Yes, that was it. It was a little confusing at first, and then I did
look at the couch internal logs and saw that it was a reduce function
paired to a different map function.

Maybe a suggestion for the couch message would be to include the
function name that caused the error.

On Fri, May 7, 2010 at 6:52 PM, Sebastian Cohnen
<sebastiancohnen@googlemail.com> wrote:
> is there any other reduce function in this design document?
>
> On 06.05.2010, at 20:54, Ryan Ramage wrote:
>
>> I have a function like the following (which is 'by the book' in the
>> Finding Your Data With Views chapter)
>>
>> function(doc) {
>>  if(doc.tags.length > 0) {
>>    for(var idx in doc.tags) {
>>      emit(doc.tags[idx], null);
>>    }
>>  }
>> }
>>
>> There is no reduce function.
>> But when I run on a very small data set (hundreds) I get the following error:
>>
>> {"error":"reduce_overflow_error","reason":"Reduce output must shrink
>> more rapidly: Current output:
>> '[[[\"Last_A8Sj9sj\",\"Person_c7114514770d9cf2a28fb5732a1a1c32\",\"Audio_Post_Processing\",\"Calgary\",\"Idea\"'...
>> (first 100 of 452 bytes)"}
>>
>>
>> Each doc has no duplicate tags, and no more than 3 tags. I am running
>> couchdb 0.11 on windows.
>>
>> Any ideas?
>
>