Skip to content

QuadkeyPandas module

QuadkeyPandas

Source code in vgridpandas/quadkeypandas.py 
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
@pd.api.extensions.register_dataframe_accessor("quadkey")
class QuadkeyPandas:
    def __init__(self, df: DataFrame):
        self._df = df

    # quadkey API
    # These methods simply mirror the Vgrid quadkey API and apply quadkey functions to all rows

    def latlon2quadkey(
        self,
        resolution: int,
        lat_col: str = "lat",
        lon_col: str = "lon",
        set_index: bool = False,
    ) -> AnyDataFrame:
        """Adds quadkey ID to (Geo)DataFrame.

        pd.DataFrame: uses `lat_col` and `lon_col` (default `lat` and `lon`)
        gpd.GeoDataFrame: uses `geometry`

        Assumes coordinates in epsg=4326.

        Parameters
        ----------
        resolution : int
            quadkey resolution
        lat_col : str
            Name of the latitude column (if used), default 'lat'
        lon_col : str
            Name of the longitude column (if used), default 'lon'
        set_index : bool
            If True, the columns with quadkey ID is set as index, default 'True'

        Returns
        -------
        (Geo)DataFrame with quadkey IDs added
        """

        if isinstance(self._df, gpd.GeoDataFrame):
            lons = self._df.geometry.x
            lats = self._df.geometry.y
        else:
            lons = self._df[lon_col]
            lats = self._df[lat_col]

        quadkey_ids = [
            latlon_to_quadkey(lat, lon, resolution) for lat, lon in zip(lats, lons)
        ]

        quadkey_col = QUADKEY_COL
        assign_arg = {quadkey_col: quadkey_ids, f"{quadkey_col}_res": resolution}
        df = self._df.assign(**assign_arg)
        if set_index:
            return df.set_index(quadkey_col)
        return df

    def quadkey2geo(self, quadkey_col: str = None) -> GeoDataFrame:
        """Add geometry with QUADKEY geometry to the DataFrame."""
        if quadkey_col is not None:
            if quadkey_col not in self._df.columns:
                raise ValueError(f"Column '{quadkey_col}' not found in DataFrame")
            ids = self._df[quadkey_col]
        else:
            if QUADKEY_COL not in self._df.columns:
                raise ValueError(f"Column '{QUADKEY_COL}' not found in DataFrame")
            ids = self._df[QUADKEY_COL]
        return dggs_ids_to_geodataframe(self._df, ids, quadkey_to_geo)

    def polyfill(
        self,
        resolution: int,
        predicate: str = None,
        compact: bool = False,
        explode: bool = False,
    ) -> AnyDataFrame:
        """
        Parameters
        ----------
        resolution : int
            Quadkey resolution
        predicate : str, optional
            Spatial predicate to apply ('intersect', 'within', 'centroid_within', 'largest_overlap')
        compact : bool, optional
            Whether to compact the Quadkey IDs
        explode : bool
            If True, will explode the resulting list vertically.
            All other columns' values are copied.
            Default: False
        """

        result = self._df.geometry.apply(
            lambda geom: polyfill_row(geom, resolution, predicate, compact)
        )

        if not explode:
            assign_args = {QUADKEY_COL: result}
            return self._df.assign(**assign_args)

        result = result.explode().to_frame(QUADKEY_COL)
        return self._df.join(result)

    def quadkeybin(
        self,
        resolution: int,
        stats: str = "count",
        numeric_col: str = None,
        category_col: str = None,
        lat_col: str = "lat",
        lon_col: str = "lon",
    ) -> GeoDataFrame:
        """
        Bin points into quadkey cells and compute statistics.
        """
        quadkey_col = QUADKEY_COL
        df = self.latlon2quadkey(resolution, lat_col, lon_col)
        result = aggregate_bin(df, quadkey_col, stats, numeric_col, category_col)
        return result.quadkey.quadkey2geo(quadkey_col=quadkey_col)

latlon2quadkey(resolution, lat_col='lat', lon_col='lon', set_index=False)

Adds quadkey ID to (Geo)DataFrame.

pd.DataFrame: uses lat_col and lon_col (default lat and lon) gpd.GeoDataFrame: uses geometry

Assumes coordinates in epsg=4326.

Parameters

resolution : int quadkey resolution lat_col : str Name of the latitude column (if used), default 'lat' lon_col : str Name of the longitude column (if used), default 'lon' set_index : bool If True, the columns with quadkey ID is set as index, default 'True'

Returns

(Geo)DataFrame with quadkey IDs added

Source code in vgridpandas/quadkeypandas.py 
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
def latlon2quadkey(
    self,
    resolution: int,
    lat_col: str = "lat",
    lon_col: str = "lon",
    set_index: bool = False,
) -> AnyDataFrame:
    """Adds quadkey ID to (Geo)DataFrame.

    pd.DataFrame: uses `lat_col` and `lon_col` (default `lat` and `lon`)
    gpd.GeoDataFrame: uses `geometry`

    Assumes coordinates in epsg=4326.

    Parameters
    ----------
    resolution : int
        quadkey resolution
    lat_col : str
        Name of the latitude column (if used), default 'lat'
    lon_col : str
        Name of the longitude column (if used), default 'lon'
    set_index : bool
        If True, the columns with quadkey ID is set as index, default 'True'

    Returns
    -------
    (Geo)DataFrame with quadkey IDs added
    """

    if isinstance(self._df, gpd.GeoDataFrame):
        lons = self._df.geometry.x
        lats = self._df.geometry.y
    else:
        lons = self._df[lon_col]
        lats = self._df[lat_col]

    quadkey_ids = [
        latlon_to_quadkey(lat, lon, resolution) for lat, lon in zip(lats, lons)
    ]

    quadkey_col = QUADKEY_COL
    assign_arg = {quadkey_col: quadkey_ids, f"{quadkey_col}_res": resolution}
    df = self._df.assign(**assign_arg)
    if set_index:
        return df.set_index(quadkey_col)
    return df

polyfill(resolution, predicate=None, compact=False, explode=False)

Parameters

resolution : int Quadkey resolution predicate : str, optional Spatial predicate to apply ('intersect', 'within', 'centroid_within', 'largest_overlap') compact : bool, optional Whether to compact the Quadkey IDs explode : bool If True, will explode the resulting list vertically. All other columns' values are copied. Default: False

Source code in vgridpandas/quadkeypandas.py 
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
def polyfill(
    self,
    resolution: int,
    predicate: str = None,
    compact: bool = False,
    explode: bool = False,
) -> AnyDataFrame:
    """
    Parameters
    ----------
    resolution : int
        Quadkey resolution
    predicate : str, optional
        Spatial predicate to apply ('intersect', 'within', 'centroid_within', 'largest_overlap')
    compact : bool, optional
        Whether to compact the Quadkey IDs
    explode : bool
        If True, will explode the resulting list vertically.
        All other columns' values are copied.
        Default: False
    """

    result = self._df.geometry.apply(
        lambda geom: polyfill_row(geom, resolution, predicate, compact)
    )

    if not explode:
        assign_args = {QUADKEY_COL: result}
        return self._df.assign(**assign_args)

    result = result.explode().to_frame(QUADKEY_COL)
    return self._df.join(result)

quadkey2geo(quadkey_col=None)

Add geometry with QUADKEY geometry to the DataFrame.

Source code in vgridpandas/quadkeypandas.py 
162
163
164
165
166
167
168
169
170
171
172
def quadkey2geo(self, quadkey_col: str = None) -> GeoDataFrame:
    """Add geometry with QUADKEY geometry to the DataFrame."""
    if quadkey_col is not None:
        if quadkey_col not in self._df.columns:
            raise ValueError(f"Column '{quadkey_col}' not found in DataFrame")
        ids = self._df[quadkey_col]
    else:
        if QUADKEY_COL not in self._df.columns:
            raise ValueError(f"Column '{QUADKEY_COL}' not found in DataFrame")
        ids = self._df[QUADKEY_COL]
    return dggs_ids_to_geodataframe(self._df, ids, quadkey_to_geo)

quadkeybin(resolution, stats='count', numeric_col=None, category_col=None, lat_col='lat', lon_col='lon')

Bin points into quadkey cells and compute statistics.

Source code in vgridpandas/quadkeypandas.py 
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
def quadkeybin(
    self,
    resolution: int,
    stats: str = "count",
    numeric_col: str = None,
    category_col: str = None,
    lat_col: str = "lat",
    lon_col: str = "lon",
) -> GeoDataFrame:
    """
    Bin points into quadkey cells and compute statistics.
    """
    quadkey_col = QUADKEY_COL
    df = self.latlon2quadkey(resolution, lat_col, lon_col)
    result = aggregate_bin(df, quadkey_col, stats, numeric_col, category_col)
    return result.quadkey.quadkey2geo(quadkey_col=quadkey_col)

poly2quadkey(geometry, resolution, predicate=None, compact=False)

Convert polygon geometries (Polygon, MultiPolygon) to Quadkey grid cells.

Parameters:

Name Type Description Default
resolution int

Quadkey resolution level [1..10]

 required
geometry Polygon or MultiPolygon

Polygon geometry to convert

 required
predicate str

Spatial predicate to apply ('intersect', 'within', 'centroid_within', 'largest_overlap')

 None

Returns:

Name Type Description
list Set[str]

List of quadkey ids intersecting the polygon
Example

from shapely.geometry import Polygon poly = Polygon([(-122.5, 37.7), (-122.3, 37.7), (-122.3, 37.9), (-122.5, 37.9)]) cells = poly2quadkey(poly, 10, predicate="intersect", compact=True) len(cells) > 0 True

Source code in vgridpandas/quadkeypandas.py 
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
def poly2quadkey(
    geometry: MultiPolyOrPoly,
    resolution: int,
    predicate: str = None,
    compact: bool = False,
) -> Set[str]:
    """
    Convert polygon geometries (Polygon, MultiPolygon) to Quadkey grid cells.

    Args:
        resolution (int): Quadkey resolution level [1..10]
        geometry (shapely.geometry.Polygon or shapely.geometry.MultiPolygon): Polygon geometry to convert
        predicate (str, optional): Spatial predicate to apply ('intersect', 'within', 'centroid_within', 'largest_overlap')

    Returns:
        list: List of quadkey ids intersecting the polygon

    Example:
        >>> from shapely.geometry import Polygon
        >>> poly = Polygon([(-122.5, 37.7), (-122.3, 37.7), (-122.3, 37.9), (-122.5, 37.9)])
        >>> cells = poly2quadkey(poly, 10, predicate="intersect", compact=True)
        >>> len(cells) > 0
        True
    """

    resolution = validate_quadkey_resolution(resolution)
    if isinstance(geometry, (Polygon, LineString)):
        polys = [geometry]
    elif isinstance(geometry, (MultiPolygon, MultiLineString)):
        polys = list(geometry.geoms)
    else:
        return []

    quadkey_ids = []
    for poly in polys:
        min_lon, min_lat, max_lon, max_lat = poly.bounds
        tiles = mercantile.tiles(min_lon, min_lat, max_lon, max_lat, resolution)
        for tile in tiles:
            z, x, y = tile.z, tile.x, tile.y
            bounds = mercantile.bounds(x, y, z)
            min_lat, min_lon = bounds.south, bounds.west
            max_lat, max_lon = bounds.north, bounds.east
            quadkey_id = mercantile.quadkey(tile)
            cell_polygon = Polygon(
                [
                    [min_lon, min_lat],
                    [max_lon, min_lat],
                    [max_lon, max_lat],
                    [min_lon, max_lat],
                    [min_lon, min_lat],
                ]
            )
            if check_predicate(cell_polygon, poly, predicate):
                quadkey_ids.append(quadkey_id)

    if compact:
        return quadkey_compact(quadkey_ids)
    return quadkey_ids

polyfill_row(geometry, resolution, predicate=None, compact=False)

Return cell ids covering a single row geometry.

Source code in vgridpandas/quadkeypandas.py 
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
def polyfill_row(geometry, resolution, predicate=None, compact=False) -> list:
    """Return cell ids covering a single row geometry."""
    if isinstance(geometry, (Polygon, MultiPolygon)):
        tokens = set(poly2quadkey(geometry, resolution, predicate, compact))
    elif isinstance(geometry, (LineString, MultiLineString)):
        tokens = set(
            poly2quadkey(geometry, resolution, predicate="intersect", compact=False)
        )
    else:
        raise TypeError(f"Unknown type {type(geometry)}")
    return list(tokens)