make --recordings accept paths to recordings even if they do not exist locally

Author: lucasgautheron

ChildProject/pipelines/processors.py

@@ -78,7 +78,9 @@ def process_recording(self, recording):
         pass
 
     def process(self, parameters):
-        recordings = self.project.get_recordings_from_list(self.recordings)
+        recordings = self.project.get_recordings_from_list(
+            self.recordings, self.input_profile
+        )
 
         os.makedirs(name=self.output_directory(), exist_ok=True)
 
@@ -94,10 +96,9 @@ def process(self, parameters):
 
         if not len(self.converted):
             return
-        
+
         self.converted.set_index(
-            ["original_filename", "converted_filename"],
-            inplace = True
+            ["original_filename", "converted_filename"], inplace=True
         )
         self.converted = self.converted.assign(parameters=parameters)
 
@@ -207,7 +208,7 @@ def process_recording(self, recording):
         success = proc.returncode == 0
 
         if not success:
-            print(stderr, file = sys.stderr)
+            print(stderr, file=sys.stderr)
 
             return pd.DataFrame(
                 [

ChildProject/projects.py

@@ -1,4 +1,5 @@
 import datetime
+from functools import partial
 import glob
 import numpy as np
 import os
@@ -7,7 +8,7 @@
 import subprocess
 
 from .tables import IndexTable, IndexColumn, is_boolean
-from .utils import get_audio_duration
+from .utils import get_audio_duration, path_is_parent
 
 
 class ChildProject:
@@ -384,7 +385,10 @@ def validate(self, ignore_files: bool = False) -> tuple:
                         self.warnings.append(message)
 
             # child id refers to an existing child in the children table
-            if str(row["child_id"]) not in self.children["child_id"].astype(str).tolist():
+            if (
+                str(row["child_id"])
+                not in self.children["child_id"].astype(str).tolist()
+            ):
                 self.errors.append(
                     "child_id '{}' in recordings on line {} cannot be found in the children table.".format(
                         row["child_id"], index
@@ -483,19 +487,41 @@ def get_converted_recording_filename(
             self.converted_recordings_hashtable[key] = None
             return None
 
-    def get_recordings_from_list(self, recordings: list):
+    def recording_from_path(self, path: str, profile: str = None) -> str:
+        if profile:
+            media_path = os.path.join(self.path, self.CONVERTED_RECORDINGS, profile)
+        else:
+            media_path = os.path.join(self.path, self.RAW_RECORDINGS)
+
+        if not path_is_parent(media_path, path):
+            return None
+
+        recording = os.path.relpath(
+            path, media_path
+        )
+
+        return recording
+        
+
+    def get_recordings_from_list(self, recordings: list, profile: str = None) -> pd.DataFrame:
+        """Recover recordings metadata from a list of recordings or path to recordings.
+
+        :param recordings: list of recording names or paths
+        :type recordings: list
+        :return: matching recordings
+        :rtype: pd.DataFrame
+        """
         _recordings = self.recordings.copy()
 
         if recordings is not None:
             # if the user provided paths,
             # transform those paths into recording_filename values
-            if all(map(os.path.exists, recordings)):
-                recordings = [
-                    os.path.relpath(
-                        recording, os.path.join(self.path, self.RAW_RECORDINGS)
-                    )
-                    for recording in recordings
-                ]
+            recordings_from_paths = [
+                self.recording_from_path(recording, profile) for recording in recordings
+            ]
+
+            if None not in recordings_from_paths:
+                recordings = recordings_from_paths
 
             _recordings = _recordings[
                 _recordings["recording_filename"].isin(recordings)

ChildProject/utils.py

@@ -1,5 +1,15 @@
 import os
 
+def path_is_parent(parent_path: str, child_path: str):
+    # Smooth out relative path names, note: if you are concerned about symbolic links, you should use os.path.realpath too
+    parent_path = os.path.abspath(parent_path)
+    child_path = os.path.abspath(child_path)
+
+    # Compare the common path of the parent and child path with the common path of just the parent path. Using the commonpath method on just the parent path will regularise the path name in the same way as the comparison that deals with both paths, removing any trailing path separator
+    return os.path.commonpath([parent_path]) == os.path.commonpath(
+        [parent_path, child_path]
+    )
+
 
 class Segment:
     def __init__(self, start, stop):