.readme-partials.yml

introduction: |-
  [Google Cloud Logging](https://cloud.google.com/logging/docs) allows you to store, search, analyze,
  monitor, and alert on log data and events from Google Cloud Platform and Amazon Web Services.

  If you require lightweight dependencies, an experimental, minified version of
  this library is available at [@google-cloud/logging-min](https://www.npmjs.com/package/@google-cloud/logging-min).
  Note: `logging-min` is experimental, and its feature surface is subject to change. 
  To install `@google-cloud/logging-min` library run the following command:

  ```bash
  npm install @google-cloud/logging-min
  ```

  For an interactive tutorial on using the client library in a Node.js application, click Guide Me:

  [![Guide Me](https://raw.githubusercontent.com/googleapis/nodejs-logging/main/_static/guide-me.svg)](https://console.cloud.google.com/?walkthrough_id=logging__logging-nodejs)
body: |-
  ## Batching Writes

  High throughput applications should avoid awaiting calls to the logger:

  ```js
  await log.write(logEntry1);
  await log.write(logEntry2);
  ```

  Rather, applications should use a _fire and forget_ approach:

  ```js
  log.write(logEntry1);
  log.write(logEntry2);
  ```

  The `@google-cloud/logging` library will handle batching and dispatching
  these log lines to the API.

  ## Writing to Stdout

  The `LogSync` class helps users easily write context-rich structured logs to
  `stdout` or any custom transport. It extracts additional log properties like
  trace context from HTTP headers and can be used as an on/off toggle between
  writing to the API or to `stdout` during local development.

  Logs written to `stdout` are then picked up, out-of-process, by a Logging
  agent in the respective GCP environment. Logging agents can add more
  properties to each entry before streaming it to the Logging API.

  Read more about [Logging agents](https://cloud.google.com/logging/docs/agent/logging).

  Serverless applications like Cloud Functions, Cloud Run, and App Engine
  are highly recommended to use the `LogSync` class as async logs may be dropped
  due to lack of CPU.

  Read more about [structured logging](https://cloud.google.com/logging/docs/structured-logging).

  ```js
  // Optional: Create and configure a client
  const logging = new Logging();
  await logging.setProjectId()
  await logging.setDetectedResource()

  // Create a LogSync transport, defaulting to `process.stdout`
  const log = logging.logSync(logname);
  const meta = { // optional field overrides here };
  const entry = log.entry(meta, 'Your log message');
  log.write(entry);

  // Syntax sugar for logging at a specific severity
  log.alert(entry);
  log.warning(entry);
  ```

  ## Populating Http request metadata

  Metadata about Http request is a part of the [structured log info](https://cloud.google.com/logging/docs/structured-logging)
  that can be captured within each log entry. It can provide a context for the application logs and
  is used to group multiple log entries under the load balancer request logs. See the [sample](https://github.com/googleapis/nodejs-logging/blob/master/samples/http-request.js)
  how to populate the Http request metadata for log entries.

  If you already have a "raw" Http `request` object you can assign it to `entry.metadata.httpRequest` directly. More information about
  how the `request` is interpreted as raw can be found in the [code](https://github.com/googleapis/nodejs-logging/blob/15849160116a814ab71113138cb211c2e0c2d4b4/src/entry.ts#L224-L238).

  ## Automatic Trace/Span ID Extraction
  Cloud Logging libraries use [trace fields within LogEntry](https://cloud.google.com/logging/docs/reference/v2/rest/v2/LogEntry#FIELDS.trace) to capture trace contexts, which enables the [correlation of logs and traces](https://cloud.google.com/logging/docs/view/correlate-logs), and distributed tracing troubleshooting. 
  These tracing fields, including [trace](https://cloud.google.com/logging/docs/reference/v2/rest/v2/LogEntry#FIELDS.trace), [spanId](https://cloud.google.com/logging/docs/reference/v2/rest/v2/LogEntry#FIELDS.span_id), and [traceSampled](https://cloud.google.com/logging/docs/reference/v2/rest/v2/LogEntry#FIELDS.trace_sampled), define the trace context for a `LogEntry`.

  If not provided explicitly in a LogEntry, the Cloud Logging library automatically populates `trace`, `span_id`, and `trace_sampled` fields from detected OpenTelemetry span contexts, or from HTTP request headers.

  ### Extracting Trace/Span ID from OpenTelemetry Context
  If you are using OpenTelemetry and there is an active span in the OpenTelemetry Context, the `trace`, `span_id`, and `trace_sampled` fields in the log entry are automatically populated from the active span. More information about OpenTelemetry can be found [here](https://opentelemetry.io/docs/languages/js/).

  ### Extracting Trace/Span ID from HTTP Headers
  If tracing fields are not provided explicitly and no OpenTelemetry context is detected, the `trace` / `span_id` fields are extracted automatically from HTTP headers. 
  Trace information can be automatically populated from either the [W3C Traceparent](https://www.w3.org/TR/trace-context) or [X-Cloud-Trace-Context](https://cloud.google.com/trace/docs/trace-context#legacy-http-header) headers.

  ## Error handling with logs written or deleted asynchronously

  The `Log` class provide users the ability to write and delete logs asynchronously. However, there are cases when log entries
  cannot be written or deleted and error is thrown - if error is not handled properly, it could crash the application.
  One possible way to catch the error is to `await` the log write/delete calls and wrap it with `try/catch` like in example below:

  ```js
      // Write log entry and and catch any errors
      try {
        await log.write(entry);
      } catch (err) {
        console.log('Error is: ' + err);
      }
  ```

  However, awaiting for every `log.write` or `log.delete` calls may introduce delays which could be avoided by 
  simply adding a callback like in the example below. This way the log entry can be queued for processing and code
  execution will continue without further delays. The callback will be called once the operation is complete:

  ```js
      // Asynchronously write the log entry and handle respone or any errors in provided callback
      log.write(entry, err => {
        if (err) {
          // The log entry was not written.
          console.log(err.message);
        } else {
          console.log('No error in write callback!');
        }
      });
  ```

  Adding a callback to every `log.write` or `log.delete` calls could be a burden, especially if code
  handling the error is always the same. For this purpose we introduced an ability to provide a default callback
  for `Log` class which could be set through `LogOptions` passed to `Log` constructor as in example below - this
  way you can define a global callback once for all `log.write` and `log.delete` calls and be able to handle errors:

  ```js
    const {Logging} = require('@google-cloud/logging');
    const logging = new Logging();
    
    // Create options with default callback to be called on every write/delete response or error
    const options = {
      defaultWriteDeleteCallback: function (err) {
        if (err) {
          console.log('Error is: ' + err);
        } else {
          console.log('No error, all is good!');
        }
      },
    };

    const log = logging.log('my-log', options);
  ``` 
  See the full sample in `writeLogWithCallback` function [here](https://github.com/googleapis/nodejs-logging/blob/master/samples/logs.js).