{ "cells": [ { "cell_type": "markdown", "id": "07f5174c", "metadata": { "tags": [] }, "source": [ "## Inserting Data Lab Content into Organizer Topics" ] }, { "cell_type": "markdown", "id": "dea57e0f", "metadata": {}, "source": [ "Data Lab Functions can be used to insert Data Lab content directly into Organizer Topic. Data Lab Functions use the cells in a Jupyter Notebook as an API endpoint. " ] }, { "cell_type": "markdown", "id": "838d9aad", "metadata": {}, "source": [ "## How To Define A Data Lab Functions Notebook\n", "### Notebook Cells\n", "#### Context Cell\n", "\n", "- A notebook cell that is not decorated with a top-line comment resembling a RESTful endpoint. Context cells appear sequentially at the beginning of a Jupyter notebook and are executed only once in a new kernel.\n", "\n", "#### Endpoint Cell\n", "\n", "- A notebook cell that is decorated with a top-line comment resembling a RESTful endpoint. For example:\n", "\n", "```\n", "# GET /items\n", "items = spy.search({\n", " 'Path': 'Example >> Cooling Tower 1 >> Area A',\n", " 'Name': 'Compressor'\n", "})\n", "items.to_json()\n", "```\n", "\n", "- Endpoint cells must appear after context cells in a Jupyter notebook\n", "\n", "- Allowed HTTP methods are GET, POST, PATCH, PUT, DELETE\n", "\n", "- Endpoints can include nested paths. For example, # GET /items/area-a/compressor\n", "\n", "- Endpoints do not support path parameters. Instead, both query parameters and a request body are available in the REQUEST object which is instantiated in the kernel for every Data Lab Functions request.\n", "\n", "- Endpoint cells must be unique. That is, they cannot duplicate HTTP method + endpoint name\n", "\n", "- Endpoint cell response content-types\n", "\n", " - application/json (default)\n", "\n", " - Cell response should be Python objects or literals. Data Lab Functions responses performs json.dumps() on endpoint cell output.\n", "\n", " - image/*\n", "\n", " - text/plain\n", "\n", " - text/html\n", "`\n", "### Reserved Variables\n", "#### REQUEST\n", "\n", "On each request, Data Lab Functions injects a REQUEST variable of type Dict directly into the kernel. The REQUEST dictionary contains all the relevant HTTP Request information, namely:\n", "\n", "| REQUEST Variable | Type | Description |\n", "| --- | --- ||\n", "|info | Dict||\n", "|id|str| Unique ID of the request|\n", "|user_id|str| The Seeq user UUID invoking the request|\n", "|notebook |str| The path and name of the API notebook|\n", "|method| str |HTTP method|\n", "|endpoint|str| Endpoint name|\n", "|body|str/Dict|Parsed from JSON or Form-Encoded name-value pairs|\n", "|args|Dict|Parsed query string name-value pairs |\n", "|headers|Dict|HTTP Header name-value pairs |\n", "\n" ] }, { "cell_type": "markdown", "id": "c9b165cc", "metadata": {}, "source": [ "An example url for an endpoint in this notebook would look like this." ] }, { "cell_type": "raw", "id": "a44dc9ce-8e38-41c0-af45-4428f06e2c28", "metadata": {}, "source": [ "data-lab//functions/notebooks//endpoints/static_image" ] }, { "cell_type": "markdown", "id": "cb525589-2ac3-431f-8223-8ef3a67b2cb8", "metadata": {}, "source": [ "### An optional common definition that could be called by various endpoints" ] }, { "cell_type": "code", "execution_count": 1, "id": "48d8bf93-204b-45c9-912f-f4d5673f9d39", "metadata": {}, "outputs": [], "source": [ "from io import StringIO\n", "import plotly.express as px\n", "\n", "#def get_query_params(*names):\n", "# args = REQUEST['args']\n", "# return {name: args.get(name, None) for name in names}\n", "\n", "def fig_to_html(fig):\n", " buffer = StringIO()\n", " fig.write_html(buffer, full_html=True, include_plotlyjs=True)\n", " return {'headers': {'Content-Type': 'text/html'}, 'data': buffer.getvalue()}" ] }, { "cell_type": "markdown", "id": "5e687fac", "metadata": {}, "source": [ "## Insert Static Content" ] }, { "cell_type": "code", "execution_count": null, "id": "93f97003", "metadata": {}, "outputs": [], "source": [ "# GET /static_image\n", "# data-lab//functions/notebooks/example/endpoints/static_image\n", "import plotly.graph_objects as go\n", "\n", "fig = go.Figure(go.Indicator(\n", " mode = \"gauge+number\",\n", " value = 450,\n", " title = {'text': \"Speed\"},\n", " domain = {'x': [0, 1], 'y': [0, 1]}\n", "))\n", "\n", "#{'headers': {'Content-Type': 'text/html'}, 'data': fig.to_html()}\n", "fig_to_html(fig)" ] }, { "cell_type": "markdown", "id": "3a80a2f7", "metadata": {}, "source": [ "## Insert Dynamic Content to update with Asset Selection and Date Range" ] }, { "cell_type": "code", "execution_count": 3, "id": "b5ea7f96-70d2-4ab2-81ab-c096cfdcee07", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "\u001b[0;31mNameError: \u001b[0mname 'REQUEST' is not defined\n", "\n", "Error found at \u001b[0;36mline 7\u001b[0m in \u001b[0;32mcell 3\u001b[0m.\n" ] }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "4946de30465946b6a8dd13442494c61f", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Button(description='Click to show stack trace', layout=Layout(height='auto', width='auto'), style=ButtonStyle(…" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# GET /TBL2\n", "# data-lab//functions/notebooks/example/endpoints/TBL2\n", "\n", "from seeq import spy\n", "from datetime import datetime\n", "\n", "asset=REQUEST['args']['asset']\n", "#our epoch timestamps come back in milliseconds, so we have to convert them\n", "startdate=datetime.fromtimestamp(int(REQUEST['args']['start'])//1000)\n", "enddate=datetime.fromtimestamp(int(REQUEST['args']['end'])//1000)\n", "\n", "#get item\n", "#search for Temperature signal\n", "item = spy.search({\n", " 'Asset': asset,\n", " 'Name': 'Temperature'\n", "})\n", "\n", "#pull for date range\n", "data = spy.pull(item, start=(startdate), end=(enddate),header='Name') #later return start and end from request parameters\n", "{'headers': {'Content-Type': 'text/html'}, 'data': data}" ] }, { "cell_type": "markdown", "id": "c17f7eca-42be-4d60-af58-9923e4869f6d", "metadata": {}, "source": [ "## Insert Dynamic Text that displays recent signal value" ] }, { "cell_type": "code", "execution_count": null, "id": "1de0ccaa-e9d8-4f2f-9a1a-7050db27f3c4", "metadata": {}, "outputs": [], "source": [ "# GET /text_string\n", "# Update the URL from this project into the partial link below, then use that \n", "# in the URL entry field in Organizer (excluding the #)\n", "# data-lab//functions/notebooks/example/endpoints/text_string\n", "\n", "import rich\n", "from seeq import spy\n", "from rich.console import Console\n", "from rich.text import Text\n", "from datetime import datetime, timedelta\n", "\n", "console = Console()\n", "\n", "#Current time - weird timezone stuff so I am shifting \"now\" to have data not NAN at the end of the df\n", "current_datetime=datetime.now()-timedelta(hours=8)\n", "#1 hr earlier\n", "hour_prior=current_datetime - timedelta(hours=10)\n", "#find string signal you want to display in Organizer\n", "signal=spy.search({'Name':'Area A_Compressor Stage','Datasource Name':'Example Data'})\n", "#pull data for representative time frame\n", "data=spy.pull(signal,start=hour_prior, end=current_datetime,grid='1h',tz_convert='UTC')\n", "\n", "# Get the dataframe's last row of text for the signal displayed\n", "last_row_text = Text(data['Area A_Compressor Stage'].iloc[-1])\n", "\n", "# Print that strig with bold formatting\n", "console.print(last_row_text, style=\"bold\" )" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.8.18" } }, "nbformat": 4, "nbformat_minor": 5 }