Added a comment

doc/todo/Retire_Esqueleto_as_a_dependency/comment_2_3a3143db7124fe8790b23fdc820a7eed._comment

@@ -0,0 +1,32 @@
+[[!comment format=mdwn
+ username="seantparsons"
+ avatar="http://cdn.libravatar.org/avatar/616fb81847630239dd1ab099138cb685"
+ subject="comment 2"
+ date="2018-11-06T22:50:25Z"
+ content="""
+I've pushed the fix needed to my branch: [[https://github.com/seanparsons/git-annex/tree/remove-esqueleto]]
+
+    This seems certianly wrong; it's lost the NOT.
+
+Well spotted, I even remember double checking the not equal syntax, that should be fixed now.
+
+    Is there any way to verify the same or equivilant SQL gets generated?
+
+I don't believe so, but I'm not an expert in either project I mostly just turned the \"fix compile errors until there are none\" crank.
+
+    Not clear from persistent's documentation if deleteWhere with two Filters combines them with AND or OR?
+
+From the docs relating to the `Filter` type ([[http://hackage.haskell.org/package/persistent-2.9.0/docs/Database-Persist.html#v:-124--124-.]]):
+
+    If you are looking for an (&&.) operator to do (A AND B AND (C OR D)) you can use the (++) operator instead as there is no (&&.).
+
+That seems pretty conclusive multiple entries in a list are combined with `FilterAnd`.
+
+    Do we know that an empty Filter list matches everything as used below instead of nothing? Not clear to me from the documentation.
+
+Yup, the type hint indicates that it relates to the table: https://www.yesodweb.com/book/persistent#persistent_delete
+
+    it would be good to expand the test suite to test the SQL code in isolation
+
+I can probably look at adding some more tests this weekend.
+"""]]