docs: modernize multimodal tutorials and migrate legacy blob APIs#16918
docs: modernize multimodal tutorials and migrate legacy blob APIs#16918
Conversation
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request updates several Jupyter notebooks to demonstrate BigQuery DataFrames features, including the anywidget display mode, generative AI functions for image analysis, and vector search capabilities. Feedback focuses on fixing a broken dependency in the vector search notebook caused by a removed cell, correcting a variable name typo, avoiding bare exception blocks, and replacing hardcoded internal project IDs with placeholders to ensure the tutorials are functional and portable for public users.
| "cell_type": "code", | ||
| "execution_count": null, | ||
| "source": [ | ||
| "# Code calling .blob.audio_transcribe() was removed to satisfy the goal of removing public Blob APIs." |
There was a problem hiding this comment.
Removing the code that populates flattened['Transcription'] while keeping subsequent cells that depend on this column (e.g., lines 574, 608, 639) will cause the notebook to fail with a KeyError. The legacy .blob call should be replaced with a modern equivalent (e.g., using bbq.ai.generate with multimodal support) rather than simply being removed to ensure the tutorial remains functional.
There was a problem hiding this comment.
Reimplemented the broken audio transcription workflow using modern bbq.obj APIs and bbq.ai.generate to properly populate flattened["Transcription"] with the exactly-matched output struct (content & status) required by downstream evaluation cells.
| "source": [ | ||
| "import bigframes.pandas as bpd\n", | ||
| "\n", | ||
| "MY_RPOJECT_ID = \"bigframes-dev\" # @param {type:\"string\"}\n", |
There was a problem hiding this comment.
There is a typo in the variable name MY_RPOJECT_ID. It should be MY_PROJECT_ID. This variable is also used in subsequent cells (e.g., lines 115 and 233), which should also be updated to maintain consistency.
| " if str(uri).lower().endswith((\".png\", \".jpg\", \".jpeg\", \".webp\")):\n", | ||
| " return f'<img src=\"{url}\" width=\"{width}\">'\n", | ||
| " return f'<a href=\"{url}\" target=\"_blank\">{uri}</a>'\n", | ||
| " except: return \"Format Error\"\n", |
There was a problem hiding this comment.
Avoid using a bare except: block. It is better to catch specific exceptions (such as json.JSONDecodeError or KeyError) or at least Exception to avoid catching system-level signals like KeyboardInterrupt.
except Exception: return \"Format Error\"
There was a problem hiding this comment.
Rewrote the cell formatter exception block from a raw except: to except Exception: to adhere to best practice
| " SELECT\n", | ||
| " AI.GENERATE(\n", | ||
| " prompt=>(\"Extract the values.\", OBJ.GET_ACCESS_URL(OBJ.FETCH_METADATA(OBJ.MAKE_REF(gcs_path, \"us.conn\")), \"r\")),\n", | ||
| " connection_id=>\"bigframes-dev.us.bigframes-default-connection\",\n", |
There was a problem hiding this comment.
The connection_id is hardcoded to an internal project (bigframes-dev). For a public tutorial, this should be a placeholder or use a variable so that users can configure it for their own environment.
| "bpd.options.bigquery.location = \"US\"\n", | ||
| "\n", | ||
| "# Set to your GCP project ID.\n", | ||
| "bpd.options.bigquery.project = \"swast-scratch\"" |
There was a problem hiding this comment.
The project ID is hardcoded as "swast-scratch". In a tutorial notebook, it is best practice to use a placeholder or a variable that prompts the user for their own project ID.
| "import bigframes.bigquery as bbq\n", | ||
| "\n", | ||
| "vector_search_results = bbq.vector_search(\n", | ||
| " base_table=f\"swast-scratch.scipy2025.national_jukebox\",\n", |
There was a problem hiding this comment.
The base_table parameter uses a hardcoded project and dataset (swast-scratch.scipy2025). This should be updated to use the project ID variable configured earlier in the notebook.
1 participant
Overview
This PR modernizes the tutorial notebooks across the BigFrames ecosystem to align with the latest public API state and successfully completes the migration path away from the deprecated internal .blob accessors and private loading mechanisms (e.g. _from_glob_path).
The core objective is ensuring user-facing documentation perfectly tracks active package behavior without invoking retirement warnings or triggering internal attribute exceptions.
Key Changes
Fixes #< 478952827 > 🦕