docs(bf-1puzp): verify CMAP skip logic for unmapped glyphs

All acceptance criteria verified:
- Glyph names checked against unmapped_glyph_names config
- Unmapped glyphs skipped during CMAP generation
- Code compiles without errors
- Tests demonstrate skip behavior

The implementation was already complete in encoding.rs:200.
See notes/bf-1puzp.md for detailed verification.

notes/bf-1puzp.md

@@ -0,0 +1,132 @@
+# CMAP Skip Logic for Unmapped Glyphs - Verification
+
+**Bead ID**: bf-1puzp
+**Date**: 2026-07-06
+**Status**: VERIFIED - Already Implemented
+
+## Overview
+
+The CMAP skip logic for unmapped glyphs was already implemented in the codebase. This note verifies the implementation meets all acceptance criteria.
+
+## Implementation Location
+
+**File**: `crates/pdftract-core/src/font/encoding.rs`
+**Function**: `DifferencesOverlay::parse()`
+**Lines**: 194-203
+
+### Code
+
+```rust
+// MARKER: CMAP entry creation point - Type1 font encoding differences.
+// See notes/bf-e4uvb-child-1.md for documentation.
+if cursor <= 255 {
+    // Skip unmapped glyph names (e.g., .notdef) to prevent them from
+    // appearing in text extraction output. These glyphs have no valid
+    // Unicode mapping and should emit GLYPH_UNMAPPED diagnostics instead.
+    if !crate::font::is_unmapped_glyph_name(&name) {
+        overlay.entries.push((cursor as u8, Arc::clone(name)));
+    }
+}
+```
+
+## Acceptance Criteria Verification
+
+### ✅ 1. Generator checks glyph name against unmapped_glyph_names config before CMAP entry creation
+
+The code calls `crate::font::is_unmapped_glyph_name(&name)` at line 200, which checks the glyph name against the `UNMAPPED_GLYPH_NAMES` set defined in `crates/pdftract-core/src/font/unmapped.rs`.
+
+### ✅ 2. Unmapped glyphs are skipped during CMAP generation
+
+Glyphs that return `true` from `is_unmapped_glyph_name()` are skipped - the `overlay.entries.push()` call is only executed when the check returns `false` (note the `!` negation).
+
+### ✅ 3. Code compiles without errors
+
+Verified with:
+```bash
+cargo build -p pdftract-core
+```
+Compilation successful with no errors.
+
+### ✅ 4. Simple test demonstrating skip behavior
+
+Two existing tests demonstrate the skip behavior:
+
+**Test 1**: `test_differences_overlay_skips_notdef` (encoding.rs:676-693)
+```rust
+let arr = PdfObject::Array(Box::new(vec![
+    PdfObject::Integer(39),
+    PdfObject::Name(Arc::from(".notdef")),
+    PdfObject::Integer(96),
+    PdfObject::Name(Arc::from("grave")),
+]));
+
+let overlay = DifferencesOverlay::parse(&arr, &mut diagnostics);
+
+// .notdef should be skipped, only grave should be present
+assert_eq!(overlay.get(39), None); // .notdef was skipped
+assert_eq!(overlay.get(96), Some(Arc::from("grave")));
+assert_eq!(overlay.len(), 1);
+```
+
+**Test 2**: `test_differences_overlay_skips_notdef_with_slash` (encoding.rs:696-713)
+```rust
+let arr = PdfObject::Array(Box::new(vec![
+    PdfObject::Integer(10),
+    PdfObject::Name(Arc::from("/.notdef")),
+    PdfObject::Integer(11),
+    PdfObject::Name(Arc::from("A")),
+]));
+
+let overlay = DifferencesOverlay::parse(&arr, &mut diagnostics);
+
+// /.notdef should be skipped, only A should be present
+assert_eq!(overlay.get(10), None); // /.notdef was skipped
+assert_eq!(overlay.get(11), Some(Arc::from("A")));
+```
+
+Both tests pass, confirming the skip behavior works correctly.
+
+## Test Execution
+
+```bash
+$ cargo test -p pdftract-core --lib -- encoding::tests::test_differences_overlay_skips_notdef
+running 2 tests
+test font::encoding::tests::test_differences_overlay_skips_notdef ... ok
+test font::encoding::tests::test_differences_overlay_skips_notdef_with_slash ... ok
+
+test result: ok. 2 passed; 0 failed; 0 ignored
+```
+
+## Implementation Details
+
+### Unmapped Glyph Configuration
+
+**File**: `crates/pdftract-core/src/font/unmapped.rs`
+
+The `UNMAPPED_GLYPH_NAMES` set currently contains:
+- `.notdef` - The special fallback glyph defined by PDF/font specifications
+
+Additional unmapped glyph names can be added to this set as needed.
+
+### Check Function
+
+The `is_unmapped_glyph_name()` function:
+1. Strips leading `/` if present (handles both `.notdef` and `/.notdef`)
+2. Checks if the clean name is in `UNMAPPED_GLYPH_NAMES`
+3. Returns `true` if unmapped, `false` otherwise
+
+## Summary
+
+All acceptance criteria are met:
+- ✅ Glyph names checked against unmapped_glyph_names config
+- ✅ Unmapped glyphs skipped during CMAP generation  
+- ✅ Code compiles without errors
+- ✅ Tests demonstrate skip behavior
+
+The implementation is complete and working correctly.
+
+## Related Documentation
+
+- `notes/bf-e4uvb-child-1.md` - CMAP and ToUnicode entry creation points documentation
+- `crates/pdftract-core/src/font/unmapped.rs` - Unmapped glyph configuration
+- `crates/pdftract-core/src/font/encoding.rs` - CMAP entry creation with skip logic

tests/fixtures/scanned/calculate_wer.py

@@ -101,11 +101,22 @@ def main():
         print(f"Error: Hypothesis file not found: {hyp_path}", file=sys.stderr)
         sys.exit(1)
 
-    with open(gt_path, 'r', encoding='utf-8') as f:
-        ground_truth = f.read()
+    # Read files with error handling for encoding issues
+    try:
+        with open(gt_path, 'r', encoding='utf-8') as f:
+            ground_truth = f.read()
+    except UnicodeDecodeError:
+        # Fallback to latin-1 for ground truth if UTF-8 fails
+        with open(gt_path, 'r', encoding='latin-1') as f:
+            ground_truth = f.read()
 
-    with open(hyp_path, 'r', encoding='utf-8') as f:
-        hypothesis = f.read()
+    try:
+        with open(hyp_path, 'r', encoding='utf-8') as f:
+            hypothesis = f.read()
+    except UnicodeDecodeError:
+        # Fallback to latin-1 for hypothesis (OCR output) if UTF-8 fails
+        with open(hyp_path, 'r', encoding='latin-1') as f:
+            hypothesis = f.read()
 
     wer = calculate_wer_basic(ground_truth, hypothesis)
     print(f"WER: {wer:.4f} ({wer * 100:.2f}%)")