Support for 2 semconv registries

crates/weaver_cache/Cargo.toml

@@ -30,10 +30,11 @@ zip = "2.2.2"
 thiserror.workspace = true
 serde.workspace = true
 miette.workspace = true
-once_cell.workspace = true
-regex.workspace = true
 url.workspace = true
 ureq.workspace = true
+schemars.workspace = true
+once_cell.workspace = true
+regex.workspace = true
 
 [dev-dependencies]
 walkdir.workspace = true

crates/weaver_cache/src/lib.rs

@@ -2,6 +2,8 @@
 
 //! A Semantic Convention Repository abstraction for OTel Weaver.
 
+pub mod registry_path;
+
 use std::default::Default;
 use std::fs::{create_dir_all, File};
 use std::io;
@@ -19,11 +21,8 @@
 use url::Url;
 
 use weaver_common::diagnostic::{DiagnosticMessage, DiagnosticMessages};
-
-use crate::registry_path::RegistryPath;
 use crate::Error::{GitError, InvalidRegistryArchive, UnsupportedRegistryArchive};
-
-pub mod registry_path;
+use crate::registry_path::RegistryPath;
 
 /// The extension for a tar gz archive.
 const TAR_GZ_EXT: &str = ".tar.gz";
@@ -543,8 +542,10 @@
 
 #[cfg(test)]
 mod tests {
+    use weaver_common::registry_path::RegistryPath;
     use super::*;
     use weaver_common::test::ServeStaticFiles;
+    use crate::registry_path::RegistryPath;
 
     fn count_yaml_files(repo_path: &Path) -> usize {
         let count = walkdir::WalkDir::new(repo_path)

crates/weaver_cache/src/registry_path.rs

@@ -4,10 +4,10 @@
 
 use std::fmt::Display;
 use std::str::FromStr;
-
+use schemars::JsonSchema;
 use once_cell::sync::Lazy;
 use regex::Regex;
-
+use serde::{Deserialize, Serialize};
 use crate::Error;
 
 /// Regex to parse a registry path supporting the following formats:
@@ -22,7 +22,9 @@ static REGISTRY_REGEX: Lazy<Regex> = Lazy::new(|| {
 
 /// Path to a semantic convention registry.
 /// The path can be a local directory or a Git URL.
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
+#[serde(try_from = "String")]
+#[serde(into = "String")]
 pub enum RegistryPath {
     /// Local folder path pointing to a semantic convention registry.
     LocalFolder {
@@ -54,6 +56,24 @@ pub enum RegistryPath {
     },
 }
 
+/// Implement `TryFrom<String>` for `RegistryPath`, so that it can be used to
+/// deserialize a `RegistryPath` via serde.
+impl TryFrom<String> for RegistryPath {
+    type Error = Error;
+
+    fn try_from(s: String) -> Result<Self, Self::Error> {
+        s.parse()
+    }
+}
+
+/// Implement `From<RegistryPath>` for String, so that it can be serialized to a
+/// string via serde.
+impl From<RegistryPath> for String {
+    fn from(path: RegistryPath) -> Self {
+        path.to_string()
+    }
+}
+
 /// Implement the `FromStr` trait for `RegistryPath`, so that it can be used as
 /// a command-line argument.
 impl FromStr for RegistryPath {

crates/weaver_common/src/lib.rs

@@ -8,7 +8,6 @@ pub mod in_memory;
 pub mod quiet;
 pub mod result;
 pub mod test;
-
 use std::sync::atomic::{AtomicBool, AtomicUsize, Ordering};
 use std::sync::{Arc, Mutex};
 

crates/weaver_semconv/src/manifest.rs

@@ -13,6 +13,7 @@ use crate::Error::{InvalidRegistryManifest, RegistryManifestNotFound};
 use schemars::JsonSchema;
 use serde::{Deserialize, Serialize};
 use std::path::PathBuf;
+use weaver_cache::registry_path::RegistryPath;
 use weaver_common::error::handle_errors;
 
 /// Represents the information of a semantic convention registry manifest.
@@ -36,6 +37,20 @@ pub struct RegistryManifest {
 
     /// The base URL where the registry's schema files are hosted.
     pub schema_base_url: String,
+
+    /// List of the registry's dependencies.
+    /// Note: In the current phase, we only support zero or one dependency.
+    /// See this GH issue for more details: https://github.com/open-telemetry/weaver/issues/604
+    pub dependencies: Option<Vec<Dependency>>,
+}
+
+/// Represents a dependency of a semantic convention registry.
+#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)]
+pub struct Dependency {
+    /// The name of the dependency.
+    pub name: String,
+    /// The registry path of the dependency.
+    pub registry_path: RegistryPath,
 }
 
 impl RegistryManifest {

crates/weaver_semconv/src/registry.rs

@@ -146,6 +146,7 @@ impl SemConvRegistry {
                 description: None,
                 semconv_version,
                 schema_base_url: "".to_owned(),
+                dependencies: None,
             });
         }
 

tests/custom_registry/registry.yaml

@@ -0,0 +1,28 @@
+groups:
+  - id: "registry.group1"
+    type: "attribute_group"
+    name: "registry.group1"
+    brief: "registry.group1"
+    attributes:
+      - id: "attr0"
+        type: boolean
+        brief: "brief0"
+        requirement_level: "recommended"
+      - id: "attr1"
+        type: boolean
+        brief: "brief1"
+        requirement_level: "recommended"
+      - id: "attr2"
+        type: string
+        brief: "brief2"
+        requirement_level: "recommended"
+      - id: attr3
+        type: int
+        brief: brief3
+        requirement_level: recommended
+      - id: attr5
+        type: double
+        brief: brief5
+        requirement_level: recommended
+        deprecated:
+          reason: obsoleted

tests/custom_registry/registry_manifest.yaml

@@ -0,0 +1,7 @@
+name: acme
+description: This registry contains the semantic conventions for the Acme vendor.
+semconv_version: 0.1.0
+schema_base_url: https://acme.com/schemas/
+dependencies:
+  - name: otel
+    registry_path: https://github.com/open-telemetry/semantic-conventions/archive/refs/tags/v1.30.0.zip[model]