python3: read xml

clean_words_merged.py

#!/usr/bin/python3
# -*- coding: utf-8 -*-

from lxml import etree
import re

file = open("words_merged.xml", "r")
dom = etree.parse(file)
wl = dom.xpath('//wordlist/w')

# Remove all words that have not allowed characters (specified in letter_allowed)
# we will unify all the letters into a REGEX
letter_allowed = "ABDCĈEFGĜHĤIJĴKLNMOPRSŜTUŬVZabdcĉefgĝhĥijĵklnmoprsŝtuŭvz'-" # REGEX to search any other letter
letter_allowed = "[^"+re.escape(letter_allowed)+"]+"
vowels_allowed = "[AEIOUaeiou]+"

for w in wl:
    conditions = (
        re.search(letter_allowed, w.text) != None or        # letter not allowed
        len(w.text)==1 or                                   # lengh = 1
        re.search(vowels_allowed, w.text) == None or        # there is no vowel
        (len(w.text)==2 and re.search("['\-]+", w.text))    # length = 2 and has ' or -
        )
    if conditions:
        w.getparent().remove(w) # remove the node

dom.write('clean_words_merged.xml', pretty_print=True, xml_declaration=True, encoding="utf-8") # save the output into clean_words_merged.xml

# Rename clean_words_merged.xml into words_merged.xml and use it

clean_words_merged.xml

<?xml version='1.0' encoding='UTF-8'?>
<ws>
    <w f="1">ks</w>
    <w f="65">la</w>
</ws>

words_merged.xml

<?xml version="1.0" encoding="UTF-8"?>
<ws>
    <w f="1">Veuchey</w>
    <w f="1">ks</w>
    <w f="65">la</w>
    <w f="65">ñ</w>
</ws>

kodea sinplifikatu dizut eta badaezpada "-" eskapatu "-" regex-ak egiteko letra berezia delako

beno dena jarri dizut if batean or-ekin oraindik kodea laburragoa da :D

Hasieran in nahi dituten ta nola ez dakiten bi baldintza daude.
Azpian ondo dabilen beste baldintza bat gehituet. Optimizatu daiteke baño nahiago det lenaoko funtzioaz aparte jarri, errexo kendu ahal izateko.

#!/usr/bin/python3
# -*- coding: utf-8 -*-

# ', " edo - hizkiakin hasi ta bukatze dianei hoiek kendu: 'hitza' edo "hitza" -hitza-  --> hitza
# ' hizkiakin haste dianei hoiek kendu: 'hitza  --> hitza

from lxml import etree
import re

file = open("words_merged.xml", "r")
dom = etree.parse(file)
wl = dom.xpath('//wordlist/w')

# REGEX
letter_allowed = "ABDCĈEFGĜHĤIJĴKLNMOPRSŜTUŬVZabdcĉefgĝhĥijĵklnmoprsŝtuŭvz'-"
letter_allowed = "[^"+re.escape(letter_allowed)+"]+"
vowels_allowed = "[AEIOUaeiou]+"
biCharacter_allowed = ['aĉ', 'aĥ', 'aj', 'al', 'aŭ', 'ba', 'be', 'ci', 'ĉe', 'ĉi', 'ĉu', 'da', 'de', 'do', 'du', 'eĉ', 'el', 'en', 'fi', 'fu', 'ĝi', 'ha', 'hm', 'ho', 'hu', 'je', 'ke', 'la', 'li', 'lo', 'mi', 'ne', 'ni', 'nu', 'oj', 'ok', 'ol', 'ri', 'se', 'si', 'ŝi', 'vi', 'uf', 'uk' 'ŭa']

for w in wl:
    conditions = (
        re.search(letter_allowed, w.text) != None or                    # letter not allowed
        len(w.text)==1 or                                               # lengh = 1
        re.search(vowels_allowed, w.text) == None or                    # there is no vowel
        (len(w.text)==2 and re.search("['\-]+", w.text)) or             # length = 2 and has ' or -
        (len(w.text)==2 and (w.text).lower not in biCharacter_allowed)  # length = 2 and not allowed bi-character
        )
    if conditions:
        w.getparent().remove(w) # remove the node

dom.write('clean_words_merged.xml', pretty_print=True, xml_declaration=True, encoding="utf-8") # save the output into clean_words_merged.xml

# Rename clean_words_merged.xml into words_merged.xml and use it

Gehitueten azkenak ez dit ondo funtzionatu, zerrendan dauden hitzak ere ezabatu dizkit ta

arazoa "lower" ez dela propietatea, metodoa baizik, beraz erabili "lower()"

#!/usr/bin/python3
# -*- coding: utf-8 -*-

import sys
from lxml import etree
import re

file = open("words_merged.xml", "r")
dom = etree.parse(file)

# hasierako eta bukaerako # eta ' kendu

letters_to_remove = "'#"

wl = dom.xpath('//wordlist/w')
for w in wl:
	conditions = (
		re.search("^["+letters_to_remove+"]{1}", w.text) != None and re.search("["+letters_to_remove+"]{1}$", w.text) != None
		)
	if conditions:
		w.text = re.sub("^['#]{1}", '', w.text)
		w.text = re.sub("['#]{1}$", '', w.text)

# errepikatuak kendu

all = {};
repeated = [];
wl = dom.xpath('//wordlist/w')
for w in wl:
	text = w.text
	if text in all:
		all[text] = all[text] + 1
		if text not in repeated:
			repeated.append(text)
	else:
		all[text] = 1

for repeat in repeated:
	while True:
		expr = '//wordlist/w[.= $var]'
		result = dom.xpath(expr, var = repeat)
		result_len = len(result)
		if result_len > 1:
			remove_index = result_len - 1
			remove_node = result[remove_index]
			remove_node.getparent().remove(remove_node)
		else:
			break

dom.write('clean_words_merged.xml', pretty_print=True, xml_declaration=True, encoding="utf-8") # save the output into clean_words_merged.xml

# Rename clean_words_merged.xml into words_merged.xml and use it