So there *is* Tree.get_file_mtime() and for RevisionTrees it is defined as:
def get_file_mtime(self, file_id, path=None):
ie = self._inventory[file_id]
revision = self._repository.get_revision(ie.revision)
return revision.timestamp
^- Notice that this deserializes a new Revision object for every request, and does *no* caching. So for 10,000 files in a tree, it does 10,000 deserializations.
So there *is* Tree.get_ file_mtime( ) and for RevisionTrees it is defined as: mtime(self, file_id, path=None): [file_id] y.get_revision( ie.revision)
def get_file_
ie = self._inventory
revision = self._repositor
return revision.timestamp
^- Notice that this deserializes a new Revision object for every request, and does *no* caching. So for 10,000 files in a tree, it does 10,000 deserializations.
Which is probably why we don't use it at present.