graphly/plotly.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## plotly implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "(require '[clojupyter.misc.display :as display]\n",
    "         '[clojure.java.io :as io]\n",
    "         '[hiccup.core :as hiccup]\n",
    "         '[cheshire.core :as cheshire]\n",
    "         '[clj-http.client :as client]\n",
    "         '[clojure.repl :refer :all])  ;; to have all the repl goodies"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To access plotly api, we need an api key.  The standard way is to store the api key in `$HOME/.plotly/.credentials` in a json object as defined in https://plot.ly/python/getting-started/.  For that you will need to create a plotly account.\n",
    "\n",
    "(For whatever reason, I believe plotly's python api does not load the credentials automatically from this location.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "#'user/plotly-cred"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(defn get-plotly-credentials\n",
    "    \"Returs the plotly credentials in $HOME/.plotly/.credentials\"\n",
    "    []\n",
    "    (let [home (-> \"HOME\" System/getenv io/as-file)\n",
    "          credential-file (io/file home \".plotly\" \".credentials\")]\n",
    "        (if (.exists credential-file)\n",
    "            (-> credential-file slurp (cheshire/parse-string true))\n",
    "            {})))\n",
    "\n",
    "(def plotly-cred  (get-plotly-credentials))\n",
    ";; (println \"credentials are now loaded\" plotly-cred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "Next, we need to create \"grids\".  Grids are dataset stored on plotly that we can then plot.  The plotting is done in two phases, uploding the datasets and then plotting the datasets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "#'user/plot-response"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(defn post-to-endpoint\n",
    "  \"General method to POST to a plotly endpoint.\"\n",
    "  [endpoint payload cred]\n",
    "  (let [url endpoint\n",
    "        headers {:Plotly-Client-Platform \"curl\"}\n",
    "        auth [(:username cred) (:api_key cred)]]\n",
    "    (client/post url {:basic-auth auth\n",
    "                      :headers headers\n",
    "                      :throw-entire-message? true\n",
    "                      :form-params payload \n",
    "                      :content-type :json})))\n",
    "\n",
    "(defn post-to-grid [grid cred] \n",
    "    (post-to-endpoint \"https://api.plot.ly/v2/grids\" grid cred))\n",
    "\n",
    "(defn post-to-plot [grid cred] \n",
    "    (post-to-endpoint \"https://api.plot.ly/v2/plots\" grid cred))\n",
    "\n",
    "(defn plot-response [plotly-data]\n",
    "    (let [to-iframe (fn [url] [:iframe {:src url :width \"100%\" :height 200 :scrolling \"no\" :frameborder \"0\"}])]\n",
    "      (-> plotly-data\n",
    "          (:file)\n",
    "          (:embed_url)\n",
    "          to-iframe \n",
    "          (hiccup/html)\n",
    "          (display/make-html))))\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<iframe frameborder=\"0\" height=\"200\" scrolling=\"no\" src=\"https://plot.ly/~didier/78.embed\" width=\"100%\"></iframe>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(def my-grid {:data \n",
    "              {:cols\n",
    "               {\n",
    "                   :height {:data [10 203 102 404] :order 0}\n",
    "                   :weight1 {:data [112 234 321 34] :order 1}\n",
    "                   :weight2 {:data [112 234 321 34] :order 2}\n",
    "                   :weight3 {:data [112 234 321 34] :order 3}\n",
    "                   :weight4 {:data [112 234 321 34] :order 4}}}})\n",
    "\n",
    "(def resp-data (-> my-grid\n",
    "                   (post-to-grid plotly-cred)\n",
    "                   (:body)\n",
    "                   (cheshire/decode true)))\n",
    "(def plotly-data (plot-response resp-data))\n",
    "plotly-data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<iframe frameborder=\"0\" height=\"200\" scrolling=\"no\" src=\"https://plot.ly/~didier/79.embed\" width=\"100%\"></iframe>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(defn post-to-plot [grid cred] \n",
    "    (post-to-endpoint \"https://api.plot.ly/v2/plots\" grid cred))\n",
    "(def bar-char {:data [\n",
    "                         {\n",
    "                          :xsrc \"jackp:16731:7a9501\"  ;; someone else data.  but we could very well use the above grid using reference\n",
    "                          :ysrc \"jackp:16731:948d9c\"  ;; someone else data.  but we could very well use the above grid using reference\n",
    "                          :hoverinfo \"none\"\n",
    "                          :type \"bar\"\n",
    "                          :marker {:color \"rgb(227, 119, 194)\"}}]})\n",
    "\n",
    "(def payload (assoc {:world_readable true} :figure bar-char))\n",
    "\n",
    "(def plot-data (-> payload\n",
    "                   (post-to-plot  plotly-cred)\n",
    "                   (:body)\n",
    "                   (cheshire/decode true)\n",
    "                   (plot-response)))\n",
    "plot-data"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Lein-Clojure",
   "language": "clojure",
   "name": "lein-clojure"
  },
  "language_info": {
   "file_extension": ".clj",
   "mimetype": "text/x-clojure",
   "name": "clojure",
   "version": "1.9.0-RC1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}