clean up and documentation for doc2tosca

#!/bin/python2.7

#!/bin/python3.5

'''

Generate tosca definitions from Docx specfication

'''

import sys

from io import BytesIO as StringIO

import docx

from docx.table import Table

from docx.text.paragraph import Paragraph

BASE_FILENAME = "try-tosca-export_{}.yaml"

TOSCA_VERSION = "tosca_simple_yaml_1_2"

SPEC_VERSION = "2.5.1"

MODEL_NAMES = ['vnfd', 'nsd', 'pnfd', 'common']

HDR = '''tosca_definitions_version: {tosca_version}

description: ETSI NFV SOL 001 {model} types definitions version {spec_version}

imports:

{imports}

data_types:

'''

def  is_tosca_def(table):

    '''

    Returns true when a table contains TOSCA definitions, i.e.

    the table contains just one cell and text starts with an

    empty space ' '' 

    empty space ' '

    '''

    txt = table.rows[0].cells[0].text[0]

    return \

    len(table.columns) == 1 and \

    txt.startswith(' ')

try:

    SOL001_FN = sys.argv[1] 

except:

    print 'Error: Filename missing or filename not a docx document'

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

    sys.exit(1)

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)

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_VNFD ='''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:

def tosca_model_info(name, imports):

    '''

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:

    Returns a dictionary for the information on the model

    '''

    return {

        'name' : name,

        'fn' : BASE_FILENAME.format(name),

        'fd' : None,

        'imports' : imports,

        'buf' :  StringIO()

    }

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:

def get_content(doc):

    '''

HDR_COMM ='''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:

    Returns a list of all paragraphs and tables in the Document

    '''

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

    parag_count = 0

    table_count = 0

    for element in body._element:

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

            ret.append(Paragraph(element, body))

            par = par + 1

            parag_count = parag_count + 1

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

            ret.append(Table(element, body))

            tables = tables + 1

            table_count = table_count + 1

        else:

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

    print "pars " + str(par)

    print "tables " + str(tables)

            print "Non paragraph or table " +  str(type(element))

    print "Paragraphs: " + str(parag_count)

    print "Tables: " + str(table_count)

    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):

    '''

    Returns the index in the doc_content list to the first paragraph

    or heading of the section with title sect_to_find,

    starting the research from start_idx

    '''

    while start_idx < len(doc_content):

        my_elem = content[start_idx]

        my_elem = doc_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)

def write_table_to_file(tab, buf):

    '''

    Writes content of table t in utf-8 encoding to file F

    '''

    buf.write(tab.rows[0].cells[0].text.encode('utf-8'))

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

def generate_tables_between(a_id, b_id, content, buf):

    '''

    Loops over content and writes all tosca definitions to the

    fdesc file. Returns the number of written definitions

    '''

    definitions_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, buf)

            definitions_count = definitions_count + 1

    return definitions_count

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

def dump_header(model_name, buf, imports):

    '''

    Writes the header to the file for a specific model

    '''

    buf.write(HDR.format(

        tosca_version=TOSCA_VERSION,

        model=model_name,

        spec_version=SPEC_VERSION,

        imports=imports))

MODELS = {}

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

for mn in MODEL_NAMES:

    MODELS[mn] = tosca_model_info(mn, '- etsi_nfv_sol001_common_types.yaml')

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')

if __name__ == "__main__":

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 

    try:

        SOL001_FN = sys.argv[1]

    except:

        print 'Error: Filename missing or filename not a docx document'

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

        sys.exit(1)

    print "Opening " + SOL001_FN

    SOL001 = docx.Document(SOL001_FN)

    for m in MODELS:

        dump_header(MODELS[m]['name'], MODELS[m]['buf'], MODELS[m]['imports'])

    p_id = 0

    cur_sect = "0"

    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"

    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)

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")

    count = generate_tables_between(sect_6_id, sect_7_id, CONTENT, MODELS['vnfd']['buf'])

    print "Printed " + str(count) + " types to " + "VNFD"

F_VNFD.write('\n')

F_NSD.write('\n')

F_PNFD.write('\n')

F_COMM.write('\n')

    count = generate_tables_between(sect_7_id, sect_8_id, CONTENT, MODELS['nsd']['buf'])

    print "Printed " + str(count) + " types to " + "NSD"

F_VNFD.close()

F_NSD.close()

F_PNFD.close()

F_COMM.close()

    count = generate_tables_between(sect_8_id, annex_a_id, CONTENT, MODELS['pnfd']['buf'])

    print "Printed " + str(count) + " types to " + "PNFD"

    for m in MODELS:

        MODELS[m]['fd'] = open(MODELS[m]['fn'], 'w')

        MODELS[m]['buf'].seek(0)

        MODELS[m]['fd'].write(MODELS[m]['buf'].read())

        MODELS[m]['fd'].write('\n')

        MODELS[m]['fd'].close()