Vérificateur d'actus Python + Vocal

crontab

*/5 * * * * export DISPLAY=:0 && ~/ih_news_checker/ih_news_checker.py 2>&1 | /usr/bin/logger -t ih_news_checker

ih.py

#!/usr/bin/env python
# -*- coding: utf-8 -*

import xml.etree.ElementTree as tree
import urllib2
import json
import os

FILE_GUID = os.path.expanduser("~") + "/.ih_news_guid_list.txt"
URL_FEED = "https://api-v1.inpact-hardware.com/rss/news.xml"

def _check_last_items():	
	_items = tree.parse(urllib2.urlopen(URL_FEED)).getroot().iter("item")

	_news = list()
	_guids = list()
	for i in _items:
		_news.append({	"title":i.findtext("title"),
				"link":i.findtext("link"),
				"link_comments":i.findtext("comments"),
				"guid":i.findtext("guid"),
				"description":i.findtext("description"),
				"creator":i.findtext("{http://purl.org/dc/elements/1.1/}creator"),
				"date":i.findtext("pubDate"),
				"image":i.find("enclosure").attrib["url"]})

		_guids.append(i.findtext("guid"))

	return {"news":_news,"guids":_guids}

def _read_already_seen_guids():
	result = list()

	if os.path.isfile(FILE_GUID):
		with open(FILE_GUID,"r") as f: result = f.readlines()

	return result

def _extract_news(_feed_content, _guids_list):
	_feed_guids_with_newline = ["%s\n" % guid for guid in _feed_content["guids"]]
	_new_guids = list(set(_feed_guids_with_newline) - set(_guids_list))
	
	_result = list()
	for n in _feed_content["news"]:
		if n["guid"] + "\n" in _new_guids: _result.append(n)

	return _result

def check_news():
	_feed_items = _check_last_items()
	_already_seen_guids = _read_already_seen_guids()

	_new_news = list()	
	if len(_already_seen_guids) == 0:
		_already_seen_guids = ["%s\n" % guid for guid in _feed_items["guids"]]
		with open(FILE_GUID,"w") as f: f.writelines(_already_seen_guids)
	else:
		_new_news = _extract_news(_feed_items, _already_seen_guids)
		for n in _new_news:
			with open(FILE_GUID,"a") as f: f.writelines(n["guid"] + "\n")

	if len(_already_seen_guids) > 500:
		with open(FILE_GUID,"w") as f: f.writelines(["%s\n" % guid for guid in _feed_items["guids"]])

	return {"feed":_feed_items["news"],"new":_new_news}

if __name__ == "__main__": print json.dumps(check_news())

ih_news_checker.py

#!/usr/bin/env python
# -*- coding: utf-8 -*

import tts
import ih
import os 

news = ih.check_news()["new"]
text_to_read = None

if len(news) == 1:
	text_to_read = "INpact Hardware a publié une nouvelle actualité"
elif len(news) > 1 :
	text_to_read = "INpact Hardware a publié {} nouvelles actualités".format(len(news))

if text_to_read:
	tts.speak(text_to_read.decode("utf-8"))
	print text_to_read + " :"

	for n in news:
		tts.speak(n["title"])
		print n["title"].encode("utf-8")
		print n["link"].encode("utf-8")

	os.system("xdg-open https://www.inpact-hardware.com")

tts.py

#!/usr/bin/env python
# -*- coding: utf-8 -*

import os
MODULE_TO_USE = "gtts"
TMP_FILE = "/tmp/gTTS_temp_sample"

def _gtts(phrase_to_read):
	from gtts import gTTS
	tts = gTTS(text=phrase_to_read, lang="fr")
	tts.save(TMP_FILE + ".mp3")
	os.system("mpg321 -q {}{}".format(TMP_FILE, ".mp3"))

def _pico2wave(phrase_to_read): 
	os.system("pico2wave -l fr-FR -w {}{} \"{}\"".format(TMP_FILE, ".wav", phrase_to_read.encode("utf-8")))
	os.system("aplay -q {}{}".format(TMP_FILE, ".wav")) 

def _pyttsx(phrase_to_read):
	import pyttsx3
	engine = pyttsx3.init()
	engine.setProperty("voice", "french")
	engine.setProperty("rate", 100)
	engine.say(phrase_to_read)
	engine.runAndWait()

def speak(phrase_to_pass):
	if MODULE_TO_USE == "gtts": _gtts(phrase_to_pass)
	elif MODULE_TO_USE == "pico": _pico2wave(phrase_to_pass)
	else: _pyttsx(phrase_to_pass)