Re: [Chicken-janitors] #905: Unreliable behavior of hash-table-copy with symbols as keys (regression wrt 4.7.4)

[Chicken Trac]

#905: Unreliable behavior of hash-table-copy with symbols as keys (regression 
wrt
4.7.4)
-----------------------------+----------------------------------------------
  Reporter:  iraikov         |       Owner:  sjamaan 
      Type:  defect          |      Status:  assigned
  Priority:  critical        |   Milestone:  4.8.0   
 Component:  core libraries  |     Version:  4.8.x   
Resolution:                  |    Keywords:          
-----------------------------+----------------------------------------------

Comment(by sjamaan):

 That isn't possible, but I've posted the patch proposed by Mario which
 still creates a hash table the regular way, but then overwrites the newly
 randomized hashing function with the old one from the original hash table.
 This is the best we can do without adding yet another slot with a
 randomization value (possibly breaking the internal "API" yet again).

 I think it's a good enough solution. We might consider rewriting {{{hash-
 table-copy}}} so it creates the vector directly instead of calling
 {{{*make-hash-table}}} and then overwriting many values.

 Ivan: Could you give this a try?  The simplified test Felix posted
 succeeds for me with this patch.

-- 
Ticket URL: <http://bugs.call-cc.org/ticket/905#comment:14>
Chicken Scheme <http://www.call-with-current-continuation.org/>
Chicken Scheme is a compiler for the Scheme programming language.