{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "quick start: http://docs.h5py.org/en/latest/quick.html\n",
    "\n",
    "Dimension Scales: http://docs.h5py.org/en/latest/high/dims.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import h5py"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "f = h5py.File('test.hdf5', mode='w')  # can also be used as context manager"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'/'"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f.name"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "f['a'] = 12"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['a']"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(f.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<HDF5 dataset \"a\": shape (), type \"<i8\">"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = f['a']\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(12)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[...]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "()"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(data.dims)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype('int64')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.dtype"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "()"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.maxshape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'/a'"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.name"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.attrs['extra info'] = 'super data'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'extra info': 'super data'}"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dict(data.attrs.items())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'a': <HDF5 dataset \"a\": shape (), type \"<i8\">}"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dict(f.items())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Dimension Scales\n",
    "\n",
    "http://docs.h5py.org/en/latest/high/dims.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "numpy_array = np.array([[1., 2, 3], [4, 5, 6]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<HDF5 dataset \"b\": shape (2, 3), type \"<f8\">"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f['b'] = numpy_array\n",
    "f['b']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<HDF5 dataset \"c\": shape (2, 3), type \"<f4\">"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f.create_dataset('c', dtype='float32', data=numpy_array)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'a': <HDF5 dataset \"a\": shape (), type \"<i8\">,\n",
       " 'b': <HDF5 dataset \"b\": shape (2, 3), type \"<f8\">,\n",
       " 'c': <HDF5 dataset \"c\": shape (2, 3), type \"<f4\">}"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dict(f.items())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "f['b'].dims[1].label = 'nice second dimension!'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "f['myscale'] = [1, 2, 3, 4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# this seems to be only necessary for setting a label:\n",
    "f['c'].dims.create_scale(f['myscale'], 'my scale label')\n",
    "#h5py.h5ds.set_scale(f['myscale'].id, b'my scale label')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "f['c'].dims[0].attach_scale(f['myscale'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['', 'nice second dimension!']"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[dim.label for dim in f['b'].dims]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<\"\" dimension 0 of HDF5 dataset at 140378890863152>,\n",
       " <\"\" dimension 1 of HDF5 dataset at 140378890863152>]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(f['c'].dims.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['my scale label']"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(f['c'].dims[0].keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<HDF5 dataset \"myscale\": shape (4,), type \"<i8\">"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f['c'].dims[0][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<HDF5 dataset \"myscale\": shape (4,), type \"<i8\">"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f['c'].dims[0]['my scale label']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "f.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HDF5 \"test.hdf5\" {\r\n",
      "GROUP \"/\" {\r\n",
      "   DATASET \"a\" {\r\n",
      "      DATATYPE  H5T_STD_I64LE\r\n",
      "      DATASPACE  SCALAR\r\n",
      "      DATA {\r\n",
      "      (0): 12\r\n",
      "      }\r\n",
      "      ATTRIBUTE \"extra info\" {\r\n",
      "         DATATYPE  H5T_STRING {\r\n",
      "            STRSIZE H5T_VARIABLE;\r\n",
      "            STRPAD H5T_STR_NULLTERM;\r\n",
      "            CSET H5T_CSET_UTF8;\r\n",
      "            CTYPE H5T_C_S1;\r\n",
      "         }\r\n",
      "         DATASPACE  SCALAR\r\n",
      "         DATA {\r\n",
      "         (0): \"super data\"\r\n",
      "         }\r\n",
      "      }\r\n",
      "   }\r\n",
      "   DATASET \"b\" {\r\n",
      "      DATATYPE  H5T_IEEE_F64LE\r\n",
      "      DATASPACE  SIMPLE { ( 2, 3 ) / ( 2, 3 ) }\r\n",
      "      DATA {\r\n",
      "      (0,0): 1, 2, 3,\r\n",
      "      (1,0): 4, 5, 6\r\n",
      "      }\r\n",
      "      ATTRIBUTE \"DIMENSION_LABELS\" {\r\n",
      "         DATATYPE  H5T_STRING {\r\n",
      "            STRSIZE H5T_VARIABLE;\r\n",
      "            STRPAD H5T_STR_NULLTERM;\r\n",
      "            CSET H5T_CSET_ASCII;\r\n",
      "            CTYPE H5T_C_S1;\r\n",
      "         }\r\n",
      "         DATASPACE  SIMPLE { ( 2 ) / ( 2 ) }\r\n",
      "         DATA {\r\n",
      "         (0): NULL, \"nice second dimension!\"\r\n",
      "         }\r\n",
      "      }\r\n",
      "   }\r\n",
      "   DATASET \"c\" {\r\n",
      "      DATATYPE  H5T_IEEE_F32LE\r\n",
      "      DATASPACE  SIMPLE { ( 2, 3 ) / ( 2, 3 ) }\r\n",
      "      DATA {\r\n",
      "      (0,0): 1, 2, 3,\r\n",
      "      (1,0): 4, 5, 6\r\n",
      "      }\r\n",
      "      ATTRIBUTE \"DIMENSION_LIST\" {\r\n",
      "         DATATYPE  H5T_VLEN { H5T_REFERENCE { H5T_STD_REF_OBJECT }}\r\n",
      "         DATASPACE  SIMPLE { ( 2 ) / ( 2 ) }\r\n",
      "         DATA {\r\n",
      "         (0): (DATASET 6344 /myscale ), ()\r\n",
      "         }\r\n",
      "      }\r\n",
      "   }\r\n",
      "   DATASET \"myscale\" {\r\n",
      "      DATATYPE  H5T_STD_I64LE\r\n",
      "      DATASPACE  SIMPLE { ( 4 ) / ( 4 ) }\r\n",
      "      DATA {\r\n",
      "      (0): 1, 2, 3, 4\r\n",
      "      }\r\n",
      "      ATTRIBUTE \"CLASS\" {\r\n",
      "         DATATYPE  H5T_STRING {\r\n",
      "            STRSIZE 16;\r\n",
      "            STRPAD H5T_STR_NULLTERM;\r\n",
      "            CSET H5T_CSET_ASCII;\r\n",
      "            CTYPE H5T_C_S1;\r\n",
      "         }\r\n",
      "         DATASPACE  SCALAR\r\n",
      "         DATA {\r\n",
      "         (0): \"DIMENSION_SCALE\"\r\n",
      "         }\r\n",
      "      }\r\n",
      "      ATTRIBUTE \"NAME\" {\r\n",
      "         DATATYPE  H5T_STRING {\r\n",
      "            STRSIZE 15;\r\n",
      "            STRPAD H5T_STR_NULLTERM;\r\n",
      "            CSET H5T_CSET_ASCII;\r\n",
      "            CTYPE H5T_C_S1;\r\n",
      "         }\r\n",
      "         DATASPACE  SCALAR\r\n",
      "         DATA {\r\n",
      "         (0): \"my scale label\"\r\n",
      "         }\r\n",
      "      }\r\n",
      "      ATTRIBUTE \"REFERENCE_LIST\" {\r\n",
      "         DATATYPE  H5T_COMPOUND {\r\n",
      "            H5T_REFERENCE { H5T_STD_REF_OBJECT } \"dataset\";\r\n",
      "            H5T_STD_I32LE \"dimension\";\r\n",
      "         }\r\n",
      "         DATASPACE  SIMPLE { ( 1 ) / ( 1 ) }\r\n",
      "         DATA {\r\n",
      "         (0): {\r\n",
      "               DATASET 1672 /c ,\r\n",
      "               0\r\n",
      "            }\r\n",
      "         }\r\n",
      "      }\r\n",
      "   }\r\n",
      "}\r\n",
      "}\r\n"
     ]
    }
   ],
   "source": [
    "!h5dump test.hdf5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "with h5py.File('two_scales.hdf5', mode='w') as f:\n",
    "    f['a'] = [1, 2, 3]\n",
    "    f['scale1'] = 1\n",
    "    f['scale2'] = 2\n",
    "    f['a'].dims.create_scale(f['scale1'], 'scale one')\n",
    "    h5py.h5ds.set_scale(f['scale2'].id, b'scale two')\n",
    "    f['a'].dims[0].attach_scale(f['scale1'])\n",
    "    f['a'].dims[0].attach_scale(f['scale2'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HDF5 \"two_scales.hdf5\" {\r\n",
      "GROUP \"/\" {\r\n",
      "   DATASET \"a\" {\r\n",
      "      DATATYPE  H5T_STD_I64LE\r\n",
      "      DATASPACE  SIMPLE { ( 3 ) / ( 3 ) }\r\n",
      "      DATA {\r\n",
      "      (0): 1, 2, 3\r\n",
      "      }\r\n",
      "      ATTRIBUTE \"DIMENSION_LIST\" {\r\n",
      "         DATATYPE  H5T_VLEN { H5T_REFERENCE { H5T_STD_REF_OBJECT }}\r\n",
      "         DATASPACE  SIMPLE { ( 1 ) / ( 1 ) }\r\n",
      "         DATA {\r\n",
      "         (0): (DATASET 1400 /scale1 , DATASET 1672 /scale2 )\r\n",
      "         }\r\n",
      "      }\r\n",
      "   }\r\n",
      "   DATASET \"scale1\" {\r\n",
      "      DATATYPE  H5T_STD_I64LE\r\n",
      "      DATASPACE  SCALAR\r\n",
      "      DATA {\r\n",
      "      (0): 1\r\n",
      "      }\r\n",
      "      ATTRIBUTE \"CLASS\" {\r\n",
      "         DATATYPE  H5T_STRING {\r\n",
      "            STRSIZE 16;\r\n",
      "            STRPAD H5T_STR_NULLTERM;\r\n",
      "            CSET H5T_CSET_ASCII;\r\n",
      "            CTYPE H5T_C_S1;\r\n",
      "         }\r\n",
      "         DATASPACE  SCALAR\r\n",
      "         DATA {\r\n",
      "         (0): \"DIMENSION_SCALE\"\r\n",
      "         }\r\n",
      "      }\r\n",
      "      ATTRIBUTE \"NAME\" {\r\n",
      "         DATATYPE  H5T_STRING {\r\n",
      "            STRSIZE 10;\r\n",
      "            STRPAD H5T_STR_NULLTERM;\r\n",
      "            CSET H5T_CSET_ASCII;\r\n",
      "            CTYPE H5T_C_S1;\r\n",
      "         }\r\n",
      "         DATASPACE  SCALAR\r\n",
      "         DATA {\r\n",
      "         (0): \"scale one\"\r\n",
      "         }\r\n",
      "      }\r\n",
      "      ATTRIBUTE \"REFERENCE_LIST\" {\r\n",
      "         DATATYPE  H5T_COMPOUND {\r\n",
      "            H5T_REFERENCE { H5T_STD_REF_OBJECT } \"dataset\";\r\n",
      "            H5T_STD_I32LE \"dimension\";\r\n",
      "         }\r\n",
      "         DATASPACE  SIMPLE { ( 1 ) / ( 1 ) }\r\n",
      "         DATA {\r\n",
      "         (0): {\r\n",
      "               DATASET 800 /a ,\r\n",
      "               0\r\n",
      "            }\r\n",
      "         }\r\n",
      "      }\r\n",
      "   }\r\n",
      "   DATASET \"scale2\" {\r\n",
      "      DATATYPE  H5T_STD_I64LE\r\n",
      "      DATASPACE  SCALAR\r\n",
      "      DATA {\r\n",
      "      (0): 2\r\n",
      "      }\r\n",
      "      ATTRIBUTE \"CLASS\" {\r\n",
      "         DATATYPE  H5T_STRING {\r\n",
      "            STRSIZE 16;\r\n",
      "            STRPAD H5T_STR_NULLTERM;\r\n",
      "            CSET H5T_CSET_ASCII;\r\n",
      "            CTYPE H5T_C_S1;\r\n",
      "         }\r\n",
      "         DATASPACE  SCALAR\r\n",
      "         DATA {\r\n",
      "         (0): \"DIMENSION_SCALE\"\r\n",
      "         }\r\n",
      "      }\r\n",
      "      ATTRIBUTE \"NAME\" {\r\n",
      "         DATATYPE  H5T_STRING {\r\n",
      "            STRSIZE 10;\r\n",
      "            STRPAD H5T_STR_NULLTERM;\r\n",
      "            CSET H5T_CSET_ASCII;\r\n",
      "            CTYPE H5T_C_S1;\r\n",
      "         }\r\n",
      "         DATASPACE  SCALAR\r\n",
      "         DATA {\r\n",
      "         (0): \"scale two\"\r\n",
      "         }\r\n",
      "      }\r\n",
      "      ATTRIBUTE \"REFERENCE_LIST\" {\r\n",
      "         DATATYPE  H5T_COMPOUND {\r\n",
      "            H5T_REFERENCE { H5T_STD_REF_OBJECT } \"dataset\";\r\n",
      "            H5T_STD_I32LE \"dimension\";\r\n",
      "         }\r\n",
      "         DATASPACE  SIMPLE { ( 1 ) / ( 1 ) }\r\n",
      "         DATA {\r\n",
      "         (0): {\r\n",
      "               DATASET 800 /a ,\r\n",
      "               0\r\n",
      "            }\r\n",
      "         }\r\n",
      "      }\r\n",
      "   }\r\n",
      "}\r\n",
      "}\r\n"
     ]
    }
   ],
   "source": [
    "!h5dump two_scales.hdf5"
   ]
  }
 ],
 "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}