#!/usr/bin/env python

# Imports an XML file into a PostgreSQL database

# Format: see http://vegbank.org/vegdocs/xml/vegbank_example_ver1.0.2.xml

import os

import os.path

import psycopg2

from psycopg2.extensions import ISOLATION_LEVEL_SERIALIZABLE

import sys

import xml.dom.minidom

sys.path.append(os.path.dirname(__file__)+"/lib")

import xml_db

def env_flag(name): return name in os.environ and os.environ[name] != ''

def main():

    prog_name = sys.argv.pop(0)

    try:

        db_config = {}

        for name in ['host', 'user', 'password', 'database']:

            if os.environ[name] != '': db_config[name] = os.environ[name]

    except KeyError: raise Exception('Usage: env host=... user=... password=...'

        ' database=... [commit=1] '+prog_name+' <dataset')

    commit = env_flag('commit')

    # Process dataset

    db = psycopg2.connect(**db_config)

    db.set_isolation_level(ISOLATION_LEVEL_SERIALIZABLE)

    try:

        doc = xml.dom.minidom.parse(sys.stdin)

        row_ct_ref = [0]

        xml_db.xml2db(db, doc.documentElement, row_ct_ref)

        print 'Inserted '+str(row_ct_ref[0])+' rows'

        if commit: db.commit()

    finally:

        db.rollback()

        db.close()

main()

#!/usr/bin/env python

# Converts a CSV dataset to XML using a mappings spreadsheet

import csv

import os.path

import re

import sys

from copy import deepcopy

from xml.dom.minidom import getDOMImplementation

sys.path.append(os.path.dirname(__file__)+"/lib")

import xpath

def main():

    prog_name = sys.argv.pop(0)

    try:

        dest = sys.argv.pop(0)

        mappings_path = sys.argv.pop(0)

    except IndexError: raise Exception('Usage: '+prog_name

        +' dest_mappings_column mappings_path <dataset >output')

    # Get mappings

    mappings = {}

    has_types = False # whether outer elements are type containiners

    stream = open(mappings_path, 'rb')

    reader = csv.reader(stream, delimiter=',', quotechar='"')

    fieldnames = reader.next()

    src = fieldnames[0]

    dest_idx = fieldnames.index(dest)

    for row in reader:

        name = row[0]

        path = row[dest_idx]

        if name != '' and path != '':

            if path.startswith('/*s/'): has_types = True # *s used for type elem

            path = path.replace('<name>', name)

            mappings[name] = xpath.XpathParser(path).parse()

    stream.close()

    # Process dataset

    doc = getDOMImplementation().createDocument(None, dest, None)

    stream = sys.stdin

    reader = csv.reader(stream, delimiter=',', quotechar='"')

    fieldnames = reader.next()

    row_idx = 0

    for row in reader:

        row_id = str(row_idx)

        for idx, name in enumerate(fieldnames):

            value = row[idx]

            if value != '' and name in mappings:

                path = deepcopy(mappings[name]) # don't modify main value!

                xpath.set_id(path, row_id, has_types)

                xpath.set_value(path, value)

                xpath.get(doc, path, True)

        row_idx += 1

    doc.writexml(sys.stdout, addindent='    ', newl='\n')

main()