Merge pull request #818 from nipy/enh-one-README

Do not create README if there is other allowed README.md or alike

Author: yarikoptic

heudiconv/bids.py

@@ -176,12 +176,12 @@ def populate_bids_templates(
     if op.exists(op.dirname(sourcedata_README)):
         create_file_if_missing(
             sourcedata_README,
-            (
-                "TODO: Provide description about source data, e.g. \n"
-                "Directory below contains DICOMS compressed into tarballs per "
-                "each sequence, replicating directory hierarchy of the BIDS dataset"
-                " itself."
-            ),
+            "TODO: Provide description about source data, e.g. \n"
+            "Directory below contains DICOMS compressed into tarballs per "
+            "each sequence, replicating directory hierarchy of the BIDS dataset"
+            " itself.",
+            # TODO: get from schema
+            glob_suffixes=[".md", ".txt", ".rst", ""],
         )
     create_file_if_missing(
         op.join(path, "CHANGES"),
@@ -196,6 +196,8 @@ def populate_bids_templates(
         op.join(path, "README"),
         "TODO: Provide description for the dataset -- basic details about the "
         "study, possibly pointing to pre-registration (if public or embargoed)",
+        # TODO: get from schema
+        glob_suffixes=[".md", ".txt", ".rst", ""],
     )
     create_file_if_missing(
         op.join(path, "scans.json"), json_dumps(SCANS_FILE_FIELDS, sort_keys=False)

heudiconv/tests/test_main.py

@@ -44,14 +44,35 @@ def test_main_version(std: StringIO) -> None:
 
 
 def test_create_file_if_missing(tmp_path: Path) -> None:
-    tf = tmp_path / "README.txt"
+    tf_base = tmp_path / "README"
+    tf = tf_base.with_suffix(".txt")
     assert not tf.exists()
-    create_file_if_missing(str(tf), "content")
+    assert create_file_if_missing(str(tf), "content")
     assert tf.exists()
     assert tf.read_text() == "content"
-    create_file_if_missing(str(tf), "content2")
+    assert not create_file_if_missing(str(tf), "content2")
     # nothing gets changed
     assert tf.read_text() == "content"
+    # matching glob - no change
+    for g in [".md", ".txt", ".json"], [".*"], [".*", ".txt"]:
+        assert not create_file_if_missing(str(tf_base), "content2", glob_suffixes=g)
+        assert tf.exists()
+        assert tf.read_text() == "content"
+        assert not tf_base.exists()  # was not created, since there is .md
+
+    # non-matching glob - change
+    for g in [".md", ".json"], [".d*", ".*d"]:
+        assert create_file_if_missing(str(tf_base), "content3", glob_suffixes=g)
+        assert tf_base.read_text() == "content3"
+        # now that we have suffix less README, we do not match, so we keep creating it
+        assert create_file_if_missing(str(tf_base), "content4", glob_suffixes=g)
+        assert tf_base.read_text() == "content4"
+        # unless we list it explicitly
+        assert not create_file_if_missing(
+            str(tf_base), "content5", glob_suffixes=g + [""]
+        )
+        assert tf_base.read_text() == "content4"
+        tf_base.unlink()
 
 
 def test_populate_bids_templates(tmp_path: Path) -> None:

heudiconv/utils.py

@@ -164,10 +164,20 @@ def anonymize_sid(sid: AnyStr, anon_sid_cmd: str) -> AnyStr:
     return anon_sid
 
 
-def create_file_if_missing(filename: str, content: str) -> bool:
+def create_file_if_missing(
+    filename: str, content: str, glob_suffixes: list[str] | None = None
+) -> bool:
     """Create file if missing, so we do not
-    override any possibly introduced changes"""
-    if op.lexists(filename):
+    override any possibly introduced changes.
+
+    Note: if glob_suffixes list is used, and it is desired
+    also to allow for original filename to 'match', add an empty
+    string within glob_suffixes.
+    """
+    if glob_suffixes:
+        if any(glob(filename + s) for s in glob_suffixes):
+            return False
+    elif op.lexists(filename):
         return False
     dirname = op.dirname(filename)
     if not op.exists(dirname):