#! /usr/bin/env python -w
# -*- coding: utf-8 -*- 
#
# Script for extracting information from Oahpa -lexicons.
# Usage: python lcat.py -h


import sys
import re
from xml.dom import minidom as _dom
from optparse import OptionParser


parser = OptionParser()

parser.add_option("-f", "--file", dest="infile",
                  help="lexicon file name")
parser.add_option("-b", "--book", dest="book",
                  help="book name")
parser.add_option("-l", "--lemma", dest="lemma",
                  help="Lemma string")
parser.add_option("-v", "--valency", dest="valency",
                  help="Valency class")
parser.add_option("-s", "--semantics", dest="semclass",
                  help="Semantic class")
parser.add_option("-t", "--stem", dest="stemclass",
                  help="Stem class")
parser.add_option("-d", "--dialect", dest="dialect",
                  help="Dialect class")
parser.add_option("-r", "--regex", dest="regex",
                  action="store_true", default=False,
                  help="Search for match for lemma and book information")
parser.add_option("-a", "--all", dest="print_all",
                  action="store_true", default=False,
                  help="print the whole entry, not just lemma")

(options, args) = parser.parse_args()

# Store options recieved from user
option={}
entries={}
all=[]
for opt, value in options.__dict__.items():
    if value and opt != "infile" and opt != "print_all" and opt!="regex":
        option[opt] = value
        entries[opt] = []

# The options give conditions for the entry they are combined with
# AND operation, thus forming an intersection of all the entries.
# This does not work with multiple options of the same type, is is still OR.
def print_entries(entries, all):
    allset = set(all)
    for key in entries.keys():
        elist = entries[key]
        eset = set(elist)
        allset =  allset & eset

    for e in list(allset):
        if options.print_all:
            print e.toxml(encoding="utf-8")
        else:
            ltext=e.getElementsByTagName("lemma")[0].firstChild.data
            print ltext.encode('utf8')

xmlfile=file(options.infile)
tree = _dom.parse(options.infile)

# Go through all the entries and check if they satisfy the search criteria.
# Store entries that fill different search criteria to separate dictionary entries
# The search results are combined in function print_entries.
for e in tree.getElementsByTagName("entry"):
    all.append(e)
    
    if options.book:
        s=e.getElementsByTagName("sources")[0]
        for b in  s.getElementsByTagName("book"):
            if options.regex:
                matchObj = re.search(options.book, b.getAttribute("name").encode('utf8'))
                if matchObj:
                    entries['book'].append(e)
            else:
                if b.getAttribute("name").encode('utf8') == options.book:                
                    entries['book'].append(e)
                            
    if options.lemma:
        ltext=e.getElementsByTagName("lemma")[0].firstChild.data
        if options.regex:
            matchObj = re.search(options.lemma, ltext)
            if matchObj:
                entries['lemma'].append(e)
        else:
            if ltext and ltext == options.lemma:
                entries['lemma'].append(e)
                    
    if options.valency:
        v=e.getElementsByTagName("valency")[0]
        for b in v.getElementsByTagName("val"):
            if b.getAttribute("class").encode('utf8') == options.valency:
                entries['valency'].append(e)

    if options.semclass:
        s=e.getElementsByTagName("semantics")[0]
        for b in  s.getElementsByTagName("sem"):
            if b.getAttribute("class").encode('utf8') == options.semclass:
                entries['semclass'].append(e)

    if options.stemclass:
        s=e.getElementsByTagName("stem")[0]
        if s.getAttribute("class").encode('utf8') == options.stemclass:
            entries['stemclass'].append(e)

    if options.dialect:
        s=e.getElementsByTagName("dialect")[0]
        if s.getAttribute("class").encode('utf8') == options.dialect:
            entries['dialect'].append(e)
                    
  
print_entries(entries,all)