path: fix unicode path problems in path.relative

monoblaine · monoblaine · commit 9acdfcde5810 · 2020-07-08T00:43:09.000+03:00
This commit changes the way two paths are compared in path.relative: Instead of comparing each char code in path strings one by one, which causes problems when the number of char codes in lowercased path string does not match the original one (e.g. path contains certain Unicode characters like 'İ'), it now splits the path string by backslash and compares the parts instead. Fixes: #27534
diff --git a/lib/path.js b/lib/path.js
@@ -464,86 +464,67 @@ const win32 = {
            from.charCodeAt(fromEnd - 1) === CHAR_BACKWARD_SLASH) {
       fromEnd--;
     }
-    const fromLen = fromEnd - fromStart;
+    if (fromStart > 0 || fromEnd < from.length)
+      from = from.slice(fromStart, fromEnd);
 
     // Trim any leading backslashes
+    const toLenBeforeTrim = to.length;
     let toStart = 0;
-    while (toStart < to.length &&
+    while (toStart < toLenBeforeTrim &&
            to.charCodeAt(toStart) === CHAR_BACKWARD_SLASH) {
       toStart++;
     }
     // Trim trailing backslashes (applicable to UNC paths only)
-    let toEnd = to.length;
+    let toEnd = toLenBeforeTrim;
     while (toEnd - 1 > toStart &&
            to.charCodeAt(toEnd - 1) === CHAR_BACKWARD_SLASH) {
       toEnd--;
     }
-    const toLen = toEnd - toStart;
+    if (toStart > 0 || toEnd < toLenBeforeTrim)
+      to = to.slice(toStart, toEnd);
+
+    // Even the device roots are different. Return the original `to`.
+    if (from.charCodeAt(0) !== to.charCodeAt(0))
+      return toOrig;
+
+    // Split paths
+    const fromParts = from.split('\\');
+    const toParts = to.split('\\');
 
-    // Compare paths to find the longest common path from root
-    const length = fromLen < toLen ? fromLen : toLen;
-    let lastCommonSep = -1;
     let i = 0;
-    for (; i < length; i++) {
-      const fromCode = from.charCodeAt(fromStart + i);
-      if (fromCode !== to.charCodeAt(toStart + i))
-        break;
-      else if (fromCode === CHAR_BACKWARD_SLASH)
-        lastCommonSep = i;
-    }
+    const partLen = fromParts.length < toParts.length ?
+      fromParts.length :
+      toParts.length;
 
-    // We found a mismatch before the first common path separator was seen, so
-    // return the original `to`.
-    if (i !== length) {
-      if (lastCommonSep === -1)
-        return toOrig;
-    } else {
-      if (toLen > length) {
-        if (to.charCodeAt(toStart + i) === CHAR_BACKWARD_SLASH) {
-          // We get here if `from` is the exact base path for `to`.
-          // For example: from='C:\\foo\\bar'; to='C:\\foo\\bar\\baz'
-          return toOrig.slice(toStart + i + 1);
-        }
-        if (i === 2) {
-          // We get here if `from` is the device root.
-          // For example: from='C:\\'; to='C:\\foo'
-          return toOrig.slice(toStart + i);
-        }
-      }
-      if (fromLen > length) {
-        if (from.charCodeAt(fromStart + i) === CHAR_BACKWARD_SLASH) {
-          // We get here if `to` is the exact base path for `from`.
-          // For example: from='C:\\foo\\bar'; to='C:\\foo'
-          lastCommonSep = i;
-        } else if (i === 2) {
-          // We get here if `to` is the device root.
-          // For example: from='C:\\foo\\bar'; to='C:\\'
-          lastCommonSep = 3;
-        }
-      }
-      if (lastCommonSep === -1)
-        lastCommonSep = 0;
-    }
+    // Find the number of common ancestors
+    while (i < partLen && fromParts[i] === toParts[i])
+      i++;
 
+    const numOfCommonAncestors = i;
+    let numOfStepsToWalkUp = fromParts.length - numOfCommonAncestors;
     let out = '';
-    // Generate the relative path based on the path difference between `to` and
-    // `from`
-    for (i = fromStart + lastCommonSep + 1; i <= fromEnd; ++i) {
-      if (i === fromEnd || from.charCodeAt(i) === CHAR_BACKWARD_SLASH) {
-        out += out.length === 0 ? '..' : '\\..';
-      }
-    }
 
-    toStart += lastCommonSep;
+    // Walk up the directory tree to get the nearest common ancestor (if any)
+    // and then append the "different" parts (if any)
+    while (numOfStepsToWalkUp-- > 0)
+      out += out.length === 0 ? '..' : '\\..';
 
-    // Lastly, append the rest of the destination (`to`) path that comes after
-    // the common path parts
-    if (out.length > 0)
-      return `${out}${toOrig.slice(toStart, toEnd)}`;
+    let partIx = 0;
+
+    // Find the starting position of the remaining part
+    const toOrigEnd = toOrig.length - toLenBeforeTrim + toEnd;
+    for (i = toStart; partIx < numOfCommonAncestors && i <= toOrigEnd; i++)
+      if (toOrig.charCodeAt(i) === CHAR_BACKWARD_SLASH)
+        partIx++;
+
+    if (partIx > 0 && i <= toOrigEnd) {
+      if (out.length > 0)
+        out += '\\';
+
+      out += toOrig.slice(i, toOrigEnd);
+    }
 
-    if (toOrig.charCodeAt(toStart) === CHAR_BACKWARD_SLASH)
-      ++toStart;
-    return toOrig.slice(toStart, toEnd);
+    return out;
   },
 
   toNamespacedPath(path) {
diff --git a/test/parallel/test-path-relative.js b/test/parallel/test-path-relative.js
@@ -31,7 +31,9 @@ const relativeTests = [
      ['\\\\foo\\baz-quux', '\\\\foo\\baz', '..\\baz'],
      ['\\\\foo\\baz', '\\\\foo\\baz-quux', '..\\baz-quux'],
      ['C:\\baz', '\\\\foo\\bar\\baz', '\\\\foo\\bar\\baz'],
-     ['\\\\foo\\bar\\baz', 'C:\\baz', 'C:\\baz']
+     ['\\\\foo\\bar\\baz', 'C:\\baz', 'C:\\baz'],
+     ['c:\\a\\İ', 'c:\\a\\İ\\test.txt', 'test.txt'],
+     ['c:\\İ\\a\\İ', 'c:\\İ\\b\\İ\\test.txt', '..\\..\\b\\İ\\test.txt']
     ]
   ],
   [ path.posix.relative,
