Skip to content

rslearn.dataset.storage.file

file

The default file-based window storage backend.

FileWindowStorage

Bases: WindowStorage

The default file-backed window storage.

Source code in rslearn/dataset/storage/file.py 
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
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
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
class FileWindowStorage(WindowStorage):
    """The default file-backed window storage."""

    def __init__(self, path: UPath):
        """Create a new FileWindowStorage.

        Args:
            path: the path to the dataset.
        """
        self.path = path

    @override
    def get_window_root(self, group: str, name: str) -> UPath:
        return Window.get_window_root(self.path, group, name)

    @override
    def get_windows(
        self,
        groups: list[str] | None = None,
        names: list[str] | None = None,
        show_progress: bool = False,
        workers: int = 0,
    ) -> list[Window]:
        """Load the windows in the dataset.

        Args:
            groups: an optional list of groups to filter loading
            names: an optional list of window names to filter loading
            show_progress: whether to show tqdm progress bar
            workers: number of parallel workers, default 0 (use main thread only to load windows)
        """
        # Avoid directory does not exist errors later.
        if not (self.path / "windows").exists():
            return []

        window_dirs = []
        if not groups:
            groups = []
            for group_dir in iter_nonhidden_subdirs(self.path / "windows"):
                groups.append(group_dir.name)
        for group in groups:
            group_dir = self.path / "windows" / group
            if not group_dir.exists():
                logger.warning(
                    f"Skipping group directory {group_dir} since it does not exist"
                )
                continue
            if not group_dir.is_dir():
                logger.warning(
                    f"Skipping group path {group_dir} since it is not a directory"
                )
                continue
            if names:
                for window_name in names:
                    window_dir = group_dir / window_name
                    if not window_dir.is_dir():
                        continue
                    window_dirs.append(window_dir)
            else:
                # We use iter_nonhidden here instead of iter_nonhidden_subdirs since
                # iter_nonhidden_subdirs is slow for large directories, and the group
                # directories could contain many windows. There should not be
                # non-hidden files in the group directory anyway.
                for window_dir in iter_nonhidden(group_dir):
                    window_dirs.append(window_dir)

        with make_pool_and_star_imap_unordered(
            workers,
            load_window,
            [dict(storage=self, window_dir=window_dir) for window_dir in window_dirs],
        ) as outputs:
            if show_progress:
                outputs = tqdm.tqdm(
                    outputs, total=len(window_dirs), desc="Loading windows"
                )
            windows = list(outputs)

        return windows

    def _validate_layer_name(self, layer_name: str) -> None:
        """Raise if the layer name contains '.', which is reserved for group index."""
        if "." in layer_name:
            raise ValueError(f"Layer name must not contain '.': got '{layer_name}'.")

    @override
    def create_or_update_window(self, window: Window) -> None:
        window_path = self.get_window_root(window.group, window.name)
        window_path.mkdir(parents=True, exist_ok=True)
        metadata_path = window_path / "metadata.json"
        logger.debug(f"Saving window metadata to {metadata_path}")
        metadata = {
            "projection": window.projection.serialize(),
            "bounds": window.bounds,
            "time_range": (
                [window.time_range[0].isoformat(), window.time_range[1].isoformat()]
                if window.time_range
                else None
            ),
            "options": window.options,
        }
        with open_atomic(metadata_path, "w") as f:
            json.dump(metadata, f)

    @override
    def get_layer_datas(self, group: str, name: str) -> dict[str, WindowLayerData]:
        window_path = self.get_window_root(group, name)
        items_fname = window_path / "items.json"
        if not items_fname.exists():
            return {}

        with items_fname.open() as f:
            layer_datas = [
                WindowLayerData.deserialize(layer_data) for layer_data in json.load(f)
            ]

        return {layer_data.layer_name: layer_data for layer_data in layer_datas}

    @override
    def save_layer_datas(
        self, group: str, name: str, layer_datas: dict[str, WindowLayerData]
    ) -> None:
        window_path = self.get_window_root(group, name)
        json_data = [layer_data.serialize() for layer_data in layer_datas.values()]
        items_fname = window_path / "items.json"
        logger.info(f"Saving window items to {items_fname}")
        with open_atomic(items_fname, "w") as f:
            json.dump(json_data, f)

    @override
    def list_completed_layers(self, group: str, name: str) -> list[tuple[str, int]]:
        window_path = self.get_window_root(group, name)
        layers_directory = window_path / LAYERS_SUBDIR
        if not layers_directory.exists():
            return []

        completed_item_groups = []
        for layer_dir in iter_nonhidden_subdirs(layers_directory):
            layer_name, group_idx = get_layer_and_group_from_dir_name(layer_dir.name)
            if not self.is_layer_completed(group, name, layer_name, group_idx):
                continue
            completed_item_groups.append((layer_name, group_idx))

        return completed_item_groups

    @override
    def is_layer_completed(
        self, group: str, name: str, layer_name: str, group_idx: int = 0
    ) -> bool:
        self._validate_layer_name(layer_name)
        window_path = self.get_window_root(group, name)
        layer_dir = _file_layer_dir(
            window_path,
            layer_name,
            group_idx,
        )
        return (layer_dir / "completed").exists()

    @override
    def mark_layer_completed(
        self, group: str, name: str, layer_name: str, group_idx: int = 0
    ) -> None:
        self._validate_layer_name(layer_name)
        window_path = self.get_window_root(group, name)
        layer_dir = _file_layer_dir(window_path, layer_name, group_idx)
        layer_dir.mkdir(parents=True, exist_ok=True)
        (layer_dir / "completed").touch()

get_windows 

get_windows(groups: list[str] | None = None, names: list[str] | None = None, show_progress: bool = False, workers: int = 0) -> list[Window]

Load the windows in the dataset.

Parameters:

Name Type Description Default
groups list[str] | None

an optional list of groups to filter loading

 None
names list[str] | None

an optional list of window names to filter loading

 None
show_progress bool

whether to show tqdm progress bar

 False
workers int

number of parallel workers, default 0 (use main thread only to load windows)

 0
Source code in rslearn/dataset/storage/file.py 
101
102
103
104
105
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
@override
def get_windows(
    self,
    groups: list[str] | None = None,
    names: list[str] | None = None,
    show_progress: bool = False,
    workers: int = 0,
) -> list[Window]:
    """Load the windows in the dataset.

    Args:
        groups: an optional list of groups to filter loading
        names: an optional list of window names to filter loading
        show_progress: whether to show tqdm progress bar
        workers: number of parallel workers, default 0 (use main thread only to load windows)
    """
    # Avoid directory does not exist errors later.
    if not (self.path / "windows").exists():
        return []

    window_dirs = []
    if not groups:
        groups = []
        for group_dir in iter_nonhidden_subdirs(self.path / "windows"):
            groups.append(group_dir.name)
    for group in groups:
        group_dir = self.path / "windows" / group
        if not group_dir.exists():
            logger.warning(
                f"Skipping group directory {group_dir} since it does not exist"
            )
            continue
        if not group_dir.is_dir():
            logger.warning(
                f"Skipping group path {group_dir} since it is not a directory"
            )
            continue
        if names:
            for window_name in names:
                window_dir = group_dir / window_name
                if not window_dir.is_dir():
                    continue
                window_dirs.append(window_dir)
        else:
            # We use iter_nonhidden here instead of iter_nonhidden_subdirs since
            # iter_nonhidden_subdirs is slow for large directories, and the group
            # directories could contain many windows. There should not be
            # non-hidden files in the group directory anyway.
            for window_dir in iter_nonhidden(group_dir):
                window_dirs.append(window_dir)

    with make_pool_and_star_imap_unordered(
        workers,
        load_window,
        [dict(storage=self, window_dir=window_dir) for window_dir in window_dirs],
    ) as outputs:
        if show_progress:
            outputs = tqdm.tqdm(
                outputs, total=len(window_dirs), desc="Loading windows"
            )
        windows = list(outputs)

    return windows

FileWindowStorageFactory

Bases: WindowStorageFactory

Factory class for FileWindowStorage.

Source code in rslearn/dataset/storage/file.py 
254
255
256
257
258
259
260
class FileWindowStorageFactory(WindowStorageFactory):
    """Factory class for FileWindowStorage."""

    @override
    def get_storage(self, ds_path: UPath) -> FileWindowStorage:
        """Get a FileWindowStorage for the given dataset path."""
        return FileWindowStorage(ds_path)

get_storage 

get_storage(ds_path: UPath) -> FileWindowStorage

Get a FileWindowStorage for the given dataset path.

Source code in rslearn/dataset/storage/file.py 
257
258
259
260
@override
def get_storage(self, ds_path: UPath) -> FileWindowStorage:
    """Get a FileWindowStorage for the given dataset path."""
    return FileWindowStorage(ds_path)

load_window 

load_window(storage: FileWindowStorage, window_dir: UPath) -> Window

Load the window from its directory by reading metadata.json.

The group and window name are derived from the filesystem path.

Parameters:

Name Type Description Default
storage FileWindowStorage

the underlying FileWindowStorage.

 required
window_dir UPath

the path where the window is stored.

 required

Returns:

Type Description
Window

the window object.
Source code in rslearn/dataset/storage/file.py 
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
def load_window(
    storage: FileWindowStorage,
    window_dir: UPath,
) -> Window:
    """Load the window from its directory by reading metadata.json.

    The group and window name are derived from the filesystem path.

    Args:
        storage: the underlying FileWindowStorage.
        window_dir: the path where the window is stored.

    Returns:
        the window object.
    """
    metadata_fname = window_dir / "metadata.json"
    with metadata_fname.open() as f:
        metadata = json.load(f)

    group = window_dir.parent.name
    name = window_dir.name

    if len(metadata["bounds"]) != 4:
        raise ValueError(
            f"expected bounds to have 4 elements but got {len(metadata['bounds'])}"
        )
    bounds = (
        metadata["bounds"][0],
        metadata["bounds"][1],
        metadata["bounds"][2],
        metadata["bounds"][3],
    )

    time_range = None
    if metadata.get("time_range"):
        time_range = (
            datetime.fromisoformat(metadata["time_range"][0]),
            datetime.fromisoformat(metadata["time_range"][1]),
        )

    return Window(
        storage=storage,
        group=group,
        name=name,
        projection=Projection.deserialize(metadata["projection"]),
        bounds=bounds,
        time_range=time_range,
        options=metadata.get("options", {}),
    )