try to figure out collation in couchdb

collation.py

from couchdb.client import Server

server = Server()
db = server['collation-test']
try:
    for idx in range(65535):
        db.save( {'unichr':unichr(idx), 'code': idx, 'hex': str(hex(idx))} )
        print "{0}\n".format(hex(idx))
except Exception, e:
    raise e

map.js

function(doc) {
  log(doc.code + " hex:"+doc.hex);
  emit(doc.unichr, doc.hex);
}

bulk_update works faster(;

docs = []
for idx in range(65535):
    docs.append({'unichr':unichr(idx), 'code': idx, 'hex': str(hex(idx))})
db.update(docs)

Not really the point... it seems there's just a bug in the indexer this map exposes. If I had done the bulk update, I wouldn't have known if the insert failed.

…

On Feb 27, 2012, at 6:50 PM, kxepal wrote: bulk_update works faster(; docs = [] for idx in range(65535): docs.append({'unichr':unichr(idx), 'code': idx, 'hex': str(hex(idx))}) db.update(docs) --- Reply to this email directly or view it on GitHub: https://gist.github.com/1904807

It's easy to handle insert failure:

docs = []
for idx in range(65535):
    docs.append({'unichr':unichr(idx), 'code': idx, 'hex': str(hex(idx))})
for success, docid, rev_or_exc in db.update(docs, all_or_nothing=True):
    if success:
        print docid, rev_or_exc
    else:
        raise rev_or_exc