Clippy warn `uninlined_format_args` & `redundant_else` by ShaharNaveh · Pull Request #7873 · RustPython/RustPython

crates/stdlib/src/unicodedata.rs (1)

111-118: ⚠️ Potential issue | 🔴 Critical | ⚡ Quick win

Dead code: the Some branch at line 113 is unreachable.

The match at lines 112-118 is redundant. If result.downcast_ref::<PyComplex>() succeeds, the code returns early at line 109. Therefore, by the time execution reaches line 112, result cannot be a PyComplex, and the match will always take the None branch. The Some(complex_obj) arm is dead code.

Remove this entire match block—the error case it's trying to handle (non-complex return from __complex__) is already covered by the fallthrough to line 122 and beyond.

🐛 Proposed fix: remove redundant match

             return Ok(Some((ret.value, true)));
         }
-
-        return match result.downcast_ref::<PyComplex>() {
-            Some(complex_obj) => Ok(Some((complex_obj.value, true))),
-            None => Err(vm.new_type_error(format!(
-                "__complex__ returned non-complex (type '{}')",
-                result.class().name()
-            ))),
-        };
     }
     // `complex` does not have a `__complex__` by default, so subclasses might not either,

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@crates/stdlib/src/unicodedata.rs` around lines 111 - 118, Remove the
redundant match over result.downcast_ref::<PyComplex>() and its
Some(complex_obj) arm: the code path that handles PyComplex already returns
early (via the earlier downcast) so the match's Some branch is unreachable;
delete the entire match block and keep the existing fallthrough/error handling
after it (i.e., rely on the prior early return and the subsequent non-complex
handling), ensuring references to result and c remain consistent in the
enclosing function.

crates/stdlib/src/binascii.rs (1)

690-694: ⚠️ Potential issue | 🔴 Critical | ⚡ Quick win

rlecode_hqx exits early and truncates output on first RUN_CHAR.

The return Ok(out_data); in the RUN_CHAR escape path stops processing the rest of the buffer, so encoded output is incomplete whenever 0x90 appears before the end.

Suggested fix

                 if ch == RUN_CHAR {
                     out_data.push(RUN_CHAR);
                     out_data.push(0);
-                    return Ok(out_data);
+                    idx += 1;
+                    continue;
                 }

Also applies to: 696-709

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@crates/stdlib/src/binascii.rs` around lines 690 - 694, The rlecode_hqx
implementation prematurely returns from the RUN_CHAR escape path, truncating the
encoded output; in the rlecode_hqx function, replace the early "return
Ok(out_data)" in the branch that handles RUN_CHAR (and mirror the same fix for
the similar block around lines 696–709) so that after pushing RUN_CHAR and the
zero sentinel you continue processing the remainder of the input instead of
returning—i.e., remove the early return and let the function loop/process the
remaining bytes, preserving existing behavior for run-length encoding and
ensuring out_data is returned only after the whole input is processed.

crates/vm/src/builtins/complex.rs (1)

97-118: ⚠️ Potential issue | 🟠 Major | 🏗️ Heavy lift

Unreachable code: the match on line 112 is redundant.

The downcast_ref::<PyComplex>() call on line 97 already checks if result is a PyComplex. If it matches, the method returns on line 109. Therefore, when execution reaches the match on line 112, the same downcast_ref call will always return None (because it would have returned Some on line 97 and exited early).

This means:

• The Some(complex_obj) branch on line 113 is unreachable
• The None branch on line 114-117 will always execute after line 97 fails

This suggests incomplete refactoring. If the intent is to handle subclasses differently from exact types, consider using downcast_ref_if_exact for one of the checks. Otherwise, remove the redundant match and raise the TypeError directly after line 110.

Proposed fix: remove redundant match and raise TypeError directly

             if let Some(ret) = result.downcast_ref::<PyComplex>() {
                 _warnings::warn(
                     vm.ctx.exceptions.deprecation_warning,
                     format!(
                         "__complex__ returned non-complex (type {ret_class}).  \
                     The ability to return an instance of a strict subclass of complex \
                     is deprecated, and may be removed in a future version of Python."
                     ),
                     1,
                     vm,
                 )?;

                 return Ok(Some((ret.value, true)));
             }
-
-            return match result.downcast_ref::<PyComplex>() {
-                Some(complex_obj) => Ok(Some((complex_obj.value, true))),
-                None => Err(vm.new_type_error(format!(
-                    "__complex__ returned non-complex (type '{}')",
-                    result.class().name()
-                ))),
-            };
+
+            return Err(vm.new_type_error(format!(
+                "__complex__ returned non-complex (type '{}')",
+                result.class().name()
+            )));
         }

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@crates/vm/src/builtins/complex.rs` around lines 97 - 118, The code currently
calls result.downcast_ref::<PyComplex>() twice in the __complex__ result
handling in complex.rs, making the second match unreachable; remove the
redundant match and return the TypeError directly after the warning path.
Specifically, inside the block that checks result.downcast_ref::<PyComplex>()
and emits the deprecation warning, keep the Ok(Some((ret.value, true))) return
for the exact-match case and replace the subsequent match on
result.downcast_ref::<PyComplex>() with a direct Err(vm.new_type_error(...)) to
raise "__complex__ returned non-complex (type '{}')" using
result.class().name().