multi file generation based on the sections of the document

import sys

import docx

from docx.table import Table

from docx.text.paragraph import Paragraph

def  is_tosca_def(table):

    '''

    print 'Usage: doc2tosca <docx-with-tosca-definitions>'

    sys.exit(1)

OUT_FN = 'try-tosca-export.yaml'

OUT_FN_VNFD = 'try-tosca-export_vnfd.yaml'

OUT_FN_NSD  = 'try-tosca-export_nsd.yaml'

OUT_FN_PNFD = 'try-tosca-export_pnfd.yaml'

OUT_FN_COMM = 'try-tosca-export_comm.yaml'

SOL001 = docx.Document(SOL001_FN)

DEFINITIONS = [t for t in SOL001.tables if is_tosca_def(t)]

F = open(OUT_FN, 'w')

F_VNFD = open(OUT_FN_VNFD, 'w')

F_NSD = open(OUT_FN_NSD, 'w')

F_PNFD = open(OUT_FN_PNFD, 'w')

F_COMM = open(OUT_FN_COMM, 'w')

HDR='''tosca_definitions_version: tosca_simple_yaml_1_2

HDR_VNFD ='''tosca_definitions_version: tosca_simple_yaml_1_2

description: ETSI NFV SOL 001 nsd types definitions version 2.5.1

imports:

data_types:

'''

F.write(HDR)

HDR_NSD ='''tosca_definitions_version: tosca_simple_yaml_1_2

description: ETSI NFV SOL 001 nsd types definitions version 2.5.1

imports:

  - etsi_nfv_sol001_vnfd_2_5_1_types.yaml

data_types:

'''

HDR_PNFD ='''tosca_definitions_version: tosca_simple_yaml_1_2

description: ETSI NFV SOL 001 nsd types definitions version 2.5.1

imports:

  - etsi_nfv_sol001_vnfd_2_5_1_types.yaml

data_types:

'''

HDR_COMM ='''tosca_definitions_version: tosca_simple_yaml_1_2

description: ETSI NFV SOL 001 nsd types definitions version 2.5.1

for t in DEFINITIONS:

imports:

  - etsi_nfv_sol001_vnfd_2_5_1_types.yaml

data_types:

'''

F_VNFD.write(HDR_VNFD)

F_NSD.write(HDR_NSD)

F_PNFD.write(HDR_PNFD)

F_COMM.write(HDR_COMM)

p_id = 0

cur_sect = "0"

from sets import Set

def get_content(doc):

    ret = []

    body = doc._body

    par = 0

    tables = 0

    for element in body._element:

        if isinstance(element, docx.oxml.text.paragraph.CT_P):

            ret.append(Paragraph(element, body))

            par = par + 1

        elif isinstance(element, docx.oxml.table.CT_Tbl):

            ret.append(Table(element, body))

            tables = tables + 1

        else:

            print "FOUND " +  str(type(element))

    print "pars " + str(par)

    print "tables " + str(tables)

    return ret

content = get_content(SOL001)

tables=0

while p_id < len(content):

    elem = content[p_id]

    if isinstance(elem, Paragraph) and elem.text == "Foreword":

        break

    p_id = p_id + 1

if p_id >= len(content):

    print "FOREWORD NOT FOUND"

def find_sect(sect_to_find, start_idx, doc_content):

    while start_idx < len(doc_content):

        my_elem = content[start_idx]

        if isinstance(my_elem, Paragraph) and my_elem.text == sect_to_find:

            break

        start_idx = start_idx + 1

    print "FOUND " + sect_to_find + " at " + str(start_idx)

    return start_idx

sect_6_id = find_sect("6\tVNFD TOSCA model", p_id, content)

sect_7_id = find_sect("7\tNSD TOSCA model", sect_6_id + 1, content)

sect_8_id = find_sect("8\tPNFD TOSCA model", sect_7_id + 1, content)

annex_a_id = find_sect("Annex A (informative):", sect_7_id + 1, content)

def write_table_to_file(t, F):

    F.write(t.rows[0].cells[0].text.encode('utf-8'))

    F.write('\n# -------------------- #\n')

F.write('\n')

F.close()

def generate_tables_between(a_id, b_id, FDESC, fn):

    count=0

    for idx in range(a_id, b_id):

        tmp_elem = content[idx]

        if isinstance(tmp_elem, Table) and is_tosca_def(tmp_elem):

            write_table_to_file(tmp_elem, FDESC)

            count = count + 1

    print "Printed " + str(count) + " type to " + fn 

generate_tables_between(sect_6_id, sect_7_id, F_VNFD, "VNFD")

generate_tables_between(sect_7_id, sect_8_id, F_NSD, "NSD")

generate_tables_between(sect_8_id, annex_a_id, F_PNFD, "PNFD")

F_VNFD.write('\n')

F_NSD.write('\n')

F_PNFD.write('\n')

F_COMM.write('\n')

F_VNFD.close()

F_NSD.close()

F_PNFD.close()

F_COMM.close()