split_shp_grid.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Divide Shapefile into Tiles\n",
    "This notebook divides an input shapefile by the coordinate grid into individual GeoJSONs or shapefiles. `coord_prec` denotes the edge length (in decimal coordinates) of each output tile. For each tile another output tile is produced with a buffer of `coord_prec / 100`, for use in generating spatial features rasters."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import json\n",
    "import geopandas as gpd\n",
    "import numpy as np\n",
    "import fiona\n",
    "from shapely.geometry import mapping, Polygon\n",
    "from matplotlib import pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Inputs and Directories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# set up directories\n",
    "INPUT_DIR = '../../data/ghana/GHA_adm0/'\n",
    "OUTPUT_DIR = '../../data/ghana/tiles'\n",
    "\n",
    "# set up inputs/outputs\n",
    "input_shp = 'GHA_adm0.shp'\n",
    "coord_prec = 1\n",
    "output_prefix = 'ghana'\n",
    "\n",
    "# read inputs\n",
    "in_shp = gpd.read_file(os.path.join(INPUT_DIR, input_shp))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Extract Bounds of Shapefile"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-3.,  5.,  1., 11.])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bounds_raw = in_shp.geometry.total_bounds\n",
    "bounds = np.array([bounds_raw[0].round(),\n",
    "                   bounds_raw[1].round(),\n",
    "                   bounds_raw[2].round(),\n",
    "                   bounds_raw[3].round()])\n",
    "bounds"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create Polygon Tiles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "tiles = []\n",
    "tiles_l = []\n",
    "numCols = (bounds[2] - bounds[0])\n",
    "numRows = (bounds[3] - bounds[1])\n",
    "\n",
    "# function to allow steps < 1\n",
    "def frange(start, stop, step):\n",
    "    i = start\n",
    "    while i < stop:\n",
    "        yield i\n",
    "        i += step\n",
    "\n",
    "# create tiles as Polygons with coord_prec edge length\n",
    "for k in frange(0, (numRows), coord_prec):\n",
    "    for i in frange(0, (numCols), coord_prec):\n",
    "        tile = Polygon([(bounds[0]+i,   bounds[1]+k),\n",
    "                        (bounds[0]+i+coord_prec, bounds[1]+k),\n",
    "                        (bounds[0]+i+coord_prec, bounds[1]+k+coord_prec),\n",
    "                        (bounds[0]+i,   bounds[1]+k+coord_prec)])\n",
    "        tiles.append(tile)\n",
    "\n",
    "# create larger Polygon tiles\n",
    "for tile in tiles:\n",
    "    l_buff = coord_prec / 100\n",
    "    tile_l = tile.buffer(l_buff).envelope\n",
    "    tiles_l.append(tile_l)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Display"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x1008 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "numTiles = int(numCols * numRows * (1 / coord_prec) ** 2)\n",
    "fig, ax = plt.subplots(figsize=(6,14))\n",
    "in_shp.plot(alpha=0.5, ax=ax)\n",
    "plt.title(output_prefix + ' divided into ' + str(numTiles) + ' tiles')\n",
    "for j in range(0, numTiles):\n",
    "    x, y = tiles_l[j].exterior.xy\n",
    "    cx, cy = tiles_l[j].centroid.xy\n",
    "    plt.annotate(str(j), (cx[0], cy[0]))\n",
    "    plt.plot(x,y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Save as GeoJSONs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "l = 0\n",
    "for m in range(0, len(tiles)):\n",
    "    tile_s_fn = output_prefix + '_s_' + str(l) + '.json'\n",
    "    tile_l_fn = output_prefix + '_l_' + str(l) + '.json'\n",
    "    tile_s_data = gpd.GeoSeries(tiles[m]).__geo_interface__\n",
    "    tile_l_data = gpd.GeoSeries(tiles_l[m]).__geo_interface__\n",
    "    with open(os.path.join(OUTPUT_DIR, tile_s_fn), 'w') as tile_s_file:\n",
    "        json.dump(tile_s_data, tile_s_file)\n",
    "    with open(os.path.join(OUTPUT_DIR, tile_l_fn), 'w') as tile_l_file:\n",
    "        json.dump(tile_l_data, tile_l_file)\n",
    "    l += 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Save as SHPs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "../../data/ghana/tiles/ghana_s_0\n",
      "../../data/ghana/tiles/ghana_l_0\n",
      "../../data/ghana/tiles/ghana_s_1\n",
      "../../data/ghana/tiles/ghana_l_1\n",
      "../../data/ghana/tiles/ghana_s_2\n",
      "../../data/ghana/tiles/ghana_l_2\n",
      "../../data/ghana/tiles/ghana_s_3\n",
      "../../data/ghana/tiles/ghana_l_3\n",
      "../../data/ghana/tiles/ghana_s_4\n",
      "../../data/ghana/tiles/ghana_l_4\n",
      "../../data/ghana/tiles/ghana_s_5\n",
      "../../data/ghana/tiles/ghana_l_5\n",
      "../../data/ghana/tiles/ghana_s_6\n",
      "../../data/ghana/tiles/ghana_l_6\n",
      "../../data/ghana/tiles/ghana_s_7\n",
      "../../data/ghana/tiles/ghana_l_7\n",
      "../../data/ghana/tiles/ghana_s_8\n",
      "../../data/ghana/tiles/ghana_l_8\n",
      "../../data/ghana/tiles/ghana_s_9\n",
      "../../data/ghana/tiles/ghana_l_9\n",
      "../../data/ghana/tiles/ghana_s_10\n",
      "../../data/ghana/tiles/ghana_l_10\n",
      "../../data/ghana/tiles/ghana_s_11\n",
      "../../data/ghana/tiles/ghana_l_11\n",
      "../../data/ghana/tiles/ghana_s_12\n",
      "../../data/ghana/tiles/ghana_l_12\n",
      "../../data/ghana/tiles/ghana_s_13\n",
      "../../data/ghana/tiles/ghana_l_13\n",
      "../../data/ghana/tiles/ghana_s_14\n",
      "../../data/ghana/tiles/ghana_l_14\n",
      "../../data/ghana/tiles/ghana_s_15\n",
      "../../data/ghana/tiles/ghana_l_15\n",
      "../../data/ghana/tiles/ghana_s_16\n",
      "../../data/ghana/tiles/ghana_l_16\n",
      "../../data/ghana/tiles/ghana_s_17\n",
      "../../data/ghana/tiles/ghana_l_17\n",
      "../../data/ghana/tiles/ghana_s_18\n",
      "../../data/ghana/tiles/ghana_l_18\n",
      "../../data/ghana/tiles/ghana_s_19\n",
      "../../data/ghana/tiles/ghana_l_19\n",
      "../../data/ghana/tiles/ghana_s_20\n",
      "../../data/ghana/tiles/ghana_l_20\n",
      "../../data/ghana/tiles/ghana_s_21\n",
      "../../data/ghana/tiles/ghana_l_21\n",
      "../../data/ghana/tiles/ghana_s_22\n",
      "../../data/ghana/tiles/ghana_l_22\n",
      "../../data/ghana/tiles/ghana_s_23\n",
      "../../data/ghana/tiles/ghana_l_23\n"
     ]
    }
   ],
   "source": [
    "# define a polygon feature geometry with one attribute\n",
    "schema = {\n",
    "    'geometry': 'Polygon',\n",
    "    'properties': {'id': 'int'},\n",
    "}\n",
    "\n",
    "# save as SHP using fiona\n",
    "l = 0\n",
    "for m in range(0, len(tiles)):\n",
    "    tile_s_fn = output_prefix + '_s_' + str(l)\n",
    "    tile_l_fn = output_prefix + '_l_' + str(l)\n",
    "    tile_s_fp = os.path.join(OUTPUT_DIR, tile_s_fn)\n",
    "    tile_l_fp = os.path.join(OUTPUT_DIR, tile_l_fn)\n",
    "    if not (os.path.isdir('tile_s_fp') or os.path.isdir('tile_l_fp')):\n",
    "        os.makedirs(tile_s_fp)\n",
    "        os.makedirs(tile_l_fp)\n",
    "    with fiona.open(os.path.join(tile_s_fp,tile_s_fn+'.shp'), 'w', 'ESRI Shapefile', schema) as out_s_shp:\n",
    "        out_s_shp.write({\n",
    "            'geometry': mapping(tiles[l]),\n",
    "            'properties': {'id': 123}\n",
    "        })\n",
    "    print(tile_s_fp)\n",
    "    with fiona.open(os.path.join(tile_l_fp,tile_l_fn+'.shp'), 'w', 'ESRI Shapefile', schema) as out_l_shp:\n",
    "        out_l_shp.write({\n",
    "            'geometry': mapping(tiles_l[l]),\n",
    "            'properties': {'id': 123}\n",
    "        })\n",
    "    print(tile_l_fp)\n",
    "    l+=1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "falco",
   "language": "python",
   "name": "falco"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}