Bug fixes; remove references to uncommitted work-in-progress

notebooks/google_drive.ipynb

@@ -9,17 +9,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 2,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "\u001b[32m2025-04-19 18:17:19.845\u001b[0m | \u001b[34m\u001b[1mDEBUG   \u001b[0m | \u001b[36mctp_slack_bot.core.config\u001b[0m:\u001b[36m__init__\u001b[0m:\u001b[36m14\u001b[0m - \u001b[34m\u001b[1mCreated Settings\u001b[0m\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "from functools import partial\n",
     "from html import escape\n",
@@ -43,15 +35,15 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "\u001b[32m2025-04-19 18:17:19.850\u001b[0m | \u001b[34m\u001b[1mDEBUG   \u001b[0m | \u001b[36mctp_slack_bot.core.config\u001b[0m:\u001b[36m__init__\u001b[0m:\u001b[36m14\u001b[0m - \u001b[34m\u001b[1mCreated Settings\u001b[0m\n",
-      "\u001b[32m2025-04-19 18:17:19.853\u001b[0m | \u001b[34m\u001b[1mDEBUG   \u001b[0m | \u001b[36mctp_slack_bot.services.google_drive_service\u001b[0m:\u001b[36m__init__\u001b[0m:\u001b[36m42\u001b[0m - \u001b[34m\u001b[1mCreated GoogleDriveService\u001b[0m\n"
+      "\u001b[32m2025-04-19 22:51:18.841\u001b[0m | \u001b[34m\u001b[1mDEBUG   \u001b[0m | \u001b[36mctp_slack_bot.core.config\u001b[0m:\u001b[36m__init__\u001b[0m:\u001b[36m14\u001b[0m - \u001b[34m\u001b[1mCreated Settings\u001b[0m\n",
+      "\u001b[32m2025-04-19 22:51:18.844\u001b[0m | \u001b[34m\u001b[1mDEBUG   \u001b[0m | \u001b[36mctp_slack_bot.services.google_drive_service\u001b[0m:\u001b[36m__init__\u001b[0m:\u001b[36m42\u001b[0m - \u001b[34m\u001b[1mCreated GoogleDriveService\u001b[0m\n"
      ]
     }
    ],
@@ -61,7 +53,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
@@ -82,7 +74,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
     {

src/ctp_slack_bot/core/config.py

@@ -62,7 +62,7 @@ class Settings(BaseSettings):
     GOOGLE_UNIVERSE_DOMAIN: str = "googleapis.com"
 
     # File Monitoring Configuration
-    FILE_MONITOR_ROOT_PATH: Optional[str] = None
+    FILE_MONITOR_ROOT_PATH: str = ""
 
     model_config = SettingsConfigDict(
         env_file=".env",

src/ctp_slack_bot/models/google_drive.py

@@ -2,8 +2,6 @@ from datetime import datetime
 from pydantic import BaseModel, ConfigDict
 from typing import Self
 
-from ctp_slack_bot.models import FileContent
-
 
 class GoogleDriveMetadata(BaseModel):
     """Represents Google Drive file or folder metadata."""
@@ -17,7 +15,7 @@ class GoogleDriveMetadata(BaseModel):
     model_config = ConfigDict(frozen=True)
 
     @classmethod
-    def from_folder_path_and_dict(cls: type["GoogleDriveMetadata"], folder_path: str, dict: dict) -> Self:
+    def from_folder_path_and_dict(cls, folder_path: str, dict: dict) -> Self:
         id = dict["id"]
         name = dict["name"]
         modified_time = datetime.fromisoformat(dict["modifiedTime"])

src/ctp_slack_bot/models/webvtt.py

@@ -4,15 +4,19 @@ from itertools import starmap
 from json import dumps
 from more_itertools import windowed
 from pydantic import BaseModel, ConfigDict, Field, PositiveInt, PrivateAttr
+from re import compile as compile_re
 from types import MappingProxyType
 from typing import Any, Dict, Literal, Mapping, Optional, Self, Sequence
 from webvtt import Caption, WebVTT
 
 from ctp_slack_bot.models.base import Chunk, Content
 
+
 CHUNK_FRAMES_OVERLAP = 1
 CHUNK_FRAMES_WINDOW = 5
 SPEAKER_SPEECH_TEXT_SEPARATOR = ": "
+ISO_DATE_TIME_PATTERN = compile_re(r"Start time: (\d{4}-\d{2}-\d{2}(?: \d{2}:\d{2}:\d{2}(?:Z|[+-]\d{2}:\d{2})?)?)")
+
 
 class WebVTTFrame(BaseModel):
     """Represents a WebVTT frame"""
@@ -26,7 +30,7 @@ class WebVTTFrame(BaseModel):
     model_config = ConfigDict(frozen=True)
 
     @classmethod
-    def from_webvtt_caption(cls: type["WebVTTFrame"], index: int, caption: Caption) -> Self:
+    def from_webvtt_caption(cls, caption: Caption, index: int) -> Self:
         identifier = caption.identifier if caption.identifier else str(index)
         start = timedelta(**caption.start_time.__dict__)
         end = timedelta(**caption.end_time.__dict__)
@@ -42,6 +46,7 @@ class WebVTTContent(Content):
 
     id: str
     metadata: Mapping[str, Any] = Field(default_factory=dict)
+    start_time: Optional[datetime]
     frames: Sequence[WebVTTFrame]
 
     def get_id(self: Self) -> str:
@@ -57,9 +62,9 @@ class WebVTTContent(Content):
                            parent_id=self.get_id(),
                            chunk_id=f"{frames[0].identifier}-{frames[-1].identifier}",
                            metadata={
-                               "start": str(frames[0].start), # TODO: This is a harder problem: to get the offsets to become real datetimes so that they can be queryable using MongoDB.
-                               "end": str(frames[-1].end),
-                               "speakers": [frame.speaker for frame in frames if frame.speaker]
+                               "start": self.start_time + frames[0].start if self.start_time else None,
+                               "end": self.start_time + frames[-1].end if self.start_time else None,
+                               "speakers": tuple(frame.speaker for frame in frames if frame.speaker)
                            })
                      for frames
                      in windows)
@@ -68,6 +73,19 @@ class WebVTTContent(Content):
         return MappingProxyType(self.metadata)
 
     @classmethod
-    def from_bytes(cls: type["WebVTTContent"], id: str, metadata: Mapping[str, Any], buffer: bytes) -> Self:
-        frames = tuple(starmap(WebVTTFrame.from_webvtt_caption, enumerate(WebVTT.from_buffer(BytesIO(buffer)).captions, 1)))
-        return WebVTTContent(id=id, metadata=MappingProxyType(metadata), frames=frames)
+    def __get_start_time(cls, web_vtt: WebVTT) -> Optional[datetime]:
+        try:
+            return next(datetime.fromisoformat(result[0])
+                        for result
+                        in map(ISO_DATE_TIME_PATTERN.findall, web_vtt.header_comments)
+                        if result)
+        except ValueError:
+            return None
+
+    @classmethod
+    def from_bytes(cls, id: str, metadata: Mapping[str, Any], buffer: bytes) -> Self:
+        web_vtt = WebVTT.from_buffer(BytesIO(buffer))
+        frames = tuple(WebVTTFrame.from_webvtt_caption(caption, index)
+                       for index, caption
+                       in enumerate(web_vtt.captions, 1))
+        return WebVTTContent(id=id, metadata=MappingProxyType(metadata), start_time=cls.__get_start_time(web_vtt), frames=frames)