Refactor

deepl.py

@@ -0,0 +1,163 @@
+import os
+import time
+import logging
+
+from selenium import webdriver
+from selenium.webdriver.common.keys import Keys
+
+from srt import SRT
+
+
+class translator:
+
+    languages = {
+        'auto': 'auto',
+        'chinese': 'zh',
+        'dutch': 'nl',
+        'english': 'en',
+        'english-us': 'en-us',
+        'english-uk': 'en-gb',
+        'french': 'fr',
+        'german': 'de',
+        'italian': 'it',
+        'japanese': 'ja',
+        'polish': 'pl',
+        'portuguese': 'pt',
+        'portuguese-br': 'pt-br',
+        'russian': 'ru',
+        'spanish': 'es'
+    }
+
+    def __init__(self):
+
+        ### Opening browser
+        logging.info('Opening browser')
+        geckodriver_path = os.path.abspath(os.path.join(os.path.dirname(__file__), 'geckodriver'))
+        self.browser = webdriver.Firefox(executable_path=geckodriver_path)
+        self.browser.maximize_window()
+
+        ### Going to deepl.com
+        logging.info('Going to deepl.com')
+        self.browser.get('https://www.deepl.com/translator')
+
+        ### Getting source and target inputs
+        logging.info('Getting source and target inputs')
+        self.inputElement = self.browser.find_element_by_class_name('lmt__source_textarea')
+        self.targetElement = self.browser.find_element_by_class_name('lmt__target_textarea')
+
+        # Need to hide the cookiebanner so that way it doesnt obscure the language options
+        # Maybe there's a better way, but this works for now. Feel free to change it
+        cookieBanner = self.browser.find_element_by_id('dl_cookieBanner')
+        self.browser.execute_script("arguments[0].style.visibility='hidden'", cookieBanner)
+
+
+    def is_translated(self, text, traslation):
+
+        if  len(traslation) == 0 or \
+            len(text.splitlines()) != len(traslation.splitlines()) or \
+            '[...]' in traslation:
+
+            return False
+
+        return True
+
+
+    def choose_language(self, languageSelect, lang):
+
+        def is_the_right_language(button, language):
+            button_dl_test_attrivute = button.get_attribute("dl-test").lower()
+            return button_dl_test_attrivute.endswith(self.languages[language])
+
+        languageSelect.click()
+
+        languages_menu = languageSelect.find_element_by_class_name('lmt__language_select__menu')
+        languagesButtons = languages_menu.find_elements_by_tag_name('button')
+        languageButton = next( (button for button in languagesButtons if is_the_right_language(button, lang)), None )
+
+        if not languageButton:
+            print(f'ERROR: Language {lang} was not found. Maybe you choose the same idiom on both')
+            self.close()
+            os._exit(os.EX_OK)
+            return
+
+        languageButton.click()
+
+
+    def translate_srt(self, file, lang_from, lang_to):
+
+        logging.info(f'Traslating file {file}')
+        srt_file = SRT(file)
+
+        sub_id = 0  # ID of initial subtitle
+        while sub_id < srt_file.n_subtitles:
+
+            # Clearing input
+            logging.info('Clearing input')
+            self.inputElement.clear()
+
+            while len(self.targetElement.get_attribute('value')) != 0:
+                time.sleep(1)
+
+            # Portion to of srt to translate
+            logging.info('Getting portion of srt to translate')
+            text, sub_id_f = srt_file.extract_portion(sub_id)
+
+            # Sending text
+            logging.info('Writing portion on input')
+            self.inputElement.send_keys(text)
+
+            # Getting traslation
+            logging.info('Traslating portion')
+            while not self.is_translated(text, self.targetElement.get_attribute('value')):
+                time.sleep(1)
+
+            traslation = self.targetElement.get_attribute('value').splitlines()
+
+            # Updating text on SRT
+            logging.info('Saving portion')
+            srt_file.update_text(sub_id, traslation)
+
+            # Getting next portion
+            sub_id = sub_id_f
+
+        # Wraping lines
+        logging.info(f'Wraping lines')
+        srt_file.wrap_lines()
+
+        # Saving file
+        filename = os.path.splitext(file)[0]
+        file_out = f'{filename}_{lang_to}.srt'
+
+        logging.info(f'saving {file_out}')
+        srt_file.save(file_out)
+
+
+    def translate(self, file_paths, lang_from, lang_to):
+
+        ### Preparing page
+        languageToSelect = self.browser.find_element_by_class_name('lmt__language_select--target')
+        languageFromSelect = self.browser.find_element_by_class_name('lmt__language_select--source')
+
+        self.choose_language(languageFromSelect, lang_from)
+        self.choose_language(languageToSelect, lang_to)
+
+
+        for path in file_paths:
+
+            if not os.path.exists(path):
+                print(f"INFO: File {path} doesn't exist, skipping...")
+                continue
+
+            self.translate_srt(path, lang_from, lang_to)
+
+            # if os.path.isdir(path):
+            #     files_in_dir = os.listdir(path)
+            #     for file_path in [os.path.join(path, file_path) for file_path in files_in_dir]:
+            #         if not os.path.isdir(file_path):
+            #             translate_srt(file_path, lang_from, lang_to)
+            # else:
+            #    translate_srt(path, lang_from, lang_to)
+
+
+    def close(self):
+        self.browser.quit()

main.py

@@ -1,15 +1,12 @@
 import os
-import time
 import argparse
+import logging
 
-from selenium import webdriver
-from selenium.webdriver.common.keys import Keys
-
-from srt import SRT
+import deepl
 
 
 parser = argparse.ArgumentParser(
-    description='Converts plain text files to docx and viceversa'
+    description='Translates .STR files using DeepL.com'
 )
 
 parser.add_argument(
@@ -19,123 +16,60 @@
     nargs='+',
     help='File to convert'
 )
+
 parser.add_argument(
-    "-v", "--verbose",
-    help="increase output verbosity",
-    action="store_true"
+    '-i',
+    '--input-lang',
+    type=str,
+    default='auto',
+    choices=('auto', 'chinese', 'dutch', 'english', 'french', 'german',
+            'italian', 'japanese', 'polish', 'portuguese', 'russian', 'spanish'),
+    help='Language to translate from'
 )
 
+parser.add_argument(
+    '-o',
+    '--output-lang',
+    type=str,
+    default='spanish',
+    choices=('chinese', 'dutch', 'english-us', 'english-uk', 'french', 'german',
+            'italian', 'japanese', 'polish', 'portuguese', 'portuguese-br', 'russian', 'spanish'),
+    help='Language to translate to'
+)
 
-args = parser.parse_args()
-
-if args.verbose:
-    def verboseprint(*args):
-        for arg in args:
-           print(arg)
-else:   
-    verboseprint = lambda *a: None      # do-nothing function
-
-
-verboseprint("Opening browser")
-geckodriver_path = os.path.abspath(os.path.join(os.path.dirname(__file__), "geckodriver"))
-
-os.environ['MOZ_HEADLESS'] = '1'
-browser = webdriver.Firefox(executable_path=geckodriver_path)
-
-verboseprint("Going to deepl.com")
-browser.get("https://www.deepl.com/translator")
-
-verboseprint("Getting source and target inputs")
-inputElement = browser.find_element_by_class_name("lmt__source_textarea")
-targetElement = browser.find_element_by_class_name("lmt__target_textarea")
-
-def traslate_srt(file_in, file_out):
-
-    verboseprint("\n" + "=" * 50)
-    verboseprint(f"File to traslate {file_in}")
-    verboseprint("=" * 50)
-    # Opening file
-    srt_file = SRT(file_in)
-
-    sub_id = 0  # ID of initial subtitle
-    while sub_id < srt_file.n_subtitles:
-
-        verboseprint("Clearing input")
-
-        # Clearing input
-        inputElement.clear()
-
-        traslation = targetElement.get_attribute('value')
-
-        while len(traslation) != 0:
-            time.sleep(1)
-            traslation = targetElement.get_attribute('value')
-
-
-        verboseprint("Getting next portion")
-
-        # Portion to of srt to traslate
-        text, sub_id_f = srt_file.extract_portion(sub_id)
-
-
-        verboseprint("Writing portion on input")
-
-        # Sending text
-        inputElement.send_keys(text)
-
-        verboseprint("Traslating portion")
-
-        # Getting traslation
-        traslation = targetElement.get_attribute('value')
-
-        while not traslated(text, traslation):
-            time.sleep(1)
-            traslation = targetElement.get_attribute('value')
-
-        traslation = targetElement.get_attribute('value').splitlines()
-
-        verboseprint("Saving portion")
-
-        # Updation text on SRT
-        srt_file.update_text(sub_id, traslation)
-
-        # Getting next portion
-        sub_id = sub_id_f
-
-    # Saving file
-    verboseprint("Wraping lines")
-    srt_file.wrap_lines()
-    verboseprint(f"Saving {file_out}")
-    srt_file.save(file_out)
-
-
-def traslated(text, traslation):
-
-    if len(traslation) == 0:
-        return False
-    elif len(text.splitlines()) != len(traslation.splitlines()):
-        return False
-    elif "[...]" in traslation:
-        return False
-
-    return True
+parser.add_argument(
+    '-v',
+    '--verbose',
+    action="store_const",
+    dest="loglevel",
+    const=logging.INFO,
+    help="Increase output verbosity",
+)
 
+parser.add_argument(
+    '-vv',
+    '--debug',
+    action="store_const",
+    dest="loglevel",
+    const=logging.DEBUG,
+    default=logging.WARNING,
+    help="Increase output verbosity for debugging",
+)
 
-def traslate_all(file_paths):
+parser.add_argument(
+    '-s',
+    dest="show_gui",
+    action='store_false',
+    help='Show browser window'
+)
 
-    for path in file_paths:
 
-        if not os.path.exists(path):
-            print(f"File '{path}' doesn't exist, skipping...")
-            continue
-
-        if os.path.isdir(path):
-            for file_path in [os.path.join(path, file_path) for file_path in os.listdir(path)]:
-                if not os.path.isdir(file_path):
-                    traslate_srt(file_path, os.path.splitext(file_path)[0] + "_traslated.srt")
-        else:
-            traslate_srt(path, os.path.splitext(path)[0] + ".srt")
+args = parser.parse_args()
+logging.basicConfig(level=args.loglevel)
 
-traslate_all(args.filepath)
+if args.show_gui:
+    os.environ['MOZ_HEADLESS'] = '1'
 
-browser.quit()
+translator = deepl.translator()
+translator.translate(args.filepath, args.input_lang, args.output_lang)
+translator.close()

readme.md

@@ -1,17 +1,23 @@
-# SRT-DeepL translator using selenium
+# SRT-DeepL translator using Selenium
 
 ## Usage
 ```
-main.py [-h] [-v] path [path ...]
+main.py [-h] [-v] [-vv] [-s] [-i {language}] [-o {language}] path [path ...]
 
 Traslate a .SRT file using DeepL and Selenium
 
 positional arguments:
     path: File to convert
 
 optional arguments:
-    -h, --help: show this help message and exit
-    -v, --verbose: increase output verbosity
+    -h, --help: Show this help message and exit
+    -v, --verbose: Increase output verbosity
+    -vv, --debug: Increase output verbosity for debugging
+    -s: Show browser window
+    -i, --input-lang: Language to translate from
+        choices: { auto, chinese, dutch, english, french, german, italian, japanese, polish, portuguese, russian, spanish }
+    -o, --output-lang: Language to translate to
+        choices: { chinese, dutch, english-us, english-uk, french, german, italian, japanese, polish, portuguese, portuguese-br, russian, spanish}
 ```
 
 ## Setup
@@ -22,5 +28,5 @@ use ```pip install -r requirements.txt``` or ```execute setup.sh```
 Geckodriver -v0.24.0
 
 ## Features to add
-- Set language to traslate before start translation
-- Use voice recognition and generate the subtitles
+- Use voice recognition to generate the subtitles
+- Use OCR to extract forced subtitles