Add unit tests

tests/datasets/test_utils.py

@@ -3,10 +3,12 @@
 
 import builtins
 import glob
+import math
 import os
 import pickle
 import shutil
 import sys
+from datetime import datetime
 from pathlib import Path
 from typing import Any, Generator, Tuple
 
@@ -19,6 +21,7 @@
 from torchgeo.datasets.utils import (
     BoundingBox,
     collate_dict,
+    disambiguate_timestamp,
     download_and_extract_archive,
     download_radiant_mlhub,
     extract_archive,
@@ -206,6 +209,62 @@ def test_invalid_t(self) -> None:
             BoundingBox(0, 1, 2, 3, 5, 4)
 
 
+@pytest.mark.parametrize(
+    "date_string,format,min_datetime,max_datetime",
+    [
+        ("", "", datetime(1970, 1, 1).timestamp(), datetime.max.timestamp()),
+        (
+            "2021",
+            "%Y",
+            datetime(2021, 1, 1, 0, 0, 0, 0).timestamp(),
+            datetime(2021, 12, 31, 23, 59, 59, 999999).timestamp(),
+        ),
+        (
+            "2021-09",
+            "%Y-%m",
+            datetime(2021, 9, 1, 0, 0, 0, 0).timestamp(),
+            datetime(2021, 9, 30, 23, 59, 59, 999999).timestamp(),
+        ),
+        (
+            "2021-09-13",
+            "%Y-%m-%d",
+            datetime(2021, 9, 13, 0, 0, 0, 0).timestamp(),
+            datetime(2021, 9, 13, 23, 59, 59, 999999).timestamp(),
+        ),
+        (
+            "2021-09-13 17",
+            "%Y-%m-%d %H",
+            datetime(2021, 9, 13, 17, 0, 0, 0).timestamp(),
+            datetime(2021, 9, 13, 17, 59, 59, 999999).timestamp(),
+        ),
+        (
+            "2021-09-13 17:21",
+            "%Y-%m-%d %H:%M",
+            datetime(2021, 9, 13, 17, 21, 0, 0).timestamp(),
+            datetime(2021, 9, 13, 17, 21, 59, 999999).timestamp(),
+        ),
+        (
+            "2021-09-13 17:21:53",
+            "%Y-%m-%d %H:%M:%S",
+            datetime(2021, 9, 13, 17, 21, 53, 0).timestamp(),
+            datetime(2021, 9, 13, 17, 21, 53, 999999).timestamp(),
+        ),
+        (
+            "2021-09-13 17:21:53:000123",
+            "%Y-%m-%d %H:%M:%S:%f",
+            datetime(2021, 9, 13, 17, 21, 53, 123).timestamp(),
+            datetime(2021, 9, 13, 17, 21, 53, 123).timestamp(),
+        ),
+    ],
+)
+def test_disambiguate_timestamp(
+    date_string: str, format: str, min_datetime: float, max_datetime: float
+) -> None:
+    mint, maxt = disambiguate_timestamp(date_string, format)
+    assert math.isclose(mint, min_datetime)
+    assert math.isclose(maxt, max_datetime)
+
+
 def test_collate_dict() -> None:
     samples = [
         {

torchgeo/datasets/geo.py

@@ -225,8 +225,8 @@ def __init__(
                     # Skip files that rasterio is unable to read
                     continue
                 else:
-                    mint: float = datetime.min.timestamp()
-                    maxt: float = datetime.max.timestamp()
+                    mint = datetime(1970, 1, 1).timestamp()
+                    maxt = datetime.max.timestamp()
                     if "date" in match.groupdict():
                         date = match.group("date")
                         mint, maxt = disambiguate_timestamp(date, self.date_format)
@@ -449,7 +449,7 @@ def __init__(
                 # Skip files that fiona is unable to read
                 continue
             else:
-                mint = datetime.min.timestamp()
+                mint = datetime(1970, 1, 1).timestamp()
                 maxt = datetime.max.timestamp()
                 coords = (minx, maxx, miny, maxy, mint, maxt)
                 self.index.insert(i, coords, filepath)

torchgeo/datasets/utils.py

@@ -10,6 +10,7 @@
 import os
 import tarfile
 import zipfile
+from calendar import monthrange
 from datetime import datetime
 from typing import Any, Dict, Iterator, List, Optional, Tuple, Union
 
@@ -276,28 +277,29 @@ def disambiguate_timestamp(date_str: str, format: str) -> Tuple[float, float]:
 
     # TODO: This is really tedious, is there a better way to do this?
 
-    if not any([f"{c}%" in format for c in "yYcxG"]):
+    if not any([f"%{c}" in format for c in "yYcxG"]):
         # No temporal info
-        mint = datetime.min
+        mint = datetime(1970, 1, 1)
         maxt = datetime.max
-    elif not any([f"{c}%" in format for c in "bBmjUWcxV"]):
+    elif not any([f"%{c}" in format for c in "bBmjUWcxV"]):
         # Year resolution
         maxt = datetime(mint.year, 12, 31, 23, 59, 59, 999999)
-    elif not any([f"{c}%" in format for c in "djcx"]):
+    elif not any([f"%{c}" in format for c in "djcx"]):
         # Month resolution
-        maxt = datetime(mint.year, mint.month, 31, 23, 59, 59, 999999)
-    elif not any([f"{c}%" in format for c in "HIcX"]):
+        maxday = monthrange(mint.year, mint.month)[1]
+        maxt = datetime(mint.year, mint.month, maxday, 23, 59, 59, 999999)
+    elif not any([f"%{c}" in format for c in "HIcX"]):
         # Day resolution
         maxt = datetime(mint.year, mint.month, mint.day, 23, 59, 59, 999999)
-    elif not any([f"{c}%" in format for c in "McX"]):
+    elif not any([f"%{c}" in format for c in "McX"]):
         # Hour resolution
         maxt = datetime(mint.year, mint.month, mint.day, mint.hour, 59, 59, 999999)
-    elif not any([f"{c}%" in format for c in "ScX"]):
+    elif not any([f"%{c}" in format for c in "ScX"]):
         # Minute resolution
         maxt = datetime(
             mint.year, mint.month, mint.day, mint.hour, mint.minute, 59, 999999
         )
-    elif not any([f"{c}%" in format for c in "f"]):
+    elif not any([f"%{c}" in format for c in "f"]):
         # Second resolution
         maxt = datetime(
             mint.year, mint.month, mint.day, mint.hour, mint.minute, mint.second, 999999