src/utils/misc/type.rs (3)

7-10: Regex cleanup: avoid zero-length global matches; fix typo; keep intent.

• ((... )|\s)* can match empty strings; replace_all will still terminate, but this does extra work.
• Rename multi_understore_pattern → multi_underscore_pattern.

Lean regexes with the same behavior:

-    let prefix_pattern = lazy_regex::regex!(r#"(([^,:<>]+::)|\s)*"#);
+    // Remove every `segment::` path prefix anywhere (keeps the final type identifiers)
+    let prefix_pattern = lazy_regex::regex!(r#"(?:[A-Za-z_][A-Za-z0-9_]*)::"#);
@@
-    let multi_understore_pattern = lazy_regex::regex!(r#"_{2,}"#);
+    let multi_underscore_pattern = lazy_regex::regex!(r#"_{2,}"#);
@@
-    let n = multi_understore_pattern.replace_all(n.as_ref(), "_");
+    let n = multi_underscore_pattern.replace_all(n.as_ref(), "_");

Also applies to: 14-14

---

12-18: Harden invariants for Prometheus: non-empty and non-digit-leading names.

If this feeds metric/label names (not just values), enforce Prometheus constraints: only [A-Za-z0-9_], not empty, and not starting with a digit. Current code can yield an empty string for types like () (becomes _ → trimmed to "").

Apply tail-end normalization:

-    n.as_ref()
-        .strip_suffix("_")
-        .unwrap_or(n.as_ref())
-        .to_string()
+    {
+        // Trim both ends (underscores can accumulate at boundaries)
+        let mut s = n.as_ref().trim_matches('_').to_string();
+        if s.is_empty() {
+            s.push_str("Unknown");
+        }
+        if s.starts_with(|c: char| c.is_ascii_digit()) {
+            s.insert(0, '_');
+        }
+        s
+    }

Confirm intended target (metric name vs label value). If it’s a metric or label name, consider an assertion test ensuring:

• starts with [A-Za-z_] (metric name can also start with ‘_’),
• only [A-Za-z0-9_],
• non-empty.

---

30-31: Tests updated appropriately; consider adding invariants coverage.

LGTM on changed expectations.

Add a couple of edge-case tests to lock in Prometheus-safety invariants:

#[test]
fn test_short_type_name_edge_cases() {
    // Unit type should not produce empty
    let u = short_type_name::<()>();
    assert!(!u.is_empty());

    // If used as (label/metric) name, ensure allowed charset and leading char
    let name = short_type_name::<Option<&'static str>>();
    assert!(name.chars().all(|c| c.is_ascii_alphanumeric() || c == '_'));
    // Metric/label name rule (skip if this is only a label value):
    assert!(name.starts_with(|c: char| c.is_ascii_alphabetic() || c == '_'));
}

Also applies to: 34-34, 36-36, 39-39

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

Learnt from: hanabi1224
PR: ChainSafe/forest#5930
File: build.rs:64-77
Timestamp: 2025-08-13T09:43:20.301Z
Learning: hanabi1224 prefers hard compile-time errors in build scripts rather than runtime safeguards or collision detection, believing it's better to fail fast and fix root causes of issues like malformed snapshot names.

src/state_manager/cache.rs (2)

31-33: Sanitize identifiers before constructing metric names.

Call sites currently pass safe literals, but this API allows arbitrary &str. Guard against regressions by normalizing to [a-z0-9_], lowering case, and replacing any other char with '' before composing "tipset_state{...}".

Apply this diff to enforce sanitization at the point of name construction:

-    fn cache_name(cache_identifier: &str) -> String {
-        format!("tipset_state_{cache_identifier}")
-    }
+    fn cache_name(cache_identifier: &str) -> String {
+        let mut out = String::with_capacity("tipset_state_".len() + cache_identifier.len());
+        out.push_str("tipset_state_");
+        for ch in cache_identifier.chars() {
+            match ch {
+                'a'..='z' | '0'..='9' | '_' => out.push(ch),
+                'A'..='Z' => out.push(ch.to_ascii_lowercase()),
+                _ => out.push('_'),
+            }
+        }
+        out
+    }

---

47-58: API change looks good; consider documenting allowed identifier charset.

Add a brief rustdoc noting the identifier should be [a-z0-9_]+ to remain Prometheus-safe.

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

Learnt from: hanabi1224
PR: ChainSafe/forest#5930
File: build.rs:64-77
Timestamp: 2025-08-13T09:43:20.301Z
Learning: hanabi1224 prefers hard compile-time errors in build scripts rather than runtime safeguards or collision detection, believing it's better to fail fast and fix root causes of issues like malformed snapshot names.