Re: TouchDB iOS examples with reduce function

[Jens Alfke]

On Nov 8, 2012, at 9:58 AM, Aleksey Dobrushin <alex.do...@gmail.com> wrote:

> It it correct? How do I call it ro get actually rows count?

It looks correct to me. If you query the view, by calling [design queryViewNamed: @“casesByCreateDateCount”], you’ll get back a single row of results, whose value is the count.

—Jens

[Aleksey Dobrushin]

Unfortunately, I receive all documents, as there is no REDUCE block defined...

        CouchQuery* query = [design queryViewNamed: @"casesByCreateDateCount"];

        for (CouchQueryRow* row in query.rows) {

            NSLog(row.documentID);

            id value = row.value;

            NSDictionary* content = [row.document properties];

            for (id key in content) {

                id value = [content objectForKey:key];

                NSLog(@"key: %@, value: %@", key, value);

[Aleksey Dobrushin]

I've played with this last night.

There is a groupLevel property, if I set this property to 1 grouping works fine, and each of the result rows contain different emitted key, the value actually is the result of reduce function. 

So I can use grouping to get count of cases, if I emit for each case always the same key, like this:

[design defineViewNamed: @"casesByCreateDateCount" mapBlock: MAPBLOCK({

        id date = [doc objectForKey: @"createDate"];

        NSString* docType = [doc objectForKey: @"type"];

        if(date && docType && [docType isEqualToString:@"case"])

            emit(@"casesByCreateDateCount", NULL);

    }) reduceBlock: REDUCEBLOCK({

        return [NSNumber numberWithInt:values.count];

    }) version: @"1.0"];

CouchQuery* query = [design queryViewNamed: @"casesByCreateDateCount"];

[query setGroupLevel:1];

for (CouchQueryRow* row in query.rows) {

      id rowCount = row.value;

       NSLog(@"Number of cases: %@", rowCount);

}

I still don't know how good or bad is it in for performance.

I still don't understand, why REDUCE block is ignored if I trying to get count as it described in all NoSQL manuals.