dcnum 0.18.0__tar.gz → 0.19.1__tar.gz

{dcnum-0.18.0 → dcnum-0.19.1}/CHANGELOG

@@ -1,3 +1,9 @@
+0.19.1
+ - enh: support steps when specifying data slices in `index_mapping`
+0.19.0
+ - enh: elevate `HDF5Data`s `index_mapping` to pipeline identifier status
+   (this changes the pipeline identifier)
+ - enh: improve sanity checks for `BackgroundRollMed`
 0.18.0
  - BREAKING CHANGE: mask postprocessing did a morphological opening instead
    of a morphological closing, failing to remove spurious noise

{dcnum-0.18.0/src/dcnum.egg-info → dcnum-0.19.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: dcnum
-Version: 0.18.0
+Version: 0.19.1
 Summary: numerics toolbox for imaging deformability cytometry
 Author: Maximilian Schlögel, Paul Müller
 Maintainer-email: Paul Müller <dev@craban.de>

{dcnum-0.18.0 → dcnum-0.19.1}/src/dcnum/_version.py

@@ -12,5 +12,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '0.18.0'
-__version_tuple__ = version_tuple = (0, 18, 0)
+__version__ = version = '0.19.1'
+__version_tuple__ = version_tuple = (0, 19, 1)

{dcnum-0.18.0 → dcnum-0.19.1}/src/dcnum/feat/feat_background/bg_roll_median.py

@@ -119,7 +119,7 @@ class BackgroundRollMed(Background):
         """Check user-defined properties of this class
 
         This method primarily exists so that the CLI knows which
-        keyword arguements can be passed to this class.
+        keyword arguments can be passed to this class.
 
         Parameters
         ----------
@@ -132,7 +132,8 @@ class BackgroundRollMed(Background):
             `kernel_size` will not increase computation speed. Larger
             values lead to a higher memory consumption.
         """
-        assert kernel_size > 0
+        assert kernel_size > 0, "kernel size must be positive number"
+        assert kernel_size % 2 == 0, "kernel size must be even number"
         assert batch_size > kernel_size
 
     def get_slices_for_batch(self, batch_index=0):

{dcnum-0.18.0 → dcnum-0.19.1}/src/dcnum/logic/ctrl.py

@@ -1,11 +1,9 @@
 import collections
 import datetime
-import hashlib
 import json
 import logging
 from logging.handlers import QueueListener
 import multiprocessing as mp
-import numbers
 import os
 import pathlib
 import platform
@@ -16,7 +14,6 @@ import traceback
 import uuid
 
 import h5py
-import numpy as np
 
 from ..feat.feat_background.base import get_available_background_methods
 from ..feat.queue_event_extractor import QueueEventExtractor
@@ -313,7 +310,17 @@ class DCNumJobRunner(threading.Thread):
             # Whether pipeline hash is invalid.
             ppid.compute_pipeline_hash(**datdict) != dathash
             # Whether the input file is the original output of the pipeline.
-            or len(self.draw) != evyield)
+            or len(self.draw) != evyield
+            # If index mapping is defined, then we always redo the pipeline.
+            # If the pipeline hashes are identical and index mapping is not
+            # None, then both pipelines were done with index mapping.
+            # But applying the same pipeline with index mapping in series
+            # will lead to a different result in the second run (e.g. 1st
+            # pipeline run: take every 2nd event; 2nd pipeline run: take
+            # every second event -> results in every 4th event in output of
+            # second pipeline run).
+            or self.draw.index_mapping is not None
+        )
         # Do we have to recompute the background data? In addition to the
         # hash sanity check above, check the generation, input data,
         # and background pipeline identifiers.
@@ -387,21 +394,7 @@ class DCNumJobRunner(threading.Thread):
             hw.h5.attrs["pipeline:dcnum yield"] = self.event_count
             # index mapping information
             im = self.job.kwargs["data_kwargs"].get("index_mapping", None)
-            if im is None:
-                dim = "0"
-            elif isinstance(im, numbers.Number):
-                dim = f"{im}"
-            elif isinstance(im, slice):
-                dim = (f"{im.start if im.start is not None else 'n'}"
-                       + f"-{im.stop if im.stop is not None else 'n'}"
-                       + f"-{im.step if im.step is not None else 'n'}"
-                       )
-            elif isinstance(im, (list, np.ndarray)):
-                idhash = hashlib.md5(
-                    np.array(im, dtype=np.uint32).tobytes()).hexdigest()
-                dim = f"h-{idhash[:8]}"
-            else:
-                dim = "unknown"
+            dim = HDF5Data.get_ppid_index_mapping(im)
             hw.h5.attrs["pipeline:dcnum mapping"] = dim
             # regular metadata
             hw.h5.attrs["experiment:event count"] = self.event_count

{dcnum-0.18.0 → dcnum-0.19.1}/src/dcnum/read/hdf5_data.py

@@ -1,7 +1,9 @@
 from __future__ import annotations
 
+import hashlib
 import io
 import json
+import numbers
 import pathlib
 import tempfile
 from typing import Dict, BinaryIO, List
@@ -293,7 +295,9 @@ class HDF5Data:
         self.h5.close()
 
     def get_ppid(self):
-        return self.get_ppid_from_ppkw({"pixel_size": self.pixel_size})
+        return self.get_ppid_from_ppkw(
+            {"pixel_size": self.pixel_size,
+             "index_mapping": self.index_mapping})
 
     @classmethod
     def get_ppid_code(cls):
@@ -304,10 +308,34 @@ class HDF5Data:
         # Data does not really fit into the PPID scheme we use for the rest
         # of the pipeline. This implementation here is custom.
         code = cls.get_ppid_code()
+        # pixel size
         ppid_ps = f"{kwargs['pixel_size']:.8f}".rstrip("0")
-        kwid = "^".join([f"p={ppid_ps}"])
+        # index mapping
+        ppid_im = cls.get_ppid_index_mapping(kwargs.get("index_mapping", None))
+        kwid = "^".join([f"p={ppid_ps}", f"i={ppid_im}"])
         return ":".join([code, kwid])
 
+    @staticmethod
+    def get_ppid_index_mapping(index_mapping):
+        """Return the pipeline identifier part for index mapping"""
+        im = index_mapping
+        if im is None:
+            dim = "0"
+        elif isinstance(im, numbers.Integral):
+            dim = f"{im}"
+        elif isinstance(im, slice):
+            dim = (f"{im.start if im.start is not None else 'n'}"
+                   + f"-{im.stop if im.stop is not None else 'n'}"
+                   + f"-{im.step if im.step is not None else 'n'}"
+                   )
+        elif isinstance(im, (list, np.ndarray)):
+            idhash = hashlib.md5(
+                np.array(im, dtype=np.uint32).tobytes()).hexdigest()
+            dim = f"h-{idhash[:8]}"
+        else:
+            dim = "unknown"
+        return dim
+
     @staticmethod
     def get_ppkw_from_ppid(dat_ppid):
         # Data does not fit in the PPID scheme we use, but we still
@@ -321,6 +349,20 @@ class HDF5Data:
             var, val = item.split("=")
             if var == "p":
                 kwargs["pixel_size"] = float(val)
+            elif var == "i":
+                if val.startswith("h-") or val == "unknown":
+                    raise ValueError(f"Cannot invert index mapping {val}")
+                elif val == "0":
+                    kwargs["index_mapping"] = None
+                elif val.count("-"):
+                    start, stop, step = val.split("-")
+                    kwargs["index_mapping"] = slice(
+                        None if start == "n" else int(start),
+                        None if stop == "n" else int(stop),
+                        None if step == "n" else int(step)
+                    )
+                else:
+                    kwargs["index_mapping"] = int(val)
             else:
                 raise ValueError(f"Invalid parameter '{var}'!")
         return kwargs

{dcnum-0.18.0 → dcnum-0.19.1}/src/dcnum/read/mapped.py

@@ -54,12 +54,10 @@ def _get_mapping_indices_cached(
         return np.arange(index_mapping)
     elif isinstance(index_mapping, tuple):
         im_slice = slice(*index_mapping)
-        if im_slice.step is not None:
-            raise NotImplementedError("Slices with step not implemented yet")
         if im_slice.stop is None or im_slice.start is None:
             raise NotImplementedError(
                 "Slices must have start and stop defined")
-        return np.arange(im_slice.start, im_slice.stop)
+        return np.arange(im_slice.start, im_slice.stop, im_slice.step)
     elif isinstance(index_mapping, list):
         return np.array(index_mapping, dtype=np.uint32)
     else:

{dcnum-0.18.0 → dcnum-0.19.1/src/dcnum.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: dcnum
-Version: 0.18.0
+Version: 0.19.1
 Summary: numerics toolbox for imaging deformability cytometry
 Author: Maximilian Schlögel, Paul Müller
 Maintainer-email: Paul Müller <dev@craban.de>

{dcnum-0.18.0 → dcnum-0.19.1}/tests/test_logic_pipeline.py

@@ -59,17 +59,12 @@ def test_chained_pipeline():
                == "sparsemed:k=250^s=1^t=0^f=0.8^o=1"
 
 
-@pytest.mark.parametrize("index_mapping,size,mapping_out", [
-    (None, 395, "0"),
-    (5, 11, "5"),
-    (slice(3, 5, None), 6, "3-5-n"),
-    ([3, 5, 6, 7], 7, "h-6e582938"),
-])
-def test_duplicate_pipeline(index_mapping, size, mapping_out):
+def test_duplicate_pipeline():
     """Test running the same pipeline twice
 
-    When the pipeline is run on a file with the same pipeline
-    identifier, data are just copied over. Nothing much fancy else.
+    When the pipeline is run on a file that has been run with the same
+    pipeline identifier, then we do not run the pipeline. Instead, we
+    copy the data from the first file.
     """
     path_orig = retrieve_data("fmt-hdf5_cytoshot_full-features_2023.zip")
     path = path_orig.with_name("input.rtdc")
@@ -79,13 +74,12 @@ def test_duplicate_pipeline(index_mapping, size, mapping_out):
     job = logic.DCNumPipelineJob(
         path_in=path,
         path_out=path2,
-        data_kwargs={"index_mapping": index_mapping},
         background_code="copy",
         segmenter_code="thresh",
         segmenter_kwargs={"thresh": -6,
                           "kwargs_mask": {"closing_disk": 0}},
         debug=True)
-    assert job.kwargs["data_kwargs"]["index_mapping"] == index_mapping
+    assert job.kwargs["data_kwargs"].get("index_mapping") is None
 
     # perform the initial pipeline
     with logic.DCNumJobRunner(job=job) as runner:
@@ -102,16 +96,12 @@ def test_duplicate_pipeline(index_mapping, size, mapping_out):
 
     # get the first image for reference
     with h5py.File(path) as h5:
-        if index_mapping is None:
-            idx0 = 0
-        else:
-            idx0 = read.get_mapping_indices(index_mapping)[0]
-        im0 = h5["/events/image"][idx0]
+        im0 = h5["/events/image"][0]
 
     # remove all logs just to be sure nothing interferes
     with h5py.File(path2, "a") as h5:
-        assert h5.attrs["pipeline:dcnum mapping"] == mapping_out
-        assert len(h5["events/deform"]) == size
+        assert h5.attrs["pipeline:dcnum mapping"] == "0"
+        assert len(h5["events/deform"]) == 395
         del h5["logs"]
 
     # now when we do everything again, not a thing should be done
@@ -140,11 +130,227 @@ def test_duplicate_pipeline(index_mapping, size, mapping_out):
         assert "deform" in h5["events"]
         assert "image" in h5["events"]
         assert "image_bg" in h5["events"]
-        assert len(h5["events/deform"]) == size
+        assert len(h5["events/deform"]) == 395
         assert h5.attrs["pipeline:dcnum mapping"] == "0"
         assert np.all(h5["events/image"][0] == im0)
 
 
+def test_duplicate_pipeline_redo_index_mapping():
+    """Test running the same pipeline twice
+
+    When the pipeline is run on a file that has been run with the same
+    pipeline identifier, then we do not run the pipeline. Instead, we
+    copy the data from the first file.
+
+    However, if something is odd, such as index mapping defined in the
+    pipeline then redo the computations.
+    This is the purpose of this test.
+    """
+    path_orig = retrieve_data("fmt-hdf5_cytoshot_full-features_2023.zip")
+    path = path_orig.with_name("input.rtdc")
+    path2 = path.with_name("path_intermediate.rtdc")
+    with read.concatenated_hdf5_data(5 * [path_orig], path_out=path):
+        pass
+    job = logic.DCNumPipelineJob(
+        path_in=path,
+        path_out=path2,
+        data_kwargs={"index_mapping": 10},
+        background_code="copy",
+        segmenter_code="thresh",
+        segmenter_kwargs={"thresh": -6,
+                          "kwargs_mask": {"closing_disk": 0}},
+        debug=True)
+    assert job.kwargs["data_kwargs"].get("index_mapping") == 10
+
+    # perform the initial pipeline
+    with logic.DCNumJobRunner(job=job) as runner:
+        runner.run()
+    # Sanity checks for initial job
+    with read.HDF5Data(job["path_out"]) as hd:
+        # Check the logs
+        logdat = " ".join(get_log(hd, time.strftime("dcnum-log-%Y")))
+        assert "Starting background computation" in logdat
+        assert "Finished background computation" in logdat
+        assert "Starting segmentation and feature extraction" in logdat
+        assert "Flushing data to disk" in logdat
+        assert "Finished segmentation and feature extraction" in logdat
+
+    with h5py.File(path2, "a") as h5:
+        # sanity checks
+        assert h5.attrs["pipeline:dcnum mapping"] == "10"
+        assert len(h5["events/deform"]) == 24
+        assert h5.attrs["pipeline:dcnum yield"] == 24
+        # remove all logs just to be sure nothing interferes
+        del h5["logs"]
+        # Modify the yield, triggering a new pipeline run
+        h5.attrs["pipeline:dcnum yield"] = 111111
+
+    # now when we do everything again, not a thing should be done
+    job2 = logic.DCNumPipelineJob(
+        path_in=path2,
+        path_out=path2.with_name("final_out.rtdc"),
+        no_basins_in_output=True,
+        data_kwargs={"index_mapping": 10},
+        background_code="copy",
+        segmenter_code="thresh",
+        segmenter_kwargs={"thresh": -6,
+                          "kwargs_mask": {"closing_disk": 0}},
+        debug=True)
+    with logic.DCNumJobRunner(job=job2) as runner2:
+        runner2.run()
+    # Real check for second run (not the `not`s [sic]!)
+    with read.HDF5Data(job2["path_out"]) as hd:
+        # Check the logs
+        logdat = " ".join(get_log(hd, time.strftime("dcnum-log-%Y")))
+        # Background computation is not repeated
+        assert "Starting background computation" not in logdat
+        assert "Finished background computation" not in logdat
+        # Segmentation is repeated
+        assert "Starting segmentation and feature extraction" in logdat
+        assert "Flushing data to disk" in logdat
+        assert "Finished segmentation and feature extraction" in logdat
+
+    with h5py.File(job2["path_out"]) as h5:
+        assert "deform" in h5["events"]
+        assert "image" in h5["events"]
+        assert "image_bg" in h5["events"]
+        # We have not 24 here, because the index mapping enumerates events,
+        # not frames.
+        assert len(h5["events/deform"]) == 11
+        assert h5.attrs["pipeline:dcnum mapping"] == "10"
+        assert h5.attrs["pipeline:dcnum yield"] == 11
+
+
+def test_duplicate_pipeline_redo_yield():
+    """Test running the same pipeline twice
+
+    When the pipeline is run on a file that has been run with the same
+    pipeline identifier, then we do not run the pipeline. Instead, we
+    copy the data from the first file.
+
+    However, if something is odd, such as the yield of the pipeline not
+    matching the data in the output file, then redo the computations.
+    This is the purpose of this test.
+    """
+    path_orig = retrieve_data("fmt-hdf5_cytoshot_full-features_2023.zip")
+    path = path_orig.with_name("input.rtdc")
+    path2 = path.with_name("path_intermediate.rtdc")
+    with read.concatenated_hdf5_data(5 * [path_orig], path_out=path):
+        pass
+    job = logic.DCNumPipelineJob(
+        path_in=path,
+        path_out=path2,
+        background_code="copy",
+        segmenter_code="thresh",
+        segmenter_kwargs={"thresh": -6,
+                          "kwargs_mask": {"closing_disk": 0}},
+        debug=True)
+    assert job.kwargs["data_kwargs"].get("index_mapping") is None
+
+    # perform the initial pipeline
+    with logic.DCNumJobRunner(job=job) as runner:
+        runner.run()
+    # Sanity checks for initial job
+    with read.HDF5Data(job["path_out"]) as hd:
+        # Check the logs
+        logdat = " ".join(get_log(hd, time.strftime("dcnum-log-%Y")))
+        assert "Starting background computation" in logdat
+        assert "Finished background computation" in logdat
+        assert "Starting segmentation and feature extraction" in logdat
+        assert "Flushing data to disk" in logdat
+        assert "Finished segmentation and feature extraction" in logdat
+
+    with h5py.File(path2, "a") as h5:
+        # sanity checks
+        assert h5.attrs["pipeline:dcnum mapping"] == "0"
+        assert len(h5["events/deform"]) == 395
+        assert h5.attrs["pipeline:dcnum yield"] == 395
+        # remove all logs just to be sure nothing interferes
+        del h5["logs"]
+        # Modify the yield, triggering a new pipeline run
+        h5.attrs["pipeline:dcnum yield"] = 111111
+
+    # now when we do everything again, not a thing should be done
+    job2 = logic.DCNumPipelineJob(
+        path_in=path2,
+        path_out=path2.with_name("final_out.rtdc"),
+        no_basins_in_output=True,
+        background_code="copy",
+        segmenter_code="thresh",
+        segmenter_kwargs={"thresh": -6,
+                          "kwargs_mask": {"closing_disk": 0}},
+        debug=True)
+    with logic.DCNumJobRunner(job=job2) as runner2:
+        runner2.run()
+    # Real check for second run (not the `not`s [sic]!)
+    with read.HDF5Data(job2["path_out"]) as hd:
+        # Check the logs
+        logdat = " ".join(get_log(hd, time.strftime("dcnum-log-%Y")))
+        # Background computation is not repeated
+        assert "Starting background computation" not in logdat
+        assert "Finished background computation" not in logdat
+        # Segmentation is repeated
+        assert "Starting segmentation and feature extraction" in logdat
+        assert "Flushing data to disk" in logdat
+        assert "Finished segmentation and feature extraction" in logdat
+
+    with h5py.File(job2["path_out"]) as h5:
+        assert "deform" in h5["events"]
+        assert "image" in h5["events"]
+        assert "image_bg" in h5["events"]
+        assert len(h5["events/deform"]) == 395
+        assert h5.attrs["pipeline:dcnum mapping"] == "0"
+        assert h5.attrs["pipeline:dcnum yield"] == 395
+
+
+@pytest.mark.parametrize("index_mapping,size,mapping_out", [
+    (5, 11, "5"),
+    (slice(3, 5, None), 6, "3-5-n"),
+    ([3, 5, 6, 7], 7, "h-6e582938"),
+])
+def test_index_mapping_pipeline(index_mapping, size, mapping_out):
+    """Test running the same pipeline twice
+
+    When the pipeline is run on a file with the same pipeline
+    identifier, data are just copied over. Nothing much fancy else.
+    """
+    path_orig = retrieve_data("fmt-hdf5_cytoshot_full-features_2023.zip")
+    path = path_orig.with_name("input.rtdc")
+    path2 = path.with_name("path_intermediate.rtdc")
+    with read.concatenated_hdf5_data(5 * [path_orig], path_out=path):
+        pass
+    job = logic.DCNumPipelineJob(
+        path_in=path,
+        path_out=path2,
+        data_kwargs={"index_mapping": index_mapping},
+        background_code="copy",
+        segmenter_code="thresh",
+        segmenter_kwargs={"thresh": -6,
+                          "kwargs_mask": {"closing_disk": 0}},
+        debug=True)
+    assert job.kwargs["data_kwargs"]["index_mapping"] == index_mapping
+
+    # perform the initial pipeline
+    with logic.DCNumJobRunner(job=job) as runner:
+        runner.run()
+    # Sanity checks for initial job
+    with read.HDF5Data(job["path_out"]) as hd:
+        # Check the logs
+        logdat = " ".join(get_log(hd, time.strftime("dcnum-log-%Y")))
+        assert "Starting background computation" in logdat
+        assert "Finished background computation" in logdat
+        assert "Starting segmentation and feature extraction" in logdat
+        assert "Flushing data to disk" in logdat
+        assert "Finished segmentation and feature extraction" in logdat
+
+    with h5py.File(job["path_out"]) as h5:
+        assert "deform" in h5["events"]
+        assert "image" in h5["events"]
+        assert "image_bg" in h5["events"]
+        assert len(h5["events/deform"]) == size
+        assert h5.attrs["pipeline:dcnum mapping"] == mapping_out
+
+
 def test_duplicate_transfer_basin_data():
     """task_transfer_basin_data should not copy basin data from input"""
     path_orig = retrieve_data("fmt-hdf5_cytoshot_full-features_2023.zip")
@@ -335,7 +541,7 @@ def test_simple_pipeline(debug):
 
     # this is the default pipeline
     gen_id = ppid.DCNUM_PPID_GENERATION
-    dat_id = "hdf:p=0.2645"
+    dat_id = "hdf:p=0.2645^i=0"
     bg_id = "sparsemed:k=200^s=1^t=0^f=0.8^o=1"
     seg_id = "thresh:t=-6:cle=1^f=1^clo=0"
     feat_id = "legacy:b=1^h=1^v=1"
@@ -402,7 +608,7 @@ def test_simple_pipeline_no_offset_correction(debug):
 
     # this is the default pipeline
     gen_id = ppid.DCNUM_PPID_GENERATION
-    dat_id = "hdf:p=0.2645"
+    dat_id = "hdf:p=0.2645^i=0"
     bg_id = "sparsemed:k=200^s=1^t=0^f=0.8^o=0"
     seg_id = "thresh:t=-6:cle=1^f=1^clo=0"
     feat_id = "legacy:b=1^h=1^v=1"
@@ -474,7 +680,7 @@ def test_simple_pipeline_in_thread():
 @pytest.mark.parametrize("attr,oldval,newbg", [
     # Changes that trigger computation of new background
     ["pipeline:dcnum generation", "1", True],
-    ["pipeline:dcnum data", "hdf:p=0.2656", True],
+    ["pipeline:dcnum data", "hdf:p=0.2656^i=0", True],
     ["pipeline:dcnum background", "sparsemed:k=100^s=1^t=0^f=0.8^o=1", True],
     # Changes that don't trigger background computation
     ["pipeline:dcnum segmenter", "thresh:t=-1:cle=1^f=1^clo=2", False],
@@ -505,7 +711,7 @@ def test_recomputation_of_background_metadata_changed(attr, oldval, newbg):
 
         # Set the default values
         h5.attrs["pipeline:dcnum generation"] = ppid.DCNUM_PPID_GENERATION
-        h5.attrs["pipeline:dcnum data"] = "hdf:p=0.2645"
+        h5.attrs["pipeline:dcnum data"] = "hdf:p=0.2645^i=0"
         h5.attrs["pipeline:dcnum background"] = \
             "sparsemed:k=200^s=1^t=0^f=0.8^o=1"
         h5.attrs["pipeline:dcnum segmenter"] = "thresh:t=-6:cle=1^f=1^clo=2"
@@ -553,7 +759,7 @@ def test_task_background():
 
     # this is the default pipeline
     gen_id = ppid.DCNUM_PPID_GENERATION
-    dat_id = "hdf:p=0.2645"
+    dat_id = "hdf:p=0.2645^i=0"
     bg_id = "sparsemed:k=200^s=1^t=0^f=0.8^o=1"
     seg_id = "thresh:t=-6:cle=1^f=1^clo=2"
     feat_id = "legacy:b=1^h=1^v=1"

{dcnum-0.18.0 → dcnum-0.19.1}/tests/test_meta_ppid_base.py

@@ -32,13 +32,13 @@ class ExampleClass:
 def test_compute_pipeline_hash():
     pp_hash = ppid.compute_pipeline_hash(
         gen_id="7",
-        dat_id="hdf:p=0.34",
+        dat_id="hdf:p=0.34^i=0",
         bg_id="sparsemed:k=200^s=1^t=0^f=0.8^o=1",
         seg_id="thresh:t=-3:cle=1^f=1^clo=2",
         feat_id="legacy:b=1^h=0^v=1",
         gate_id="norm:o=0^s=11",
     )
-    assert pp_hash == "2e56aa93fcb264381c90a8fd181b3fbc"
+    assert pp_hash == "4f3a850410b9801393ab5738afe69e9a"
 
 
 @pytest.mark.parametrize("in_list,out_list", [

{dcnum-0.18.0 → dcnum-0.19.1}/tests/test_meta_ppid_data.py

@@ -1,5 +1,7 @@
 from dcnum.read import HDF5Data
 
+import pytest
+
 
 def test_ppid_decoding_dat_check_kwargs():
     dat_ppid = "hdf:p=0.2658"
@@ -7,17 +9,30 @@ def test_ppid_decoding_dat_check_kwargs():
     assert kwargs["pixel_size"] == 0.2658
 
 
+@pytest.mark.parametrize("imppid,value", [
+    ["0", None],
+    ["10", 10],
+    ["10-20-n", slice(10, 20)],
+    ["10-20-2", slice(10, 20, 2)],
+    ["n-n-2", slice(None, None, 2)],
+])
+def test_ppid_decoding_dat_check_kwargs_index_mapping(imppid, value):
+    dat_ppid = f"hdf:p=0.2658^i={imppid}"
+    kwargs = HDF5Data.get_ppkw_from_ppid(dat_ppid)
+    assert kwargs["index_mapping"] == value
+
+
 def test_ppid_encoding_dat_check_kwargs():
     kwargs = {"pixel_size": 0.34}
     ppid = HDF5Data.get_ppid_from_ppkw(kwargs)
-    assert ppid == "hdf:p=0.34"
+    assert ppid == "hdf:p=0.34^i=0"
 
 
 def test_ppid_encoding_dat_check_kwargs_acc():
     # accuracy for pixel_size is 8 digits after the decimal point
     kwargs = {"pixel_size": 0.3400000036}
     ppid = HDF5Data.get_ppid_from_ppkw(kwargs)
-    assert ppid == "hdf:p=0.34"
+    assert ppid == "hdf:p=0.34^i=0"
 
 
 def test_ppid_required_method_definitions():

{dcnum-0.18.0 → dcnum-0.19.1}/tests/test_read_hdf5.py

@@ -49,10 +49,10 @@ def test_get_ppid():
         "fmt-hdf5_cytoshot_full-features_legacy_allev_2023.zip")
 
     with read.HDF5Data(path) as hd:
-        assert hd.get_ppid() == "hdf:p=0.2645"
+        assert hd.get_ppid() == "hdf:p=0.2645^i=0"
 
     with read.HDF5Data(path, pixel_size=0.49) as hd:
-        assert hd.get_ppid() == "hdf:p=0.49"
+        assert hd.get_ppid() == "hdf:p=0.49^i=0"
 
 
 def test_get_ppkw_from_ppid_error_bad_code():
@@ -64,7 +64,7 @@ def test_get_ppkw_from_ppid_error_bad_code():
 def test_get_ppkw_from_ppid_error_bad_parameter():
     with pytest.raises(ValueError,
                        match="Invalid parameter 'k'"):
-        read.HDF5Data.get_ppkw_from_ppid("hdf:k=0.44")
+        read.HDF5Data.get_ppkw_from_ppid("hdf:k=0.44^i=0")
 
 
 def test_get_ppkw_from_ppid_pixel_size():
@@ -158,6 +158,25 @@ def test_image_cache_iter_chunks(size, chunks, tmp_path):
         assert list(hic.iter_chunks()) == list(range(chunks))
 
 
+@pytest.mark.parametrize("index_mapping,result_data", [
+    [2, [0, 1]],
+    [slice(1, 10, 2), [1, 3, 5, 7, 9]],
+    [slice(1, 11, 3), [1, 4, 7, 10]],
+    [slice(10, 11), [10]],
+    [slice(1, 3, None), [1, 2]],
+])
+def test_index_mapping(index_mapping, result_data):
+    path = retrieve_data(
+        "fmt-hdf5_cytoshot_full-features_legacy_allev_2023.zip")
+    with h5py.File(path, "a") as h5:
+        size = len(h5["events/image"])
+        assert size == 11
+        h5["events/temp"] = np.arange(size, dtype=np.float64)
+
+    with read.HDF5Data(path, index_mapping=index_mapping) as hd:
+        assert np.allclose(hd["temp"], result_data)
+
+
 def test_keyerror_when_image_is_none(tmp_path):
     path = tmp_path / "test.hdf5"
     with h5py.File(path, "w") as hw: