xedocs 0.2.37__tar.gz → 0.2.39__tar.gz

{xedocs-0.2.37 → xedocs-0.2.39}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: xedocs
-Version: 0.2.37
+Version: 0.2.39
 Summary: Top-level package for xedocs.
 Author: Yossi Mosbacher
 Author-email: joe.mosbacher@gmail.com

{xedocs-0.2.37 → xedocs-0.2.39}/pyproject.toml

@@ -1,7 +1,7 @@
 [tool]
 [tool.poetry]
 name = "xedocs"
-version = "0.2.37"
+version = "0.2.39"
 homepage = "https://github.com/XENONnT/xedocs"
 description = "Top-level package for xedocs."
 authors = ["Yossi Mosbacher <joe.mosbacher@gmail.com>"]

{xedocs-0.2.37 → xedocs-0.2.39}/xedocs/__init__.py

@@ -2,7 +2,7 @@
 
 __author__ = """Yossi Mosbacher"""
 __email__ = "joe.mosbacher@gmail.com"
-__version__ = "0.2.37"
+__version__ = "0.2.39"
 
 
 import logging

{xedocs-0.2.37 → xedocs-0.2.39}/xedocs/dispatchers.py

@@ -54,8 +54,8 @@ def read_csv_files(path, **kwargs) -> List[dict]:
     docs =[]
     
     with fsspec.open_files(path, **kwargs) as fs:
-         kwargs = rframe.utils.filter_kwargs(pd.read_csv, kwargs)
-         for f in fs:
+        kwargs = rframe.utils.filter_kwargs(pd.read_csv, kwargs)
+        for f in fs:
             ds = pd.read_csv(f, **kwargs).to_dict(orient='records')
             docs.extend(ds)
     return docs

{xedocs-0.2.37 → xedocs-0.2.39}/xedocs/utils.py

@@ -134,20 +134,38 @@ class LazyFileAccessor(DataAccessor):
     def iter_path_records(self, ignore_paths=(), **labels):
         for path in self.urlpaths:
             glob_patttern = self.format_to_glob(path)
-            fs, _, fpaths = fsspec.get_fs_token_paths(glob_patttern,
-                                                   storage_options=self.storage_options)
-            pattern = path.replace(f"{fs.protocol}://", "")
+            fs, _, fpaths = fsspec.get_fs_token_paths(glob_patttern, storage_options=self.storage_options)
+
+            # Always define original_protocol
+            original_protocol = fs.protocol if fs.protocol else 'file'
+            if isinstance(original_protocol, tuple):
+                original_protocol = original_protocol[0]
+            else:
+                original_protocol = original_protocol
+
+            # Debug prints
+            print(f"\n[DEBUG]")
+            print(f"Processing path: {path}")
+            print(f"fs.protocol: {fs.protocol}")
+            print(f"Original protocol: {original_protocol}")
+            print(f"Glob pattern: {glob_patttern}")
+            print(f"File paths: {fpaths}")
+    
+            pattern = path.replace(f"{original_protocol}://", "")
             pattern = parse.compile(self.glob_to_format(pattern))
+    
+            # More debug prints
+            print(f"Parse pattern: {pattern}")
+    
             loaded = set(ignore_paths)
-
+    
             for fpath in fpaths:
                 if fpath in loaded:
                     continue
                 r = pattern.parse(fpath)
                 if r is None:
                     continue
-                for k,vs in labels.items():
-                    
+                for k, vs in labels.items():
                     if vs is None:
                         continue
                     if not isinstance(vs, list):
@@ -164,10 +182,11 @@ class LazyFileAccessor(DataAccessor):
                     if label not in vs:
                         break
                 else:
-                    records = read_files(fpath, protocol=fs.protocol, **fs.storage_options)
+                    records = read_files(fpath, protocol=original_protocol, **fs.storage_options)
                     yield fpath, records
                     self.loaded.add(fpath)
 
+
     def load_files(self, **labels):
         index_fields = list(self.schema.get_index_fields())
         if len(index_fields) == 1: