_cachekey.py _artifactelement.py: Create new is_key() and use it

To more cleanly check if a given string could be a cache key, create
a new _cachekey.is_key() function and use that in _artifactelement's
verify_artifact_ref() function.

Signed-off-by: Daniel Silverstone <daniel.silverstone@codethink.co.uk>

2 files changed, 27 insertions, 2 deletions

diff --git a/buildstream/_artifactelement.py b/buildstream/_artifactelement.py
index 8210b4d97..d65d46173 100644
--- a/buildstream/_artifactelement.py
+++ b/buildstream/_artifactelement.py
@@ -83,8 +83,8 @@ class ArtifactElement(Element):
 def verify_artifact_ref(ref):
     try:
         project, element, key = ref.split('/', 2)  # This will raise a Value error if unable to split
-        # Explicitly raise a ValueError if the key lenght is not as expected
-        if len(key) != len(_cachekey.generate_key(_yaml.new_empty_node())):
+        # Explicitly raise a ValueError if the key length is not as expected
+        if not _cachekey.is_key(key):
             raise ValueError
     except ValueError:
         raise ArtifactElementError("Artifact: {} is not of the expected format".format(ref))
diff --git a/buildstream/_cachekey.py b/buildstream/_cachekey.py
index 953263aa9..e56b582fa 100644
--- a/buildstream/_cachekey.py
+++ b/buildstream/_cachekey.py
@@ -24,6 +24,31 @@ import ujson
 
 from . import _yaml
 
+# Internal record of the size of a cache key
+_CACHEKEY_SIZE = len(hashlib.sha256().hexdigest())
+
+
+# Hex digits
+_HEX_DIGITS = "0123456789abcdef"
+
+
+# is_key()
+#
+# Check if the passed in string *could be* a cache key.  This basically checks
+# that the length matches a sha256 hex digest, and that the string does not
+# contain any non-hex characters and is fully lower case.
+#
+# Args:
+#    key (str): The string to check
+#
+# Returns:
+#    (bool): Whether or not `key` could be a cache key
+#
+def is_key(key):
+    if len(key) != _CACHEKEY_SIZE:
+        return False
+    return not any(ch not in _HEX_DIGITS for ch in key)
+
 
 # generate_key()
 #