[Feature] Implement any_of and include constraints support

crates/weaver_resolved_schema/src/attribute.rs

@@ -74,7 +74,7 @@ pub struct Attribute {
 }
 
 /// An unresolved attribute definition.
-#[derive(Debug, Clone)]
+#[derive(Debug, Deserialize, Clone)]
 pub struct UnresolvedAttribute {
     /// The attribute specification.
     pub spec: AttributeSpec,

crates/weaver_resolved_schema/src/catalog.rs

@@ -3,7 +3,7 @@
 //! Defines the catalog of attributes, metrics, and other telemetry items
 //! that are shared across multiple signals in the Resolved Telemetry Schema.
 
-use crate::attribute::Attribute;
+use crate::attribute::{Attribute, AttributeRef};
 use crate::metric::Metric;
 use serde::{Deserialize, Serialize};
 use std::fmt::Debug;
@@ -31,3 +31,13 @@ pub enum Stability {
     /// A stable definition.
     Stable,
 }
+
+impl Catalog {
+    /// Returns the attribute name from an attribute ref if it exists
+    /// in the catalog or None if it does not exist.
+    pub fn attribute_name(&self, attribute_ref: &AttributeRef) -> Option<String> {
+        self.attributes
+            .get(attribute_ref.0 as usize)
+            .map(|attr| attr.name.clone())
+    }
+}

crates/weaver_resolved_schema/src/registry.rs

@@ -6,6 +6,7 @@
 
 use crate::attribute::AttributeRef;
 use serde::{Deserialize, Serialize};
+use std::collections::HashSet;
 
 use crate::catalog::Stability;
 use crate::lineage::GroupLineage;
@@ -137,3 +138,36 @@ pub struct Constraint {
     /// not already defined in the current semantic convention.
     pub include: Option<String>,
 }
+
+impl Group {
+    /// Returns true if the group contains at least one `include` constraint.
+    pub fn has_include(&self) -> bool {
+        self.constraints.iter().any(|c| c.include.is_some())
+    }
+
+    /// Import attributes from the provided slice that do not exist in the
+    /// current group.
+    pub fn import_attributes_from(&mut self, attributes: &[AttributeRef]) {
+        for attr in attributes {
+            if !self.attributes.contains(attr) {
+                self.attributes.push(*attr);
+            }
+        }
+    }
+
+    /// Update the group constraints according to the provided constraints to
+    /// add and the `include` constraints to remove.
+    pub fn update_constraints(
+        &mut self,
+        constraints_to_add: Vec<Constraint>,
+        include_to_remove: HashSet<String>,
+    ) {
+        // Add the new constraints
+        self.constraints.extend(constraints_to_add);
+
+        // Remove the include constraints
+        self.constraints.retain(|c| {
+            c.include.is_none() || !include_to_remove.contains(c.include.as_ref().unwrap())
+        });
+    }
+}

crates/weaver_resolver/README.md

@@ -1,6 +1,8 @@
 # Telemetry Schema Resolution Process
 
-Status: **Work-In-Progress**
+Resolution Process Status:
+- Semantic Convention Registry: **Fully Implemented**, **Partially Tested**
+- Application Telemetry Schema: **Partially Implemented**
 
 This crate describes the resolution process for the OpenTelemetry telemetry
 schema. The resolution process takes a telemetry schema and/or a semantic
@@ -36,7 +38,7 @@ involves the following steps:
   - Resolve iteratively all `extends` parent/child clauses until no more
     resolvable `extends` are found. The extended entity inherits prefix,
     attributes, and constraints from the parent entity.
-- Apply constraints `any_of` and `include` (not yet supported).
+- Apply constraints `any_of` and `include`.
 - Validate the resolved semantic conventions
   - No more unresolved `ref` or `extends` clauses. The unresolved list should
     be empty.

crates/weaver_resolver/data/registry-test-7-spans/expected-registry.json

@@ -2477,29 +2477,44 @@
         "events": []
       },
       "brief": "Semantic convention group for specific technologies",
-      "constraints": [
-        {
-          "any_of": [],
-          "include": "db.cassandra"
-        },
-        {
-          "any_of": [],
-          "include": "db.redis"
-        },
-        {
-          "any_of": [],
-          "include": "db.mongodb"
-        },
-        {
-          "any_of": [],
-          "include": "db.sql"
-        },
-        {
-          "any_of": [],
-          "include": "db.cosmosdb"
-        }
+      "constraints": [],
+      "attributes": [
+        62,
+        63,
+        64,
+        65,
+        66,
+        67,
+        68,
+        69,
+        70,
+        71,
+        72,
+        73,
+        74,
+        75,
+        77,
+        78,
+        79,
+        80,
+        81,
+        82,
+        83,
+        84,
+        87,
+        88,
+        89,
+        99,
+        100,
+        101,
+        102,
+        103,
+        104,
+        105,
+        106,
+        107,
+        108
       ],
-      "attributes": [],
       "lineage": {
         "provenance": "data/registry-test-7-spans/registry/trace-database.yaml"
       }

crates/weaver_resolver/data/registry-test-7-spans/registry/trace-database.yaml

@@ -70,7 +70,7 @@ groups:
         tag: call-level-tech-specific-cassandra
         brief: >
           The keyspace name in Cassandra.
-        examples: ["mykeyspace"]
+        examples: [ "mykeyspace" ]
         note: For Cassandra the `db.name` should be set to the Cassandra keyspace name.
       - ref: db.cassandra.page_size
         tag: call-level-tech-specific-cassandra
@@ -97,7 +97,7 @@ groups:
         tag: call-level-tech-specific
         brief: >
           The HBase namespace.
-        examples: ['mynamespace']
+        examples: [ 'mynamespace' ]
         note: For HBase the `db.name` should be set to the HBase namespace.
 
   - id: db.couchdb
@@ -110,7 +110,7 @@ groups:
         tag: call-level-tech-specific
         brief: >
           The HTTP method + the target REST route.
-        examples: ['GET /{db}/{docid}']
+        examples: [ 'GET /{db}/{docid}' ]
         note: >
           In **CouchDB**, `db.operation` should be set to the HTTP method +
           the target REST route according to the API reference documentation.
@@ -132,7 +132,7 @@ groups:
         tag: call-level-tech-specific
         brief: >
           The full syntax of the Redis CLI command.
-        examples: ["HMSET myhash field1 'Hello' field2 'World'"]
+        examples: [ "HMSET myhash field1 'Hello' field2 'World'" ]
         note: >
           For **Redis**, the value provided for `db.statement` SHOULD correspond to the syntax of the Redis CLI.
           If, for example, the [`HMSET` command](https://redis.io/commands/hmset) is invoked, `"HMSET myhash field1 'Hello' field2 'World'"` would be a suitable value for `db.statement`.
@@ -227,7 +227,7 @@ groups:
           `failover_information` : Generated key to determine if region failover enabled.
              Format Reg-{D (Disabled discovery)}-S(application region)|L(List of preferred regions)|N(None, user did not configure it).
              Default value is "NS".
-        examples: ['cosmos-netstandard-sdk/3.23.0\|3.23.1\|1\|X64\|Linux 5.4.0-1098-azure 104 18\|.NET Core 3.1.32\|S\|']
+        examples: [ 'cosmos-netstandard-sdk/3.23.0\|3.23.1\|1\|X64\|Linux 5.4.0-1098-azure 104 18\|.NET Core 3.1.32\|S\|' ]
         tag: call-level-tech-specific
       - ref: db.cosmosdb.connection_mode
         requirement_level:
@@ -260,4 +260,5 @@ groups:
       - include: 'db.redis'
       - include: 'db.mongodb'
       - include: 'db.sql'
-      - include: 'db.cosmosdb'
+      - include: 'db.cosmosdb'
+      - any_of: ['db.cassandra.page_size', 'db.cassandra.consistency_level', 'db.cassandra.table', 'db.cassandra.idempotence', 'db.cassandra.speculative_execution_count', 'db.cassandra.coordinator.id', 'db.cassandra.coordinator.dc']

crates/weaver_resolver/src/attribute.rs

@@ -56,6 +56,17 @@ impl AttributeCatalog {
         attributes.into_iter().map(|(attr, _)| attr).collect()
     }
 
+    /// Returns a list of indexed attribute names ordered by their references.
+    pub fn attribute_name_index(&self) -> Vec<String> {
+        let mut attributes: Vec<(&attribute::Attribute, &AttributeRef)> =
+            self.attribute_refs.iter().collect();
+        attributes.sort_by_key(|(_, attr_ref)| attr_ref.0);
+        attributes
+            .iter()
+            .map(|(attr, _)| attr.name.clone())
+            .collect()
+    }
+
     /// Tries to resolve the given attribute spec (ref or id) from the catalog.
     /// Returns `None` if the attribute spec is a ref and it does not exist yet
     /// in the catalog.

crates/weaver_resolver/src/lib.rs

@@ -17,15 +17,16 @@ use regex::Regex;
 use url::Url;
 use walkdir::DirEntry;
 
-use crate::attribute::AttributeCatalog;
 use weaver_cache::Cache;
 use weaver_logger::Logger;
+use weaver_resolved_schema::attribute::AttributeRef;
 use weaver_resolved_schema::catalog::Catalog;
 use weaver_resolved_schema::ResolvedTelemetrySchema;
 use weaver_schema::{SemConvImport, TelemetrySchema};
 use weaver_semconv::{ResolverConfig, SemConvRegistry, SemConvSpec, SemConvSpecWithProvenance};
 use weaver_version::VersionChanges;
 
+use crate::attribute::AttributeCatalog;
 use crate::events::resolve_events;
 use crate::metrics::{resolve_metrics, semconv_to_resolved_metric};
 use crate::registry::resolve_semconv_registry;
@@ -67,6 +68,15 @@ pub enum UnresolvedReference {
         /// The provenance of the reference (URL or path).
         provenance: String,
     },
+    /// An unresolved `include` reference.
+    IncludeRef {
+        /// The id of the group containing the `include` reference.
+        group_id: String,
+        /// The unresolved `include` reference.
+        include_ref: String,
+        /// The provenance of the reference (URL or path).
+        provenance: String,
+    },
 }
 
 /// An error that can occur while resolving a telemetry schema.
@@ -136,6 +146,24 @@ pub enum Error {
         /// The error that occurred.
         message: String,
     },
+
+    /// The `any_of` constraints are unsatisfied for a group.
+    #[error("The `any_of` constraints are unsatisfied for the group '{group_id}'.\nGroup attributes: {group_attributes:#?}\n`any_of` constraints: {any_of_constraints:#?}")]
+    UnsatisfiedAnyOfConstraint {
+        /// The id of the group containing the unsatisfied `any_of` constraint.
+        group_id: String,
+        /// The attributes of the group.
+        group_attributes: Vec<String>,
+        /// The `any_of` constraints of the group.
+        any_of_constraints: Vec<Vec<String>>,
+    },
+
+    /// Attribute ref not found in the catalog.
+    #[error("Attribute ref `{attribute_ref:?}` not found in the catalog.")]
+    UnresolvedAttribute {
+        /// The unresolved attribute reference.
+        attribute_ref: AttributeRef,
+    },
 }
 
 impl SchemaResolver {
@@ -380,14 +408,16 @@ impl SchemaResolver {
             .map(semconv_to_resolved_metric)
             .collect();
 
+        let catalog = Catalog {
+            attributes: attr_catalog.drain_attributes(),
+            metrics,
+        };
+
         let resolved_schema = ResolvedTelemetrySchema {
             file_format: "1.0.0".to_string(),
             schema_url: "".to_string(),
             registries: vec![resolved_registry],
-            catalog: Catalog {
-                attributes: attr_catalog.drain_attributes(),
-                metrics,
-            },
+            catalog,
             resource: None,
             instrumentation_library: None,
             dependencies: vec![],