From a76eb938ff19da01652605d59dceadbaaff2dcde Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Paul=20M=C3=BCller?= <dev@craban.de>
Date: Fri, 24 Nov 2023 23:28:12 +0100
Subject: [PATCH] feat: implement HDF5Writer.store_log

---
 CHANGELOG                  |  1 +
 src/dcnum/write/writer.py  | 27 +++++++++++++++++++++++++++
 tests/test_write_writer.py | 29 +++++++++++++++++++++++++++++
 3 files changed, 57 insertions(+)

diff --git a/CHANGELOG b/CHANGELOG
index 8ac4c5b..56c24ba 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -2,6 +2,7 @@
  - BREAKING CHANGE: Remove preselection capabilities, because it is not
    well integrated into the pipeline. For more information, please see
    issue #15.
+ - feat: implement HDF5Writer.store_log
  - enh: add Segmenter.hardware_processor property
  - enh: introduce pipeline identifier for data pixel size
  - enh: reduce pixel_size accuracy to 8 digits after the decimal point
diff --git a/src/dcnum/write/writer.py b/src/dcnum/write/writer.py
index fc25d23..86b8b5e 100644
--- a/src/dcnum/write/writer.py
+++ b/src/dcnum/write/writer.py
@@ -136,6 +136,33 @@ def store_feature_chunk(self, feat, data):
         ds.resize(offset + dsize, axis=0)
         ds[offset:offset + dsize] = data
 
+    def store_log(self,
+                  log: str,
+                  data: List[str],
+                  override: bool = False):
+        """Store log data
+
+        Store the log data under the key `log`. The `data`
+        kwarg must be a list of strings. If the log entry
+        already exists, `ValueError` is raised unless
+        `override` is set to True.
+        """
+        logs = self.h5.require_group("logs")
+        if log in logs:
+            if override:
+                del logs[log]
+            else:
+                raise ValueError(
+                    f"Log '{log}' already exists in {self.h5.filename}!")
+        logs.create_dataset(
+            name=log,
+            data=data,
+            shape=(len(data),),
+            # maximum line length
+            dtype=f"S{max([len(ll) for ll in data])}",
+            chunks=True,
+            **self.ds_kwds)
+
 
 def create_with_basins(
         path_out: str | pathlib.Path,
diff --git a/tests/test_write_writer.py b/tests/test_write_writer.py
index 764675c..aca579d 100644
--- a/tests/test_write_writer.py
+++ b/tests/test_write_writer.py
@@ -216,3 +216,32 @@ def test_writer_basin_file_relative():
         assert data_dict["type"] == "file"
         assert data_dict["format"] == "hdf5"
         assert data_dict["features"] == ["deform", "area_um"]
+
+
+def test_writer_logs(tmp_path):
+    path_test = tmp_path / "test.h5"
+    # We basically create a file that consists only of the metadata.
+    with write.HDF5Writer(path_test) as hw:
+        hw.store_log("peter", ["McNulty", "Freamon", "Omar"])
+
+    with read.HDF5Data(path_test) as hd:
+        assert hd.logs["peter"] == ["McNulty", "Freamon", "Omar"]
+
+
+def test_writer_logs_override(tmp_path):
+    path_test = tmp_path / "test.h5"
+    # We basically create a file that consists only of the metadata.
+    with write.HDF5Writer(path_test) as hw:
+        hw.store_log("peter", ["McNulty", "Freamon", "Omar"])
+
+    with read.HDF5Data(path_test) as hd:
+        assert hd.logs["peter"] == ["McNulty", "Freamon", "Omar"]
+
+    with write.HDF5Writer(path_test) as hw:
+        with pytest.raises(ValueError, match="peter"):
+            hw.store_log("peter", ["Omar", "McNulty", "Freamon"])
+        hw.store_log("peter", ["Omar", "McNulty", "Freamon"],
+                     override=True)
+
+    with read.HDF5Data(path_test) as hd:
+        assert hd.logs["peter"] == ["Omar", "McNulty", "Freamon"]