Merge pull request #2 from sentenzo/dev

`dev` => `master` (some features + refactoring)

config.yml

@@ -4,21 +4,33 @@ bot:
   token:  # 0123456789:will_be_taken_from_.env
   allowed_usernames: ["durov"] # will be enriched from .env
 
-utils:
+thirdparty:
   balcon:
     path: c:\bin\balcon.exe
+    defaults:
+      ru:
+        voice: Microsoft Irina Desktop
+        speed: 8
+        volume: 100
+        pitch: 0
+      en:
+        voice: Microsoft Zira Desktop
+        speed: 4
+        volume: 100
+        pitch: 0
   ffmpeg:
     path: c:\bin\ffmpeg.exe
   blb2txt:
     path: c:\bin\blb2txt.exe
 
-to_txt:
+text:
   max_input_size: 5242880 # 5 * 2 ^ 20 == 5 MiB 
+  input_formats: [".epub", ".fb2", ".fb3", ".md", ".txt", ".doc", ".docx", ".rtf", ".html", ".htm"]
 
-url_parser:
+web_parser:
   sites:
     habr:
-      url_re: https\://habr\.com/..(/company/[-\w]+/blog|/post)/\d+
+      url_re: https\://habr\.com/..(/company/[-\w]+/blog|/post|/news/t)/\d+
       lang: ru
       re:
         title: main h1[data-test-id="articleTitle"] span

narrator/exceptions.py

@@ -4,6 +4,9 @@ class NarratorException(Exception):
     """
 
 
+###
+
+
 class UrlParserException(NarratorException):
     pass
 
@@ -16,5 +19,12 @@ class UrlUnreachable(UrlParserException):
     pass
 
 
-class TxtTransformerException(NarratorException):
+class ParsingRulesNotFound(UrlParserException):
+    pass
+
+
+###
+
+
+class TextException(NarratorException):
     pass

narrator/telegram/worker.py

@@ -5,17 +5,9 @@
 from aiogram.types import Message, Document
 
 from narrator.exceptions import UrlParserException
-import narrator.url_parser as url_parser
-import narrator.to_txt as to_txt
-from narrator.text import Text
-from narrator.sub_utils import (
-    balcon,
-    blb2txt,
-    ffmpeg__to_mp3,
-    add_suffix,
-    crop_suffix,
-    make_filename,
-)
+import narrator.text.web_parser as web_parser
+from narrator.text.text import Text
+from narrator.utils import make_filename
 
 
 class ValidityCheckResult(NamedTuple):
@@ -39,10 +31,10 @@ async def produce_audio_file(self, directory: str) -> str:
 class UrlWorker(BaseWorker):
     def __init__(self, bot: Bot, message: Message) -> None:
         super().__init__(bot, message)
-        self._url = url_parser.Url(self._message.text)
+        self._url = web_parser.Url(self._message.text)
 
     def check_validity(self) -> ValidityCheckResult:
-        url: url_parser.Url = self._url
+        url: web_parser.Url = self._url
         if not url.is_valid:
             description = "Not a valid url"
             return ValidityCheckResult(False, description, description)
@@ -63,10 +55,7 @@ def check_validity(self) -> ValidityCheckResult:
 
     async def produce_audio_file(self, directory: str) -> str:
         text: Text = self._url.parse()
-        # balcon.exe only works with UTF-8-BOM (or "utf-8-sig")
-        txt_path = text.save_to_txt(directory, encoding="utf-8-sig")
-        wav_path = balcon(txt_path)
-        mp3_path = ffmpeg__to_mp3(wav_path)
+        mp3_path = text.save_to_mp3(directory)
         return mp3_path
 
 
@@ -85,15 +74,15 @@ def __init__(self, bot: Bot, message: Message) -> None:
 
     def check_validity(self) -> ValidityCheckResult:
         doc = self._doc
-        if not to_txt.has_proper_extention(doc.file_name):
+        if not Text.has_proper_extention(doc.file_name):
             description = "The file extention is not supported"
             user_description = "The file extention is not supported\n"
             user_description += "List of supported extentions:\n"
-            user_description += "  " + ", ".join(to_txt.INPUT_FORMATS)
+            user_description += "  " + ", ".join(Text.INPUT_FORMATS)
             return ValidityCheckResult(False, description, user_description)
-        elif doc.file_size > to_txt.MAX_INPUT_SIZE:
+        elif doc.file_size > Text.MAX_INPUT_SIZE:
             s_cur_mib = DocWorker._bytes_to_mib_str(doc.file_size)
-            s_max_mib = DocWorker._bytes_to_mib_str(to_txt.MAX_INPUT_SIZE)
+            s_max_mib = DocWorker._bytes_to_mib_str(Text.MAX_INPUT_SIZE)
             description = f"The file is too big: the file size is {s_cur_mib} MiB, and the max size allowed is {s_max_mib} MiB)"
             return ValidityCheckResult(False, description, description)
         return ValidityCheckResult(True)
@@ -102,7 +91,6 @@ async def produce_audio_file(self, directory: str) -> str:
         filename = make_filename(self._doc.file_name)
         file_path = os.path.join(directory, filename)
         await self._bot.download(self._doc, file_path)
-        txt_path = blb2txt(file_path)
-        wav_path = balcon(txt_path)
-        mp3_path = ffmpeg__to_mp3(wav_path)
+        text = Text.from_file(file_path)
+        mp3_path = text.save_to_mp3(directory)
         return mp3_path

narrator/text/language.py

@@ -0,0 +1,25 @@
+import string
+from enum import Enum
+
+
+class Language(Enum):
+    RU: str = "ru"
+    EN: str = "en"
+
+    @staticmethod
+    def get_val(ch: str) -> str | None:
+        for lang_val in ALPHABETS:
+            if ch in ALPHABETS[lang_val]:
+                return lang_val
+        return None
+
+
+EN_ALPHABET = string.ascii_letters
+
+RU_ALPHABET = "абвгдеёжзийклмнопрстуфхцчшщъыьэюя"
+RU_ALPHABET += RU_ALPHABET.upper()
+
+ALPHABETS = {
+    Language.RU.value: RU_ALPHABET,
+    Language.EN.value: EN_ALPHABET,
+}