<< Back to previous view

[CLJ-270] defn-created fns inherit old metadata from the Var they are assigned to Created: 13/Feb/10  Updated: 24/Aug/10

Status: Open
Project: Clojure
Component/s: None
Affects Version/s: None
Fix Version/s: Approved Backlog

Type: Defect
Reporter: Anonymous Assignee: Rich Hickey
Resolution: Unresolved Votes: 0
Labels: None


 Description   
  • What (small set of) steps will reproduce the problem?

user> (def #^{:foo "bar"} x 5)
#'user/x
user> (meta #'x)
{:ns #<Namespace user>, :name x, :file "NO_SOURCE_FILE", :line 1, :foo "bar"}
user> (defn x [] 5)
#'user/x
user> (meta #'x)
{:ns #<Namespace user>, :name x, :file "NO_SOURCE_FILE", :line 1, :arglists ([])}
user> (meta x)
{:ns #<Namespace user>, :name x, :file "NO_SOURCE_FILE", :line 1, :foo "bar"}

  • What is the expected output? What do you see instead?

I expect (meta #'x) to evaluate to the value of the final (meta x) in the above.

  • What version are you using?

Current master (commit 61202d2ff6925002400a9843e8fbd080f3bef3a5).

  • Was this discussed on the group? If so, please provide a link to the discussion.

http://groups.google.com/group/clojure/browse_thread/thread/4c7151aa9c4d919c/d1b033ef5a13dd89?lnk=gst&q=off-by-one#d1b033ef5a13dd89

Prompted by discussion at

http://groups.google.com/group/clojure/browse_thread/thread/6553d48c981019eb/3c55b0bd43a5d8e9?lnk=gst&q=off-by-one#3c55b0bd43a5d8e9

  • Initial attempt at a diagnosis:

I think this is due to DefExpr's eval method binding the Var to init.eval() first and attaching the supplied metadata to the Var later – see Compiler.java lines 341-352. (Note the Var is always already in place when init.eval() is called, regardless of whether it existed prior to the evaluation of the def / defn.) Thus the init expression supplied by defn sees the old (and wrong) metadata on the Var.

 Comments   
Comment by Assembla Importer [ 24/Aug/10 6:23 AM ]

Converted from http://www.assembla.com/spaces/clojure/tickets/270
Attachments:
dont-copy-val-metadata-onto-new-var-value-in-defn.patch - https://www.assembla.com/spaces/clojure/documents/dwK4yssayr37y_eJe5d-aX/download/dwK4yssayr37y_eJe5d-aX
0001-set-meta-on-vars-before-evaluating-their-init-see-27.patch - https://www.assembla.com/spaces/clojure/documents/arrhbiAI4r35lQeJe5cbLr/download/arrhbiAI4r35lQeJe5cbLr

Comment by Assembla Importer [ 24/Aug/10 6:23 AM ]

stu said: [file:dwK4yssayr37y_eJe5d-aX]

Comment by Assembla Importer [ 24/Aug/10 6:23 AM ]

stu said: The problem happens with defn, but not with def+fn, so I think the original diagnosis is incorrect.

Another stab at diagnosis: The defn macro copies metadata from a var onto its new value, so if you defn a var that already exists, the old var metadata becomes metadata on the new value. I don't know why this would be the right thing to do, and if you eliminate this behavior (see patch) all the Clojure and Contrib tests still pass.

If this is correct, please assign back to me and I will write tests. If this is wrong, please tell me what's going on here so I know how to write the tests.

Comment by Assembla Importer [ 24/Aug/10 6:23 AM ]

cgrand said: Child association with ticket #363 was added

Comment by Assembla Importer [ 24/Aug/10 6:23 AM ]

cgrand said: I think the original diagnosis is correct. Note that the behavior differ when def iscompiled:

user=> (def #^{:foo "bar"} x 5)
#'user/x
user=> (let [] (defn x [] 5))
#'user/x
user=> (meta x)
{:ns #<Namespace user>, :name x, :file "NO_SOURCE_PATH", :line 83, :arglists ([])}
user=> (meta #'x)
{:ns #<Namespace user>, :name x, :file "NO_SOURCE_PATH", :line 83, :arglists ([])}

See also http://groups.google.com/group/clojure/browse_thread/thread/6afd81896ca368b2#

Comment by Assembla Importer [ 24/Aug/10 6:23 AM ]

cgrand said: [file:arrhbiAI4r35lQeJe5cbLr]

Comment by Assembla Importer [ 24/Aug/10 6:23 AM ]

cgrand said: My patch aligns DefExpr.eval with DefExpr.emit (first set meta then eval the init value).

Comment by Assembla Importer [ 24/Aug/10 6:23 AM ]

cgrand said: Forget it, my current patch is broken.

Comment by Assembla Importer [ 24/Aug/10 6:23 AM ]

cgrand said: My current patch is broken because of #352, what is the rational for #352?

Generated at Sat May 18 14:32:16 CDT 2013 using JIRA 4.4#649-r158309.