Merge pull request #59 from HenestrosaDev/dev

Make the necessary changes to compile the program for macOS

README.md

@@ -639,7 +639,7 @@ When you click on the `Save transcription` button, you'll be prompted for a file
 
 #### Autosave
 
-If checked, the transcription will automatically be saved in the root of the folder where the file to transcribe is stored. If there are already existing files with the same name, they won't be overwritten. To do that, you'll need to check the `Overwrite existing files` option (see below).
+Unchecked by default. If checked, the transcription will automatically be saved in the root of the folder where the file to transcribe is stored. If there are already existing files with the same name, they won't be overwritten. To do that, you'll need to check the `Overwrite existing files` option (see below).
 
 Note that if you create a transcription using the `Microphone` or `YouTube` audio sources with the `Autosave` action enabled, the transcription files will be saved in the root of the `audiotext-vX.X.X` directory.
 

audiotext.spec

@@ -1,46 +1,65 @@
 # -*- mode: python ; coding: utf-8 -*-
-from os.path import join
-from platform import system
-from PyInstaller.utils.hooks import copy_metadata
-from PyInstaller.utils.hooks import collect_data_files
-from shutil import copyfile
+from pathlib import Path
+from PyInstaller.compat import is_darwin, is_win
+import shutil
+
+import sys ; sys.setrecursionlimit(sys.getrecursionlimit() * 5)
+
+def find_site_packages(venv_dir = "venv"):
+    venv_path = Path(venv_dir)
+    for site_packages in venv_path.rglob("site-packages"):
+        if site_packages.is_dir():
+            return site_packages
+
+    return None
+
+site_packages_path = find_site_packages()
 
 datas = [
-    (r'venv/Lib/site-packages/customtkinter', 'customtkinter'),
-    (r'venv/Lib/site-packages/transformers', 'transformers'),
-    (r'venv/Lib/site-packages/lightning', 'lightning'),
-    (r'venv/Lib/site-packages/lightning_fabric', 'lightning_fabric'),
-    (r'venv/Lib/site-packages/speechbrain', 'speechbrain'),
-    (r'venv/Lib/site-packages/pyannote', 'pyannote'),
-    (r'venv/Lib/site-packages/asteroid_filterbanks', 'asteroid_filterbanks'),
-    (r'venv/Lib/site-packages/whisperx', 'whisperx'),
-    ('res', 'res'),
-    ('config.ini', '.'),
-    ('.env', '.'),
+    (f"{site_packages_path}/customtkinter", "customtkinter"),
+    (f"{site_packages_path}/transformers", "transformers"),
+    (f"{site_packages_path}/lightning", "lightning"),
+    (f"{site_packages_path}/lightning_fabric", "lightning_fabric"),
+    (f"{site_packages_path}/speechbrain", "speechbrain"),
+    (f"{site_packages_path}/pyannote", "pyannote"),
+    (f"{site_packages_path}/asteroid_filterbanks", "asteroid_filterbanks"),
+    (f"{site_packages_path}/whisperx", "whisperx"),
+    (f"{site_packages_path}/librosa", "librosa"),
+    ("res", "res"),
+    ("config.ini", "."),
+    (".env", "."),
 ]
 
-datas += copy_metadata('torch')
-datas += copy_metadata('tqdm', recursive=True)
-datas += copy_metadata('regex')
-datas += copy_metadata('requests')
-datas += copy_metadata('packaging')
-datas += copy_metadata('filelock')
-datas += copy_metadata('numpy')
-datas += copy_metadata('tokenizers')
-datas += copy_metadata('pillow')
-datas += copy_metadata('huggingface_hub')
-datas += copy_metadata('safetensors')
-datas += copy_metadata('pyyaml')
-datas += collect_data_files('librosa')
+hiddenimports = [
+    "huggingface_hub.repository",
+    "sklearn.utils._cython_blas",
+    "sklearn.neighbors.quad_tree",
+    "sklearn.tree",
+    "sklearn.tree._utils",
+]
 
 block_cipher = None
 
+is_debug = False
+if is_debug:
+    options = [("v", None, "OPTION")]
+else:
+    options = []
+
+binaries = [
+    (shutil.which("ffmpeg"), "."),
+    (shutil.which("ffprobe"), "."),
+]
+
+if flac_path := shutil.which("flac"):
+    binaries += [(flac_path, ".")]
+
 a = Analysis(
-    ['src/app.py'],
-    pathex=[],
-    binaries=[],
+    ["src/app.py"],
+    pathex=[site_packages_path],
+    binaries=binaries,
     datas=datas,
-    hiddenimports=['huggingface_hub.repository', 'pytorch', 'sklearn.utils._cython_blas', 'sklearn.neighbors.typedefs', 'sklearn.neighbors.quad_tree', 'sklearn.tree', 'sklearn.tree._utils'],
+    hiddenimports=hiddenimports,
     hookspath=[],
     hooksconfig={},
     runtime_hooks=[],
@@ -51,76 +70,53 @@ a = Analysis(
     noarchive=False,
 )
 
-# Filter out unused and/or duplicate shared libs
-torch_lib_paths = {
-    join('torch', 'lib', 'libtorch_cuda.so'),
-    join('torch', 'lib', 'libtorch_cpu.so'),
-}
-a.datas = [entry for entry in a.datas if not entry[0] in torch_lib_paths]
+pyz = PYZ(a.pure, a.zipped_data, cipher=block_cipher)
 
-os_path_separator = '\\' if system() == 'Windows' else '/'
-a.datas = [entry for entry in a.datas if not f'torch{os_path_separator}_C.cp' in entry[0]]
-a.datas = [entry for entry in a.datas if not f'torch{os_path_separator}_dl.cp' in entry[0]]
+macos_icon = "res/macos/icon.icns"
+windows_icon = "res/windows/icon.ico"
 
-pyz = PYZ(a.pure, a.zipped_data, cipher=block_cipher)
+exe = EXE(
+    pyz,
+    a.scripts,
+    options,
+    exclude_binaries=True,
+    name="Audiotext",
+    debug=is_debug,
+    bootloader_ignore_signals=False,
+    strip=False,
+    upx=True,
+    upx_exclude=[],
+    runtime_tmpdir=None,
+    console=is_debug,
+    disable_windowed_traceback=False,
+    argv_emulation=False,
+    target_arch=None,
+    codesign_identity=None,
+    entitlements_file="res/macos/entitlements.plist" if is_darwin else None,
+    icon=windows_icon if is_win else macos_icon,
+)
 
-if system() == 'Darwin':  # macOS
-    exe = EXE(
-        pyz,
-        a.scripts,
-        a.binaries,
-        a.zipfiles,
-        a.datas,
-        [],
-        name='Audiotext',
-        debug=False,
-        bootloader_ignore_signals=False,
-        strip=False,
-        upx=True,
-        upx_exclude=[],
-        runtime_tmpdir=None,
-        console=False,
-        disable_windowed_traceback=False,
-        argv_emulation=False,
-        target_arch='x86_64',
-        codesign_identity=None,
-        entitlements_file=None,
-        icon=['res/img/icon.icns'],
-    )
-
-    # BUNDLE statement is used to create a macOS application bundle (.app) for the program
-    app = BUNDLE(
-        exe,
-        name='Audiotext.app',
-        icon=['res/img/icon.icns'],
-        bundle_identifier=None,
-    )
-else:
-    exe = EXE(
-        pyz,
-        a.scripts,
-        [],
-        exclude_binaries=True,
-        name='Audiotext',
-        debug=False,
-        bootloader_ignore_signals=False,
-        strip=False,
-        upx=True,
-        console=False,
-        disable_windowed_traceback=False,
-        argv_emulation=False,
-        target_arch='x86_64',
-        codesign_identity=None,
-        entitlements_file=None,
-        icon=['res/img/icon.ico'],
-    )
-    coll = COLLECT(
-        exe,
-        a.binaries,
-        a.zipfiles,
-        a.datas,
-        strip=False,
-        upx=True,
-        upx_exclude=[],
-        name='audiotext',
-    )
+coll = COLLECT(
+    exe,
+    a.binaries,
+    a.zipfiles,
+    a.datas,
+    strip=False,
+    upx=True,
+    upx_exclude=[],
+    name="Audiotext",
+)
+
+app = BUNDLE(
+    coll,
+    name="Audiotext.app",
+    icon=macos_icon,
+    bundle_identifier="com.henestrosadev.audiotext",
+    version="2.3.0",
+    info_plist={
+        "NSPrincipalClass": "NSApplication",
+        "NSAppleScriptEnabed": False,
+        "NSHighResolutionCapable": True,
+        "NSMicrophoneUsageDescription": "Allow Audiotext to record audio from your microphone to generate transcriptions.",
+    }
+)

config.ini

@@ -10,7 +10,7 @@ appearance_mode = System
 language = English
 audio_source = File
 method = WhisperX
-autosave = True
+autosave = False
 overwrite_files = False
 
 [whisper_api]

res/macos/entitlements.plist

@@ -0,0 +1,15 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
+<plist version="1.0">
+  <dict>
+    <!-- These are required for binaries built by PyInstaller -->
+    <key>com.apple.security.cs.allow-jit</key>
+    <true/>
+    <key>com.apple.security.cs.allow-unsigned-executable-memory</key>
+    <true/>
+    <key>com.apple.security.cs.disable-library-validation</key>
+    <true/>
+    <key>com.apple.security.device.audio-input</key>
+    <true/>
+  </dict>
+</plist>

src/app.py

@@ -1,5 +1,4 @@
 import customtkinter as ctk
-import torch
 import utils.config_manager as cm
 import utils.constants as c
 import utils.path_helper as ph
@@ -23,7 +22,7 @@ def __init__(self) -> None:
         ctk.set_default_color_theme("blue")
 
         self.title(c.APP_NAME)
-        self.wm_iconbitmap(ph.ROOT_PATH / ph.IMG_RELATIVE_PATH / "icon.ico")
+        self.wm_iconbitmap(ph.ROOT_PATH / "res/windows/icon.ico")
 
         # Initial size of the window
         width = 1000
@@ -35,6 +34,9 @@ def __init__(self) -> None:
         min_height = 250
         self.minsize(min_width, min_height)
 
+        # Place the torch import here to avoid the "No ffmpeg exe could be found" error
+        import torch
+
         # Check GPU
         cm.ConfigManager.modify_value(
             section=ConfigWhisperX.Key.SECTION,

src/handlers/whisperx_handler.py

@@ -112,6 +112,7 @@ def save_transcription(
                 writer = whisperx.transcribe.get_writer(output_type, str(output_dir))
 
                 # https://github.com/m-bain/whisperX/issues/455#issuecomment-1707547704
-                self._whisperx_result["language"] = "en"  # type: ignore[index]
+                if self._whisperx_result:
+                    self._whisperx_result["language"] = "en"
 
                 writer(self._whisperx_result, file_path, vars(config_subtitles))

src/utils/env_keys.py

@@ -2,9 +2,10 @@
 from enum import Enum
 from typing import Optional
 
-from dotenv import find_dotenv, load_dotenv
+import utils.path_helper as ph
+from dotenv import load_dotenv
 
-load_dotenv(find_dotenv())
+load_dotenv(ph.ROOT_PATH / ".env")
 
 
 class EnvKeys(Enum):

src/utils/path_helper.py

@@ -1,3 +1,4 @@
+import os
 import sys
 from pathlib import Path
 
@@ -8,13 +9,20 @@ def get_root_path() -> Path:
 
     Taken from the [PyInstaller docs](https://pyinstaller.org/en/stable/runtime-information.html)
 
-    :return: The absolute path to the file or directory specified by the relative path.
+    :return: The absolute path to the program directory.
     :rtype: Path
     """
     if getattr(sys, "frozen", False) and hasattr(sys, "_MEIPASS"):
-        return Path(sys._MEIPASS)
+        root_path = Path(sys._MEIPASS)
     else:
-        return Path(__file__).parent.parent.parent
+        root_path = Path(__file__).parent.parent.parent
+
+    # Add the root path to the PATH in order to allow macOS to pick up the binaries,
+    # since `.app` bundles launched from Finder get a very limited shell environment.
+    # https://github.com/orgs/pyinstaller/discussions/8773
+    os.environ["PATH"] += os.pathsep + str(root_path)
+
+    return root_path
 
 
 IMG_RELATIVE_PATH = "res/img"