java-topology/tools/scans/duckdb.sh
russell@unturf.com db29a08762 undefect. CWE-407 — 92 sites, 42 ecosystems
B&W print-friendly diagrams + tinkerpop-0001 + wave-3 proof sections.
Squash of 94 local commits onto remote master.
2026-03-26 19:48:18 -04:00

39 lines
1.6 KiB
Bash
Executable file

#!/usr/bin/env bash
# CWE-407 scan — DuckDB (analytical in-process database)
# C++ query optimizer with join ordering, expression analysis, DAG execution
# Strategy: C++ membership patterns (std::find, .count() on vector) in graph/plan context
set -euo pipefail
echo "# scan=duckdb host=$(hostname) date=$(date -u +%Y-%m-%dT%H:%M:%SZ)"
cd /tmp && git clone --depth 1 https://github.com/duckdb/duckdb duckdb 2>&1 | tail -1
# O(n) membership: std::find on vector, or .count() on std::map (O(log n) but worth flagging)
CPP_MEM='std::find\s*\(|\.find\s*\(.*\.end\s*\(\)|vector.*\.count\s*\(|\bfind_if\s*\('
# Loop constructs
CPP_LOOP='for\s*\(|while\s*\(|for\s*\(auto\b'
for dir in \
src/optimizer \
src/planner \
src/execution \
src/common/graph; do
[ -d /tmp/duckdb/$dir ] || continue
echo "# roots: /tmp/duckdb/$dir"
find /tmp/duckdb/$dir -name "*.cpp" -o -name "*.hpp" 2>/dev/null | sort | while IFS= read -r f; do
ml=$(grep -nE "$CPP_MEM" "$f" 2>/dev/null | cut -d: -f1 || true)
[ -z "$ml" ] && continue
while IFS= read -r ln; do
s=$(( ln > 20 ? ln - 20 : 1 )); e=$(( ln + 20 ))
ctx=$(sed -n "${s},${e}p" "$f" 2>/dev/null || true)
graph_hit=$(echo "$ctx" | grep -iE 'graph|depend|topo|visit|cycle|dag|sort|stack|order|reachab|ancestor|transit|join|plan' | head -1 | sed 's/^\s*//' || true)
[ -z "$graph_hit" ] && continue
loop_hit=$(echo "$ctx" | grep -iE "$CPP_LOOP" | head -1 | sed 's/^\s*//' || true)
[ -n "$loop_hit" ] && {
mem_line=$(sed -n "${ln}p" "$f" | sed 's/^\s*//')
echo "CANDIDATE\t${f}:${ln}\t${mem_line} [loop: ${loop_hit}]"
}
done <<< "$ml"
done
done
echo "# scan complete"