From 7d1e5dace398e65cc75d61cf689067c41ec4ca08 Mon Sep 17 00:00:00 2001
From: daxid <david.rouquet@tetras-libre.fr>
Date: Mon, 27 Feb 2023 14:55:15 +0000
Subject: [PATCH] First selfContainedDemo working
---
selfContainedDemo.ipynb | 302 ++++++++++++++++++++++++++++++++++++++++
1 file changed, 302 insertions(+)
create mode 100644 selfContainedDemo.ipynb
diff --git a/selfContainedDemo.ipynb b/selfContainedDemo.ipynb
new file mode 100644
index 0000000..67cfb43
--- /dev/null
+++ b/selfContainedDemo.ipynb
@@ -0,0 +1,302 @@
+{
+ "cells": [
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "0d91640d-23ea-4079-b765-2eea030926c5",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import importlib.util\n",
+ "import re\n",
+ "import amrlib\n",
+ "from amrlib.graph_processing.amr_plot import AMRPlot\n",
+ "import uuid\n",
+ "from IPython.display import SVG, display\n",
+ "import os\n",
+ "import shutil\n",
+ "import subprocess\n",
+ "from subprocess import Popen, PIPE, STDOUT\n",
+ "from glob import glob\n",
+ "import sys\n",
+ "import os\n",
+ "TENET_PATH = \"/opt/dashboards/TetrasMARS/tenet/\"\n",
+ "sys.path.insert(0, os.path.abspath(TENET_PATH))\n",
+ "import tenet\n",
+ "from IPython.display import HTML,IFrame\n",
+ "import ipywidgets\n",
+ "import dash_bootstrap_components as dbc\n",
+ "from dash import dcc, html, Input, Output\n",
+ "from jupyter_dash import JupyterDash as Dash\n",
+ "from dash.dependencies import Input, Output, State\n",
+ "import base64\n",
+ "\n",
+ "MEDIA_PATH = \"/opt/dashboards/media/17/\"\n",
+ "MEDIA_URL = \"https://unsel.tetras-lab.io/dashboard/17/media/\"\n",
+ "ROOT_PATH = \"/opt/dashboards/TetrasMARS/tetras-mars-demo/\"\n",
+ "AMRLD_PATH = \"/opt/dashboards/TetrasMARS/tetras-mars-demo/lib/amrld/\"\n",
+ "owl2vowlPath = '/opt/dashboards/tools/owl2vowl_0.3.7/owl2vowl.jar'\n",
+ "WEBVOWL_PATH = '/opt/webvowl/'\n",
+ "onto_prefix=\"ontologyTarget\"\n",
+ "\n",
+ "# The following is basically `import tenet`\n",
+ "#spec=importlib.util.spec_from_file_location(\"tenet\",TENET_PATH+'tenet/__init__.py')\n",
+ "#tenet = importlib.util.module_from_spec(spec)\n",
+ "#spec.loader.exec_module(tenet)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "295e4aef-bbd8-40f0-8d84-0b8032b7b039",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "stog = amrlib.load_stog_model(model_dir=\"/opt/dashboards/TetrasMARS/corpus/cm-tool/amrModel/model_parse_xfm_bart_large-v0_1_0\") \n",
+ " \n",
+ "uuidStr = str(uuid.uuid4())\n",
+ "uuidDirPath = \"/opt/data/tmp/demo-tetras-mars/\"+uuidStr+'/'\n",
+ "os.mkdir(uuidDirPath)\n",
+ "prefixPath = uuidDirPath+\"file\"\n",
+ "penmanPath = prefixPath+\".amr.penman\"\n",
+ "svgPath = prefixPath+\".amr.svg\"\n",
+ "ttlFilePath = uuidDirPath+onto_prefix+\"-0/\"+onto_prefix+\"_factoid.ttl\"\n",
+ "webvowlFileName = ttlFilePath.split('/')[-1].replace('ttl','json')\n",
+ "webvowlFilepath = WEBVOWL_PATH+uuidStr+'_'+webvowlFileName\n",
+ "uuidZipPath = MEDIA_PATH+uuidStr # without the .zip extention\n",
+ "uuidZipUrl = MEDIA_URL+uuidStr+\".zip\""
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "a410a6b3-865d-441f-9b83-90a1badae291",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def clean_sting(string):\n",
+ " \"\"\" Sentence cleanup as needed \"\"\"\n",
+ " return re.sub(\"(\\.)*\\\\n\", \"\", string)\n",
+ "\n",
+ "def string2amr(string,stog):\n",
+ " stog_result = stog.parse_sents([clean_sting(string)], add_metadata=True) \n",
+ " return stog_result[0] \n",
+ "\n",
+ "def show_svg(path):\n",
+ " display(SVG(filename=path))\n",
+ " \n",
+ "def add_id_in_penman_if_needed(penmanStr,uuidStr):\n",
+ " if not(penmanStr.startswith('# ::id')):\n",
+ " penmanStr = '# ::id '+uuidStr+'\\n'+penmanStr\n",
+ " return penmanStr\n",
+ "\n",
+ "def owl2vowl(ttlFilePath, uuid='', importList=[]):\n",
+ " # Run java parser\n",
+ " if importList == []:\n",
+ " cmd = ['java', '-jar', owl2vowlPath,\n",
+ " '-file', ttlFilePath] \n",
+ " else:\n",
+ " cmd = ['java', '-jar', owl2vowlPath,\n",
+ " '-file', ttlFilePath,\n",
+ " '-dependencies'] + importList \n",
+ " with Popen(cmd, stdout=PIPE, stderr=STDOUT) as p:\n",
+ " p.wait()\n",
+ " p.stdout.flush()\n",
+ " if p.returncode != 0:\n",
+ " print(\"Error in owl2vowl: \\n\\n\"+p.stdout.read().decode())\n",
+ " os.rename(webvowlFileName, webvowlFilepath)\n",
+ " \n",
+ "def localImage2htmlImg(imgPath):\n",
+ " with open(imgPath, \"rb\") as image_file:\n",
+ " img_data = base64.b64encode(image_file.read())\n",
+ " img_data = img_data.decode()\n",
+ " img_data = \"data:image/svg+xml;base64,{}\".format(img_data)\n",
+ " # ...\n",
+ " return html.Img(id=\"tag_id\", src=img_data, width=\"100%\", height=\"100%\", className=\"img_class\")#, alt=\"my image\""
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "5fd9cf0c-990a-4776-b206-8cc94f87c7be",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def processStr(input):\n",
+ " penmanStr = string2amr(input,stog)\n",
+ " format = 'svg'\n",
+ " penmanStr = add_id_in_penman_if_needed(penmanStr,uuidStr)\n",
+ " penmanFile = open(penmanPath,\"w\")\n",
+ " penmanFile.write(penmanStr)\n",
+ " penmanFile.close()\n",
+ " plot = AMRPlot(uuidDirPath+\"/file.amr\", format) \n",
+ " plot.build_from_graph(penmanStr)\n",
+ " plot.graph.render()\n",
+ " amrldWorkPenmanFilepath = AMRLD_PATH+\"/wk/\"+uuidStr+\".amr.penman\"\n",
+ " amrldWorkNtFilepath = AMRLD_PATH+\"/wk/\"+uuidStr+\".amr.nt\"\n",
+ "\n",
+ " amrNtPath = prefixPath+\".amr.nt\" \n",
+ " amrTtlPath = prefixPath+\".amr.ttl\" \n",
+ " os.chdir(AMRLD_PATH)\n",
+ "\n",
+ " amrld_process = [\"python3\", \"amr_to_rdf.py\", \n",
+ " \"-i\", penmanPath, \n",
+ " \"-o\", amrTtlPath,\n",
+ " \"-f\", \"ttl\" ]\n",
+ " subprocess.run(amrld_process) \n",
+ "\n",
+ " # Besoin de se mettre dans le répertoire tenet jusqu'à résolution du ticket https://gitlab.tetras-libre.fr/tetras-mars/tenet/-/issues/133\n",
+ " os.chdir(TENET_PATH+'tenet/')\n",
+ " factoids = tenet.create_ontology_from_amrld_file(amrTtlPath,\n",
+ " onto_prefix=onto_prefix, # \"https://tenet.tetras-libre.fr/demo/\",\n",
+ " out_file_path=uuidDirPath+\"factoid.ttl\",\n",
+ " technical_dir_path=uuidDirPath)\n",
+ " webvowlFilepath = owl2vowl(ttlFilePath,uuid=uuidStr)\n",
+ " shutil.make_archive(uuidZipPath, 'zip', uuidDirPath)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "0cd10e8b-cf7a-4fd4-b8ac-540fcb943325",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "##################################################################################################\n",
+ "# THE FOLLOWING PART IS SPECIFIC TO TÉTRAS LAB\n",
+ "#\n",
+ "# The _get_tl_config function gets configuration parameters for your\n",
+ "# Tétras Lab instance.\n",
+ "# Those parameters are passed when initialising the Dash app.\n",
+ "##################################################################################################\n",
+ "def _get_tl_config():\n",
+ " import socket, errno, os\n",
+ " # Find a free port\n",
+ " host = \"0.0.0.0\"\n",
+ " port = 8050\n",
+ " end = 9999\n",
+ " found = False\n",
+ " while not found:\n",
+ " with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:\n",
+ " try:\n",
+ " s.bind((host, port))\n",
+ " found = True\n",
+ " except socket.error as e:\n",
+ " if e.errno == errno.EADDRINUSE:\n",
+ " port = port + 1\n",
+ " if (port > end):\n",
+ " raise \"No available APP port\"\n",
+ " else:\n",
+ " raise e\n",
+ " if (os.getenv(\"HOST\", None) is not None):\n",
+ " proto = os.getenv(\"PROTO\")\n",
+ " actualhost = os.getenv(\"JUPYTER_HOST\", os.getenv(\"VOILA_HOST\", \"\"))\n",
+ " localport = os.getenv(\"PORT\", 80)\n",
+ " intermediatehost = os.getenv(\"HOST\", \"localhost\")\n",
+ " base_path = f\"/{actualhost}/app_proxy/{port}/\"\n",
+ " proxified= f\"{proto}://{intermediatehost}:{localport}{base_path}\"\n",
+ " localurl = f\"http://{host}:{port}\"\n",
+ " proxy = f\"{localurl}::{proxified}\"\n",
+ " return ((proxified, host, port, proxy, base_path))\n",
+ " return ((f\"http://localhost:{port}\", host, port, None, \"/\"))\n",
+ "\n",
+ "server_url, host, port, proxy, base_path = _get_tl_config()\n",
+ "\n",
+ "app = Dash(\n",
+ " server_url=server_url, \n",
+ " requests_pathname_prefix=base_path,\n",
+ ")\n",
+ "##################################################################################################\n",
+ "\n",
+ "\n",
+ "##################################################################################################\n",
+ "# THE FOLLOWING PART IS GENERIC (JUPYTER)-DASH CODE FROM https://dash.plotly.com/basic-callbacks\n",
+ "#\n",
+ "# The _get_tl_config function gets configuration parameters for your\n",
+ "# Tétras Lab instance.\n",
+ "# Those parameters are passed when initialising the Dash app.\n",
+ "##################################################################################################\n",
+ "\n",
+ "app.layout = html.Div([\n",
+ " dcc.Textarea(\n",
+ " id='textarea-state',\n",
+ " value='Jupyter is a gas giant.',\n",
+ " style={'width': '100%', 'height': 200},\n",
+ " ),\n",
+ " html.Button('Construct AMR graphs and extract ontology', id='textarea-state-button', n_clicks=0),\n",
+ " #html.Button('Download result as zip', id='download-zip-button', n_clicks=0),\n",
+ " html.A(children=\"\", href='', target=\"_blank\",id=\"download-link\"),\n",
+ " dcc.Loading(html.Div(id='my-output'), color='#5A8264')\n",
+ "])\n",
+ "\n",
+ "@app.callback(\n",
+ " #Output('textarea-state-output', 'children'),\n",
+ " Output(component_id='my-output', component_property='children'),\n",
+ " Output(component_id='download-link', component_property='children'),\n",
+ " Output(component_id='download-link', component_property='href'), \n",
+ " Input('textarea-state-button', 'n_clicks'),\n",
+ " State('textarea-state', 'value'),\n",
+ " prevent_initial_call=True,\n",
+ ")\n",
+ "def update_output(n_clicks, value):\n",
+ " if n_clicks > 0:\n",
+ " processStr(value)\n",
+ " #show_svg(svgPath) \n",
+ " #display(IFrame('''https://unsel.tetras-lab.io/webvowl/#{}\">'''.format(webvowlFilepath.replace(\"/opt/webvowl/\",\"\").replace(\".json\",\"\")),800,1200))\n",
+ " return [[\n",
+ " html.Iframe(src='''https://unsel.tetras-lab.io/webvowl/#{}'''.format(webvowlFilepath.replace(\"/opt/webvowl/\",\"\").replace(\".json\",\"\")),\n",
+ " style={\"height\": \"800px\", \"width\": \"100%\"}),\n",
+ " localImage2htmlImg(svgPath)\n",
+ " ],\n",
+ " \"Download Zip File\", uuidZipUrl\n",
+ " ]\n",
+ "\n",
+ "#@app.callback(\n",
+ "# Output(\"download-zip\", \"data\"),\n",
+ "# Input(\"download-zip-button\", \"n_clicks\"),\n",
+ "# prevent_initial_call=True,\n",
+ "#)\n",
+ "#def func(n_clicks):\n",
+ "# if n_clicks > 0:\n",
+ "# return dcc.send_file('https://unsel.tetras-lab.io/dashboard/17/media/9f7287d0-e7b2-4328-9137-7a7c44225b68.zip')\n",
+ " \n",
+ "\n",
+ " \n",
+ "app.run_server(mode=\"inline\", \n",
+ " host=host, port=port, proxy=proxy, height=2000)\n",
+ "##################################################################################################\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "4cbaf48e-3467-45b0-b931-20a758b79895",
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "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": "3.8.6"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
--
GitLab