Add blog post about the OTTL context inference feature

content/en/blog/2025/ottl-contexts-just-got-easier.md

@@ -0,0 +1,131 @@
+---
+title: OTTL contexts just got easier with context inference
+linkTitle: OTTL contexts just got easier
+date: 2025-02-17
+author: '[Edmo Vamerlatti Costa](https://github.com/edmocosta) (Elastic)'
+draft: true # TODO: remove this line once your post is ready to be published
+issue: 6289
+sig: Collector SIG
+cSpell:ignore: OTTL Vamerlatti
+---
+
+Selecting the right context for running OTTL statements can be challenging, even
+for experienced users. Choosing the correct context impacts both accuracy and
+efficiency, as using higher-level contexts can avoid unnecessary iterations
+through nested lower-level contexts.
+
+To simplify this process, the OpenTelemetry community is excited to announce
+OTTL
+[context inference](https://github.com/open-telemetry/opentelemetry-collector-contrib/blob/main/processor/transformprocessor/README.md#context-inference)
+support for the
+[transform processor](https://github.com/open-telemetry/opentelemetry-collector-contrib/blob/main/processor/transformprocessor).
+This feature removes the need to manually specify contexts, improving statement
+processing efficiency by automatically selecting the most appropriate one. This
+optimization ensures that data transformations are both accurate and performant.
+
+## How does it work?
+
+Starting with version `0.120.0`, the transform processor supports two new
+[context-inferred configuration](https://github.com/open-telemetry/opentelemetry-collector-contrib/blob/main/processor/transformprocessor/README.md#context-inferred-configurations)
+styles. The first one offers a simpler and flatter approach, while the second
+closely resembles the existing configuration format.
+
+### Flat configuration
+
+The flat configuration style simplifies configuration by allowing users to list
+all statements together, without worrying about contexts or extra configuration
+structures. This style support statements from multiple contexts and does not
+require grouping them separately.
+
+To illustrate this, compare the following configuration:
+
+```yaml
+metric_statements:
+  - context: resource
+    statements:
+      - keep_keys(attributes, ["host.name"])
+  - context: metric
+    statements:
+      - set(description, "Sum") where type == "Sum"
+      - convert_sum_to_gauge() where name == "system.processes.count"
+  - context: datapoint
+    statements:
+      - limit(attributes, 100, ["host.name"])
+```
+
+With the new flat configuration style, the same logic is expressed more
+concisely as:
+
+```yaml
+metric_statements:
+  - keep_keys(resource.attributes, ["host.name"])
+  - set(metric.description, "Sum") where metric.type == "Sum"
+  - convert_sum_to_gauge() where metric.name == "system.processes.count"
+  - limit(datapoint.attributes, 100, ["host.name"])
+```
+
+This streamlined approach enhances readability and makes configuration more
+intuitive. Please note that all paths in the statements must be prefixed with
+their respective contexts. These prefixes are required for all context-inferred
+configurations and serve as hints for selecting the best match. It also makes
+statements unambiguous and portable between components.
+
+### Structured configuration
+
+The context-inferred structured configuration style closely resembles the
+existing format and allows users to leverage the benefits of context inference
+while providing granular control over statement configurations, such as
+`error_mode` and `conditions`. For example, consider the following
+configuration:
+
+<!-- prettier-ignore-start -->
+```yaml
+metric_statements:
+  - context: datapoint
+    conditions:
+      - resource.attributes["service.name"] == "my.service"
+    statements:
+      - set(metric.description, "counter") where attributes["my.attr"] == "some"
+```
+<!-- prettier-ignore-end -->
+
+It can now be written as:
+
+<!-- prettier-ignore-start -->
+```yaml
+metric_statements:
+  - conditions:
+      - resource.attributes["service.name"] == "my.service"
+    statements:
+      - set(metric.description, "counter") where datapoint.attributes["my.attr"] == "some"
+```
+<!-- prettier-ignore-end -->
+
+In this example, the `context` value is omitted and is automatically inferred to
+`datapoint`, as it is the only context present in the statements that supports
+parsing both `datapoint` and `metric` data.
+
+If we update the above configuration removing the `datapoint` usage:
+
+<!-- prettier-ignore-start -->
+```yaml
+metric_statements:
+  - conditions:
+      - resource.attributes["service.name"] == "my.service"
+    statements:
+      - set(metric.description, "counter")
+```
+<!-- prettier-ignore-end -->
+
+The context inferrer would select the `metric` context instead, since no data
+points are accessed. Although it would be possible to run the statements using
+the `datapoint` context, `metric` is the most efficient option.
+
+## Try it out
+
+As we wrap up, we encourage users to explore this new functionality and take
+advantage of its benefits in their telemetry pipelines!
+
+If you have any questions or suggestions, we’d love to hear from you! Join the
+conversation in the `#otel-collector` channel on the
+[CNCF Slack workspace](https://slack.cncf.io/).

static/refcache.json

@@ -3435,6 +3435,10 @@
     "StatusCode": 206,
     "LastSeen": "2025-02-02T10:41:21.884395-05:00"
   },
+  "https://github.com/edmocosta": {
+    "StatusCode": 206,
+    "LastSeen": "2025-02-11T15:13:03.156135+01:00"
+  },
   "https://github.com/edsoncelio": {
     "StatusCode": 200,
     "LastSeen": "2024-11-06T19:17:39.555698Z"
@@ -5051,6 +5055,18 @@
     "StatusCode": 206,
     "LastSeen": "2025-01-16T14:34:46.89984-05:00"
   },
+  "https://github.com/open-telemetry/opentelemetry-collector-contrib/blob/main/processor/transformprocessor": {
+    "StatusCode": 206,
+    "LastSeen": "2025-02-11T17:52:52.423025+01:00"
+  },
+  "https://github.com/open-telemetry/opentelemetry-collector-contrib/blob/main/processor/transformprocessor/README.md#context-inference": {
+    "StatusCode": 206,
+    "LastSeen": "2025-02-11T15:13:04.026653+01:00"
+  },
+  "https://github.com/open-telemetry/opentelemetry-collector-contrib/blob/main/processor/transformprocessor/README.md#context-inferred-configurations": {
+    "StatusCode": 206,
+    "LastSeen": "2025-02-11T15:13:06.05203+01:00"
+  },
   "https://github.com/open-telemetry/opentelemetry-collector-contrib/blob/main/receiver/hostmetricsreceiver/README.md": {
     "StatusCode": 206,
     "LastSeen": "2025-01-16T14:34:35.525335-05:00"