Python script to parse SFDC metadata on reports to determine which fields are being used.

reportfields.py

#!/usr/bin/python
#---------------------------------------------------------------------------
# This python script traverses the reports directory of the SFDC metadata.
# It goes through each directory and reports, and pulls out fields
# used in the report. A file called reporting_fields_used.csv is generated.
#---------------------------------------------------------------------------
import os
import sys
import xml.etree.ElementTree as ET
from StringIO import StringIO

# start traversing in the directory the script is in
walk_dir = "."

# generate a file with this name
list_file_path = ('reporting_fields_used.csv')
print('Generating file:  ' + list_file_path)

# open file as output stream
with open(list_file_path, 'wb') as list_file:
    #for subdir in subdirs:
    #print('\t- subdirectory ' + subdir)

    # print the header of output file once
    list_file.write("file,report,type,field\n")

    # walk through directories using os.walk
    for root, subdirs, files in os.walk(walk_dir):

        for filename in files:
            file_path = os.path.join(root, filename)

            # only open up .report files
            if ".report" in file_path:

                with open(file_path, 'rb') as f:
                    f_content = f.read()

                    # read the XML from the file handle and then create the root from string
                    # be careful here. Since we are traversing directories, we have another variable called
                    # root for the root directory.  So we call the XML root "xmlroot".
                    xmlroot = ET.fromstring(f_content)

                    # SFDC report metadata XML looks something like this
                    #<Report xmlns="http://soap.sforce.com/2006/04/metadata">
                    #    <columns>
                    #        <field>USERS.NAME</field>
                    #    </columns>
                    #    <filter>
                    #        <booleanFilter>1 AND (2 OR 3)</booleanFilter>
                    #        <criteriaItems>
                    #            <column>Address_vod__c.Primary_vod__c</column>
                    #            <operator>equals</operator>
                    #            <value>1</value>
                    #        </criteriaItems>
                    #    <groupingsDown>
                    #        <dateGranularity>Day</dateGranularity>
                    #        <field>ACCOUNT.NAME</field>
                    #        <sortOrder>Asc</sortOrder>
                    #    </groupingsDown

                    # parse out reporting columns from columns/field
                    for columns in xmlroot.findall('./{http://soap.sforce.com/2006/04/metadata}columns'):
                        #print ('columns: %s' % columns)
                        for field in columns:

                            s = '%s,%s,%s,%s\n' % (file_path,filename, "columns|field", field.text)
                            list_file.write(s)
                            #print ('%s' % s)

                    # parse out reporting columns from GroupingsDown/field
                    for columns in xmlroot.findall('./{http://soap.sforce.com/2006/04/metadata}groupingsDown/.'):
                        #print ('columns: %s' % columns)
                        for child in columns:

                            if "field" in child.tag:
                                #print ('child %s' % child.text  )
                                s = '%s,%s,%s,%s\n' % (file_path,filename, "groupingsDown|field", child.text)
                                list_file.write(s)
                                #print ('%s' % s)

                    # parse out reporting columns from GroupingsAcross/field
                    for columns in xmlroot.findall('./{http://soap.sforce.com/2006/04/metadata}groupingsAcross/.'):
                        #print ('columns: %s' % columns)
                        for child in columns:

                            if "field" in child.tag:
                                #print ('child %s' % child.text  )
                                s = '%s,%s,%s,%s\n' % (file_path,filename, "groupingsAcross|field", child.text)
                                list_file.write(s)
                                #print ('%s' % s)

                    # parse out reporting columns from filter/criteriaItems/column
                    for filters in xmlroot.findall('./{http://soap.sforce.com/2006/04/metadata}filter/.'):
                        # print ('columns: %s' % filters)
                        for criteriaItems in filters:
                            if "criteriaItems" in criteriaItems.tag:
                                # print ('criteria item: %s , %s' % (criteriaItems.tag , criteriaItems.text))
                                for child in criteriaItems:
                                    # if column
                                    if "column" in child.tag:
                                        # print ('child %s' % child.text)
                                        s = '%s,%s,%s,%s\n' % (file_path,filename, "filter|criteriaItem|column", child.text)
                                        list_file.write(s)
                                        # print ('%s' % s)