Metacat

#!/usr/bin/python

#

# TODO:

#  validate

#  getNextDocid

#  getNextRevision(docid)

#  getDocids(keyword=None)

#  query(keyword, returnfields)

#  queryDict => same as above but returns python dictionary data structure

import httplib, urllib

class MetacatClient:

    def __init__(self, server="pmark.msi.ucsb.edu:8180", urlPath="/knb_test/metacat"):

        self.metacatUrlPath = urlPath

        self.metacatServer = server

        self.sessionid = None

    def getMetacatUrl(self):

        return "http://" + self.metacatServer +  self.metacatUrlPath

    def login(self, username, password, organization=None):

        if organization == 'NCEAS':

            uid = 'uid=%s,o=NCEAS,dc=ecoinformatics,dc=org' % username

        else:

            uid = username

        postdata = { 'action'   : 'login',

                     'qformat'  : 'xml',

                     'username' : uid,

                     'password' : password }

        response = self.postRequest(postdata) 

        if response.find("<login>") != -1:

            return True

        else:

            return False

    def logout(self):

        postdata = { 'action'   : 'logout',

                     'qformat'  : 'xml'}

        response = self.postRequest(postdata) 

        if response.find("<logout>") != -1:

            return True

        else:

            return False

    def read(self, docid, qformat="xml"):

        postdata = { 'action'   : 'read',

                     'qformat'  : qformat,

                     'docid'    : docid }

        response = self.postRequest(postdata) 

        # if error node returned

        if response.find("<error>") != -1:

            return False

        else:

            return response

    def insert(self, docid, doctext):

        postdata = { 'action'   : 'insert',

                     'doctext'  : doctext,

                     'docid'    : docid }

        response = self.postRequest(postdata) 

        # if error node returned

        if response.find("<error>") != -1:

            return False

        else:

            return response

    def update(self, docid, doctext):

        postdata = { 'action'   : 'update',

                     'doctext'  : doctext,

                     'docid'    : docid }

        response = self.postRequest(postdata) 

        return response

    def delete(self, docid):

        postdata = { 'action'   : 'delete',

                     'docid'    : docid }

        response = self.postRequest(postdata) 

        return response

    def squery(self, pathquery, qformat="xml"):

        postdata = { 'action'   : 'squery',

                     'qformat'  : qformat,

                     'query'    : pathquery }

        response = self.postRequest(postdata) 

        return response

    def postRequest(self, postdata):

        conn = httplib.HTTPConnection( self.metacatServer )

        params = urllib.urlencode( postdata )

        headers = { "Content-type" : "application/x-www-form-urlencoded", 

                    "Accept"       : "*/*"}

        # If we have an active session, set the cookie

        if self.sessionid is not None:

            headers['Cookie'] = self.sessionid

        conn.request( "POST", self.metacatUrlPath, params, headers )

        response = conn.getresponse()

        # If metacat responds with a new session id,

        # register it with the metacat client instance

        setcookie = response.getheader("set-cookie", None)

        if setcookie:

            jsid = setcookie.split(';')[0]

            if jsid[:11] == "JSESSIONID=":

               self.sessionid = jsid

        if response.status == 200:

           content = response.read()

        else:

           print " !!! SERVER DID NOT RETURN 'OK'.... STATUS is " + str(response.status) 

           content = ""

        conn.close()

        return content

#!/usr/bin/python

"""

 Transfers documents from one metacat server to another

 Author: matt perry

 Date: 09/28/06

"""

import metacat

import sys

try:

    docidlist = sys.argv[1]

except:

    print

    print " usage: python transferEml.py docidlist.txt "

    print "        where docidlist.txt is a line-delimited list of docids "

    print

    sys.exit(1)

print "Logging into destination server"

dest = metacat.MetacatClient('yourserver:8180','/knb/metacat')

dest.login('user','password','NCEAS')

print "Logging into source server"

src = metacat.MetacatClient('knb.ecoinformatics.org', '/knb/metacat')

src.login('user','password','NCEAS')

print "Open docid list and loop"

for docid in open(docidlist,'r').readlines():

    docid = docid.strip().replace('\n','')

    if docid != '' and docid is not None:

       print " reading docid " + docid

       eml = src.read(docid) 

       if eml:

           print " inserting docid " + docid

           response = dest.insert(docid + ".1" ,eml)

           if not response:

              print "   insert failed"

       else:

           print "   docid " + docid + " not found"

print "Logging out"

dest.logout()

src.logout()

PyMetacat 

Python library to work with Metacat XML Databases.

Author: Matthew Perry

Date: 12/06/2006

Introduction

-------------

Metacat is a flexible XML database. It can store, search and rerieve XML documents of any schema and it used primarily for storing metadata documents. This library provides a python interface to any Metacat server allowing you to add, insert, update, delete and query documents.

Installation

------------

To install pymetcat, copy the metacat.py file into your python site-packages directory (or anywhere in $PYTHONPATH ).

Examples

------------

Along with transferEml.py (a full working example of pymetacat usage), here are some basic usage examples:

##########################################################################################

import metacat

# Initialize

mc = metacat.MetacatClient('www.server.net','/context/metacat')

# Print the url

print mc.getMetacatUrl()

# http://www.server.net/context/metacat

# Login

response = mc.login('uid=user,o=ORG,dc=ecoinformatics,dc=org','PassW0Rd!')

print "login : ", response

# Read the eml 

docid = 'org.99.4.1'

eml = mc.read(docid)

#print eml

# Insert the read document as a new copy w/ new docid

newdocid = 'anotherorg.905'

response = mc.insert(newdocid + ".1",eml)

print response

# Make some change to the xml text and update it

eml.replace("Serengeti", "Some other")

response = mc.update(newdocid + ".2",eml)

print response

# Delete the document

response = mc.delete(newdocid + ".2")

print "DELETE"

print response

pathquery = """

   <pathquery version="1.0">

      <meta_file_id>unspecified</meta_file_id>

      <querytitle>unspecified</querytitle>

      <returnfield>dataset/title</returnfield>

      <returndoctype></returndoctype>

      <querygroup operator="UNION">

              <queryterm casesensitive="false" searchmode="contains">

                <value>%</value>

              </queryterm>

        <queryterm casesensitive="false" searchmode="contains">

          <value>Africa</value>

          <pathexpr>keyword</pathexpr>

        </queryterm>

      </querygroup>

    </pathquery> 

"""

response = mc.squery(pathquery)

print "==============="

print response

print "==============="

response = mc.logout()

print "logout : ", response

#########################################################################################