uxoxo/eb2ab
0
1import argparse2import filecmp3import importlib.util4import os5import shutil6import socket7import subprocess8import sys9import tempfile10 11from pathlib import Path12from lib import *13 14def check_virtual_env(script_mode):15 current_version = sys.version_info[:2] # (major, minor)16 if str(os.path.basename(sys.prefix)) == 'python_env' or script_mode == FULL_DOCKER or current_version >= min_python_version and current_version <= max_python_version:17 return True 18 error = f'''***********19Wrong launch! ebook2audiobook must run in its own virtual environment!20NOTE: If you are running a Docker so you are probably using an old version of ebook2audiobook.21To solve this issue go to download the new version at https://github.com/DrewThomasson/ebook2audiobook22If the directory python_env does not exist in the ebook2audiobook root directory,23run your command with "./ebook2audiobook.sh" for Linux and Mac or "ebook2audiobook.cmd" for Windows24to install it all automatically.25{install_info}26***********'''27 print(error)28 return False29 30def check_python_version():31 current_version = sys.version_info[:2] # (major, minor)32 if current_version < min_python_version or current_version > max_python_version:33 error = f'''***********34Wrong launch: Your OS Python version is not compatible! (current: {current_version[0]}.{current_version[1]})35In order to install and/or use ebook2audiobook correctly you must run 36"./ebook2audiobook.sh" for Linux and Mac or "ebook2audiobook.cmd" for Windows.37{install_info}38***********'''39 print(error)40 return False41 else:42 return True43 44def check_and_install_requirements(file_path):45 if not os.path.exists(file_path):46 error = f'Warning: File {file_path} not found. Skipping package check.'47 print(error)48 return False49 try:50 from importlib.metadata import version, PackageNotFoundError51 try:52 from packaging.specifiers import SpecifierSet53 except ImportError:54 subprocess.check_call([sys.executable, '-m', 'pip', 'install', '--no-cache-dir', 'packaging'])55 from packaging.specifiers import SpecifierSet56 import regex as re57 from tqdm import tqdm58 with open(file_path, 'r') as f:59 contents = f.read().replace('\r', '\n')60 packages = [61 pkg.strip()62 for pkg in contents.splitlines()63 if pkg.strip() and re.search(r'[a-zA-Z0-9]', pkg)64 ]65 missing_packages = []66 for package in packages:67 # remove extras so '[lang]==x.y' becomes 'pkg==x.y'68 clean_pkg = re.sub(r'\[.*?\]', '', package)69 pkg_name = re.split(r'[<>=]', clean_pkg, 1)[0].strip()70 try:71 installed_version = version(pkg_name)72 if pkg_name == 'num2words':73 code = "ZH_CN"74 spec = importlib.util.find_spec(f"num2words.lang_{code}")75 if spec is None:76 missing_packages.append(package)77 except PackageNotFoundError:78 error = f'{package} is missing.'79 print(error)80 missing_packages.append(package)81 else:82 # get specifier from clean_pkg, not from the raw string83 spec_str = clean_pkg[len(pkg_name):].strip()84 if spec_str:85 spec = SpecifierSet(spec_str)86 if installed_version not in spec:87 error = (f'{pkg_name} (installed {installed_version}) does not satisfy "{spec_str}".')88 print(error)89 missing_packages.append(package)90 if missing_packages:91 msg = '\nInstalling missing or upgrade packages...\n'92 print(msg)93 tmp_dir = tempfile.mkdtemp()94 os.environ['TMPDIR'] = tmp_dir95 result = subprocess.call([sys.executable, '-m', 'pip', 'cache', 'purge'])96 subprocess.check_call([sys.executable, '-m', 'pip', 'install', '--upgrade', 'pip'])97 with tqdm(total=len(packages),98 desc='Installation 0.00%',99 bar_format='{desc}: {n_fmt}/{total_fmt} ',100 unit='step') as t:101 for package in tqdm(missing_packages, desc="Installing", unit="pkg"):102 try:103 if package == 'num2words':104 pkgs = ['git+https://github.com/savoirfairelinux/num2words.git', '--force']105 else:106 pkgs = [package]107 subprocess.check_call([108 sys.executable, '-m', 'pip', 'install',109 '--no-cache-dir', '--use-pep517',110 *pkgs111 ])112 t.update(1)113 except subprocess.CalledProcessError as e:114 error = f'Failed to install {package}: {e}'115 print(error)116 return False117 msg = '\nAll required packages are installed.'118 print(msg)119 return True120 except Exception as e:121 error = f'check_and_install_requirements() error: {e}'122 raise SystemExit(error)123 return False124 125def check_dictionary():126 import unidic127 unidic_path = unidic.DICDIR128 dicrc = os.path.join(unidic_path, 'dicrc')129 if not os.path.exists(dicrc) or os.path.getsize(dicrc) == 0:130 try:131 error = 'UniDic dictionary not found or incomplete. Downloading now...'132 print(error)133 subprocess.run(['python', '-m', 'unidic', 'download'], check=True)134 except subprocess.CalledProcessError as e:135 error = f'Failed to download UniDic dictionary. Error: {e}. Unable to continue without UniDic. Exiting...'136 raise SystemExit(error)137 return False138 return True139 140def is_port_in_use(port):141 with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:142 return s.connect_ex(('0.0.0.0', port)) == 0143 144def main():145 # Argument parser to handle optional parameters with descriptions146 parser = argparse.ArgumentParser(147 description='Convert eBooks to Audiobooks using a Text-to-Speech model. You can either launch the Gradio interface or run the script in headless mode for direct conversion.',148 epilog='''149Example usage: 150Windows:151 Gradio/GUI:152 ebook2audiobook.cmd153 Headless mode:154 ebook2audiobook.cmd --headless --ebook '/path/to/file'155Linux/Mac:156 Gradio/GUI:157 ./ebook2audiobook.sh158 Headless mode:159 ./ebook2audiobook.sh --headless --ebook '/path/to/file'160 161Tip: to add of silence (1.4 seconds) into your text just use "###" or "[pause]".162 ''',163 formatter_class=argparse.RawTextHelpFormatter164 )165 options = [166 '--script_mode', '--session', '--share', '--headless', 167 '--ebook', '--ebooks_dir', '--language', '--voice', '--device', '--tts_engine', 168 '--custom_model', '--fine_tuned', '--output_format',169 '--temperature', '--length_penalty', '--num_beams', '--repetition_penalty', '--top_k', '--top_p', '--speed', '--enable_text_splitting',170 '--text_temp', '--waveform_temp',171 '--output_dir', '--version', '--workflow', '--help'172 ]173 tts_engine_list_keys = [k for k in TTS_ENGINES.keys()]174 tts_engine_list_values = [k for k in TTS_ENGINES.values()]175 all_group = parser.add_argument_group('**** The following options are for all modes', 'Optional')176 all_group.add_argument(options[0], type=str, help=argparse.SUPPRESS)177 parser.add_argument(options[1], type=str, help='''Session to resume the conversion in case of interruption, crash, 178 or reuse of custom models and custom cloning voices.''')179 gui_group = parser.add_argument_group('**** The following option are for gradio/gui mode only', 'Optional')180 gui_group.add_argument(options[2], action='store_true', help='''Enable a public shareable Gradio link.''')181 headless_group = parser.add_argument_group('**** The following options are for --headless mode only')182 headless_group.add_argument(options[3], action='store_true', help='''Run the script in headless mode''')183 headless_group.add_argument(options[4], type=str, help='''Path to the ebook file for conversion. Cannot be used when --ebooks_dir is present.''')184 headless_group.add_argument(options[5], type=str, help=f'''Relative or absolute path of the directory containing the files to convert. 185 Cannot be used when --ebook is present.''')186 headless_group.add_argument(options[6], type=str, default=default_language_code, help=f'''Language of the e-book. Default language is set 187 in ./lib/lang.py sed as default if not present. All compatible language codes are in ./lib/lang.py''')188 headless_optional_group = parser.add_argument_group('optional parameters')189 headless_optional_group.add_argument(options[7], type=str, default=None, help='''(Optional) Path to the voice cloning file for TTS engine. 190 Uses the default voice if not present.''')191 headless_optional_group.add_argument(options[8], type=str, default=default_device, choices=device_list, help=f'''(Optional) Pprocessor unit type for the conversion. 192 Default is set in ./lib/conf.py if not present. Fall back to CPU if GPU not available.''')193 headless_optional_group.add_argument(options[9], type=str, default=None, choices=tts_engine_list_keys+tts_engine_list_values, help=f'''(Optional) Preferred TTS engine (available are: {tts_engine_list_keys+tts_engine_list_values}.194 Default depends on the selected language. The tts engine should be compatible with the chosen language''')195 headless_optional_group.add_argument(options[10], type=str, default=None, help=f'''(Optional) Path to the custom model zip file cntaining mandatory model files. 196 Please refer to ./lib/models.py''')197 headless_optional_group.add_argument(options[11], type=str, default=default_fine_tuned, help='''(Optional) Fine tuned model path. Default is builtin model.''')198 headless_optional_group.add_argument(options[12], type=str, default=default_output_format, help=f'''(Optional) Output audio format. Default is set in ./lib/conf.py''')199 headless_optional_group.add_argument(options[13], type=float, default=None, help=f"""(xtts only, optional) Temperature for the model. 200 Default to config.json model. Higher temperatures lead to more creative outputs.""")201 headless_optional_group.add_argument(options[14], type=float, default=None, help=f"""(xtts only, optional) A length penalty applied to the autoregressive decoder. 202 Default to config.json model. Not applied to custom models.""")203 headless_optional_group.add_argument(options[15], type=int, default=None, help=f"""(xtts only, optional) Controls how many alternative sequences the model explores. Must be equal or greater than length penalty. 204 Default to config.json model.""")205 headless_optional_group.add_argument(options[16], type=float, default=None, help=f"""(xtts only, optional) A penalty that prevents the autoregressive decoder from repeating itself. 206 Default to config.json model.""")207 headless_optional_group.add_argument(options[17], type=int, default=None, help=f"""(xtts only, optional) Top-k sampling. 208 Lower values mean more likely outputs and increased audio generation speed. 209 Default to config.json model.""")210 headless_optional_group.add_argument(options[18], type=float, default=None, help=f"""(xtts only, optional) Top-p sampling. 211 Lower values mean more likely outputs and increased audio generation speed. Default to config.json model.""")212 headless_optional_group.add_argument(options[19], type=float, default=None, help=f"""(xtts only, optional) Speed factor for the speech generation. 213 Default to config.json model.""")214 headless_optional_group.add_argument(options[20], action='store_true', help=f"""(xtts only, optional) Enable TTS text splitting. This option is known to not be very efficient. 215 Default to config.json model.""")216 headless_optional_group.add_argument(options[21], type=float, default=None, help=f"""(bark only, optional) Text Temperature for the model. 217 Default to {default_engine_settings[TTS_ENGINES['BARK']]['text_temp']}. Higher temperatures lead to more creative outputs.""")218 headless_optional_group.add_argument(options[22], type=float, default=None, help=f"""(bark only, optional) Waveform Temperature for the model. 219 Default to {default_engine_settings[TTS_ENGINES['BARK']]['waveform_temp']}. Higher temperatures lead to more creative outputs.""")220 headless_optional_group.add_argument(options[23], type=str, help=f'''(Optional) Path to the output directory. Default is set in ./lib/conf.py''')221 headless_optional_group.add_argument(options[24], action='version', version=f'ebook2audiobook version {prog_version}', help='''Show the version of the script and exit''')222 headless_optional_group.add_argument(options[25], action='store_true', help=argparse.SUPPRESS)223 224 for arg in sys.argv:225 if arg.startswith('--') and arg not in options:226 error = f'Error: Unrecognized option "{arg}"'227 print(error)228 sys.exit(1)229 230 args = vars(parser.parse_args())231 232 if not 'help' in args:233 if not check_virtual_env(args['script_mode']):234 sys.exit(1)235 236 if not check_python_version():237 sys.exit(1)238 239 # Check if the port is already in use to prevent multiple launches240 if not args['headless'] and is_port_in_use(interface_port):241 error = f'Error: Port {interface_port} is already in use. The web interface may already be running.'242 print(error)243 sys.exit(1)244 245 args['script_mode'] = args['script_mode'] if args['script_mode'] else NATIVE246 args['session'] = 'ba800d22-ee51-11ef-ac34-d4ae52cfd9ce' if args['workflow'] else args['session'] if args['session'] else None247 args['share'] = args['share'] if args['share'] else False248 args['ebook_list'] = None249 250 print(f"v{prog_version} {args['script_mode']} mode")251 252 if args['script_mode'] == NATIVE:253 check_pkg = check_and_install_requirements(requirements_file)254 if check_pkg:255 if not check_dictionary():256 sys.exit(1)257 else:258 error = 'Some packages could not be installed'259 print(error)260 sys.exit(1)261 262 from lib.functions import SessionContext, convert_ebook_batch, convert_ebook, web_interface263 ctx = SessionContext()264 # Conditions based on the --headless flag265 if args['headless']:266 args['is_gui_process'] = False267 args['audiobooks_dir'] = os.path.abspath(args['output_dir']) if args['output_dir'] else audiobooks_cli_dir268 args['device'] = 'cuda' if args['device'] == 'gpu' else args['device']269 args['tts_engine'] = TTS_ENGINES[args['tts_engine']] if args['tts_engine'] in TTS_ENGINES.keys() else args['tts_engine'] if args['tts_engine'] in TTS_ENGINES.values() else None270 args['output_split'] = default_output_split271 args['output_split_hours'] = default_output_split_hours272 # Condition to stop if both --ebook and --ebooks_dir are provided273 if args['ebook'] and args['ebooks_dir']:274 error = 'Error: You cannot specify both --ebook and --ebooks_dir in headless mode.'275 print(error)276 sys.exit(1)277 # convert in absolute path voice, custom_model if any278 if args['voice']:279 if os.path.exists(args['voice']):280 args['voice'] = os.path.abspath(args['voice'])281 if args['custom_model']:282 if os.path.exists(args['custom_model']):283 args['custom_model'] = os.path.abspath(args['custom_model'])284 if not os.path.exists(args['audiobooks_dir']):285 error = 'Error: --output_dir path does not exist.'286 print(error)287 sys.exit(1) 288 if args['ebooks_dir']:289 args['ebooks_dir'] = os.path.abspath(args['ebooks_dir'])290 if not os.path.exists(args['ebooks_dir']):291 error = f'Error: The provided --ebooks_dir "{args["ebooks_dir"]}" does not exist.'292 print(error)293 sys.exit(1) 294 args['ebook_list'] = []295 for file in os.listdir(args['ebooks_dir']):296 if any(file.endswith(ext) for ext in ebook_formats):297 full_path = os.path.abspath(os.path.join(args['ebooks_dir'], file))298 args['ebook_list'].append(full_path)299 progress_status, passed = convert_ebook_batch(args, ctx)300 if passed is False:301 error = f'Conversion failed: {progress_status}'302 print(error)303 sys.exit(1)304 elif args['ebook']:305 args['ebook'] = os.path.abspath(args['ebook'])306 if not os.path.exists(args['ebook']):307 error = f'Error: The provided --ebook "{args["ebook"]}" does not exist.'308 print(error)309 sys.exit(1) 310 progress_status, passed = convert_ebook(args, ctx)311 if passed is False:312 error = f'Conversion failed: {progress_status}'313 print(error)314 sys.exit(1)315 else:316 error = 'Error: In headless mode, you must specify either an ebook file using --ebook or an ebook directory using --ebooks_dir.'317 print(error)318 sys.exit(1) 319 else:320 args['is_gui_process'] = True321 passed_arguments = sys.argv[1:]322 allowed_arguments = {'--share', '--script_mode'}323 passed_args_set = {arg for arg in passed_arguments if arg.startswith('--')}324 if passed_args_set.issubset(allowed_arguments):325 web_interface(args, ctx)326 else:327 error = 'Error: In non-headless mode, no option or only --share can be passed'328 print(error)329 sys.exit(1)330if __name__ == '__main__':331 main()332