fsspec: normalize checksum() to return string (#6767)

dvc/fs/hdfs.py

@@ -5,7 +5,6 @@ import threading
 
 from funcy import cached_property, wrap_prop
 
-from dvc.hash_info import HashInfo
 from dvc.scheme import Schemes
 from dvc.utils import fix_env
 
@@ -43,11 +42,7 @@ class HDFSFileSystem(CallbackMixin, FSSpecWrapper):
         return HadoopFileSystem(**self.fs_args)
 
     def checksum(self, path_info):
-        return HashInfo(
-            "checksum",
-            self._checksum(path_info),
-            size=self.getsize(path_info),
-        )
+        return self._checksum(path_info)
 
     def _checksum(self, path_info, **kwargs):
         # PyArrow doesn't natively support retrieving the

dvc/fs/webhdfs.py

@@ -2,7 +2,6 @@ import threading
 
 from funcy import cached_property, wrap_prop
 
-from dvc.hash_info import HashInfo
 from dvc.path_info import CloudURLInfo
 from dvc.scheme import Schemes
 
@@ -47,7 +46,4 @@ class WebHDFSFileSystem(CallbackMixin, FSSpecWrapper):
     def checksum(self, path_info):
         path = self._with_bucket(path_info)
         ukey = self.fs.ukey(path)
-
-        return HashInfo(
-            "checksum", ukey["bytes"], size=self.getsize(path_info)
-        )
+        return ukey["bytes"]

dvc/objects/stage.py

@@ -45,18 +45,16 @@ def _get_file_hash(path_info, fs, name):
     info = fs.info(path_info)
     if name in info:
         assert not info[name].endswith(".dir")
-        return HashInfo(name, info[name], size=info["size"])
+        hash_value = info[name]
+    elif hasattr(fs, name):
+        func = getattr(fs, name)
+        hash_value = func(path_info)
+    elif name == "md5":
+        hash_value = file_md5(path_info, fs)
+    else:
+        raise NotImplementedError
 
-    func = getattr(fs, name, None)
-    if func:
-        return func(path_info)
-
-    if name == "md5":
-        return HashInfo(
-            name, file_md5(path_info, fs), size=fs.getsize(path_info)
-        )
-
-    raise NotImplementedError
+    return HashInfo(name, hash_value, size=info["size"])
 
 
 def get_file_hash(path_info, fs, name, state=None):