#!/usr/bin/python3

from lxml import etree
from lxml.etree import tostring

tree = etree.parse('mekanikk-1999/meksme-utf8.xml')
root = tree.getroot()

#print(root)
#print(tostring(tree))

cols = (
    'topic', 'sme', 'desc-sme', 'desc-nb', 'nb', 'sv', 'fi', 'en', 'is',
    'unknown',
)

resultset = root.find("{http://www.filemaker.com/fmpxmlresult}RESULTSET")

words = []
for row in resultset.getchildren():
    d = {}
    index = 0
    for col in row.findall("{http://www.filemaker.com/fmpxmlresult}COL"):
        t = col.getchildren()[0].text
        if t:
            import re
            t = re.sub(r'\s+', ' ', t)
            d[cols[index]] = t
        index += 1
    #print(d)
    words.append(d)

def make_glossary(lang):
    print(".. glossary::")
    print()

    def langsort(e):
        if lang in e:
            return e[lang]
        else:
            return e['sme']
    for e in sorted(words, key=langsort):
        if lang in e and 'desc-%s' % lang in e:
            if 'topic' not in e:
                e['topic'] = 'n/a'
            #print(e)
            print("  %s [%s]\n     %s" % (e[lang], e['topic'], e['desc-%s' % lang]))
            print()
        else:
            # ERROR / missing definition
            pass

print("Nordsamisk")
print("==========")
print()
make_glossary(lang='sme')

print("Norsk")
print("=====")
print()
make_glossary(lang='nb')

#print("Engelsk")
#print("=====")
#print()
#make_glossary(lang='en')