Details
-
Type:
Defect
-
Status:
Closed
-
Priority:
Minor
-
Resolution: Declined
-
Affects Version/s: Release 1.2
-
Fix Version/s: None
-
Component/s: None
-
Labels:None
-
Environment:Mac OS X Java 6
Description
Very minor, but noticed while browsing that "some" is defined as:
(defn some
...
[pred coll]
(when (seq coll)
(or (pred (first coll)) (recur pred (next coll)))))
The "(seq coll)" call will potentially generate an unused seq each time.
Activity
Fogus
made changes -
| Field | Original Value | New Value |
|---|---|---|
| Assignee | Michel Fogus [ fogus ] |
Fogus
made changes -
| Status | Open [ 1 ] | Closed [ 6 ] |
| Assignee | Michel Fogus [ fogus ] | |
| Resolution | Declined [ 2 ] |
Hide
Matthew Phillips
added a comment -
Thanks for the informative comment Michael. I was just watching Rich's talk on seq's, and remembered making this noob bug report: sorry! 
Show
Matthew Phillips
added a comment - Thanks for the informative comment Michael. I was just watching Rich's talk on seq's, and remembered making this noob bug report: sorry!
The `seq` function is the idiomatic way to ask the question "does this collection have something in it?" and is the correct approach in the case of `some`. If we were to use `seq?` then we would be asking an entirely different question: "does this collection implement the ISeq interface". In the case of using `seq?`, `some` would no longer work against the seq abstraction. It's tempting to look at the name `seq` and `seq?` and assume that they mean similar things. The type-based predicates are tricky.