Merge branch 'main' into wip-bump-jaq-to-2.0

.github/workflows/ci.yml

@@ -46,6 +46,8 @@ jobs:
 #        run: cargo test --workspace --all-features
       - name: No-default features
         run: cargo test --workspace --no-default-features
+      - name: History check
+        run: cargo xtask history        
   msrv:
       name: Check MSRV
       runs-on: ubuntu-latest

.gitignore

@@ -37,4 +37,7 @@ lcov.info
 
 # Ignore the output of the `weaver resolve` command
 output.json
-resolved-schema.yaml
+resolved-schema.yaml
+
+# Ignore the temporary repo dir created by `cargo xtask history` command
+history-temp-repo/

CODEOWNERS

@@ -15,6 +15,7 @@
 # Global owners, will be the owners for everything in the repo
 
 * @open-telemetry/weaver-maintainers
+* @open-telemetry/weaver-approvers
 
 # Semantic conventions schema
 /schemas/semconv.schema.json @open-telemetry/specs-semconv-approvers

Cargo.toml

@@ -37,7 +37,7 @@ publish = false
 rust-version = "1.81.0"
 
 [workspace.dependencies]
-serde = { version = "1.0.215", features = ["derive"] }
+serde = { version = "1.0.216", features = ["derive"] }
 serde_yaml = "0.9.34"
 serde_json = { version = "1.0.133"}
 thiserror = "2.0.6"

crates/xtask/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "xtask"
-version = "0.1.0"
-publish = false             # xtask is not meant to be published
+version = "0.1.1"
+publish = false               # xtask is not meant to be published
 authors.workspace = true
 edition.workspace = true
 repository.workspace = true
@@ -10,6 +10,16 @@ rust-version.workspace = true
 
 [dependencies]
 anyhow.workspace = true
+assert_cmd = "2.0.14"
+gix = { version = "0.68.0", default-features = false, features = [
+    "comfort",
+    "blocking-http-transport-reqwest",
+    "max-performance-safe",
+    "worktree-mutation",
+    "blocking-http-transport-reqwest-rust-tls",
+] }
+semver = "1.0.23"
+serde_json.workspace = true
 toml = "0.8.19"
 
 #[lints]

crates/xtask/src/history.rs

@@ -0,0 +1,108 @@
+// SPDX-License-Identifier: Apache-2.0
+
+use assert_cmd::Command;
+use gix::clone::PrepareFetch;
+use gix::create::{self, Kind};
+use gix::{open, progress, Repository};
+use semver::Version;
+use std::fs;
+use std::sync::atomic::AtomicBool;
+use std::time::Duration;
+
+const REPO_URL: &str = "https://github.com/open-telemetry/semantic-conventions.git";
+const ARCHIVE_URL: &str =
+    "https://github.com/open-telemetry/semantic-conventions/archive/refs/tags/";
+const START_TAG: &str = "v1.22.0";
+const TEMP_REPO_DIR: &str = "history-temp-repo";
+
+/// Get the git repository, no checkout
+#[cfg(not(tarpaulin_include))]
+fn get_repo() -> anyhow::Result<Repository> {
+    let _ = fs::remove_dir_all(TEMP_REPO_DIR);
+    let _ = fs::create_dir(TEMP_REPO_DIR);
+    let mut fetch = PrepareFetch::new(
+        REPO_URL,
+        TEMP_REPO_DIR,
+        Kind::WithWorktree,
+        create::Options {
+            destination_must_be_empty: true,
+            fs_capabilities: None,
+        },
+        open::Options::isolated(),
+    )?;
+
+    let (repo, _) = fetch.fetch_only(progress::Discard, &AtomicBool::new(false))?;
+    Ok(repo)
+}
+
+/// Get the list of tags from the git repository filtered by the start version
+#[cfg(not(tarpaulin_include))]
+fn get_versions_from_git(repo: &Repository, start_semver: Version) -> anyhow::Result<Vec<String>> {
+    let tags: Vec<String> = repo
+        .references()?
+        .tags()?
+        .filter_map(|reference| {
+            let reference = reference.ok()?;
+            let tag = reference.name().shorten().to_string();
+
+            // Ignore tags with a lower version than the start tag
+            let version_str = tag.trim_start_matches(char::is_alphabetic);
+            if let Ok(version) = Version::parse(version_str) {
+                if version >= start_semver {
+                    Some(tag)
+                } else {
+                    None
+                }
+            } else {
+                None
+            }
+        })
+        .collect();
+    Ok(tags)
+}
+
+/// Run registry check on every semconv archive starting from start_version
+#[cfg(not(tarpaulin_include))]
+pub fn run(start_version: Option<String>) -> anyhow::Result<()> {
+    use anyhow::Context;
+
+    let start_version = start_version.unwrap_or(START_TAG.to_string());
+    let start_semver = Version::parse(start_version.trim_start_matches(char::is_alphabetic))
+        .context(format!(
+            "The provided version `{start_version}` is not a valid semver."
+        ))?;
+    let repo = get_repo().context("Failed to fetch the semconv repo.")?;
+    let versions =
+        get_versions_from_git(&repo, start_semver).context("Failed to get the tag list.")?;
+    let _ = fs::remove_dir_all(TEMP_REPO_DIR);
+    println!("Checking versions: {:?}", versions);
+    let mut failed = false;
+    for version in versions {
+        let mut cmd =
+            Command::cargo_bin("weaver").context("Failed to create the cargo command.")?;
+        let output = cmd
+            .arg("--quiet")
+            .arg("registry")
+            .arg("check")
+            .arg("-r")
+            .arg(format!("{ARCHIVE_URL}{version}.zip[model]"))
+            .timeout(Duration::from_secs(60))
+            .output()
+            .context("Failed to execute the weaver process.")?;
+
+        if output.status.success() {
+            println!("Success for: {}", version);
+        } else {
+            failed = true;
+            println!("Failure for: {}", version);
+            println!(
+                "{}",
+                String::from_utf8(output.stdout).context("Invalid UTF-8")?
+            );
+        }
+    }
+    if failed {
+        anyhow::bail!("Some versions failed the registry check.");
+    }
+    Ok(())
+}

crates/xtask/src/main.rs

@@ -10,6 +10,7 @@
 #![allow(clippy::print_stdout)]
 #![allow(clippy::print_stderr)]
 
+mod history;
 mod validate;
 
 #[cfg(not(tarpaulin_include))]
@@ -20,6 +21,7 @@ fn main() -> anyhow::Result<()> {
         None => print_help(),
         Some(task) => match task.as_str() {
             "validate" => validate::run(),
+            "history" => history::run(std::env::args().nth(2)),
             "help" => print_help(),
             _ => {
                 eprintln!("Unknown task: {}", task);
@@ -38,6 +40,8 @@ Usage: Execute the command using `cargo xtask <task>`, e.g., `cargo xtask valida
 
 Tasks:
   - validate: Validate the entire structure of the weaver project.
+  - history: Run registry check on semconv models within back compatibility range.
+             Optionally provide a start semver e.g. `history 1.29.0`.
 "
     );
     Ok(())

justfile

@@ -22,6 +22,7 @@ pre-push-check:
     cargo clippy --workspace --all-targets -- -D warnings --allow deprecated
     rm -rf crates/weaver_forge/observed_output/*
     cargo nextest run --all
+    cargo xtask history
     # [workaround] removed --all-features due to an issue in one of the dependency in Tantity (zstd-safe)
     # [ToDo LQ] Re-enable --all-features once the issue is resolved
     # cargo doc --workspace --all-features --no-deps --document-private-items