
ctxt.xpathRegisterNs("gcda", namespace_name)
result = ctxt.xpathEval(given_xpath)
for node in result:
    given_name = node.content
    
result = ctxt.xpathEval(family_xpath)
for node in result:
    family_name = node.content
    
result = ctxt.xpathEval(docID_xpath)
for node in result:
    docID = node.content

# Construct new file names out of the pieces
#
new_filename = family_name + "_" + given_name + "_" + docID + ".xml"
make_directory = "data/patient_data/" + family_name + "_" + given_name + "_" + docID+ "/"
green_filename = green_filename_prefix + new_filename
cda_filename = cda_filename_prefix + new_filename

#
# Create a directory for the patient
#
os.makedirs(make_directory)
print "Just created " + make_directory + " directory for patient data storage"
print "######################################################################################"
print

#
# Write out the XML document and the new patient pkl
#
print "######################################################################################"
print "Generating patient file " + new_filename + " from input file " + input
new_file = open(make_directory + new_filename, 'w')
new_file.write(patient_doc.serialize())
new_file.close()
print "Generating patient file " + new_filename[0:len(new_filename)-3] + "pkl from the tmp pickle"
#pickle.dump(patient, file(new_filename[0:len(new_filename)-3] + 'pkl','wb'))
print "Done. HR4E patient " + new_filename + " generated along with patient's pickle"
print "#######################################################################################"


# Build the command line to remove hr4e elements from options to this script
#
print "Transforming HR4E patient data in file " + new_filename + "..."     
command = "java -jar saxon/saxon9he.jar " + " -s:" + make_directory + new_filename + " -xsl:" + hr4e_xslt + " > " + make_directory + green_filename
#print command

# 
# Fire off a process to invoke java and run the HR4E stylesheet
#       
p = subprocess.Popen(command, shell=True)
retval = p.wait()
print "Done. green CCD patient data file " + green_filename + " generated with hr4e namespace elements removed."

#
# Now built the command line to transform to CDA document
#
print "Transforming green patient data in file " + green_filename + "..."     
command = "java -jar saxon/saxon9he.jar " + " -s:" + make_directory + green_filename + " -xsl:" + green_xslt + " > " + make_directory +cda_filename
#print command

# 
# Fire off a process to invoke java and run the the green CCD stylesheet
#       
p = subprocess.Popen(command, shell=True)
retval = p.wait()
print "Done. CDA patient data file " + cda_filename + " generated."
print
print