#!/usr/bin/python3

from lxml import etree
from lxml.etree import tostring

filemakerxml = 'meksme-utf8.xml'

tree = etree.parse(filemakerxml)
root = tree.getroot()

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

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

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 langsort(lang, e):
    if lang in e:
        return e[lang]
    else:
        return e['sme']

def make_glossary_docbook(lang):
    import lxml.builder
    E = lxml.builder.ElementMaker(
        nsmap={
#            'xi': "http://www.w3.org/2001/XInclude",
        }
    )

    langcodes = ('en', 'nb', 'sme', 'sv', 'fi',)

    def indexit(entry, wlist, lang=None):
        for w in wlist.split(","):
            if "" != w:
                if lang and '[' not in w:
                    w += "[%s]" % lang
                entry.append(E.indexterm(E.primary(w)))
    glossary = E.glosslist()
    for e in sorted(words, key=lambda x: langsort(lang, x)):
        if 'topic' not in e:
            e['topic'] = 'n/a'
        if lang in e and 'desc-%s' % lang in e:
            entry = E.glossentry(
                E.glossterm('%s [%s]' % (e[lang], e['topic'])),
            )
            indexit(entry, e[lang])
            lstr = ""
            for l in langcodes:
                if l != lang and l in e:
                    lstr += "%s (%s) " % (e[l], l)
                    # Add foreign words to index, split on comma
                    indexit(entry, e[l], l)
            entry.append(E.glossdef(E.para(e['desc-%s' % lang])))
            if "" != lstr:
                entry.append(E.glossdef(E.para(lstr)))
            glossary.append(entry)

    content = lxml.etree.tostring(glossary,
                                  pretty_print=True,
                                  xml_declaration=True,
                                  encoding='UTF-8')
#    print(content)
    with open('glossary.xml', 'wb') as f:
        f.write(content)

def make_glossary(lang):
    make_glossary_docbook(lang)

if True:
    print("Norsk/bokmål")
    print()
    make_glossary(lang='nb')
else:
    print("Nordsamisk")
    print()
    make_glossary(lang='sme')

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