jedarden 9c955822c4 docs(bf-1iefu): categorize tests/fixtures/ generator scripts

Survey completed and categorized all 17 generator scripts across
7 fixture subdirectories.

KEEP (10):
- Active fixture generators for scanned, encoding, forms, security,
  malformed, and vector test fixtures
- All actively maintained (May-July 2025)
- Core infrastructure for OCR, encoding, and security testing

DELETE (5):
- Redundant Python/Rust duplicates (prefer Python for encoding,
  Rust for security)
- Orphaned Makefile dependency file
- Superseded implementations

RELOCATE (2):
- convert_to_scanned.sh → tools/ (general-purpose utility)
- regenerate.sh → tools/ or DELETE (incomplete stub)

Full categorization with rationale: /tmp/generator_categorization.md
Closes: bf-1iefu

2026-07-05 20:12:34 -04:00

3.2 KiB

Raw Blame History

bf-5ucbr: Capture pdftract debug logs on JavaScript PDF

Task Summary

Capture pdftract debug logs on JavaScript PDF fixture and preserve the output for verification.

Execution

PDF File Used

File: tests/fixtures/security/embedded-js.pdf
Size: 1.1K
Contains: JavaScript action with app.alert("pwn") in catalog.openaction

Commands Executed

# Attempt 1: With debug logging and combined output
RUST_LOG=debug cargo run --bin pdftract -- extract tests/fixtures/security/embedded-js.pdf --json - 2>&1 | tee notes/bf-5ucbr-pdftract-debug.log

# Attempt 2: With trace logging, separated stdout/stderr  
RUST_LOG=trace cargo run --bin pdftract -- extract tests/fixtures/security/embedded-js.pdf --json - > /tmp/js_output.json 2> notes/bf-5ucbr-pdftract-debug-stderr.log

Results

pdftract Version: Built from current main branch (dev profile, unoptimized debuginfo)

Execution Status: ✅ SUCCESS - pdftract ran successfully on the JavaScript PDF fixture

JavaScript Detection: Successfully detected 1 JavaScript action:

{
  "javascript_actions": [
    {
      "code_excerpt": "app.alert(\"pwn\")",
      "location": "catalog.openaction"
    }
  ],
  "metadata": {
    "diagnostics": [
      "Detected 1 JavaScript action(s) in PDF document. JavaScript was NOT executed."
    ]
  }
}

Logging Observation: ⚠️ RUST_LOG environment variable does not produce debug output. The pdftract CLI (main.rs) does not initialize any logging framework (env_logger, tracing, etc.). While serve.rs uses tracing for HTTP server logging, the CLI extraction commands don't implement structured logging.

Output Captured:

JSON extraction result: /tmp/js_output.json
Log capture: notes/bf-5ucbr-pdftract-debug.log (27 lines - JSON output only)
Stderr log: notes/bf-5ucbr-pdftract-debug-stderr.log (0 lines - no debug output)

Key Findings

JavaScript Detection Working: pdftract successfully identifies JavaScript in PDFs and reports code excerpts and locations
No Execution: JavaScript is detected but NOT executed (security feature confirmed working)
Logging Gap: The CLI lacks debug logging infrastructure - RUST_LOG has no effect outside of serve mode
Extraction Success: The PDF extraction pipeline completes successfully on JavaScript-containing files

Verification

✅ pdftract runs successfully on JavaScript PDF fixture
✅ Output captured to files (notes/bf-5ucbr-pdftract-debug.log, /tmp/js_output.json)
✅ Debug logging level (RUST_LOG=debug) was attempted (though not implemented in CLI)
✅ Log capture preserved for verification

Recommendations for Future Debug Logging

If debug logging is needed for CLI commands, consider:

Add env_logger::init() in main.rs conditioned on RUST_LOG environment variable
Or migrate CLI to use tracing framework (already used in serve.rs)
Add debug log statements at key extraction pipeline points

Files Generated

notes/bf-5ucbr-pdftract-debug.log - Combined stdout/stderr capture (JSON output)
notes/bf-5ucbr-pdftract-debug-stderr.log - Stderr-only capture (empty - no debug logs)
/tmp/js_output.json - JSON extraction result (temporary file)

3.2 KiB Raw Blame History