{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Export your design to GDS"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For convenience, let's begin by enabling [automatic reloading of modules](https://ipython.readthedocs.io/en/stable/config/extensions/autoreload.html?highlight=autoreload) when they change."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Will export_to_gds() add to my GDS file?   \n",
    "Most likely, yes!\n",
    "\n",
    "The cells below describe what is in an exported gds file. Geometries that an user wants to add for a component, can be put on a layer chosen by user. What export_to_gds does in the background is not finalized and will most likely be changed, based on a format, that is usable by multiple types of users.  Layers represent manufacturing intention, thus proprietary information, which cannot be published outside scientific papers or fab-customer relationships. While on our roadmap, we have the enablement, of a techfile-driven setup for layers. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import Qiskit Metal"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import qiskit_metal as metal\n",
    "from qiskit_metal import designs, draw\n",
    "from qiskit_metal import MetalGUI, Dict, Headings\n",
    "from qiskit_metal.qlibrary.qubits.transmon_pocket import TransmonPocket\n",
    "from qiskit_metal.qlibrary.qubits.transmon_cross import TransmonCross"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "design = designs.DesignPlanar()\n",
    "gui = MetalGUI(design)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "design.overwrite_enabled = True\n",
    "design.delete_all_components() \n",
    "gui.rebuild() # refresh"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "     <h1 style=\"         background-color: #d4418e;         background-image: linear-gradient(315deg, #d4418e 0%, #0652c5 74%);         margin-top: 50px;         border-style: outset;         padding-top:100px;         padding-bottom:50px;         padding-left:25px;         color: white;     \"> Populate QDesign to demonstrate exporting to GDS format. <h1>     "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Headings.h1('Populate QDesign to demonstrate exporting to GDS format.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "from qiskit_metal.qlibrary.qubits.transmon_pocket import TransmonPocket\n",
    "\n",
    "# Allow running the same cell here multiple times to overwrite changes.\n",
    "design.overwrite_enabled = True\n",
    "\n",
    "## Custom options for all the transmons.\n",
    "options = dict(\n",
    "    # Some options we want to modify from the defaults.\n",
    "    # (see below for defaults)\n",
    "    pad_gap = '80 um',\n",
    "    pad_width = '425 um', \n",
    "    pocket_height = '650um',\n",
    "    # Adding 4 connectors (see below for defaults)\n",
    "    connection_pads=dict( \n",
    "        a = dict(loc_W=+1,loc_H=+1), \n",
    "        b = dict(loc_W=-1,loc_H=+1, pad_height='30um'),\n",
    "        c = dict(loc_W=+1,loc_H=-1, pad_width='200um'),\n",
    "        d = dict(loc_W=-1,loc_H=-1, pad_height='50um')\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Note repeated from \"Introduction to QRenderers.ipynb\":  \n",
    "The cell name denoted by, \"gds_cell_name\" will be the selected cell   \n",
    "from design.renderers.gds.options['path_filename']   \n",
    "when design.renderers.gds.export_to_gds() is executed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Create 4 TransmonPockets\n",
    "\n",
    "# For variety and demonstartion, use different gds_cell_names.\n",
    "q1 = TransmonPocket(design, 'Q1', options = dict(\n",
    "    pos_x='+2.55mm', pos_y='+0.0mm', gds_cell_name='FakeJunction_02', **options))\n",
    "q2 = TransmonPocket(design, 'Q2', options = dict(\n",
    "    pos_x='+0.0mm', pos_y='-0.9mm', orientation = '90', gds_cell_name='FakeJunction_01', **options))\n",
    "q3 = TransmonPocket(design, 'Q3', options = dict(\n",
    "    pos_x='-2.55mm', pos_y='+0.0mm', gds_cell_name='FakeJunction_01',**options))\n",
    "q4 = TransmonPocket(design, 'Q4', options = dict(\n",
    "    pos_x='+0.0mm', pos_y='+0.9mm', orientation = '90', gds_cell_name='my_other_junction', **options))\n",
    "\n",
    "## Rebuild the design\n",
    "gui.rebuild()\n",
    "gui.autoscale()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Connecting QPins with coplanar waveguides (CPWs) as described in earlier notebooks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from qiskit_metal.qlibrary.tlines.meandered import RouteMeander\n",
    "RouteMeander.get_template_options(design)\n",
    "\n",
    "options = Dict(\n",
    "    meander=Dict(\n",
    "        lead_start='0.1mm',\n",
    "        lead_end='0.1mm',\n",
    "        asymmetry='0 um')\n",
    ")\n",
    "\n",
    "\n",
    "def connect(component_name: str, component1: str, pin1: str, component2: str, pin2: str,\n",
    "            length: str, asymmetry='0 um', flip=False, fillet='50um'):\n",
    "    \"\"\"Connect two pins with a CPW.\"\"\"\n",
    "    myoptions = Dict(\n",
    "        fillet=fillet,\n",
    "        pin_inputs=Dict(\n",
    "            start_pin=Dict(\n",
    "                component=component1,\n",
    "                pin=pin1),\n",
    "            end_pin=Dict(\n",
    "                component=component2,\n",
    "                pin=pin2)),\n",
    "        lead=Dict(\n",
    "            start_straight='0.13mm',\n",
    "            end_straight='0.13mm'\n",
    "        ),\n",
    "        total_length=length)\n",
    "    myoptions.update(options)\n",
    "    myoptions.meander.asymmetry = asymmetry\n",
    "    myoptions.meander.lead_direction_inverted = 'true' if flip else 'false'\n",
    "    return RouteMeander(design, component_name, myoptions)\n",
    "\n",
    "\n",
    "asym = 90\n",
    "\n",
    "# For variety in output, use different fillet values.\n",
    "cpw1 = connect('cpw1', 'Q1', 'd', 'Q2', 'c', '5.7 mm', f'+{asym}um', fillet='25um')\n",
    "cpw2 = connect('cpw2', 'Q3', 'c', 'Q2', 'a', '5.4 mm', f'-{asym}um', flip=True, fillet='100um')\n",
    "cpw3 = connect('cpw3', 'Q3', 'a', 'Q4', 'b', '5.3 mm', f'+{asym}um', fillet='75um')\n",
    "cpw4 = connect('cpw4', 'Q1', 'b', 'Q4', 'd', '5.5 mm', f'-{asym}um', flip=True)\n",
    "\n",
    "gui.rebuild()\n",
    "gui.autoscale()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {
      "image/png": {
       "width": 500
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "gui.screenshot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "     <h1 style=\"         background-color: #d4418e;         background-image: linear-gradient(315deg, #d4418e 0%, #0652c5 74%);         margin-top: 50px;         border-style: outset;         padding-top:100px;         padding-bottom:50px;         padding-left:25px;         color: white;     \"> Exporting to a GDS file. <h1>     "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Headings.h1('Exporting to a GDS file.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'short_segments_to_not_fillet': 'True',\n",
       " 'check_short_segments_by_scaling_fillet': '2.0',\n",
       " 'gds_unit': 0.001,\n",
       " 'ground_plane': 'True',\n",
       " 'negative_mask': {'main': []},\n",
       " 'fabricate': 'False',\n",
       " 'corners': 'circular bend',\n",
       " 'tolerance': '0.00001',\n",
       " 'precision': '0.000000001',\n",
       " 'width_LineString': '10um',\n",
       " 'path_filename': '../resources/Fake_Junctions.GDS',\n",
       " 'junction_pad_overlap': '5um',\n",
       " 'max_points': '199',\n",
       " 'cheese': {'datatype': '100',\n",
       "  'shape': '0',\n",
       "  'cheese_0_x': '25um',\n",
       "  'cheese_0_y': '25um',\n",
       "  'cheese_1_radius': '100um',\n",
       "  'view_in_file': {'main': {1: True}},\n",
       "  'delta_x': '100um',\n",
       "  'delta_y': '100um',\n",
       "  'edge_nocheese': '200um'},\n",
       " 'no_cheese': {'datatype': '99',\n",
       "  'buffer': '25um',\n",
       "  'cap_style': '2',\n",
       "  'join_style': '2',\n",
       "  'view_in_file': {'main': {1: True}}},\n",
       " 'bounding_box_scale_x': '1.2',\n",
       " 'bounding_box_scale_y': '1.2'}"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#QDesign registers GDS renderer during init of QDesign.\n",
    "a_gds = design.renderers.gds\n",
    "# An alternate way to invoke gds commands without using a_gds:\n",
    "# design.renderers.gds.export_to_gds()\n",
    "\n",
    "#Show the options for GDS\n",
    "a_gds.options"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Will export_to_gds() add datatype for Polygon and Flexpath to my GDS file. \n",
    "Yes.  If you see a datatype being added to your positive mask, datatype=10 means a Polygon was added to the layer, datatype=11 means Flexpath was added to the layer.   However, `for each layer` that has a negative mask, due to boolean subtract command done by export_to_gds, the 10 and 11 datatypes will not be in the gds file.\n",
    "\n",
    "For positive mask, datatype=0 is for the ground plane. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  To make `junction table` work correctly, GDS Renderer needs a correct path to a gds file, which has cells.  \n",
    "Each cell is a junction, to be placed, in a Transmon.  A sample gds file is provided in directory `qiskit_metal/tutorials/resources`.\n",
    "There are three cells with names \"Fake_Junction_01\", \"Fake_Junction_01\", and \"my_other_junction\".\n",
    "The default name used by GDS Render is \"my_other_junction\".  If you want to customize and select a junction, through the QComponent's options, you can pass it when a qcomponent is being added to QDesign. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## If your added junctions to a qcomponent.\n",
    "\n",
    "The junctions are imported from file at gds.options.path_filename.  The cell from path_filename which is denoted by TransmonPocket.options.gds_cell_name is imported and placed at location determined by the qcomponent developer.  The method export_to_gds will place the imported cell at the same layers which are identified in the file at gds.options.path_filename."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "a_gds.options['path_filename'] = '../resources/Fake_Junctions.GDS'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Do you want GDS Renderer to fix any short-segments in your QDesign when using fillet?'\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "#If you have a fillet_value and there are LineSegments that are shorter than 2*fillet_value, \n",
    "#When true, the short segments will not be fillet'd. \n",
    "a_gds.options['short_segments_to_not_fillet'] = 'False'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "scale_fillet = 2.0\n",
    "a_gds.options['check_short_segments_by_scaling_fillet'] = scale_fillet"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What criteria will be used for identifying a short segment?\n",
    "If a segment is smaller than (fillet length *  scale_fillet)\n",
    "\n",
    "## What if a segment of LineString has few short segments?  \n",
    "If option 'short_segments_to_not_fillet' == 'True', QGDSRenderer will\n",
    "break the LineString into shorter Linestrings to make smaller LineStrings that will be \n",
    "either fillet'd or not, based on if the segment is short."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "#If you want to have the short segments not be fillet'd.\n",
    "a_gds.options['short_segments_to_not_fillet'] = 'True'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Export to a GDS formatted file for all components in design.\n",
    "#def export_to_gds(self, file_name: str, highlight_qcomponents: list = []) -> int:\n",
    "\n",
    "a_gds.export_to_gds('GDS QRenderer Notebook.gds')\n",
    "# You can also specify a different path. Example:\n",
    "#   a_gds.export_to_gds(\"../../../gds-files/GDS QRenderer Notebook.gds\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Export a GDS file which contains only few components.\n",
    "\n",
    "# You will probably want to put the exported file in a specific directory.  \n",
    "# Please give the full path for output. \n",
    "a_gds.export_to_gds(\"four_qcomponents.gds\",\n",
    "                           highlight_qcomponents=['cpw1', 'cpw4', 'Q1', 'Q3'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Export a GDS file using explicit path and cpw1.name vs typing string.\n",
    "# You will probably want to put the exported file in a specific directory.  \n",
    "# Please give the full path for output. \n",
    "a_gds.export_to_gds(\"four_same_qcomponents.gds\",\n",
    "                           highlight_qcomponents=[cpw1.name, 'cpw4', q1.name, 'Q3'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "     <h1 style=\"         background-color: #d4418e;         background-image: linear-gradient(315deg, #d4418e 0%, #0652c5 74%);         margin-top: 50px;         border-style: outset;         padding-top:100px;         padding-bottom:50px;         padding-left:25px;         color: white;     \"> QUESTION:  Where is the geometry of a QComponent placed? <h1>     "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Headings.h1('QUESTION:  Where is the geometry of a QComponent placed?')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Answer:  QGeometry tables!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is better explained in folder \"2 Front End User/2.4 QRenderer Introduction\" in notebook, \"QRenderer Introduction \"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "     <h1 style=\"         background-color: #d4418e;         background-image: linear-gradient(315deg, #d4418e 0%, #0652c5 74%);         margin-top: 50px;         border-style: outset;         padding-top:100px;         padding-bottom:50px;         padding-left:25px;         color: white;     \"> What does GDS do with \"junction\" table? <h1>     "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Headings.h1('What does GDS do with \"junction\" table?')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The junction table is handled differently by each QRenderer.\n",
    "\n",
    "GDS QRenderer gets a cell, with the name, equal to \"gds_cell_name\" and places the cell into the QDesign before exporting the entire QDesign to GDS. In file: `a_gds.options['path_filename'] = '../resources/Fake_Junctions.GDS'`, the gds_cell_name is searched.  The cell is placed into QDesign using LINESTRING and width information.\n",
    "\n",
    "The cell within \"path_filename\", should be \"x-axis\" aligned and then GDS rotates based on LineString.  The LineString should be two vertexes and it denotes two things.\n",
    "1. The midpoint of segment is the center of cell. \n",
    "2. The angle made by (second tuple - fist tuple), for delta y/ delta x, is used to rotate the cell. \n",
    "\n",
    "When the cell from default_options.path_filename does not fit the width of LineString,\n",
    "- QGDSRender will create two pads and add to cell, which is denoted in junction table, to fill the width of LineString. The length of the additional pads is the value of \"width\" from the junction table. \n",
    "- The option `a_gds.options[\"junction_pad_overlap\"]='5um'` is the amount the new pads will overlap the cell. The final width of the cell plus two pads is equal to the magnitude of LineString. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>component</th>\n",
       "      <th>name</th>\n",
       "      <th>geometry</th>\n",
       "      <th>layer</th>\n",
       "      <th>subtract</th>\n",
       "      <th>helper</th>\n",
       "      <th>chip</th>\n",
       "      <th>width</th>\n",
       "      <th>hfss_inductance</th>\n",
       "      <th>hfss_capacitance</th>\n",
       "      <th>hfss_resistance</th>\n",
       "      <th>hfss_mesh_kw_jj</th>\n",
       "      <th>q3d_inductance</th>\n",
       "      <th>q3d_capacitance</th>\n",
       "      <th>q3d_resistance</th>\n",
       "      <th>q3d_mesh_kw_jj</th>\n",
       "      <th>gds_cell_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>rect_jj</td>\n",
       "      <td>LINESTRING (2.55000 -0.04000, 2.55000 0.04000)</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>main</td>\n",
       "      <td>0.02</td>\n",
       "      <td>10nH</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>10nH</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>FakeJunction_02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>rect_jj</td>\n",
       "      <td>LINESTRING (0.04000 -0.90000, -0.04000 -0.90000)</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>main</td>\n",
       "      <td>0.02</td>\n",
       "      <td>10nH</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>10nH</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>FakeJunction_01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>rect_jj</td>\n",
       "      <td>LINESTRING (-2.55000 -0.04000, -2.55000 0.04000)</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>main</td>\n",
       "      <td>0.02</td>\n",
       "      <td>10nH</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>10nH</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>FakeJunction_01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>rect_jj</td>\n",
       "      <td>LINESTRING (0.04000 0.90000, -0.04000 0.90000)</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>main</td>\n",
       "      <td>0.02</td>\n",
       "      <td>10nH</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>10nH</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>my_other_junction</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  component     name                                          geometry  layer  \\\n",
       "0         1  rect_jj    LINESTRING (2.55000 -0.04000, 2.55000 0.04000)      1   \n",
       "1         2  rect_jj  LINESTRING (0.04000 -0.90000, -0.04000 -0.90000)      1   \n",
       "2         3  rect_jj  LINESTRING (-2.55000 -0.04000, -2.55000 0.04000)      1   \n",
       "3         4  rect_jj    LINESTRING (0.04000 0.90000, -0.04000 0.90000)      1   \n",
       "\n",
       "   subtract  helper  chip  width hfss_inductance  hfss_capacitance  \\\n",
       "0     False   False  main   0.02            10nH                 0   \n",
       "1     False   False  main   0.02            10nH                 0   \n",
       "2     False   False  main   0.02            10nH                 0   \n",
       "3     False   False  main   0.02            10nH                 0   \n",
       "\n",
       "   hfss_resistance  hfss_mesh_kw_jj q3d_inductance  q3d_capacitance  \\\n",
       "0                0         0.000007           10nH                0   \n",
       "1                0         0.000007           10nH                0   \n",
       "2                0         0.000007           10nH                0   \n",
       "3                0         0.000007           10nH                0   \n",
       "\n",
       "   q3d_resistance  q3d_mesh_kw_jj      gds_cell_name  \n",
       "0               0        0.000007    FakeJunction_02  \n",
       "1               0        0.000007    FakeJunction_01  \n",
       "2               0        0.000007    FakeJunction_01  \n",
       "3               0        0.000007  my_other_junction  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# View every entry in junction table.\n",
    "design.qgeometry.tables['junction']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>component</th>\n",
       "      <th>name</th>\n",
       "      <th>geometry</th>\n",
       "      <th>layer</th>\n",
       "      <th>subtract</th>\n",
       "      <th>helper</th>\n",
       "      <th>chip</th>\n",
       "      <th>width</th>\n",
       "      <th>hfss_inductance</th>\n",
       "      <th>hfss_capacitance</th>\n",
       "      <th>hfss_resistance</th>\n",
       "      <th>hfss_mesh_kw_jj</th>\n",
       "      <th>q3d_inductance</th>\n",
       "      <th>q3d_capacitance</th>\n",
       "      <th>q3d_resistance</th>\n",
       "      <th>q3d_mesh_kw_jj</th>\n",
       "      <th>gds_cell_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>rect_jj</td>\n",
       "      <td>LINESTRING (2.55000 -0.04000, 2.55000 0.04000)</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>main</td>\n",
       "      <td>0.02</td>\n",
       "      <td>10nH</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>10nH</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>FakeJunction_02</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  component     name                                        geometry  layer  \\\n",
       "0         1  rect_jj  LINESTRING (2.55000 -0.04000, 2.55000 0.04000)      1   \n",
       "\n",
       "   subtract  helper  chip  width hfss_inductance  hfss_capacitance  \\\n",
       "0     False   False  main   0.02            10nH                 0   \n",
       "\n",
       "   hfss_resistance  hfss_mesh_kw_jj q3d_inductance  q3d_capacitance  \\\n",
       "0                0         0.000007           10nH                0   \n",
       "\n",
       "   q3d_resistance  q3d_mesh_kw_jj    gds_cell_name  \n",
       "0               0        0.000007  FakeJunction_02  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# View the juction table for component \"q1\". \n",
    "\n",
    "q1.qgeometry_table('junction')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Geometric boundary of a QComponent?\n",
    "How to obtain the boundary box of the geometry, for example: `q1.qgeometry_bounds()`.  \n",
    "Explained in in folder \"2 Front End User/2.4 QRenderer Introduction\" in notebook, \"QRenderer Introduction \"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'short_segments_to_not_fillet': 'True',\n",
       " 'check_short_segments_by_scaling_fillet': 2.0,\n",
       " 'gds_unit': 0.001,\n",
       " 'ground_plane': 'True',\n",
       " 'negative_mask': {'main': []},\n",
       " 'fabricate': 'False',\n",
       " 'corners': 'circular bend',\n",
       " 'tolerance': '0.00001',\n",
       " 'precision': '0.000000001',\n",
       " 'width_LineString': '10um',\n",
       " 'path_filename': '../resources/Fake_Junctions.GDS',\n",
       " 'junction_pad_overlap': '5um',\n",
       " 'max_points': '199',\n",
       " 'cheese': {'datatype': '100',\n",
       "  'shape': '0',\n",
       "  'cheese_0_x': '25um',\n",
       "  'cheese_0_y': '25um',\n",
       "  'cheese_1_radius': '100um',\n",
       "  'view_in_file': {'main': {1: True}},\n",
       "  'delta_x': '100um',\n",
       "  'delta_y': '100um',\n",
       "  'edge_nocheese': '200um'},\n",
       " 'no_cheese': {'datatype': '99',\n",
       "  'buffer': '25um',\n",
       "  'cap_style': '2',\n",
       "  'join_style': '2',\n",
       "  'view_in_file': {'main': {1: True}}},\n",
       " 'bounding_box_scale_x': '1.2',\n",
       " 'bounding_box_scale_y': '1.2'}"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#The current value of all the options for GDS QRenderer.\n",
    "a_gds.options"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Number of vertices for linestrings.\n",
    "The option max_points has default of 199.  You can set it to a number no higher than 8191."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### For demo, set max_points to 8191 and look at the GDS output.\n",
    "a_gds.options['max_points'] = '8191'\n",
    "a_gds.export_to_gds('GDS QRenderer Notebook maxpoints8191.gds')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Changing options 'precision' vs 'tolerance' ratio can impact how fillet will look.\n",
    "\n",
    "The below numbers are used to create a gdspy.FlexPath when \n",
    "there is a fillet value in QGeometry table.\n",
    "\n",
    "bend_radius – Bend radii for each path when corners is ‘circular bend’. \n",
    "It has no effect for other corner types.  QGDSRender uses 'circular bend' as a default for corners.\n",
    "The fillet value used here.\n",
    "\n",
    "Ensure that tolerance > precision.\n",
    "\n",
    "tolerance (number) – Tolerance used to draw the paths and calculate joins.\n",
    "\n",
    "precision (number) – Precision for rounding the coordinates of vertices when fracturing the final polygonal boundary.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Restore previous example.\n",
    "a_gds.options['max_points'] = '199'\n",
    "\n",
    "# For Demo, change tolerance value.\n",
    "a_gds.options['tolerance'] = '0.01'\n",
    "# This exported file will not look as desired.\n",
    "a_gds.export_to_gds('GDS QRenderer Notebook change tolerance.gds')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "#change it back to was it was.\n",
    "a_gds.options['tolerance'] = '0.00001'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "     <h1 style=\"         background-color: #d4418e;         background-image: linear-gradient(315deg, #d4418e 0%, #0652c5 74%);         margin-top: 50px;         border-style: outset;         padding-top:100px;         padding-bottom:50px;         padding-left:25px;         color: white;     \"> Layer numbers: cheese and mask <h1>     "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Headings.h1('Layer numbers: cheese and mask')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What can I control by layer number?  \n",
    "\n",
    "* positive mask\n",
    "* negative mask\n",
    "* cheese\n",
    "* no_cheese\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## If you wanted cheese or no_cheese in the GDS file.\n",
    "\n",
    "\n",
    "The reason you choose to use cheesing is up to the developer.   There are considerations for both manufacturing and EM design.  This refers to the holes in the output GDS ground plane (cheese = holes). From an EM stand-point https://www.degruyter.com/document/doi/10.1515/psr-2017-8000/html.\n",
    "\n",
    "Cheesing refers to the holes added to the chip. The “cheese” output will correspond to the mask (positive or negative) selected.  Both cheese and no_cheese output can be generated for any layer.  You can choose to add the no_cheese to the exported GDS file; adding no_cheese information in GDS file, is just to show the intermediate step.\n",
    "\n",
    "In the cheese dict, edge_nocheese option denotes a buffer around the perimeter of the chip.  You can identify which layer has the cheesing by using the key view_in_file.  To identify a datatype to store the geometry for cheesing, edit the key named datatype. \n",
    "\n",
    "The option no_cheese can be thought of as a “keep-out” region for cheesing.  Based on the components on the chip, a no-cheese region is determined.  The no_cheese region is created by adding a buffer around all the components.   In no_cheese dict, there is a key called buffer, that denotes the size around all the components for no_cheese.  Even if view_in_file may be True for cheese dict, user can make it false for no_cheese dict.  If the user wants to view the no_cheese region in the GDS output, keep it as True for each layer you are interested it.  \n",
    "\n",
    "The location of cheesing is determined by\n",
    "1. generating a grid of holes for the size of chip\n",
    "2. subtract the no_cheese region\n",
    "3. subtract the edge_nocheese region\n",
    "\n",
    "\n",
    "The method export_to_gds, uses the datatype in cheese dict, default is 100, and no_cheese dict, default is 99; then uses few extra datatypes by adding one or two.   The extra datatypes are to show some intermediate steps.  If you choose to view_in_file, then you will see cheese or no-cheese in the gds file.  \n",
    "\n",
    "TOP_chipname_layernumber_one_hole is a cell with just one hole.  The rest of the holes are made as a reference from the one hole.  The hole is placed with datatype_cheese+2.  This is expected to change when we agree to some convention. \n",
    "\n",
    "TOP_chipname_layernumber_NoCheese_datatype is a cell that holds the “keep-out” region for cheesing.  \n",
    "\n",
    "TOP_chipname_layernumber_Cheese_datatype is a cell that holds the cheesing region.  \n",
    "\n",
    "TOP_chipname_layernumber holds the geometry taken from Metal GUI.   \n",
    "\n",
    "\n",
    "\n",
    "TOP_chipname_layernumber_Cheese_diff is a cell that holds the difference between grid of holes on chip, minus the no_cheese region.  It is placed in datatype_cheese+1.  This is expected to change when we agree to some convention. \n",
    "\n",
    "`For negative mask`:\n",
    "TOP_chipname_layernumber_Cheese_diff is placed under TOP_chipname_layernumber.  \n",
    "\n",
    "\n",
    "`For positive mask`:\n",
    "TOP_chipname_layernumber_Cheese_diff is subtracted from ground and placed under TOP_chipname_layernumber. \n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Changing Cheesing options\n",
    "There are two dicts in default_options for cheesing.  One for selecting the no_cheese (keep-out region) for cheesing, the other for cheesing.  Use them to determine if you would like the cheesing cell and the no_cheese cell to be added to the GDS file.\n",
    "\n",
    "To create the no-cheese region, we take the existing components and add a buffer around the perimeter.  The size of the buffer is an option.  \n",
    "\n",
    "The output of the cells can be placed with a data type.  The user can select a different data type for both the no-cheese region and the cheese region.\n",
    "\n",
    "Regarding the cheese option, there is room for expansion, presently, the shape that available for a hole is a shape=0, which is a square.  \n",
    "\n",
    "The present placement of holes allows the user to apply edge_nocheese to the perimeter of the chip.  This allows the user to place holes at a perimeter smaller than the chip size.   Then a grid of holes is made using delta_x and delta_y. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'datatype': '99',\n",
       " 'buffer': '25um',\n",
       " 'cap_style': '2',\n",
       " 'join_style': '2',\n",
       " 'view_in_file': {'main': {1: True}}}"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a_gds.options.no_cheese"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'datatype': '100',\n",
       " 'shape': '0',\n",
       " 'cheese_0_x': '25um',\n",
       " 'cheese_0_y': '25um',\n",
       " 'cheese_1_radius': '100um',\n",
       " 'view_in_file': {'main': {1: True}},\n",
       " 'delta_x': '100um',\n",
       " 'delta_y': '100um',\n",
       " 'edge_nocheese': '200um'}"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a_gds.options.cheese"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is a dict option called view_in_file that denotes if the cells are added to GDS file.  The first sub-option is the chip name, the second sub-option is layer number, the third sub-option is  bool True/False. \n",
    "\n",
    "In our example, the chip is 'main' and the layer is 1.  We allow for expansion for multiple chips, and multiple ground layers. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a_gds.options['cheese']['view_in_file']['main'][1] = True\n",
    "a_gds.options['no_cheese']['view_in_file']['main'][1] = True\n",
    "a_gds.export_to_gds(\"GDS QRender_cheese_keepout.gds\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a_gds.options['cheese']['view_in_file']['main'][1] = True\n",
    "a_gds.options['no_cheese']['view_in_file']['main'][1] = False\n",
    "a_gds.export_to_gds(\"GDS QRender_cheese_only.gds\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "#For example, just move the cpws to layer=14. \n",
    "cpw1.options.layer = 14\n",
    "cpw2.options.layer = 14\n",
    "cpw3.options.layer = 14\n",
    "cpw4.options.layer = 14\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "gui.rebuild()\n",
    "\n",
    "# Get a list of all the qcomponents in QDesign and then zoom on them.\n",
    "all_component_names = design.components.keys()\n",
    "gui.zoom_on_components(all_component_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Turn on/off keep-out and cheesing.\n",
    "a_gds.options['cheese']['view_in_file']['main'][1] = True\n",
    "a_gds.options['no_cheese']['view_in_file']['main'][1] = True\n",
    "a_gds.options['cheese']['view_in_file']['main'][14] = True\n",
    "a_gds.options['no_cheese']['view_in_file']['main'][14] = True\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## If you want a positive and/or negative masks.  \n",
    "\n",
    "User can request a positive or negative mask by chip-name and layer-number.  The default is to use a positive mask.  If user wants to use a negative mask, then just identify the layers that should be a negative mask.  The user is allowed to mix the output per chip, per layer. Particularly, a chip can have both positive and negative masks; the granularity is by layer.  This is exemplified in this notebook by editing option “negative_mask”. This feature was implemented for, potential use, for \"flip-chip\".\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Export GDS file with negative masks for layers 1 and 14.\n",
    "a_gds.options['negative_mask'] = Dict(main=[1, 14])\n",
    "a_gds.export_to_gds('GDS QRenderer Notebook_neg_mask_for_layer_1_14.gds')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Export GDS files with positive mask for layer 1 and negative mask for layer 14. \n",
    "a_gds.options['negative_mask'] = Dict(main=[14])\n",
    "a_gds.export_to_gds('GDS QRenderer Notebook_neg_mask_for_layer_14.gds')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "     <h1 style=\"         background-color: #d4418e;         background-image: linear-gradient(315deg, #d4418e 0%, #0652c5 74%);         margin-top: 50px;         border-style: outset;         padding-top:100px;         padding-bottom:50px;         padding-left:25px;         color: white;     \"> Fabrication <h1>     "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Headings.h1('Fabrication')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is no universally accepted method to prepare for fabrication of a chip.  If this flag is True, cells created as part of intermittent steps would not be in the gds file. For example cells with names similar to “TOP_main_1_NoCheese_99” , “TOP_main_1_one_hole” and  “ground_main_1”."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "#By default, the fabrication flag is False.  \n",
    "a_gds.options['fabricate'] = 'True'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Export GDS files with positive mask for layer 1 and negative mask for layer 14 with fabricate flag as True. \n",
    "a_gds.options['negative_mask'] = Dict(main=[14])\n",
    "a_gds.export_to_gds('GDS QRenderer Notebook_neg_mask_for_layer_14_fabricate.gds')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  Qiskit Metal Version"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Qiskit Metal        0.0.4\n",
      "\n",
      "Basic\n",
      "____________________________________\n",
      " Python              3.9.5 | packaged by conda-forge | (default, Jun 19 2021, 00:22:33) [MSC v.1916 64 bit (AMD64)]\n",
      " Platform            Windows AMD64\n",
      " Installation path   c:\\users\\marcofacchini\\documents\\github\\qiskit-metal-public\\qiskit_metal\n",
      "\n",
      "Packages\n",
      "____________________________________\n",
      " Numpy               1.20.3\n",
      " Qutip               4.6.2\n",
      "\n",
      "Rendering\n",
      "____________________________________\n",
      " Matplotlib          3.4.2\n",
      "\n",
      "GUI\n",
      "____________________________________\n",
      " PySide2 version     5.13.2\n",
      " Qt version          5.12.9\n",
      " SIP version         Not installed\n",
      "\n",
      "IBM Quantum Team\n"
     ]
    }
   ],
   "source": [
    "metal.about();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# If you would like, close the gui \n",
    "gui.main_window.close()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.9.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}