fix: add file size to dataset identifier hashing process

CHANGELOG

@@ -1,9 +1,9 @@
-0.10.4
- - fix: missing keyword argument `idx` in `SingleTifHolo.get_time`
- - example: add conversion script to hdf5 qpimage data
 0.10.3
- - fix: Correctly pass precision to phasics and hologram zip file
+ - fix: missing keyword argument `idx` in `SingleTifHolo.get_time`
+ - fix: correctly pass precision to phasics and hologram zip file
    formats
+ - fix: add file size to dataset identifier hashing process
+ - example: add conversion script to hdf5 qpimage data
 0.10.2
  - fix: SeriesFolder did not pass on holo_kw when loading hologram data
 0.10.1

qpformat/file_formats/dataset.py

@@ -121,6 +121,7 @@ def _identifier_data(self):
         else:
             with self.path.open("rb") as fd:
                 data.append(fd.read(50 * 1024))
+            data.append(self.path.stat().st_size)
         data += self._identifier_meta()
         return hash_obj(data)
 
@@ -398,6 +399,8 @@ def obj2bytes(data):
         tohash.append(data)
     elif isinstance(data, np.ndarray):
         tohash.append(data.tobytes())
+    elif isinstance(data, int):
+        tohash.append(bytes(data))
     else:
         msg = "No rule to convert to bytes: {}".format(data)
         raise NotImplementedError(msg)

tests/test_dataset.py

@@ -53,12 +53,12 @@ def test_identifier():
     path = datapath / "series_phasics.zip"
 
     ds1 = qpformat.load_data(path=path)
-    assert ds1.identifier == "7fc65"
+    assert ds1.identifier == "49bbc"
 
     bg_data = ds1.get_qpimage(0)
     ds2 = qpformat.load_data(path=path, bg_data=bg_data)
-    assert ds2.background_identifier == "26461"
-    assert ds2.identifier == "67213"
+    assert ds2.background_identifier == "bf13f"
+    assert ds2.identifier == "b6d1c"
 
 
 def test_meta():

tests/test_single_tif_holo.py

@@ -19,7 +19,7 @@ def test_basic():
 def test_identifier():
     path = datapath / "single_holo.tif"
     ds = qpformat.load_data(path, holo_kw={"sideband": 1})
-    assert ds.identifier == "e9d9d"
+    assert ds.identifier == "b65c8"
 
 
 def test_load_data():