Closed alexbiehl closed 10 years ago
You are right. Thanks for taking a look at the project.
I set it to maxBound for now. This should fix the problem for now. I'm not sure if the need of splitting batches is still necessary at all.
Also note: Hunt is intended to be run on a 64bit machine. I am not sure how stable it might run on a 32Bit machine.
Ah yes, thought so too, since the index is stored in-memory 32bit machines could only handle max 4gbs anyway.
(I am on GHC 7.8.2, 32 Bit)
When running the Quickstart examples I noticed when executing the document inserts hunt runs forever with:
All but the first three (equals example doc count)
InsertList
commands interpreted byexecCmd
had an empty list in them.The glitch seems to be in
toBasicCommand
. More precisely in line 265:Reducing the number to something in the range of integers like 2147483647 (or anything below MAX_INT) worked fine. It's probably because I'am on a 32bit machine.
Edit: splitEvery cannot handle negative numbers. (30000000000 becomes -2147483647)