mirror of
https://github.com/rhasspy/espeak-ng.git
synced 2026-10-31 13:38:47 +00:00
78 lines
2.4 KiB
Python
Executable File
78 lines
2.4 KiB
Python
Executable File
#!/usr/bin/python3
|
|
|
|
import os
|
|
import re
|
|
import sys
|
|
import math
|
|
import codecs
|
|
|
|
import xml.etree.ElementTree as etree
|
|
|
|
class Emoji:
|
|
def __init__(self, m):
|
|
self.emoji = m.group(1)
|
|
self.pronunciation = m.group(2)
|
|
self.codepoints = m.group(3)
|
|
self.comment = m.group(4)
|
|
|
|
def __repr__(self):
|
|
return "Emoji(emoji={0}, pronunciation={1}, codepoints={2}, comment={3})".format(
|
|
repr(self.emoji),
|
|
repr(self.pronunciation),
|
|
repr(self.codepoints),
|
|
repr(self.comment))
|
|
|
|
def __str__(self):
|
|
return "{0}{1}// [{2}]{3}".format(self.emoji, self.pronunciation, self.codepoints, self.comment)
|
|
|
|
def read_annotations(filename):
|
|
ldml = etree.parse(filename).getroot()
|
|
for annotations in ldml.findall("annotations"):
|
|
for annotation in annotations.findall("annotation"):
|
|
if annotation.attrib.get("type", "") == "tts":
|
|
yield annotation.attrib["cp"], annotation.text
|
|
|
|
def read_emoji(filename, encoding="utf-8"):
|
|
re_emoji = re.compile(r"^([^ \t]*)([^/]*)// \[([^\]]*)\](.*)$")
|
|
with codecs.open(filename, "r", encoding) as f:
|
|
for line in f:
|
|
line = line.replace("\n", "")
|
|
if line.strip() == "":
|
|
yield line # blank line
|
|
elif line.startswith("//"):
|
|
yield line # line comment
|
|
elif line.startswith("$"):
|
|
yield line # flags only
|
|
else:
|
|
m = re_emoji.match(line)
|
|
if m:
|
|
yield Emoji(m)
|
|
else:
|
|
yield line
|
|
|
|
emoji_dict = sys.argv[1]
|
|
lang = sys.argv[2]
|
|
cldr_path = sys.argv[3]
|
|
|
|
annotations = {}
|
|
filename = os.path.join(cldr_path, "common", "annotations", "{0}.xml".format(lang))
|
|
for cp, name in read_annotations(filename):
|
|
annotations[cp] = name
|
|
|
|
for entry in read_emoji(emoji_dict):
|
|
if isinstance(entry, Emoji):
|
|
translation = annotations.get(entry.emoji, None)
|
|
if translation:
|
|
length = len(entry.pronunciation.strip())
|
|
tabs = entry.pronunciation.count('\t') - 1
|
|
first_tab = 8 - (length % 8)
|
|
tab_length = length + first_tab + ((tabs - 1) * 8)
|
|
|
|
new_length = len(translation)
|
|
new_tabs = math.ceil((tab_length - new_length)/8)
|
|
|
|
entry.pronunciation = "\t{0}{1}".format(translation, "\t"*int(new_tabs))
|
|
else:
|
|
entry.comment += " (no translation)"
|
|
print(entry)
|