Prevent returning cached entry if the entry is degenerate

fsspec/implementations/dbfs.py

@@ -1,9 +1,11 @@
+from __future__ import annotations
+
 import base64
 import urllib
 
 import requests
-import requests.exceptions
 from requests.adapters import HTTPAdapter, Retry
+from typing_extensions import override
 
 from fsspec import AbstractFileSystem
 from fsspec.spec import AbstractBufferedFile
@@ -57,6 +59,24 @@ def __init__(self, instance, token, **kwargs):
 
         super().__init__(**kwargs)
 
+    @override
+    def _ls_from_cache(self, path) -> list[dict[str, str | int]] | None:
+        """Check cache for listing
+
+        Returns listing, if found (may be empty list for a directory that
+        exists but contains nothing), None if not in cache.
+        """
+        self.dircache.pop(path.rstrip("/"), None)
+
+        parent = self._parent(path)
+        if parent in self.dircache:
+            for entry in self.dircache[parent]:
+                if entry["name"] == path.rstrip("/"):
+                    if entry["type"] != "directory":
+                        return [entry]
+                    return []
+            raise FileNotFoundError(path)
+
     def ls(self, path, detail=True, **kwargs):
         """
         List the contents of the given path.
@@ -70,7 +90,15 @@ def ls(self, path, detail=True, **kwargs):
             but also additional information on file sizes
             and types.
         """
-        out = self._ls_from_cache(path)
+        try:
+            out = self._ls_from_cache(path)
+        except FileNotFoundError:
+            # This happens if the `path`'s parent was cached, but `path` is not
+            # there. This suggests that `path` is new since the parent was
+            # cached. Attempt to invalidate parent's cache before continuing.
+            self.dircache.pop(self._parent(path), None)
+            out = None
+
         if not out:
             try:
                 r = self._send_to_api(
@@ -460,7 +488,7 @@ def _fetch_range(self, start, end):
         return return_buffer
 
     def _to_sized_blocks(self, length, start=0):
-        """Helper function to split a range from 0 to total_length into bloksizes"""
+        """Helper function to split a range from 0 to total_length into blocksizes"""
         end = start + length
         for data_chunk in range(start, end, self.blocksize):
             data_start = data_chunk

fsspec/implementations/tests/cassettes/test_dbfs/test_dbfs_listing.yaml

@@ -13,7 +13,7 @@ interactions:
       Content-Type:
       - application/json
       User-Agent:
-      - python-requests/2.32.3
+      - python-requests/2.32.4
       authorization:
       - DUMMY
     method: POST
@@ -44,7 +44,7 @@ interactions:
       x-content-type-options:
       - nosniff
       x-request-id:
-      - e71bb75f-b05a-4523-b8bc-ce331812fbf0
+      - 8b8f96eb-6260-476f-9bee-5579fe95ce97
     status:
       code: 200
       message: OK
@@ -62,17 +62,17 @@ interactions:
       Content-Type:
       - application/json
       User-Agent:
-      - python-requests/2.32.3
+      - python-requests/2.32.4
       authorization:
       - DUMMY
     method: GET
     uri: https://my_instance.com/api/2.0/dbfs/list
   response:
     body:
       string: !!binary |
-        H4sIAAAAAAAEAxzMMQ7CMAwF0Lv8OVKYcwAuwIiQ1SaOMKI4ss3SqncvZX3D29DlzY5y3zCmeKIg
-        X39yCzXO3X1wpTZ3z8P0xTXIVAMJ4tTEUMK+nP4JuayMcklYtEmXOoXoh0KWU/fHfgAAAP//AwCB
-        uORhbAAAAA==
+        H4sIAAAAAAAEAxzMOw7CMAwA0Lt4jhQX8VMOwAUYURWVxBFGFEe2Wah6dyrWN7wFGr/IIN0W6JM/
+        IEG8bHJ1UYrNrFPJ9d4sdpUnFc8q4hCALVdWSK4fCv8kG38JEgaYpXLjMjnLOzvPmw6nw2444x6P
+        iLiO6w8AAP//AwBhkoZWeAAAAA==
     headers:
       access-control-allow-headers:
       - Authorization, X-Databricks-Azure-Workspace-Resource-Id, X-Databricks-Org-Id,
@@ -100,7 +100,7 @@ interactions:
       x-content-type-options:
       - nosniff
       x-request-id:
-      - 6e66514c-c883-49c7-b215-2fecc5ef8222
+      - 4dab76b2-a193-45d7-8000-642f1c003aa6
     status:
       code: 200
       message: OK
@@ -118,7 +118,7 @@ interactions:
       Content-Type:
       - application/json
       User-Agent:
-      - python-requests/2.32.3
+      - python-requests/2.32.4
       authorization:
       - DUMMY
     method: GET
@@ -149,7 +149,7 @@ interactions:
       x-content-type-options:
       - nosniff
       x-request-id:
-      - 59b8c763-be19-4402-8d37-135a7d6c7aed
+      - 9a476bcf-3900-47ca-b21d-412419e70a11
     status:
       code: 200
       message: OK
@@ -167,19 +167,19 @@ interactions:
       Content-Type:
       - application/json
       User-Agent:
-      - python-requests/2.32.3
+      - python-requests/2.32.4
       authorization:
       - DUMMY
     method: GET
     uri: https://my_instance.com/api/2.0/dbfs/list
   response:
     body:
       string: !!binary |
-        H4sIAAAAAAAEAzyOwWrDMBBEf0XoXNtx4hqSU6B1IZcY4hQKpQhFWrkqtlfdVUtDyL9XbqCnZdh5
-        M3ORQISkDFqQG3louvb58NCox7bp1L49quZl1x3lnRyBWfezZ4/C+QEEkrCewESks4Afz5EFTiLo
-        +C6Kp+To0gcKxxzAKHtyXATCjwQoQozF6Jn91Kv/kDzVWIjaDyw3rxe5jecwF84n7xH7AXTwnBsc
-        i5vMKZj8AJ9fwHE3OUwBdFPK20TWTjuoa8jK2q6yamGq7ATrZbYuYelW9aLU91ViGOj7b4mOOlHy
-        +nb9BQAA//8DADr+FHAYAQAA
+        H4sIAAAAAAAEAzyOUUvEMBCE/0rIs20VC7nek3BWuJcrtCcIIiFNtjXSduNuFI/j/rupBz4tw843
+        M2cJREjaogO5lW3dNc/trtaPTd3pQ3PU9cu+O8obOQOzGVfPAcXgJxBIwnkCG5FOAn48Rxa4iGDi
+        uyiekqNLHygG5gBWu37gIhB+JEATYixmz+yXUf+H5KnGQTR+Yrl9PcuHeApr4XryEXGcwATPucW5
+        uMqcgs1b+PwCjvtlwBRAV6W9S2R/V92qsiwzZXqTlRu1yYy5d1mvKuhLU1m1UYlhoO+/JSaaRMnL
+        2+UXAAD//wMA9nt9bhgBAAA=
     headers:
       access-control-allow-headers:
       - Authorization, X-Databricks-Azure-Workspace-Resource-Id, X-Databricks-Org-Id,
@@ -207,7 +207,7 @@ interactions:
       x-content-type-options:
       - nosniff
       x-request-id:
-      - 6fafe66e-16d3-40c4-be92-91e2f3601a54
+      - b1907444-7aba-4878-aa3d-b79eb4a9c787
     status:
       code: 404
       message: Not Found