refactor timestamps managing (#210)

* rename build files test

* add unit tests for audio file check util

* move audio file tests to specific test module

* add datetime template validation function

* add datetime regex builder

* add extract timestamp from filename with integration tests

* add tests ids and check exception messages

* move regex dictionary to module constants

* add docstrings

* add associate_timestamps function

* add strftime to osmose format

* use string interpolation in test result

* backwards compatibility of timestamp.csv labels

* use pandas.to_datetime for strptime operations

* support timezone in strptime

* use new timestamps utils in build method

* remove obsolete test module

* format with black

* rename filename to text in strptime parser util
---------

Co-authored-by: Mathieu Dupont <108517594+mathieudpnt@users.noreply.github.com>

src/OSmOSE/Dataset.py

@@ -19,10 +19,14 @@
 import pandas as pd
 import numpy as np
 from tqdm import tqdm
-from OSmOSE.utils.timestamp_utils import check_epoch
+from OSmOSE.utils.audio_utils import get_all_audio_files
+from OSmOSE.utils.timestamp_utils import (
+    check_epoch,
+    associate_timestamps,
+    strftime_osmose_format,
+)
 from OSmOSE.utils.core_utils import check_n_files, set_umask, chmod_if_needed
 from OSmOSE.utils.path_utils import make_path
-from OSmOSE.timestamps import write_timestamp
 from OSmOSE.config import DPDEFAULT, FPDEFAULT, OSMOSE_PATH, TIMESTAMP_FORMAT_AUDIO_FILE
 
 
@@ -354,20 +358,18 @@ def build(
             if original_folder is not None
             else self._find_or_create_original_folder()
         )
-        path_timestamp_formatted = path_raw_audio.joinpath("timestamp.csv")
 
-        resume_test_anomalies = path_raw_audio.joinpath("resume_test_anomalies.txt")
+        path_timestamp_formatted = path_raw_audio.joinpath("timestamp.csv")
+        user_timestamp = (
+            path_timestamp_formatted.exists()
+        )  # TODO: Formatting audio files beforehand will make this obsolete
 
-        if not path_timestamp_formatted.exists():
-            user_timestamp = False
-            write_timestamp(
-                audio_path=path_raw_audio,
-                date_template=date_template,
-                timezone=self.timezone,
-                verbose=False,
+        if not user_timestamp:
+            self._write_timestamp_csv_from_audio_files(
+                audio_path=path_raw_audio, date_template=date_template
             )
-        else:
-            user_timestamp = True
+
+        resume_test_anomalies = path_raw_audio.joinpath("resume_test_anomalies.txt")
 
         # read the timestamp.csv file
         timestamp_csv = pd.read_csv(path_timestamp_formatted)["timestamp"].values
@@ -655,6 +657,21 @@ def build(
 
             print("\n DONE ! your dataset is on OSmOSE platform !")
 
+    def _write_timestamp_csv_from_audio_files(
+        self, audio_path: Path, date_template: str
+    ):
+        supported_audio_files = [file.name for file in get_all_audio_files(audio_path)]
+        filenames_with_timestamps = associate_timestamps(
+            audio_files=supported_audio_files, datetime_template=date_template
+        )
+        filenames_with_timestamps["timestamp"] = filenames_with_timestamps[
+            "timestamp"
+        ].apply(lambda t: strftime_osmose_format(t))
+        filenames_with_timestamps.to_csv(
+            audio_path / "timestamp.csv", index=False, na_rep="NaN"
+        )
+        os.chmod(audio_path / "timestamp.csv", mode=FPDEFAULT)
+
     def _format_timestamp(
         self,
         cur_timestamp_not_formatted: str,

src/OSmOSE/__init__.py

@@ -1,5 +1,4 @@
 from OSmOSE.Dataset import Dataset
-from OSmOSE.timestamps import write_timestamp
 from OSmOSE.job import Job_builder
 from OSmOSE.Spectrogram import Spectrogram
 import OSmOSE.utils as utils
@@ -8,7 +7,6 @@
 __all__ = [
     "Auxiliary",
     "Dataset",
-    "write_timestamp",
     "Job_builder",
     "Spectrogram",
     "utils",

src/OSmOSE/utils/audio_utils.py

@@ -2,7 +2,7 @@
 from pathlib import Path
 
 
-def is_audio(filename: Path) -> bool:
+def is_supported_audio_format(filename: Path) -> bool:
     """
     Check if a given file is a supported audio file based on its extension.
 
@@ -17,7 +17,7 @@ def is_audio(filename: Path) -> bool:
         True if the file has an extension that matches a supported audio format,
         False otherwise.
     """
-    return filename.suffix in SUPPORTED_AUDIO_FORMAT
+    return filename.suffix.lower() in SUPPORTED_AUDIO_FORMAT
 
 
 def get_all_audio_files(directory: Path) -> list[Path]: