Add doc summarizing info on v2 schemas

.github/workflows/ci.yml

@@ -298,34 +298,42 @@ jobs:
         run: |
           cargo run -- --quiet registry json-schema -j semconv-definition-v2 -o /tmp/semconv.schema.v2.json
           cargo run -- --quiet registry json-schema -j resolved-registry-v2 -o /tmp/semconv.resolved.v2.json
+          cargo run -- --quiet registry json-schema -j materialized-registry-v2 -o /tmp/semconv.materialized.v2.json
+          cargo run -- --quiet registry json-schema -j diff-v2 -o /tmp/semconv.diff.v2.json
           cargo run -- --quiet registry json-schema -j policy-finding -o /tmp/policy.finding.json
+          cargo run -- --quiet registry json-schema -j publication-manifest-v2 -o /tmp/publication-manifest.v2.json
+          cargo run -- --quiet registry json-schema -j definition-manifest-v2 -o /tmp/definition-manifest.v2.json
       - name: Check generated files are up-to-date
         run: |
           failed=0
 
-          if ! diff -q docs/usage.md /tmp/usage.md > /dev/null 2>&1; then
-            echo "docs/usage.md is out of date. Run 'cargo run -- --quiet markdown-help > docs/usage.md' to regenerate it."
-            diff docs/usage.md /tmp/usage.md
-            failed=1
-          fi
-
-          if ! diff -q schemas/semconv.schema.v2.json /tmp/semconv.schema.v2.json > /dev/null 2>&1; then
-            echo "schemas/semconv.schema.v2.json is out of date. Run 'cargo run -- --quiet registry json-schema -j semconv-definition-v2 -o ./schemas/semconv.schema.v2.json' to regenerate it."
-            diff schemas/semconv.schema.v2.json /tmp/semconv.schema.v2.json
-            failed=1
-          fi
+          check_file() {
+            local repo_file="$1"
+            local tmp_file="$2"
+            local regen_cmd="$3"
+            if ! diff -q "$repo_file" "$tmp_file" > /dev/null 2>&1; then
+              echo "$repo_file is out of date. Run '$regen_cmd' to regenerate it."
+              diff "$repo_file" "$tmp_file"
+              failed=1
+            fi
+          }
 
-          if ! diff -q schemas/semconv.resolved.v2.json /tmp/semconv.resolved.v2.json > /dev/null 2>&1; then
-            echo "schemas/semconv.resolved.v2.json is out of date. Run 'cargo run -- --quiet registry json-schema -j resolved-registry-v2 -o ./schemas/semconv.resolved.v2.json' to regenerate it."
-            diff schemas/semconv.resolved.v2.json /tmp/semconv.resolved.v2.json
-            failed=1
-          fi
-
-          if ! diff -q schemas/policy.finding.json /tmp/policy.finding.json > /dev/null 2>&1; then
-            echo "schemas/policy.finding.json is out of date. Run 'cargo run -- --quiet registry json-schema -j policy-finding -o ./schemas/policy.finding.json' to regenerate it."
-            diff schemas/policy.finding.json /tmp/policy.finding.json
-            failed=1
-          fi
+          check_file docs/usage.md /tmp/usage.md \
+            "cargo run -- --quiet markdown-help > docs/usage.md"
+          check_file schemas/semconv.schema.v2.json /tmp/semconv.schema.v2.json \
+            "cargo run -- --quiet registry json-schema -j semconv-definition-v2 -o ./schemas/semconv.schema.v2.json"
+          check_file schemas/semconv.resolved.v2.json /tmp/semconv.resolved.v2.json \
+            "cargo run -- --quiet registry json-schema -j resolved-registry-v2 -o ./schemas/semconv.resolved.v2.json"
+          check_file schemas/policy.finding.json /tmp/policy.finding.json \
+            "cargo run -- --quiet registry json-schema -j policy-finding -o ./schemas/policy.finding.json"
+          check_file schemas/semconv.materialized.v2.json /tmp/semconv.materialized.v2.json \
+            "cargo run -- --quiet registry json-schema -j materialized-registry-v2 -o ./schemas/semconv.materialized.v2.json"
+          check_file schemas/semconv.diff.v2.json /tmp/semconv.diff.v2.json \
+            "cargo run -- --quiet registry json-schema -j diff-v2 -o ./schemas/semconv.diff.v2.json"
+          check_file schemas/publication-manifest.v2.json /tmp/publication-manifest.v2.json \
+            "cargo run -- --quiet registry json-schema -j publication-manifest-v2 -o ./schemas/publication-manifest.v2.json"
+          check_file schemas/definition-manifest.v2.json /tmp/definition-manifest.v2.json \
+            "cargo run -- --quiet registry json-schema -j definition-manifest-v2 -o ./schemas/definition-manifest.v2.json"
 
           exit $failed
   coverage:

CHANGELOG.md

@@ -7,6 +7,7 @@ All notable changes to this project will be documented in this file.
 - Fix ([#1323](https://github.com/open-telemetry/weaver/issues/1323)) - Live-check: treat observed `int` as compatible with expected `double` attribute type, avoiding false violations from OTLP serializers (e.g. JS) that emit `int_value` for integral numbers. ([#1331](https://github.com/open-telemetry/weaver/pull/1331) by @jerbly)
 - Fix CLI so it doesn't error out early when loading a resolved schema. ([#1304](https://github.com/open-telemetry/weaver/pull/1304) by @jsuereth)
 - 💥 BREAKING CHANGE 💥 Use `schema_url` to track registries consistently in lineage / provenance ([#1298](https://github.com/open-telemetry/weaver/pull/1298) by @jsuereth)
+- Publish and document missing JSON schemas. Add `head_schema_url`/`baseline_schema_url` to diff v2. ([#1106](https://github.com/open-telemetry/weaver/pull/1106) by @lmolkova)
 - Add `--allow-git-credentials` global flag to enable system credential helpers (e.g. `osxkeychain`, `git-credential-manager`) when cloning private registries. By default, git operations remain isolated for security. ([#1306](https://github.com/open-telemetry/weaver/pull/1306) by @jerbly)
 - MCP: Add `browse_namespace` tool and `findings_only` output mode for `live_check`. Add configurable namespace separator via `--namespace-separator`. ([#1324](https://github.com/open-telemetry/weaver/pull/1324) by @jerbly)
 

crates/weaver_resolved_schema/src/lib.rs

@@ -39,7 +39,7 @@ pub const OTEL_REGISTRY_ID: &str = "OTEL";
 
 /// Version string denoting V1 resolved schema.
 pub(crate) const V1_RESOLVED_FILE_FORMAT: &str = "resolved/1.0.0";
-/// Version string dentoing V2 resolved scehma.
+/// Version string denoting V2 resolved schema.
 pub(crate) const V2_RESOLVED_FILE_FORMAT: &str = "resolved/2.0.0";
 
 /// A Resolved Telemetry Schema.

crates/weaver_resolved_schema/src/v2/mod.rs

@@ -74,6 +74,8 @@ impl ResolvedTelemetrySchema {
     pub fn diff(&self, baseline_schema: &ResolvedTelemetrySchema) -> SchemaChanges {
         // TODO - get manifests
         SchemaChanges {
+            head_schema_url: self.schema_url.clone(),
+            baseline_schema_url: baseline_schema.schema_url.clone(),
             registry: self.registry_diff(baseline_schema),
         }
     }

crates/weaver_version/Cargo.toml

@@ -18,3 +18,4 @@ thiserror.workspace = true
 schemars.workspace = true
 
 semver = {version =  "1.0.27", features = ["serde"]}
+weaver_semconv = { path = "../weaver_semconv" }

crates/weaver_version/src/v2/mod.rs

@@ -2,18 +2,33 @@
 
 use schemars::JsonSchema;
 use serde::{Deserialize, Serialize};
+use weaver_semconv::schema_url::SchemaUrl;
 
 // V2 Leverages the same nomenclature for diff as V1.
 pub use crate::schema_changes::SchemaItemChange;
 
 /// A summary of schema changes between two versions of a schema.
-#[derive(Debug, Default, Serialize, Deserialize, JsonSchema)]
+#[derive(Debug, Serialize, Deserialize, JsonSchema)]
 #[serde(rename_all = "snake_case")]
 pub struct SchemaChanges {
+    /// Schema URL of the head (newer) schema.
+    pub head_schema_url: SchemaUrl,
+    /// Schema URL of the baseline (older) schema.
+    pub baseline_schema_url: SchemaUrl,
     /// Changes to the registry.
     pub registry: RegistryChanges,
 }
 
+impl Default for SchemaChanges {
+    fn default() -> Self {
+        Self {
+            head_schema_url: SchemaUrl::new_unknown(),
+            baseline_schema_url: SchemaUrl::new_unknown(),
+            registry: RegistryChanges::default(),
+        }
+    }
+}
+
 /// A summary of changes to the registry of signals and attributes.
 #[derive(Debug, Default, Serialize, Deserialize, JsonSchema)]
 #[serde(rename_all = "snake_case")]
@@ -31,6 +46,7 @@ pub struct RegistryChanges {
     /// Changes across the registry of spans.
     pub span_changes: Vec<SchemaItemChange>,
 }
+
 impl RegistryChanges {
     /// Returns true if there are no changes in the schema.
     /// Otherwise, it returns false.
@@ -53,3 +69,17 @@ impl SchemaChanges {
         self.registry.is_empty()
     }
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_schema_changes_default() {
+        let changes = SchemaChanges::default();
+        assert_eq!(changes.head_schema_url, SchemaUrl::new_unknown());
+        assert_eq!(changes.baseline_schema_url, SchemaUrl::new_unknown());
+        assert!(changes.registry.is_empty());
+        assert!(changes.is_empty());
+    }
+}

docs/usage.md

@@ -342,7 +342,7 @@ The produced JSON Schema can be used to generate documentation of the resolved r
     The JSON schema of the V2 definition
   - `resolved-registry-v2`:
     The JSON schema of the V2 resolved registry
-  - `forge-registry-v2`:
+  - `materialized-registry-v2`:
     The JSON schema we send to Rego / Jinja
   - `diff`:
     The JSON schema of the diff