CHANGELOG.md

@@ -13,163 +13,170 @@ project adheres to [Semantic Versioning](http://semver.org/).
 
 ### Added
 
+## [11.1.2] - 2018-09-19
+
+### Changed
+
+- Allow setting Gauge values to NaN, +Inf, and -Inf
+- Fixed `histogram` scrape performance by using `acc.push` instead of `acc.concat`. Fixes #216 with #219
+
 ## [11.1.1] - 2018-06-29
 
 ### Changed
 
-* Fixed `processOpenFileDescriptors` metric when no custom config was set
+- Fixed `processOpenFileDescriptors` metric when no custom config was set
 
 ## [11.1.0] - 2018-06-29
 
-* Added ability to set a name prefix in the default metrics
+- Added ability to set a name prefix in the default metrics
 
 ### Changed
 
-* Fixed `startTimer` utility to not mutate objects passed as `startLabels`
-* Fixed `Counter` to validate labels parameter of `inc()` against initial
+- Fixed `startTimer` utility to not mutate objects passed as `startLabels`
+- Fixed `Counter` to validate labels parameter of `inc()` against initial
   labelset
-* Fixed `AggregatorFactory` losing the aggregator method of metrics
+- Fixed `AggregatorFactory` losing the aggregator method of metrics
 
 ## [11.0.0] - 2018-03-10
 
 ### Breaking
 
-* Fixed `gauge.setToCurrentTime()` to use seconds instead of milliseconds
-  * This conforms to Prometheus
+- Fixed `gauge.setToCurrentTime()` to use seconds instead of milliseconds
+  - This conforms to Prometheus
     [best practices](https://prometheus.io/docs/practices/naming/#base-units)
-* Dropped support for node 4
+- Dropped support for node 4
 
 ## [10.2.3] - 2018-02-28
 
 ### Breaking
 
 ### Changed
 
-* Fixed issue that `registry.getMetricsAsJSON()` ignores registry default labels
+- Fixed issue that `registry.getMetricsAsJSON()` ignores registry default labels
 
 ### Added
 
 ## [10.2.2] - 2017-11-02
 
 ### Changed
 
-* Fixed invalid `process_virtual_memory_bytes` reported under linux
+- Fixed invalid `process_virtual_memory_bytes` reported under linux
 
 ## [10.2.1] - 2017-10-27
 
 ### Changed
 
-* Only resolve/reject `clusterMetrics` promise if no callback is provided
+- Only resolve/reject `clusterMetrics` promise if no callback is provided
 
 ## [10.2.0] - 2017-10-16
 
 ### Changed
 
-* Don't add event listeners if cluster module is not used.
-* Fixed issue with counters having extra records when using empty labels
+- Don't add event listeners if cluster module is not used.
+- Fixed issue with counters having extra records when using empty labels
 
 ### Added
 
-* Added `reset` to Counter and Gauge
-* Added `resetMetrics` to register to calling `reset` of all metric instances
+- Added `reset` to Counter and Gauge
+- Added `resetMetrics` to register to calling `reset` of all metric instances
 
 ## [10.1.1] - 2017-09-26
 
 ### Changed
 
-* Update TypeScript definitions and JSDoc comments to match JavaScript sources
-* Fix lexical scope of `arguments` in cluster code
+- Update TypeScript definitions and JSDoc comments to match JavaScript sources
+- Fix lexical scope of `arguments` in cluster code
 
 ## [10.1.0] - 2017-09-04
 
 ### Added
 
-* Support aggregating metrics across workers in a Node.js cluster.
+- Support aggregating metrics across workers in a Node.js cluster.
 
 ## [10.0.4] - 2017-08-22
 
 ### Changed
 
-* Include invalid values in the error messages
+- Include invalid values in the error messages
 
 ## [10.0.3] - 2017-08-07
 
 ### Added
 
-* Added registerMetric to definitions file
+- Added registerMetric to definitions file
 
 ### Changed
 
-* Fixed typing of DefaultMetricsCollectorConfiguration in definitions file
-* Don't pass timestamps through to pushgateway by default
+- Fixed typing of DefaultMetricsCollectorConfiguration in definitions file
+- Don't pass timestamps through to pushgateway by default
 
 ## [10.0.2] - 2017-07-07
 
 ### Changed
 
-* Don't poll default metrics every single tick
+- Don't poll default metrics every single tick
 
 ## [10.0.1] - 2017-07-06
 
 ### Added
 
-* Metrics should be initialized to 0 when there are no labels
+- Metrics should be initialized to 0 when there are no labels
 
 ## [10.0.0] - 2017-07-04
 
 ### Breaking
 
-* Print deprecation warning when metrics are constructed using non-objects
-* Print deprecation warning when `collectDefaultMetrics` is called with a number
+- Print deprecation warning when metrics are constructed using non-objects
+- Print deprecation warning when `collectDefaultMetrics` is called with a number
 
 ### Added
 
-* Ability to set default labels by registry
-* Allow passing in `registry` as second argument to `collectDefaultMetrics` to
+- Ability to set default labels by registry
+- Allow passing in `registry` as second argument to `collectDefaultMetrics` to
   use that instead of the default registry
 
 ### Changed
 
-* Convert code base to ES2015 code (node 4)
-  * add engines field to package.json
-  * Use object shorthand
-  * Remove `util-extend` in favor of `Object.assign`
-  * Arrow functions over binding or putting `this` in a variable
-  * Use template strings
-  * `prototype` -> `class`
+- Convert code base to ES2015 code (node 4)
+  - add engines field to package.json
+  - Use object shorthand
+  - Remove `util-extend` in favor of `Object.assign`
+  - Arrow functions over binding or putting `this` in a variable
+  - Use template strings
+  - `prototype` -> `class`
 
 ## [9.1.1] - 2017-06-17
 
 ### Changed
 
-* Don't set timestamps for metrics that are never updated
+- Don't set timestamps for metrics that are never updated
 
 ## [9.1.0] - 2017-06-07
 
 ### Added
 
-* Ability to merge registries
+- Ability to merge registries
 
 ### Changed
 
-* Correct typedefs for object constructor of metrics
+- Correct typedefs for object constructor of metrics
 
 ## [9.0.0] - 2017-05-06
 
 ### Added
 
-* Support for multiple registers
-* Support for object literals in metric constructors
-* Timestamp support
+- Support for multiple registers
+- Support for object literals in metric constructors
+- Timestamp support
 
 ### Changed
 
-* Collection of default metrics is now disabled by default. Start collection by
+- Collection of default metrics is now disabled by default. Start collection by
   running `collectDefaultMetrics()`.
 
 ### Deprecated
 
-* Creating metrics with one argument per parameter - use object literals
+- Creating metrics with one argument per parameter - use object literals
   instead.
 
 [unreleased]: https://github.com/siimon/prom-client/compare/v10.2.2...HEAD

README.md

@@ -3,13 +3,13 @@
 A prometheus client for node.js that supports histogram, summaries, gauges and
 counters.
 
-### Usage
+## Usage
 
 See example folder for a sample usage. The library does not bundle any web
 framework, to expose the metrics just return the `metrics()` function in the
 registry.
 
-#### Usage with Node.js's `cluster` module
+### Usage with Node.js's `cluster` module
 
 Node.js's `cluster` module spawns multiple processes and hands off socket
 connections to those workers. Returning metrics from a worker's local registry
@@ -32,13 +32,13 @@ registry, call
 `client.AggregatorRegistry.setRegistries(registryOrArrayOfRegistries)` from the
 worker processes.
 
-### API
+## API
 
-#### Configuration
+### Configuration
 
 All metric types has 2 mandatory parameters, name and help.
 
-#### Default metrics
+### Default metrics
 
 There are some default metrics recommended by Prometheus
 [itself](https://prometheus.io/docs/instrumenting/writing_clientlibs/#standard-and-runtime-collectors).
@@ -114,7 +114,7 @@ NOTE: `unref` is called on the `interval` internally, so it will not keep your
 node process going indefinitely if it's the only thing keeping it from shutting
 down.
 
-##### Stop polling default metrics
+#### Stop polling default metrics
 
 To stop collecting the default metrics, you have to call the function and pass
 it to `clearInterval`.
@@ -128,7 +128,7 @@ clearInterval(client.collectDefaultMetrics());
 client.register.clear();
 ```
 
-#### Counter
+### Counter
 
 Counters go up, and reset when the process restarts.
 
@@ -150,7 +150,7 @@ counter.reset();
 counter.inc(); // Inc with 1 starting from 0
 ```
 
-#### Gauge
+### Gauge
 
 Gauges are similar to Counters but Gauges value can be decreased.
 
@@ -183,7 +183,7 @@ xhrRequest(function(err, res) {
 });
 ```
 
-#### Histogram
+### Histogram
 
 Histograms track sizes and frequency of events.
 
@@ -240,7 +240,7 @@ and reinitializes the observations.
 histogram.reset();
 ```
 
-#### Summary
+### Summary
 
 Summaries calculate percentiles of observed values.
 
@@ -285,7 +285,7 @@ reinitializes the observations.
 summary.reset();
 ```
 
-#### Labels
+### Labels
 
 All metrics can take a labelNames property in the configuration object. All
 labelNames that the metric support needs to be declared here. There are 2 ways
@@ -317,7 +317,7 @@ xhrRequest(function(err, res) {
 });
 ```
 
-##### Default Labels (segmented by registry)
+#### Default Labels (segmented by registry)
 
 Static labels may be applied to every metric emitted by a registry:
 
@@ -339,7 +339,7 @@ Default labels will be overridden if there is a name conflict.
 
 `register.clear()` will clear default labels.
 
-#### Timestamps
+### Timestamps
 
 Counter and gauge metrics can take a timestamp argument after the value
 argument. This argument must be a Date or a number (milliseconds since Unix
@@ -355,7 +355,7 @@ gauge.labels('GET', '200').set(100, new Date()); // Same as above
 counter.inc(1, new Date()); // Increment counter with timestamp
 ```
 
-#### Multiple registries
+### Multiple registries
 
 By default, metrics are automatically registered to the global registry (located
 at `require('prom-client').register`). You can prevent this by setting last
@@ -399,37 +399,37 @@ AggregatorRegistry.setRegistries(registry);
 AggregatorRegistry.setRegistries([registry1, registry2]);
 ```
 
-#### Register
+### Register
 
 You can get all metrics by running `register.metrics()`, which will output a
 string for prometheus to consume.
 
 `register.metrics()` takes an optional object with a `timestamps` field. Setting
 this to false will strip timestamps from the string.
 
-##### Getting a single metric for Prometheus displaying
+#### Getting a single metric for Prometheus displaying
 
 If you need to output a single metric for Prometheus, you can use
 `register.getSingleMetricAsString(*name of metric*)`, it will output a string
 for Prometheus to consume.
 
-##### Getting a single metric
+#### Getting a single metric
 
 If you need to get a reference to a previously registered metric, you can use
 `register.getSingleMetric(*name of metric*)`.
 
-##### Removing metrics
+#### Removing metrics
 
 You can remove all metrics by calling `register.clear()`. You can also remove a
 single metric by calling `register.removeSingleMetric(*name of metric*)`.
 
-##### Resetting metrics
+#### Resetting metrics
 
 If you need to reset all metrics, you can use `register.resetMetrics()`. The
 metrics will remain present in the register and can be used without the need to
 instantiate them again, like you would need to do after `register.clear()`.
 
-##### Cluster metrics
+#### Cluster metrics
 
 You can get aggregated metrics for all workers in a node.js cluster with
 `register.clusterMetrics()`. This method both returns a promise and accepts a
@@ -453,7 +453,7 @@ register.clusterMetrics((err, metrics) => {
 });
 ```
 
-#### Pushgateway
+### Pushgateway
 
 It is possible to push metrics via a
 [Pushgateway](https://github.com/prometheus/pushgateway).
@@ -480,7 +480,7 @@ gateway.pushAdd({ jobName: 'test', groupings: { key: 'value' } }, function(
 gateway = new client.Pushgateway('http://127.0.0.1:9091', { timeout: 5000 }); //Set the request timeout to 5000ms
 ```
 
-#### Utilites
+### Utilites
 
 For convenience, there are 2 bucket generator functions - linear and
 exponential.
@@ -503,7 +503,7 @@ new client.Histogram({
 The content-type prometheus expects is also exported as a constant, both on the
 `register` and from the main file of this project, called `contentType`.
 
-### Garbage Collection
+## Garbage Collection
 
 To avoid dependencies in this module, GC stats are kept outside of it. If you
 want GC stats, you can use https://github.com/SimenB/node-prometheus-gc-stats