Remove print statements in datasets

datasets/code_x_glue_tt_text_to_text/code_x_glue_tt_text_to_text.py

@@ -60,7 +60,6 @@ def generate_urls(self, split_name):
             yield self.KEYS[i], f"{split_name}/{lang_pair}.{split_name}.{lang}"
 
     def _generate_examples(self, split_name, file_paths):
-        print(file_paths)
         # Open each file (one for source language and the other for target language)
         files = {k: open(file_paths[k], encoding="utf-8") for k in file_paths}
 

datasets/compguesswhat/create_dummy_data.py

@@ -1,5 +1,6 @@
 import gzip
 import json
+import logging
 import os
 from argparse import ArgumentParser
 
@@ -50,10 +51,10 @@ def create_dummy_data_for_split(data_path, dataset_name, dataset_version, data_f
         os.makedirs(dummy_data_path)
 
     for split_name, split_file in data_files.items():
-        print(f"Generating dummy data for split {split_name} (num. examples = {args.examples})")
+        logging.info(f"Generating dummy data for split {split_name} (num. examples = {args.examples})")
 
         split_filepath = os.path.join(data_path, full_dataset_name, dataset_version, split_file)
-        print(f"Reading split file {split_filepath}")
+        logging.info(f"Reading split file {split_filepath}")
         with gzip.open(split_filepath) as in_file:
             dummy_filepath = os.path.join(dummy_data_path, split_file)
             with gzip.open(dummy_filepath, mode="w") as out_file:
@@ -81,12 +82,12 @@ def main(args):
 
     dataset_version = dataset_info["compguesswhat-original"]["version"]["version_str"]
 
-    print(f"Creating dummy data for CompGuessWhat?! {dataset_version}")
+    logging.info(f"Creating dummy data for CompGuessWhat?! {dataset_version}")
 
-    print("Original dataset...")
+    logging.info("Original dataset...")
     create_dummy_data_for_split(args.data_path, "original", dataset_version, original_data_files)
 
-    print("Zero-shot dataset...")
+    logging.info("Zero-shot dataset...")
     create_dummy_data_for_split(args.data_path, "zero_shot", dataset_version, zs_data_files)
 
 

datasets/oscar/generate_dummy.py

@@ -3,15 +3,15 @@
 
 import fsspec as fs
 import requests
-from oscar import _BASE_CHECKSUM_FILE_NAME, Oscar
+from oscar import _BASE_CHECKSUM_FILE_NAME, Oscar, logger
 
 
 N_EXAMPLES = 2
 
 if __name__ == "__main__":
 
     for i, config in enumerate(Oscar.BUILDER_CONFIGS):
-        print(f"Loading config '{config.name}' ({i + 1}/{len(Oscar.BUILDER_CONFIGS)})")
+        logger.info(f"Loading config '{config.name}' ({i + 1}/{len(Oscar.BUILDER_CONFIGS)})")
 
         # Get data url
         checksum_filename = _BASE_CHECKSUM_FILE_NAME.format(language=config.language)
@@ -42,6 +42,6 @@
         root_dir = str(dummy_data_dir.parent)
         base_name = str(dummy_data_dir)
         base_dir = "dummy_data"
-        print(f"Compressing dummy data folder to '{base_name}.zip'")
+        logger.info(f"Compressing dummy data folder to '{base_name}.zip'")
         shutil.make_archive(base_name, "zip", root_dir, base_dir)
         shutil.rmtree(base_name)

datasets/wiki_lingua/create_dummy.py

@@ -1,4 +1,5 @@
 import itertools
+import logging
 import os
 import pickle
 import shutil
@@ -46,7 +47,7 @@ def create():
     base_path = "/Users/katnoria/dev/projects/workspaces/python/datasets"
     for key in _URLs.keys():
         # data = load_dataset('./datasets/wiki_lingua', key)
-        print(f"Finding {key}.pkl")
+        logging.info(f"Finding {key}.pkl")
         filepath = [name for name in files if name.endswith(f"{key}.pkl")][0]
         with open(filepath, "rb") as f:
             data = pickle.load(f)
@@ -55,13 +56,13 @@ def create():
         fname = sanitize_url(_URLs[key])
         dirname = pjoin(base_path, f"datasets/wiki_lingua/dummy/{key}/1.1.0/dummy_data")
         if not os.path.exists(dirname):
-            print(f"created folder {dirname}")
+            logging.info(f"created folder {dirname}")
             os.makedirs(dirname)
         fname = pjoin(dirname, fname)
-        print(f"creating for {key}:{fname}")
+        logging.info(f"creating for {key}:{fname}")
         with open(fname, "wb") as f:
             pickle.dump(data_subset, f)
-        print("SUCCESS")
+        logging.info("SUCCESS")
 
 
 def zip():
@@ -70,10 +71,10 @@ def zip():
     for key in _URLs.keys():
         # dirname = pjoin(base_path, f"datasets/wiki_lingua/dummy/{key}/1.1.0/dummy_data")
         dirname = pjoin(base_path, f"datasets/wiki_lingua/dummy/{key}/1.1.0")
-        print(f"Zipping {dirname}")
+        logging.info(f"Zipping {dirname}")
         shutil.make_archive(f"{dirname}/dummy_data", "zip", dirname, "dummy_data")
         shutil.rmtree(f"{dirname}/dummy_data")
-        print(f"Deleted folder {dirname}/dummy_data")
+        logging.info(f"Deleted folder {dirname}/dummy_data")
 
 
 # Utility script to create the dummy data and zip the contents

datasets/wmt14/wmt_utils.py

@@ -793,7 +793,6 @@ def _get_filenames(dataset):
             # +++++++++++++++++++++
 
             logger.info("Generating examples from: %s", ss_name)
-            print("Generating examples from: %s", ss_name)
             dataset = DATASET_MAP[ss_name]
             extract_dirs = extraction_map[ss_name]
             files = _get_local_paths(dataset, extract_dirs)

datasets/wmt15/wmt_utils.py

@@ -793,7 +793,6 @@ def _get_filenames(dataset):
             # +++++++++++++++++++++
 
             logger.info("Generating examples from: %s", ss_name)
-            print("Generating examples from: %s", ss_name)
             dataset = DATASET_MAP[ss_name]
             extract_dirs = extraction_map[ss_name]
             files = _get_local_paths(dataset, extract_dirs)

datasets/wmt16/wmt_utils.py

@@ -793,7 +793,6 @@ def _get_filenames(dataset):
             # +++++++++++++++++++++
 
             logger.info("Generating examples from: %s", ss_name)
-            print("Generating examples from: %s", ss_name)
             dataset = DATASET_MAP[ss_name]
             extract_dirs = extraction_map[ss_name]
             files = _get_local_paths(dataset, extract_dirs)

datasets/wmt17/wmt_utils.py

@@ -793,7 +793,6 @@ def _get_filenames(dataset):
             # +++++++++++++++++++++
 
             logger.info("Generating examples from: %s", ss_name)
-            print("Generating examples from: %s", ss_name)
             dataset = DATASET_MAP[ss_name]
             extract_dirs = extraction_map[ss_name]
             files = _get_local_paths(dataset, extract_dirs)

datasets/wmt18/wmt_utils.py

@@ -793,7 +793,6 @@ def _get_filenames(dataset):
             # +++++++++++++++++++++
 
             logger.info("Generating examples from: %s", ss_name)
-            print("Generating examples from: %s", ss_name)
             dataset = DATASET_MAP[ss_name]
             extract_dirs = extraction_map[ss_name]
             files = _get_local_paths(dataset, extract_dirs)

datasets/wmt19/wmt_utils.py

@@ -793,7 +793,6 @@ def _get_filenames(dataset):
             # +++++++++++++++++++++
 
             logger.info("Generating examples from: %s", ss_name)
-            print("Generating examples from: %s", ss_name)
             dataset = DATASET_MAP[ss_name]
             extract_dirs = extraction_map[ss_name]
             files = _get_local_paths(dataset, extract_dirs)

datasets/wmt_t2t/wmt_utils.py

@@ -793,7 +793,6 @@ def _get_filenames(dataset):
             # +++++++++++++++++++++
 
             logger.info("Generating examples from: %s", ss_name)
-            print("Generating examples from: %s", ss_name)
             dataset = DATASET_MAP[ss_name]
             extract_dirs = extraction_map[ss_name]
             files = _get_local_paths(dataset, extract_dirs)

tests/test_dataset_scripts.py

@@ -0,0 +1,64 @@
+import re
+from pathlib import Path
+from unittest import TestCase
+
+
+class TestDatasetScripts(TestCase):
+    def _no_encoding_on_file_open(self, filepath: str):
+        r"""Find all instances where a non-binary file is opened without UTF-8 encoding.
+
+        This function uses regular expressions to find instances where Python's `open()` function is used to open
+        non-binary files. See below for an explanation of the regular expression:
+
+        (?!.*\b(?:encoding|rb|w|wb|w+|wb+|ab|ab+)\b): Lookahead and discard match if `encoding` or `rb` etc are
+        arguments of `open()`.
+
+        (?<=\s): Lookbehind and match if `open()` predeceded by one whitespace.
+
+        (open)\((.*)\): Capture everything in parentheses of `open()`.
+        """
+
+        with open(filepath, "r", encoding="utf-8") as input_file:
+            regexp = re.compile(r"(?!.*\b(?:encoding|rb|w|wb|w+|wb+|ab|ab+)\b)(?<=\s)(open)\((.*)\)")
+            input_text = input_file.read()
+            match = regexp.search(input_text)
+
+        return match
+
+    def _no_print_statements(self, filepath: str):
+        r"""Find all instances where a python sctipt file contains a `print` statement.
+
+        #[^\r\n]*print\(: Match print statement inside a comment. We ignore this group.
+
+        \"[^\r\n]*print\(: Match print statement inside a string. We ignore this group.
+
+        \"\"\".*?print\(.*?\"\"\"": Match print statement inside a triple-quoted string. Uses re.DOTALL to also match newlines with ".".
+        We ignore this group.
+
+        (print\()): Match print statement.
+        """
+
+        with open(filepath, "r", encoding="utf-8") as input_file:
+            regexp = re.compile(r"#[^\r\n]*print\(|\"[^\r\n]*print\(|\"\"\".*?print\(.*?\"\"\"|(print\()", re.DOTALL)
+            input_text = input_file.read()
+            # use `re.finditer` to handle the case where the ignored groups would be matched first by `re.search`
+            matches = regexp.finditer(input_text)
+
+        matches = [match for match in matches if match is not None and match.group(1) is not None]
+        return matches[0] if matches else None
+
+    def test_no_encoding_on_file_open(self):
+        dataset_paths = Path("./datasets")
+        dataset_files = list(dataset_paths.absolute().glob("**/*.py"))
+
+        for dataset in dataset_files:
+            if self._no_encoding_on_file_open(str(dataset)):
+                raise AssertionError(f"open(...) must use utf-8 encoding in {dataset}")
+
+    def test_no_print_statements(self):
+        dataset_paths = Path("./datasets")
+        dataset_files = list(dataset_paths.absolute().glob("**/*.py"))
+
+        for dataset in dataset_files:
+            if self._no_print_statements(str(dataset)):
+                raise AssertionError(f"print statement found in {dataset}. Use datasets.logger/logging instead.")