Need help with publish-counts?
Click the “chat” button below for chat support from the developer who created it, or find similar developers for support.

About the developer

199 Stars 37 Forks MIT License 133 Commits 15 Opened issues


Meteor package to help you publish the count of a cursor, in real time

Services available


Need anything else?

Contributors list

Publish Counts

A package to help you publish the count of a cursor, in real time.

Publish-counts is designed for counting a small number of documents around an order of 100. Due to the real-time capability, this package should not be used to count all documents in large datasets. Maybe some, but not all. Otherwise you will maximize your server's CPU usage as each client connects.

Table of Contents


$ meteor add tmeasday:publish-counts


Counts.publish [server]

Counts.publish(subscription, counter-name, cursor, options)

Simply call

within a publication, passing in a name and a cursor:

Example 1

Meteor.publish('publication', function() {
  Counts.publish(this, 'name-of-counter', Posts.find());
Meteor.publish 'publication', ->
  Counts.publish this, 'name-of-counter', Posts.find()
  return undefined    # otherwise coffeescript returns a Counts.publish
                      # handle when Meteor expects a Mongo.Cursor object.


function returns the observer handle that's used to maintain the counter. You can call its
method in order to stop the observer from running.

Warning: Make sure you call collection

separately for
and the
return value, otherwise you'll get empty documents on the client.

For more info regarding the

parameter, see Options.

Counts.get [client]


Once you've subscribed to

(Ex 1), you can call
to get the value of the counter, reactively.

This function will always return an integer,

is returned if the counter is neither published nor subscribed to.

Counts.has [client]


Returns true if a counter is both published and subscribed to, otherwise returns false. This function is reactive.

Useful for validating the existence of counters.

Counts.noWarnings [server]


This function disables all development warnings on the server from publish-counts.

Not recommended for use by development teams, as warnings are meant to inform library users of potential conflicts, inefficiencies, etc in their use of publish-counts as a sanity check. Suppressing all warnings precludes this sanity check for future changes. See the

option for fine-grained warning suppression.



If you publish a count within a publication that also returns cursor(s), you probably want to pass

{noReady: true}
as a final argument to ensure that the "data" publication sets the ready state. For example, the following publication sends down 10 posts, but allows us to see how many there are in total:
Meteor.publish('posts-with-count', function() {
  Counts.publish(this, 'posts', Posts.find(), { noReady: true });
  return Posts.find({}, { limit: 10 });


If you specify

{nonReactive: true}
the cursor won't be observed and only the initial count will be sent on initially subscribing. This is useful in some cases where reactivity is not desired, and can improve performance.


allows you to specify a field to calculate the sum of its numbers across all documents. For example if we were to store page visits as numbers on a field called
{ content: 'testing', visits: 100 },
{ content: 'a comment', visits: 50 }

We could then publish them like:

Meteor.publish('posts-visits-count', function() {
  Counts.publish(this, 'posts-visits', Posts.find(), { countFromField: 'visits' });

And calling


If the counter field is deeply nested, e.g.:

{ content: 'testing', stats: { visits: 100 } },
{ content: 'a comment', stats: { visits: 50 } }

Then use an accessor function instead like:

Meteor.publish('posts-visits-count', function() {
  Counts.publish(this, 'posts-visits',
    Posts.find({}, { fields: { _id: 1, 'stats.visits': 1 }}),
    { countFromField: function (doc) { return doc.stats.visits; } }

Note that when using an accessor function, you must limit the fields fetched if desired, otherwise Counts will fetch entire documents as it updates the count.


allows you to specify a field to calculate the sum of its length across all documents. For example if we were to store the userIds in an array on a field called
{ content: 'testing', likes: ['6PNw4GQKMA8CLprZf', 'HKv4S7xQ52h6KsXQ7'] },
{ content: 'a comment', likes: ['PSmYXrxpwg276aPf5'] }

We could then publish them like:

Meteor.publish('posts-likes-count', function() {
  Counts.publish(this, 'posts-likes', Posts.find(), { countFromFieldLength: 'likes' });

If the counter field is deeply nested, e.g.:

{ content: 'testing', popularity: { likes: ['6PNw4GQKMA8CLprZf', 'HKv4S7xQ52h6KsXQ7'] } },
{ content: 'a comment', popularity: { likes: ['PSmYXrxpwg276aPf5'] } }

Then use an accessor function instead like:

Meteor.publish('posts-likes-count', function() {
  Counts.publish(this, 'posts-likes',
    Posts.find({}, { fields: { _id: 1, 'popularity.likes': 1 }}),
    { countFromFieldLength: function (doc) { return doc.popularity.likes; } }

Note that when using an accessor function, you must limit the fields fetched if desired, otherwise Counts will fetch entire documents as it updates the count.


Pass the option,

noWarnings: true
, to
to disable its warnings in a development environment.

Each call to

may print warnings to the console to inform developers of non-fatal conflicts with publish-counts. In some situations, a developer may intentionally invoke
in a way that generates a warnings. Use this option to disable warnings for a particular invocation of

This fine-grained method of warning suppression is recommended for development teams that rely on warnings with respect to future changes.

Template helpers

To easily show counter values within your templates, use the

template helper.



There are {{getPublishedCount 'posts'}} posts.

{{#if hasPublishedCount 'posts'}} There are {{getPublishedCount 'posts'}} posts. {{else}} The number of posts is loading... {{/if}}


Observer handle leak testing

The package includes a test that checks the number of observer handles opened and closed (to check for memory leaks). You need to run the

branch of
to run it however.

Why doesn't this library count directly in Mongo? or...

Why does my MongoDB connection time-out with large (1000+) datasets?

This package is designed primarily for correctness, not performance. That's why it's aimed at counting smaller datasets and keeping the count instantly up to date.

To achieve perfect correctness in Meteor data layer, we use a database observer to know immediately if a relevant change has occurred. This approach does not necessarily scale to larger datasets, as the observer needs to cache the entire matching dataset (amongst other reasons).

Counting large datasets in this manner is suspected to cause database connections to time out (see #86).

An alternative approach would be to take a .count() of the relevant cursor (or perform an aggregation in more complex use cases), and poll it regularly to update the count. Bulletproof Meteor provides a proof of concept of this approach in their bullet-counter example.

We'd love to see someone publish a package for this use case! If you do end up making such a package, let us know and we'll link it here.

Scalable Count Packages

Compatibility with Meteor < 1.3

Publish-counts 0.8.0 introduces an explicit dependency on the underscore js library which may be incompatible with versions of Meteor below 1.3. Please upgrade Meteor to the latest version or, if you cannot, continue to use publish-counts 0.7.3.

Frequently Asked Questions

More information can be found in the FAQ Section of the issue tracker.


MIT. (c) Percolate Studio

publish-counts was developed as part of the Verso project.

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.