feat(rust): make PyO3 optional to fix Python 3.14 fuzz linking (#270)

vinitkumar · ampcode-com · web-flow · commit 94a60ab0e896 · 2026-01-28T16:18:08.000+05:30
* feat(rust): make PyO3 optional to fix Python 3.14 fuzz linking - Make pyo3 dependency optional behind 'python' feature flag - Add rlib crate-type to support both cdylib (Python) and rlib (fuzzing) - Gate all Python-specific code with #[cfg(feature = "python")] - Make pure Rust functions (escape_xml, wrap_cdata, etc.) public for fuzzing - Add comprehensive unit tests for XML utility functions Fixes linker error with Python 3.14 where PyUnicode_DATA and PyUnicode_KIND symbols are now inline macros, not exported functions. Fuzz targets can now build without linking against Python. Amp-Thread-ID: https://ampcode.com/threads/T-019c0425-ac62-76d8-9d59-4d6aba3edf45 Co-authored-by: Amp <amp@ampcode.com> * ci: add Rust unit tests to CI workflow Amp-Thread-ID: https://ampcode.com/threads/T-019c0425-ac62-76d8-9d59-4d6aba3edf45 Co-authored-by: Amp <amp@ampcode.com> * refactor: reformat code with rust fmt * fix: improve fuzz test assertion for make_attr_string Use more specific pattern matching (` key="`) instead of just checking if key exists as substring. This avoids false positives with overlapping keys (e.g. 'a' vs 'aa') or malformed attribute formatting. Suggested-by: sourcery-ai Amp-Thread-ID: https://ampcode.com/threads/T-019c0425-ac62-76d8-9d59-4d6aba3edf45 Co-authored-by: Amp <amp@ampcode.com> --------- Co-authored-by: Amp <amp@ampcode.com>
diff --git a/.github/workflows/rust-ci.yml b/.github/workflows/rust-ci.yml
@@ -38,6 +38,10 @@ jobs:
         working-directory: rust
         run: cargo clippy --all-targets --all-features -- -D warnings
 
+      - name: Run Rust unit tests
+        working-directory: rust
+        run: cargo test --no-default-features
+
   rust-test:
     name: Build & Test (${{ matrix.os }}, Python ${{ matrix.python-version }})
     runs-on: ${{ matrix.os }}
diff --git a/rust/Cargo.toml b/rust/Cargo.toml
@@ -7,10 +7,14 @@ license = "Apache-2.0"
 
 [lib]
 name = "json2xml_rs"
-crate-type = ["cdylib"]
+crate-type = ["cdylib", "rlib"]
+
+[features]
+default = ["python"]
+python = ["pyo3/extension-module", "dep:pyo3"]
 
 [dependencies]
-pyo3 = { version = "0.27", features = ["extension-module"] }
+pyo3 = { version = "0.27", optional = true }
 
 [profile.release]
 lto = true
diff --git a/rust/fuzz/Cargo.toml b/rust/fuzz/Cargo.toml
@@ -0,0 +1,51 @@
+[package]
+name = "json2xml_rs-fuzz"
+version = "0.0.0"
+publish = false
+edition = "2021"
+
+[package.metadata]
+cargo-fuzz = true
+
+[dependencies]
+libfuzzer-sys = "0.4"
+arbitrary = { version = "1", features = ["derive"] }
+
+[dependencies.json2xml_rs]
+path = ".."
+default-features = false
+
+[[bin]]
+name = "fuzz_escape_xml"
+path = "fuzz_targets/fuzz_escape_xml.rs"
+test = false
+doc = false
+bench = false
+
+[[bin]]
+name = "fuzz_wrap_cdata"
+path = "fuzz_targets/fuzz_wrap_cdata.rs"
+test = false
+doc = false
+bench = false
+
+[[bin]]
+name = "fuzz_is_valid_xml_name"
+path = "fuzz_targets/fuzz_is_valid_xml_name.rs"
+test = false
+doc = false
+bench = false
+
+[[bin]]
+name = "fuzz_make_valid_xml_name"
+path = "fuzz_targets/fuzz_make_valid_xml_name.rs"
+test = false
+doc = false
+bench = false
+
+[[bin]]
+name = "fuzz_make_attr_string"
+path = "fuzz_targets/fuzz_make_attr_string.rs"
+test = false
+doc = false
+bench = false
diff --git a/rust/fuzz/corpus/fuzz_escape_xml/71853c6197a6a7f222db0f1978c7cb232b87c5ee b/rust/fuzz/corpus/fuzz_escape_xml/71853c6197a6a7f222db0f1978c7cb232b87c5ee
@@ -0,0 +1,2 @@
+
+
diff --git a/rust/fuzz/corpus/fuzz_escape_xml/a70fb6a9415d3c34a9ea5510ba7e3a055d88aef9 b/rust/fuzz/corpus/fuzz_escape_xml/a70fb6a9415d3c34a9ea5510ba7e3a055d88aef9
@@ -0,0 +1 @@
+�
diff --git a/rust/fuzz/corpus/fuzz_make_attr_string/23dbd0963ab0e20509f61d3bbab1aab8c0acc1ce b/rust/fuzz/corpus/fuzz_make_attr_string/23dbd0963ab0e20509f61d3bbab1aab8c0acc1ce
@@ -0,0 +1 @@
+]I]]
diff --git a/rust/fuzz/corpus/fuzz_make_attr_string/2cac163ebc8063f5adb15a7ede5fab5c2a9b9e23 b/rust/fuzz/corpus/fuzz_make_attr_string/2cac163ebc8063f5adb15a7ede5fab5c2a9b9e23
@@ -0,0 +1 @@
+����
diff --git a/rust/fuzz/corpus/fuzz_make_attr_string/55cdfedcb4f4369cf383ac30deee62d55cbea7f3 b/rust/fuzz/corpus/fuzz_make_attr_string/55cdfedcb4f4369cf383ac30deee62d55cbea7f3
@@ -0,0 +1 @@
+]]J(
diff --git a/rust/fuzz/corpus/fuzz_make_attr_string/71261c370e317f7ebcb6d903dba6f5dc1396384a b/rust/fuzz/corpus/fuzz_make_attr_string/71261c370e317f7ebcb6d903dba6f5dc1396384a
@@ -0,0 +1 @@
+]](
diff --git a/rust/fuzz/corpus/fuzz_make_attr_string/8206f3e251d21e9c9a1b29ff459c7fb6addd8ae6 b/rust/fuzz/corpus/fuzz_make_attr_string/8206f3e251d21e9c9a1b29ff459c7fb6addd8ae6
@@ -0,0 +1 @@
+]J(
diff --git a/rust/fuzz/corpus/fuzz_make_attr_string/de603c91038f329cce1cca8a30ea161b2271e2f6 b/rust/fuzz/corpus/fuzz_make_attr_string/de603c91038f329cce1cca8a30ea161b2271e2f6
@@ -0,0 +1 @@
+]
diff --git a/rust/fuzz/fuzz_targets/fuzz_escape_xml.rs b/rust/fuzz/fuzz_targets/fuzz_escape_xml.rs
@@ -0,0 +1,23 @@
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use json2xml_rs::escape_xml;
+
+fuzz_target!(|data: &str| {
+    let result = escape_xml(data);
+    
+    // Verify invariants:
+    // 1. Result should not contain unescaped special chars
+    assert!(!result.contains('&') || result.contains("&amp;") || result.contains("&quot;") 
+            || result.contains("&apos;") || result.contains("&lt;") || result.contains("&gt;"));
+    
+    // 2. Result should be valid (no panics occurred)
+    // 3. If input had no special chars, output equals input
+    if !data.contains('&') && !data.contains('"') && !data.contains('\'') 
+       && !data.contains('<') && !data.contains('>') {
+        assert_eq!(result, data);
+    }
+    
+    // 4. Output length should be >= input length (escaping only adds chars)
+    assert!(result.len() >= data.len());
+});
diff --git a/rust/fuzz/fuzz_targets/fuzz_is_valid_xml_name.rs b/rust/fuzz/fuzz_targets/fuzz_is_valid_xml_name.rs
@@ -0,0 +1,33 @@
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use json2xml_rs::is_valid_xml_name;
+
+fuzz_target!(|data: &str| {
+    let result = is_valid_xml_name(data);
+    
+    // Verify invariants:
+    // 1. Empty string is always invalid
+    if data.is_empty() {
+        assert!(!result);
+    }
+    
+    // 2. String starting with digit is invalid
+    if let Some(first) = data.chars().next() {
+        if first.is_ascii_digit() {
+            assert!(!result);
+        }
+    }
+    
+    // 3. String starting with "xml" (case-insensitive) is invalid
+    if data.to_lowercase().starts_with("xml") {
+        assert!(!result);
+    }
+    
+    // 4. String containing spaces is invalid
+    if data.contains(' ') {
+        assert!(!result);
+    }
+    
+    // 5. Function should never panic - reaching here means it didn't
+});
diff --git a/rust/fuzz/fuzz_targets/fuzz_make_attr_string.rs b/rust/fuzz/fuzz_targets/fuzz_make_attr_string.rs
@@ -0,0 +1,42 @@
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use arbitrary::Arbitrary;
+use json2xml_rs::make_attr_string;
+
+#[derive(Arbitrary, Debug)]
+struct AttrInput {
+    attrs: Vec<(String, String)>,
+}
+
+fuzz_target!(|input: AttrInput| {
+    let result = make_attr_string(&input.attrs);
+    
+    // Verify invariants:
+    // 1. Empty attrs should produce empty string
+    if input.attrs.is_empty() {
+        assert!(result.is_empty());
+        return;
+    }
+    
+    // 2. Result should start with space (for XML formatting)
+    assert!(result.starts_with(' '), "Attribute string should start with space");
+    
+    // 3. Each attribute should produce a ` key="value"`-like fragment.
+    //    We check for the more specific pattern ` {key}="` to avoid
+    //    passing on overlapping keys (e.g. "a" vs "aa") or malformed formatting.
+    for (key, _value) in &input.attrs {
+        let expected_fragment = format!(" {}=\"", key);
+        assert!(
+            result.contains(&expected_fragment),
+            "Attribute fragment '{}' should appear in result '{}'",
+            expected_fragment,
+            result
+        );
+    }
+    
+    // 4. Values should be escaped (no raw & < > " ' in values)
+    //    The make_attr_string calls escape_xml on values
+    
+    // 5. Function should never panic - reaching here means it didn't
+});
diff --git a/rust/fuzz/fuzz_targets/fuzz_make_valid_xml_name.rs b/rust/fuzz/fuzz_targets/fuzz_make_valid_xml_name.rs
@@ -0,0 +1,30 @@
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use json2xml_rs::make_valid_xml_name;
+
+fuzz_target!(|data: &str| {
+    let (name, attr) = make_valid_xml_name(data);
+    
+    // Verify invariants:
+    // 1. The returned name must be a valid XML name OR be "key" with an attribute
+    if name != "key" {
+        // If we didn't fall back to "key", the name should be valid
+        // (though it might have been transformed)
+        assert!(!name.is_empty(), "Name should not be empty");
+    }
+    
+    // 2. If attr is Some, name should be "key"
+    if attr.is_some() {
+        assert_eq!(name, "key", "Fallback name should be 'key'");
+        let (attr_name, _attr_value) = attr.unwrap();
+        assert_eq!(attr_name, "name", "Attribute key should be 'name'");
+    }
+    
+    // 3. Purely numeric input should get 'n' prefix
+    if !data.is_empty() && data.chars().all(|c| c.is_ascii_digit()) {
+        assert!(name.starts_with('n'), "Numeric keys should get 'n' prefix");
+    }
+    
+    // 4. Function should never panic - reaching here means it didn't
+});
diff --git a/rust/fuzz/fuzz_targets/fuzz_wrap_cdata.rs b/rust/fuzz/fuzz_targets/fuzz_wrap_cdata.rs
@@ -0,0 +1,21 @@
+#![no_main]
+
+use libfuzzer_sys::fuzz_target;
+use json2xml_rs::wrap_cdata;
+
+fuzz_target!(|data: &str| {
+    let result = wrap_cdata(data);
+    
+    // Verify invariants:
+    // 1. Result must start with CDATA opening
+    assert!(result.starts_with("<![CDATA["));
+    
+    // 2. Result must end with CDATA closing
+    assert!(result.ends_with("]]>"));
+    
+    // 3. The ]]> sequence in input must be properly escaped
+    //    (split into ]]]]><![CDATA[>)
+    
+    // 4. Result should be longer than or equal to input + CDATA wrapper (12 chars)
+    assert!(result.len() >= data.len() + 12);
+});
diff --git a/rust/src/lib.rs b/rust/src/lib.rs
