From fcc4c13169cd81baa9a6205b9f895d5d9c5c65c5 Mon Sep 17 00:00:00 2001
From: David Hassell <david.hassell@ncas.ac.uk>
Date: Mon, 28 Mar 2022 15:47:25 +0100
Subject: [PATCH 1/3] get_filenames

---
 cf/data/__init__.py  |  2 ++
 cf/data/data.py      | 45 +++++++++++++++++++++++++-------------------
 cf/test/test_Data.py |  3 +++
 3 files changed, 31 insertions(+), 19 deletions(-)

diff --git a/cf/data/__init__.py b/cf/data/__init__.py
index 6a28aa2c46..4d80e10a5e 100644
--- a/cf/data/__init__.py
+++ b/cf/data/__init__.py
@@ -1,3 +1,5 @@
+from .abstract import FileArray
+
 from .cachedarray import CachedArray
 from .netcdfarray import NetCDFArray
 from .umarray import UMArray
diff --git a/cf/data/data.py b/cf/data/data.py
index c93d48f7d4..8bece7e1fa 100644
--- a/cf/data/data.py
+++ b/cf/data/data.py
@@ -47,6 +47,7 @@
 from ..mixin_container import Container
 from ..units import Units
 from . import (  # GatheredSubarray,; RaggedContiguousSubarray,; RaggedIndexedContiguousSubarray,; RaggedIndexedSubarray,
+    FileArray,
     NetCDFArray,
     UMArray,
 )
@@ -9466,37 +9467,43 @@ def insert_dimension(self, position=0, inplace=False):
 
         return d
 
+    @daskified(_DASKIFIED_VERBOSE)
     def get_filenames(self):
         """Return the names of files containing parts of the data array.
 
         :Returns:
 
             `set`
-                The file names in normalized, absolute form. If the data
-                is are memory then an empty `set` is returned.
+                The file names in normalized, absolute form. If the
+                data is are memory then an empty `set` is returned.
 
-        **Examples:**
+        **Examples**
+
+        >>> f = cf.NetCDFArray(TODODASK)
+        >>> d = cf.Data(f)
+        >>> d.get_filenames()
+        {TODODASK}
 
-        >>> f = cf.read('../file[123]')[0]
-        >>> f.get_filenames()
-        {'/data/user/file1',
-         '/data/user/file2',
-         '/data/user/file3'}
-        >>> a = f.array
-        >>> f.get_filenames()
+        >>> d = cf.Data([1, 2, 3])
+        >>> d.get_filenames()
         set()
 
         """
-        print("TODODASK - is this still possible?")
-        out = set(
-            [
-                abspath(p.subarray.get_filename())
-                for p in self.partitions.matrix.flat
-                if p.in_file
-            ]
-        )
-        out.discard(None)
+        out = set()
 
+        dx = self._get_dask()
+        hlg = dx.dask
+        dsk = hlg.to_dict()
+        for key, value in hlg.get_all_dependencies().items():
+            if value:
+                continue
+
+            # This key has no dependencies, and so is raw data.
+            a = dsk[key]
+            if isinstance(a, FileArray):
+                out.add(abspath(a.get_filename()))
+
+        out.discard(None)
         return out
 
     @daskified(_DASKIFIED_VERBOSE)
diff --git a/cf/test/test_Data.py b/cf/test/test_Data.py
index 7e991b8f68..90994d321a 100644
--- a/cf/test/test_Data.py
+++ b/cf/test/test_Data.py
@@ -3927,6 +3927,9 @@ def test_Data_set_units(self):
         with self.assertRaises(ValueError):
             d.set_units("km")
 
+    def test_Data_get_filenames(self):
+        raise ValueError("need updated NetCDFArray to test")
+
 
 if __name__ == "__main__":
     print("Run date:", datetime.datetime.now())

From 2d960c74835fa270b9852b37c5fbf2787d8c3cb6 Mon Sep 17 00:00:00 2001
From: David Hassell <davidhassell@users.noreply.github.com>
Date: Fri, 1 Apr 2022 15:51:22 +0100
Subject: [PATCH 2/3] Typo

Co-authored-by: Sadie L. Bartholomew <sadie.bartholomew@ncas.ac.uk>
---
 cf/data/data.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/cf/data/data.py b/cf/data/data.py
index 8bece7e1fa..9f011b74be 100644
--- a/cf/data/data.py
+++ b/cf/data/data.py
@@ -9475,7 +9475,7 @@ def get_filenames(self):
 
             `set`
                 The file names in normalized, absolute form. If the
-                data is are memory then an empty `set` is returned.
+                data is in memory then an empty `set` is returned.
 
         **Examples**
 

From bb8739ceb41cb5b8371df260ab912b48a712b055 Mon Sep 17 00:00:00 2001
From: David Hassell <david.hassell@ncas.ac.uk>
Date: Fri, 1 Apr 2022 15:55:01 +0100
Subject: [PATCH 3/3] skip get_filenames test

---
 cf/test/test_Data.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/cf/test/test_Data.py b/cf/test/test_Data.py
index 90994d321a..a3d873690f 100644
--- a/cf/test/test_Data.py
+++ b/cf/test/test_Data.py
@@ -3927,8 +3927,9 @@ def test_Data_set_units(self):
         with self.assertRaises(ValueError):
             d.set_units("km")
 
+    @unittest.skipIf(TEST_DASKIFIED_ONLY, "Needs updated NetCDFArray to test")
     def test_Data_get_filenames(self):
-        raise ValueError("need updated NetCDFArray to test")
+        pass
 
 
 if __name__ == "__main__":