Add code README and automatic CHANGELOG entries

code/README.md

@@ -1,3 +1,12 @@
-All custom code goes into this directory. All scripts should be written such
-that they can be executed from the root of the dataset, and are only using
-relative paths for portability.
+# Popcorn dataset code
+
+Each script can be run stand-alone like `python code/files.py <input-dir> <output-dir>`,
+exchanging the script file for the one intended.
+
+It is suggested, however, to run the scripts using the `just` command runner from the
+dataset root, such as `just files` for the same effect as above.
+This will automatically populate the correct input and output directories.
+
+To create new `datalad` versioned output data, run `just versioned` or `just` without any arguments.
+A new commit containing the updated data will be created,
+and an automatic entry in the CHANGELOG made.

code/files.py

@@ -1,6 +1,4 @@
 import csv
-from collections.abc import Generator
-from datetime import datetime
 from pathlib import Path
 
 
@@ -29,7 +27,6 @@ def main(input: str, output: str) -> None:
     input_dir = Path(input)
     output_dir = Path(output)
     ensure_dirs(input_dir, output_dir)
-
     filesize_csv(input_dir, output_dir)
 
 

code/kernels.py

@@ -1,7 +1,5 @@
 import csv
 import json
-from collections.abc import Generator
-from datetime import datetime
 from pathlib import Path
 from typing import Any, cast
 
@@ -33,7 +31,6 @@ def package_kernel_csv(input_dir: Path, output_dir: Path) -> None:
                     p_count = cast(int, data["XuKernel"][entry])
                     p_date = date
                     writer.writerow([p_date, p_name, p_count])
-                    # print(output_file, p_date, p_name, p_count)
 
 
 def ensure_dirs(input_dir: Path, output_dir: Path):
@@ -46,9 +43,6 @@ def main(input: str, output: str) -> None:
     input_dir = Path(input)
     output_dir = Path(output)
     ensure_dirs(input_dir, output_dir)
-    # json_to_daily_pkg(Path(input) / "daily", Path(output) / "daily", force=True)
-    # json_to_unique_csv(Path(input) / "daily", Path(output), force=True)
-
     package_kernel_csv(input_dir, output_dir)
 
 

justfile

@@ -16,3 +16,4 @@ packages:
 
 versioned:
     datalad run -m "Create updated output data" -i input/ -o output/ just all
+    printf "\n## %s\n\n- auto generated\n" $(date +%F) | sed -i '/CHANGELOG/r /dev/stdin' CHANGELOG.md