start writing script for updating bids uri

Author: Remi-Gau

tools/check_intended_for.py

@@ -20,52 +20,59 @@
         "IntendedFor": ["bids::sub-01/anat/sub-01_T1w.nii.gz"]
 """
 
-from pathlib import Path
 import json
-from warnings import warn
-from rich import print
+from pathlib import Path
 
 VERBOSE = False
 
-root_dir = Path(__file__).parent.parent
 
-deprecated_formats = {}
+def root_dir():
+    return Path(__file__).parent.parent
+
 
-for json_path in root_dir.glob("**/*.json"):
+def main():
+    deprecated_formats = {}
 
-    if VERBOSE:
-        print(f"Checking {json_path.relative_to(root_dir)}")
+    for json_path in root_dir().glob("**/*.json"):
+        if VERBOSE:
+            print(f"Checking {json_path.relative_to(root_dir())}")
 
-    with open(json_path) as f:
-        content = json.load(f)
+        with open(json_path) as f:
+            content = json.load(f)
+
+        if "IntendedFor" not in content:
+            continue
 
-    if "IntendedFor" in content:
         intended_for = content["IntendedFor"]
         if isinstance(intended_for, str):
             intended_for = [intended_for]
 
         for intended_for_path in intended_for:
-            if not intended_for_path.startswith("bids"):
-                if json_path not in deprecated_formats:
-                    deprecated_formats[json_path] = []
-                deprecated_formats[json_path].append(intended_for_path)
-
-if deprecated_formats:
-
-    log_file = root_dir / "deprecated_intended_for.log"
-
-    with open(log_file, "w") as f:
-        for json_path, deprecated_paths in deprecated_formats.items():
-            f.write(f"{json_path.relative_to(root_dir)}\n")
-            print(f"{json_path.relative_to(root_dir)}")
-            for deprecated_path in deprecated_paths:
-                f.write(f"  {deprecated_path}\n")
-                print(f"  {deprecated_path}")
-
-    raise(ValueError)(
-        f"Found {len(deprecated_formats)} jsons with deprecated IntendedFor formats.\n"
-        f"See {log_file}\n"
-        "Please update them to the new format.\n"
-        "See https://bids-specification.readthedocs.io/en/latest/common-principles.html#bids-uri"
-    )
-
+            if intended_for_path.startswith("bids:"):
+                continue
+
+            if json_path not in deprecated_formats:
+                deprecated_formats[json_path] = []
+            deprecated_formats[json_path].append(intended_for_path)
+
+    if deprecated_formats:
+        log_file = root_dir() / "deprecated_intended_for.log"
+
+        with open(log_file, "w") as f:
+            for json_path, deprecated_paths in deprecated_formats.items():
+                f.write(f"{json_path.relative_to(root_dir())}\n")
+                print(f"{json_path.relative_to(root_dir())}")
+                for deprecated_path in deprecated_paths:
+                    f.write(f"  {deprecated_path}\n")
+                    print(f"  {deprecated_path}")
+
+        raise (ValueError)(
+            f"Found {len(deprecated_formats)} jsons with deprecated IntendedFor formats.\n"
+            f"See {log_file}\n"
+            "Please update them to the new format.\n"
+            "See https://bids-specification.readthedocs.io/en/latest/common-principles.html#bids-uri"
+        )
+
+
+if __name__ == "__main__":
+    main()

tools/print_dataset_listing.py

@@ -1,6 +1,7 @@
 """Take the listing of examples datasets
 and turns it into a markdown document with a series of markdown tables."""
 from pathlib import Path
+
 import pandas as pd
 from bids import BIDSLayout
 
@@ -93,7 +94,7 @@ def update_datatypes_and_suffixes(df, root):
 def add_links(df):
     print("Adding hyperlinks in table...")
     for row in df.iterrows():
-        for col in ["name","link to full data", "maintained by"]:
+        for col in ["name", "link to full data", "maintained by"]:
             if not isinstance(row[1][col], str):
                 continue
             if col == "name":
@@ -105,7 +106,6 @@ def add_links(df):
     return df
 
 
-
 def clean_previous_run(output_file: Path) -> None:
     print("Cleaning output files from previous run...")
     lines = output_file.read_text().split("\n")
@@ -117,13 +117,17 @@ def clean_previous_run(output_file: Path) -> None:
                 break
             f.write(line + "\n")
 
+
 def add_warning(f):
-    f.write("""<!--
+    f.write(
+        """<!--
 Table below is generated automatically.
 Do not edit directly.
 -->
 
-""".upper())
+""".upper()
+    )
+
 
 def add_tables(df: pd.DataFrame, output_file: Path) -> None:
     print("Writing markdown tables...")

tools/requirements.txt

@@ -1,4 +1,3 @@
 pybids  
 pandas
-tabulate
-rich
+tabulate

tools/update_intended_for.py

@@ -0,0 +1,84 @@
+"""Update jsons of all datasets for IntendedFor field to use proper format.
+
+https://bids-specification.readthedocs.io/en/latest/common-principles.html#bids-uri
+
+Example:
+
+- bad
+
+    .. code-block:: json
+
+        "IntendedFor": ["anat/sub-01_T1w.nii.gz"]
+
+
+- good
+
+    .. code-block:: json
+
+        "IntendedFor": ["bids::sub-01/anat/sub-01_T1w.nii.gz"]
+"""
+
+import json
+from pathlib import Path
+
+from rich import print
+
+VERBOSE = False
+
+
+def root_dir():
+    return Path(__file__).parent.parent
+
+
+def main():
+    for json_path in root_dir().glob("**/*.json"):
+
+        if VERBOSE:
+            print(f"Checking {json_path.relative_to(root_dir())}")
+
+        with open(json_path) as f:
+            content = json.load(f)
+
+        if "IntendedFor" not in content:
+            continue
+
+        intended_for = content["IntendedFor"]
+
+        was_str = False
+        if isinstance(intended_for, str):
+            was_str = True
+            intended_for = [intended_for]
+
+        for i, intended_for_path in enumerate(intended_for):
+            if intended_for_path.startswith("bids:"):
+                continue
+
+            if not intended_for_path.startswith(
+                ("ses-", "anat", "func", "ieeg", "fmap", "perf", "micr")
+            ):
+                if VERBOSE:
+                    print(f"[red]will not update {intended_for_path}[/red]")
+                continue
+
+            filename = Path(intended_for_path).name
+            subject = filename.split("_")[0]
+
+            if intended_for_path.startswith("ses-"):
+                session = intended_for_path.split("/")[0]
+                datatype = intended_for_path.split("/")[1]
+                uri = [subject, session, datatype, filename]
+            else:
+                datatype = intended_for_path.split("/")[0]
+                uri = [subject, datatype, filename]
+
+            intended_for[i] = f"bids::{'/'.join(uri)}"
+
+        if was_str:
+            intended_for = intended_for[0]
+
+        if VERBOSE:
+            print(intended_for)
+
+
+if __name__ == "__main__":
+    main()