Merge branch 'dev'

Author: thiswillbeyourgithub

bumpver.toml

@@ -1,6 +1,6 @@
 
 [bumpver]
-current_version = "5.1.1"
+current_version = "5.1.3"
 version_pattern = "MAJOR.MINOR.PATCH"
 commit_message = "bump version {old_version} -> {new_version}"
 tag_message = "{new_version}"

docs/source/conf.py

@@ -38,7 +38,7 @@ def setup(app):
 project = "wdoc"
 copyright = "2025, thiswillbeyourgithub"
 author = "thiswillbeyourgithub"
-release = "5.1.1"
+release = "5.1.3"
 
 # -- General configuration ---------------------------------------------------
 # https://www.sphinx-doc.org/en/master/usage/configuration.html#general-configuration

setup.py

@@ -129,7 +129,7 @@ def run(self):
 
 setup(
     name="wdoc",
-    version="5.1.1",
+    version="5.1.3",
     description="A perfect AI powered RAG for document query and summary. Supports ~all LLM and ~all filetypes (url, pdf, epub, youtube (incl playlist), audio, anki, md, docx, pptx, or any combination!)",
     long_description=long_description,
     long_description_content_type="text/markdown",
@@ -215,7 +215,7 @@ def run(self):
         "pdfplumber >= 0.11.1",
         "pdf2image >= 1.17.0",
         # URL / web loading (default, since urls are the most common filetype)
-        "playwright >= 1.45.0",  # for online_media and urls
+        "playwright >= 1.60.0",  # for online_media and urls
         "goose3 >= 3.1.20",
         "tldextract>=5.1.2",
         # online search via 'filetype=web'
@@ -232,12 +232,12 @@ def run(self):
             # audio/video transcription
             "deepgram-sdk >= 3.2.7",
             "httpx >= 0.27.0",  # to increase deepgram timeout
-            "pydub >= 0.25.1",  # extracting audio from local video
+            "pydub == 0.25.1",  # extracting audio from local video
             # audioop was removed in stdlib in Python 3.13 and pydub needs it
             # See https://github.com/jiaaro/pydub/issues/815
             "audioop-lts>=0.2.2; python_version>='3.13'",
-            "ffmpeg-python >= 0.2.0",  # extracting audio from local video
-            "torchaudio >= 2.8.0",  # silence removal from audio
+            "ffmpeg-python == 0.2.0",  # extracting audio from local video
+            "torchaudio == 2.8.0",  # silence removal from audio
         ],
         "anki": [
             "ankipandas>=0.3.15",

wdoc-skill/REFERENCE.md

@@ -327,7 +327,7 @@ These load multiple documents and can combine different sources:
 | Variable | Default | Description |
 |----------|---------|-------------|
 | `WDOC_WHISPER_ENDPOINT` | `""` | Custom Whisper API endpoint |
-| `WDOC_WHISPER_API_KEY` | `""` | Custom Whisper API key |
+| `WDOC_WHISPER_API_KEY` | `""` | Custom Whisper API key. When empty, falls back to `OPENAI_API_KEY`, then `WHISPER_API_KEY` |
 | `WDOC_WHISPER_MODEL` | `whisper-1` | Whisper model name |
 | `WDOC_WHISPER_PARALLEL_SPLITS` | `True` | Parallelize split audio transcription |
 

wdoc/docs/help.md

@@ -947,7 +947,8 @@
 * `WDOC_WHISPER_API_KEY`, default: `""`
     * If provided, sets a custom API key for Whisper transcription services. This is useful when using alternative 
     Whisper-compatible services that require their own authentication.
-    * When empty, uses the default OPENAI_API_KEY environment variable.
+    * When empty, falls back to the `OPENAI_API_KEY` environment variable, and if that is also unset, to the 
+    `WHISPER_API_KEY` environment variable. The resolution order is `WDOC_WHISPER_API_KEY` > `OPENAI_API_KEY` > `WHISPER_API_KEY`.
 
 * `WDOC_WHISPER_MODEL`, default: `"whisper-1"`
     * Specifies which Whisper model to use for audio transcription. This can be any model supported by your Whisper endpoint.

wdoc/utils/loaders/online_media.py

@@ -3,7 +3,7 @@
 from pathlib import Path
 
 import ffmpeg
-import playwright
+import playwright.sync_api
 import pydub
 import uuid6
 import yt_dlp as youtube_dl

wdoc/utils/loaders/shared_audio.py

@@ -210,13 +210,30 @@ def transcribe_audio_whisper(
         )
     else:
         assert (
-            "OPENAI_API_KEY" in os.environ
-            and not os.environ["OPENAI_API_KEY"]
-            == "REDACTED_BECAUSE_WDOC_IN_PRIVATE_MODE"
-        ) or (
-            env.WDOC_WHISPER_API_KEY
-            and not env.WDOC_WHISPER_API_KEY == "REDACTED_BECAUSE_WDOC_IN_PRIVATE_MODE"
-        ), "No environment variable OPENAI_API_KEY nor WDOC_WHISPER_API_KEY found"
+            (
+                "OPENAI_API_KEY" in os.environ
+                and not os.environ["OPENAI_API_KEY"]
+                == "REDACTED_BECAUSE_WDOC_IN_PRIVATE_MODE"
+            )
+            or (
+                env.WDOC_WHISPER_API_KEY
+                and not env.WDOC_WHISPER_API_KEY
+                == "REDACTED_BECAUSE_WDOC_IN_PRIVATE_MODE"
+            )
+            or (
+                os.environ.get("WHISPER_API_KEY")
+                and not os.environ["WHISPER_API_KEY"]
+                == "REDACTED_BECAUSE_WDOC_IN_PRIVATE_MODE"
+            )
+        ), (
+            "No environment variable OPENAI_API_KEY, WDOC_WHISPER_API_KEY nor WHISPER_API_KEY found"
+        )
+
+    # Resolve the whisper API key. WDOC_WHISPER_API_KEY takes precedence; if it
+    # is unset and OPENAI_API_KEY is also unset, fall back to WHISPER_API_KEY.
+    whisper_api_key = env.WDOC_WHISPER_API_KEY
+    if not whisper_api_key and not os.environ.get("OPENAI_API_KEY"):
+        whisper_api_key = os.environ.get("WHISPER_API_KEY", "")
 
     try:
         t1 = time.time()
@@ -238,8 +255,8 @@ def transcribe_audio_whisper(
                     f"Using custom whisper endpoint: {env.WDOC_WHISPER_ENDPOINT}"
                 )
 
-            if env.WDOC_WHISPER_API_KEY:
-                transcription_kwargs["api_key"] = env.WDOC_WHISPER_API_KEY
+            if whisper_api_key:
+                transcription_kwargs["api_key"] = whisper_api_key
                 logger.debug("Using custom whisper API key")
 
             try:
@@ -270,8 +287,8 @@ def transcribe_audio_whisper(
                     data["language"] = language
 
                 headers = {}
-                if env.WDOC_WHISPER_API_KEY:
-                    headers["Authorization"] = f"Bearer {env.WDOC_WHISPER_API_KEY}"
+                if whisper_api_key:
+                    headers["Authorization"] = f"Bearer {whisper_api_key}"
 
                 # Make the request
                 endpoint_url = (

wdoc/wdoc.py

@@ -70,7 +70,7 @@ class wdoc:
     This docstring is dynamically updated with the content of wdoc/docs/help.md
     """
 
-    VERSION: str = "5.1.1"
+    VERSION: str = "5.1.3"
     allowed_extra_args = extra_args_types
     __import_mode__: bool = True