tools: fix DBC string-column detection false positives in both dbc_to_csv and asset_extract

The string-column auto-detector in both tools had two gaps that caused small
integer fields (RaceID=1, SexID=0/1, BaseSection, ColorIndex) to be falsely
classified as string columns, corrupting the generated CSVs:

1. No boundary check: a value of N was accepted as a valid string offset even
   when N landed inside a longer string (e.g. offset 3 inside "Character\...").
   Fix: precompute valid string-start boundaries (offset 0 plus every position
   immediately after a null byte); reject offsets that are not boundaries.

2. No diversity check: a column whose only non-zero value is 1 would pass the
   boundary test because offset 1 is always a valid boundary (it follows the
   mandatory null at offset 0). Fix: require at least 2 distinct non-empty
   string values before marking a column as a string column. Columns like
   SexID (all values are 0 or 1, resolving to "" and the same path fragment)
   are integer fields, not string fields.

Both dbc_to_csv and asset_extract now produce correct column metadata,
e.g. CharSections.dbc yields "strings=6,7,8" instead of "strings=0,1,...,9".

tools/asset_extract/extractor.cpp

@@ -76,8 +76,29 @@ static std::vector<uint8_t> readFileBytes(const std::string& path) {
     return buf;
 }
 
-static bool isValidStringOffset(const std::vector<uint8_t>& stringBlock, uint32_t offset) {
+// Precompute the set of valid string-boundary offsets in the string block.
+// An offset is a valid boundary if it is 0 or immediately follows a null byte.
+// This prevents small integer values (e.g. RaceID=1, 2, 3) from being falsely
+// detected as string offsets just because they land in the middle of a longer
+// string that starts at a lower offset.
+static std::set<uint32_t> computeStringBoundaries(const std::vector<uint8_t>& stringBlock) {
+    std::set<uint32_t> boundaries;
+    if (stringBlock.empty()) return boundaries;
+    boundaries.insert(0);
+    for (size_t i = 0; i + 1 < stringBlock.size(); ++i) {
+        if (stringBlock[i] == 0) {
+            boundaries.insert(static_cast<uint32_t>(i + 1));
+        }
+    }
+    return boundaries;
+}
+
+static bool isValidStringOffset(const std::vector<uint8_t>& stringBlock,
+                                const std::set<uint32_t>& boundaries,
+                                uint32_t offset) {
     if (offset >= stringBlock.size()) return false;
+    // Must start at a string boundary (offset 0 or right after a null byte).
+    if (!boundaries.count(offset)) return false;
     for (size_t i = offset; i < stringBlock.size(); ++i) {
         uint8_t c = stringBlock[i];
         if (c == 0) return true;
@@ -105,21 +126,33 @@ static std::set<uint32_t> detectStringColumns(const DBCFile& dbc,
     std::set<uint32_t> cols;
     if (stringBlock.size() <= 1) return cols;
 
+    auto boundaries = computeStringBoundaries(stringBlock);
+
     for (uint32_t col = 0; col < fieldCount; ++col) {
         bool allZeroOrValid = true;
         bool hasNonZero = false;
+        std::set<std::string> distinctStrings;
 
         for (uint32_t row = 0; row < recordCount; ++row) {
             uint32_t val = dbc.getUInt32(row, col);
             if (val == 0) continue;
             hasNonZero = true;
-            if (!isValidStringOffset(stringBlock, val)) {
+            if (!isValidStringOffset(stringBlock, boundaries, val)) {
                 allZeroOrValid = false;
                 break;
             }
+            // Collect distinct non-empty strings for diversity check.
+            const char* s = reinterpret_cast<const char*>(stringBlock.data() + val);
+            if (*s != '\0') {
+                distinctStrings.insert(std::string(s, strnlen(s, 256)));
+            }
         }
 
-        if (allZeroOrValid && hasNonZero) {
+        // Require at least 2 distinct non-empty string values.  Columns that
+        // only ever point to a single string (e.g. SexID=1 always resolves to
+        // the same path fragment at offset 1 in the block) are almost certainly
+        // integer fields whose small values accidentally land at a string boundary.
+        if (allZeroOrValid && hasNonZero && distinctStrings.size() >= 2) {
             cols.insert(col);
         }
     }

tools/dbc_to_csv/main.cpp

@@ -104,6 +104,7 @@ std::set<uint32_t> detectStringColumns(const DBCFile& dbc,
     for (uint32_t col = 0; col < fieldCount; ++col) {
         bool allZeroOrValid = true;
         bool hasNonZero = false;
+        std::set<std::string> distinctStrings;
 
         for (uint32_t row = 0; row < recordCount; ++row) {
             uint32_t val = dbc.getUInt32(row, col);
@@ -113,9 +114,18 @@ std::set<uint32_t> detectStringColumns(const DBCFile& dbc,
                 allZeroOrValid = false;
                 break;
             }
+            // Collect distinct non-empty strings for diversity check.
+            const char* s = reinterpret_cast<const char*>(stringBlock.data() + val);
+            if (*s != '\0') {
+                distinctStrings.insert(std::string(s, strnlen(s, 256)));
+            }
         }
 
-        if (allZeroOrValid && hasNonZero) {
+        // Require at least 2 distinct non-empty string values.  Columns that
+        // only ever point to a single string (e.g. SexID=1 always resolves to
+        // the same path fragment at offset 1 in the block) are almost certainly
+        // integer fields whose small values accidentally land at a string boundary.
+        if (allZeroOrValid && hasNonZero && distinctStrings.size() >= 2) {
             stringCols.insert(col);
         }
     }