@sanity/diff-match-patch - npm Package Compare versions

Comparing version 3.0.0 to 3.0.1

dist/index.d.ts

@@ -14,14 +14,2 @@ /**
 /**
- * Takes a `patches` array as produced by diff-match-patch and adjusts the
- * `start1` and `start2` properties so that they refer to UTF-8 index instead
- * of a UCS-2 index.
- *
- * @param patches - The patches to adjust
- * @param base - The base string to use for counting bytes
- * @returns A new array of patches with adjusted indicies
- * @beta
- */
-export declare function adjustIndiciesToUtf8(patches: Patch[], base: string): Patch[]
-
-/**
  * Merge a set of patches onto the text. Returns patched text, as well as a
@@ -221,2 +209,4 @@ * list of true/false values indicating which patches were applied.
   start2: number
+  utf8Start1: number
+  utf8Start2: number
   length1: number
@@ -238,3 +228,3 @@ length2: number
 /**
- * Create a textual representation of a
+ * Create a textual representation of a patch.
  *
@@ -241,0 +231,0 @@ * @param patch - Patch to stringify

dist/index.js

@@ -956,2 +956,4 @@ function isHighSurrogate(char) {
     start2,
+    utf8Start1: start1,
+    utf8Start2: start2,
     length1: 0,
@@ -992,35 +994,2 @@ length2: 0,
 }
-function adjustIndiciesToUtf8(patches, base) {
-  let byteOffset = 0;
-  let idx = 0;
-  function advanceTo(target) {
-    for (; idx < target; idx++) {
-      const codePoint = base.codePointAt(idx);
-      if (typeof codePoint === "undefined") {
-        throw new Error("Failed to get codepoint");
-      }
-      byteOffset += utf8len(codePoint);
-      if (codePoint > 65535) {
-        idx++;
-      }
-    }
-    if (idx !== target) {
-      throw new Error("Failed to determine byte offset");
-    }
-    return byteOffset;
-  }
-  const adjusted = [];
-  for (const patch of patches) {
-    adjusted.push({
-      diffs: patch.diffs.map(diff => cloneDiff(diff)),
-      start1: advanceTo(patch.start1),
-      start2: advanceTo(patch.start2),
-      length1: patch.length1,
-      length2: patch.length2,
-      utf8Length1: patch.utf8Length1,
-      utf8Length2: patch.utf8Length2
-    });
-  }
-  return adjusted;
-}
 function adjustIndiciesToUcs2(patches, base) {
@@ -1033,3 +1002,3 @@ let byteOffset = 0;
       if (typeof codePoint === "undefined") {
-        throw new Error("Failed to get codepoint");
+        return idx;
       }
@@ -1054,2 +1023,4 @@ byteOffset += utf8len(codePoint);
       start2: advanceTo(patch.start2),
+      utf8Start1: patch.utf8Start1,
+      utf8Start2: patch.utf8Start2,
       length1: patch.length1,
@@ -1109,7 +1080,9 @@ length2: patch.length2,
   let charCount2 = 0;
+  let utf8Count1 = 0;
+  let utf8Count2 = 0;
   let prepatchText = textA;
   let postpatchText = textA;
   for (let x = 0; x < diffs.length; x++) {
-    const diffType = diffs[x][0];
-    const diffText = diffs[x][1];
+    const currentDiff = diffs[x];
+    const [diffType, diffText] = currentDiff;
     const diffTextLength = diffText.length;
@@ -1120,6 +1093,8 @@ const diffByteLength = countUtf8Bytes(diffText);
       patch.start2 = charCount2;
+      patch.utf8Start1 = utf8Count1;
+      patch.utf8Start2 = utf8Count2;
     }
     switch (diffType) {
       case DIFF_INSERT:
-        patch.diffs[patchDiffLength++] = diffs[x];
+        patch.diffs[patchDiffLength++] = currentDiff;
         patch.length2 += diffTextLength;
@@ -1132,3 +1107,3 @@ patch.utf8Length2 += diffByteLength;
         patch.utf8Length1 += diffByteLength;
-        patch.diffs[patchDiffLength++] = diffs[x];
+        patch.diffs[patchDiffLength++] = currentDiff;
         postpatchText = postpatchText.substring(0, charCount2) + postpatchText.substring(charCount2 + diffTextLength);
@@ -1138,3 +1113,3 @@ break;
         if (diffTextLength <= 2 * options.margin && patchDiffLength && diffs.length !== x + 1) {
-          patch.diffs[patchDiffLength++] = diffs[x];
+          patch.diffs[patchDiffLength++] = currentDiff;
           patch.length1 += diffTextLength;
@@ -1152,2 +1127,3 @@ patch.length2 += diffTextLength;
             charCount1 = charCount2;
+            utf8Count1 = utf8Count2;
           }
@@ -1161,5 +1137,7 @@ }
       charCount1 += diffTextLength;
+      utf8Count1 += diffByteLength;
     }
     if (diffType !== DIFF_DELETE) {
       charCount2 += diffTextLength;
+      utf8Count2 += diffByteLength;
     }
@@ -1171,3 +1149,3 @@ }
   }
-  return adjustIndiciesToUtf8(patches, textA);
+  return patches;
 }
@@ -1193,2 +1171,4 @@ function addContext(patch, text, opts) {
   }
+  const prefixLength = prefix.length;
+  const prefixUtf8Length = countUtf8Bytes(prefix);
   let suffixEnd = patch.start2 + patch.length1 + padding;
@@ -1202,8 +1182,12 @@ if (suffixEnd < text.length && isLowSurrogate(text[suffixEnd])) {
   }
-  patch.start1 -= prefix.length;
-  patch.start2 -= prefix.length;
-  patch.length1 += prefix.length + suffix.length;
-  patch.length2 += prefix.length + suffix.length;
-  patch.utf8Length1 += prefix.length + suffix.length;
-  patch.utf8Length2 += prefix.length + suffix.length;
+  const suffixLength = suffix.length;
+  const suffixUtf8Length = countUtf8Bytes(suffix);
+  patch.start1 -= prefixLength;
+  patch.start2 -= prefixLength;
+  patch.utf8Start1 -= prefixUtf8Length;
+  patch.utf8Start2 -= prefixUtf8Length;
+  patch.length1 += prefixLength + suffixLength;
+  patch.length2 += prefixLength + suffixLength;
+  patch.utf8Length1 += prefixUtf8Length + suffixUtf8Length;
+  patch.utf8Length2 += prefixUtf8Length + suffixUtf8Length;
 }
@@ -1270,2 +1254,4 @@ function levenshtein(diffs) {
     p.start2 += paddingLength;
+    p.utf8Start1 += paddingLength;
+    p.utf8Start2 += paddingLength;
   }
@@ -1278,2 +1264,4 @@ let patch = patches[0];
     patch.start2 -= paddingLength;
+    patch.utf8Start1 -= paddingLength;
+    patch.utf8Start2 -= paddingLength;
     patch.length1 += paddingLength;
@@ -1284,6 +1272,9 @@ patch.length2 += paddingLength;
   } else if (paddingLength > diffs[0][1].length) {
-    const extraLength = paddingLength - diffs[0][1].length;
-    diffs[0][1] = nullPadding.substring(diffs[0][1].length) + diffs[0][1];
+    const firstDiffLength = diffs[0][1].length;
+    const extraLength = paddingLength - firstDiffLength;
+    diffs[0][1] = nullPadding.substring(firstDiffLength) + diffs[0][1];
     patch.start1 -= extraLength;
     patch.start2 -= extraLength;
+    patch.utf8Start1 -= extraLength;
+    patch.utf8Start2 -= extraLength;
     patch.length1 += extraLength;
@@ -1482,4 +1473,4 @@ patch.length2 += extraLength;
     utf8Length2,
-    start1,
-    start2,
+    utf8Start1,
+    utf8Start2,
     diffs
@@ -1489,15 +1480,15 @@ } = patch;
   if (utf8Length1 === 0) {
-    coords1 = "".concat(start1, ",0");
+    coords1 = "".concat(utf8Start1, ",0");
   } else if (utf8Length1 === 1) {
-    coords1 = "".concat(start1 + 1);
+    coords1 = "".concat(utf8Start1 + 1);
   } else {
-    coords1 = "".concat(start1 + 1, ",").concat(utf8Length1);
+    coords1 = "".concat(utf8Start1 + 1, ",").concat(utf8Length1);
   }
   let coords2;
   if (utf8Length2 === 0) {
-    coords2 = "".concat(start2, ",0");
+    coords2 = "".concat(utf8Start2, ",0");
   } else if (utf8Length2 === 1) {
-    coords2 = "".concat(start2 + 1);
+    coords2 = "".concat(utf8Start2 + 1);
   } else {
-    coords2 = "".concat(start2 + 1, ",").concat(utf8Length2);
+    coords2 = "".concat(utf8Start2 + 1, ",").concat(utf8Length2);
   }
@@ -1541,2 +1532,3 @@ const text = ["@@ -".concat(coords1, " +").concat(coords2, " @@\n")];
       patch.start1--;
+      patch.utf8Start1--;
       patch.length1 = 1;
@@ -1549,2 +1541,3 @@ patch.utf8Length1 = 1;
       patch.start1--;
+      patch.utf8Start1--;
       patch.utf8Length1 = toInt(m[2]);
@@ -1555,2 +1548,3 @@ patch.length1 = patch.utf8Length1;
       patch.start2--;
+      patch.utf8Start2--;
       patch.length2 = 1;
@@ -1563,2 +1557,3 @@ patch.utf8Length2 = 1;
       patch.start2--;
+      patch.utf8Start2--;
       patch.utf8Length2 = toInt(m[4]);
@@ -1606,3 +1601,3 @@ patch.length2 = patch.utf8Length2;
 }
-export { DIFF_DELETE, DIFF_EQUAL, DIFF_INSERT, adjustIndiciesToUcs2, adjustIndiciesToUtf8, apply as applyPatches, cleanupEfficiency, cleanupSemantic, diff as makeDiff, make as makePatches, match, parse as parsePatch, stringifyPatch, stringify as stringifyPatches };
+export { DIFF_DELETE, DIFF_EQUAL, DIFF_INSERT, adjustIndiciesToUcs2, apply as applyPatches, cleanupEfficiency, cleanupSemantic, diff as makeDiff, make as makePatches, match, parse as parsePatch, stringifyPatch, stringify as stringifyPatches };
 //# sourceMappingURL=index.js.map

package.json

 {
   "name": "@sanity/diff-match-patch",
-  "version": "3.0.0",
+  "version": "3.0.1",
   "description": "Robust diff, match and patch algorithms to perform operations required for synchronizing plain text",
@@ -5,0 +5,0 @@ "sideEffects": false,

src/index.ts

@@ -23,3 +23,3 @@ // Diff
 
-// UTF-8 / UCS-2 utils (beta)
-export {adjustIndiciesToUcs2, adjustIndiciesToUtf8} from './utils/utf8Indices.js'
+// UCS-2 utils (beta)
+export {adjustIndiciesToUcs2} from './utils/utf8Indices.js'

src/patch/addPadding.ts

@@ -25,2 +25,4 @@ import {DIFF_EQUAL} from '../diff/diff.js'
     p.start2 += paddingLength
+    p.utf8Start1 += paddingLength
+    p.utf8Start2 += paddingLength
   }
@@ -36,2 +38,4 @@
     patch.start2 -= paddingLength // Should be 0.
+    patch.utf8Start1 -= paddingLength // Should be 0.
+    patch.utf8Start2 -= paddingLength // Should be 0.
     patch.length1 += paddingLength
@@ -43,6 +47,9 @@ patch.length2 += paddingLength
     // Grow first equality.
-    const extraLength = paddingLength - diffs[0][1].length
-    diffs[0][1] = nullPadding.substring(diffs[0][1].length) + diffs[0][1]
+    const firstDiffLength = diffs[0][1].length
+    const extraLength = paddingLength - firstDiffLength
+    diffs[0][1] = nullPadding.substring(firstDiffLength) + diffs[0][1]
     patch.start1 -= extraLength
     patch.start2 -= extraLength
+    patch.utf8Start1 -= extraLength
+    patch.utf8Start2 -= extraLength
     patch.length1 += extraLength
@@ -49,0 +56,0 @@ patch.length2 += extraLength

src/patch/apply.ts

@@ -110,3 +110,2 @@ /* eslint-disable max-depth */
       delta -= parsed[x].length2 - parsed[x].length1
-      // @todo byte count?
     } else {
@@ -113,0 +112,0 @@ // Found a match.  :)

src/patch/createPatchObject.ts

@@ -10,4 +10,8 @@ import type {Diff} from '../diff/diff.js'
   diffs: Diff[]
+
   start1: number
   start2: number
+  utf8Start1: number
+  utf8Start2: number
+
   length1: number
@@ -52,4 +56,8 @@ length2: number
     diffs: [],
+
     start1,
     start2,
+    utf8Start1: start1,
+    utf8Start2: start2,
+
     length1: 0,
@@ -56,0 +64,0 @@ length2: 0,

src/patch/make.ts

@@ -5,3 +5,3 @@ import {cleanupSemantic, cleanupEfficiency} from '../diff/cleanup.js'
 import {isLowSurrogate} from '../utils/surrogatePairs.js'
-import {adjustIndiciesToUtf8, countUtf8Bytes} from '../utils/utf8Indices.js'
+import {countUtf8Bytes} from '../utils/utf8Indices.js'
 import {MAX_BITS} from './constants.js'
@@ -98,2 +98,3 @@ import {createPatchObject, type Patch} from './createPatchObject.js'
   const patches = []
+
   let patch = createPatchObject(0, 0)
@@ -103,2 +104,5 @@ let patchDiffLength = 0 // Keeping our own length var is faster in JS.
   let charCount2 = 0 // Number of characters into the textB string.
+  let utf8Count1 = 0 // Number of utf-8 bytes into the textA string.
+  let utf8Count2 = 0 // Number of utf-8 bytes into the textB string.
+
   // Start with textA (prepatchText) and apply the diffs until we arrive at
@@ -109,5 +113,6 @@ // textB (postpatchText).  We recreate the patches one by one to determine
   let postpatchText = textA
+
   for (let x = 0; x < diffs.length; x++) {
-    const diffType = diffs[x][0]
-    const diffText = diffs[x][1]
+    const currentDiff = diffs[x]
+    const [diffType, diffText] = currentDiff
     const diffTextLength = diffText.length
@@ -120,2 +125,4 @@ const diffByteLength = countUtf8Bytes(diffText)
       patch.start2 = charCount2
+      patch.utf8Start1 = utf8Count1
+      patch.utf8Start2 = utf8Count2
     }
@@ -125,3 +132,3 @@
       case DIFF_INSERT:
-        patch.diffs[patchDiffLength++] = diffs[x]
+        patch.diffs[patchDiffLength++] = currentDiff
         patch.length2 += diffTextLength
@@ -135,3 +142,3 @@ patch.utf8Length2 += diffByteLength
         patch.utf8Length1 += diffByteLength
-        patch.diffs[patchDiffLength++] = diffs[x]
+        patch.diffs[patchDiffLength++] = currentDiff
         postpatchText =
@@ -144,3 +151,3 @@ postpatchText.substring(0, charCount2) +
           // Small equality inside a patch.
-          patch.diffs[patchDiffLength++] = diffs[x]
+          patch.diffs[patchDiffLength++] = currentDiff
           patch.length1 += diffTextLength
@@ -163,2 +170,3 @@ patch.length2 += diffTextLength
             charCount1 = charCount2
+            utf8Count1 = utf8Count2
           }
@@ -174,7 +182,10 @@ }
       charCount1 += diffTextLength
+      utf8Count1 += diffByteLength
     }
     if (diffType !== DIFF_DELETE) {
       charCount2 += diffTextLength
+      utf8Count2 += diffByteLength
     }
   }
+
   // Pick up the leftover patch if not empty.
@@ -186,3 +197,3 @@ if (patchDiffLength) {
 
-  return adjustIndiciesToUtf8(patches, textA)
+  return patches
 }
@@ -231,2 +242,5 @@
 
+  const prefixLength = prefix.length
+  const prefixUtf8Length = countUtf8Bytes(prefix)
+
   // Add the suffix.
@@ -245,11 +259,16 @@
 
+  const suffixLength = suffix.length
+  const suffixUtf8Length = countUtf8Bytes(suffix)
+
   // Roll back the start points.
-  patch.start1 -= prefix.length
-  patch.start2 -= prefix.length
+  patch.start1 -= prefixLength
+  patch.start2 -= prefixLength
+  patch.utf8Start1 -= prefixUtf8Length
+  patch.utf8Start2 -= prefixUtf8Length
 
   // Extend the lengths.
-  patch.length1 += prefix.length + suffix.length
-  patch.length2 += prefix.length + suffix.length
-  patch.utf8Length1 += prefix.length + suffix.length
-  patch.utf8Length2 += prefix.length + suffix.length
+  patch.length1 += prefixLength + suffixLength
+  patch.length2 += prefixLength + suffixLength
+  patch.utf8Length1 += prefixUtf8Length + suffixUtf8Length
+  patch.utf8Length2 += prefixUtf8Length + suffixUtf8Length
 }

src/patch/parse.ts

@@ -33,2 +33,3 @@ import {DIFF_DELETE, DIFF_EQUAL, DIFF_INSERT} from '../diff/diff.js'
       patch.start1--
+      patch.utf8Start1--
       patch.length1 = 1
@@ -41,2 +42,3 @@ patch.utf8Length1 = 1
       patch.start1--
+      patch.utf8Start1--
       // The patch itself will contain the UTF-8 length
@@ -50,2 +52,3 @@ patch.utf8Length1 = toInt(m[2])
       patch.start2--
+      patch.utf8Start2--
       patch.length2 = 1
@@ -58,2 +61,3 @@ patch.utf8Length2 = 1
       patch.start2--
+      patch.utf8Start2--
       // The patch itself will contain the UTF-8 length
@@ -60,0 +64,0 @@ patch.utf8Length2 = toInt(m[4])

src/patch/stringify.ts

@@ -16,3 +16,3 @@ import {DIFF_DELETE, DIFF_EQUAL, DIFF_INSERT} from '../diff/diff.js'
 /**
- * Create a textual representation of a
+ * Create a textual representation of a patch.
  *
@@ -24,11 +24,11 @@ * @param patch - Patch to stringify
 export function stringifyPatch(patch: Patch): string {
-  const {utf8Length1, utf8Length2, start1, start2, diffs} = patch
+  const {utf8Length1, utf8Length2, utf8Start1, utf8Start2, diffs} = patch
 
   let coords1: string
   if (utf8Length1 === 0) {
-    coords1 = `${start1},0`
+    coords1 = `${utf8Start1},0`
   } else if (utf8Length1 === 1) {
-    coords1 = `${start1 + 1}`
+    coords1 = `${utf8Start1 + 1}`
   } else {
-    coords1 = `${start1 + 1},${utf8Length1}`
+    coords1 = `${utf8Start1 + 1},${utf8Length1}`
   }
@@ -38,7 +38,7 @@
   if (utf8Length2 === 0) {
-    coords2 = `${start2},0`
+    coords2 = `${utf8Start2},0`
   } else if (utf8Length2 === 1) {
-    coords2 = `${start2 + 1}`
+    coords2 = `${utf8Start2 + 1}`
   } else {
-    coords2 = `${start2 + 1},${utf8Length2}`
+    coords2 = `${utf8Start2 + 1},${utf8Length2}`
   }
@@ -45,0 +45,0 @@

src/utils/utf8Indices.ts

@@ -26,54 +26,2 @@ import {cloneDiff} from '../diff/clone.js'
  * Takes a `patches` array as produced by diff-match-patch and adjusts the
- * `start1` and `start2` properties so that they refer to UTF-8 index instead
- * of a UCS-2 index.
- *
- * @param patches - The patches to adjust
- * @param base - The base string to use for counting bytes
- * @returns A new array of patches with adjusted indicies
- * @beta
- */
-export function adjustIndiciesToUtf8(patches: Patch[], base: string): Patch[] {
-  let byteOffset = 0
-  let idx = 0 // index into the input.
-
-  function advanceTo(target: number) {
-    for (; idx < target; idx++) {
-      const codePoint = base.codePointAt(idx)
-      if (typeof codePoint === 'undefined') {
-        throw new Error('Failed to get codepoint')
-      }
-
-      byteOffset += utf8len(codePoint)
-
-      // This is encoded as a surrogate pair.
-      if (codePoint > 0xffff) {
-        idx++
-      }
-    }
-
-    if (idx !== target) {
-      throw new Error('Failed to determine byte offset')
-    }
-
-    return byteOffset
-  }
-
-  const adjusted: Patch[] = []
-  for (const patch of patches) {
-    adjusted.push({
-      diffs: patch.diffs.map((diff) => cloneDiff(diff)),
-      start1: advanceTo(patch.start1),
-      start2: advanceTo(patch.start2),
-      length1: patch.length1,
-      length2: patch.length2,
-      utf8Length1: patch.utf8Length1,
-      utf8Length2: patch.utf8Length2,
-    })
-  }
-
-  return adjusted
-}
-
-/**
- * Takes a `patches` array as produced by diff-match-patch and adjusts the
  * `start1` and `start2` properties so that they refer to UCS-2 index instead
@@ -95,3 +43,5 @@ * of a UTF-8 index.
       if (typeof codePoint === 'undefined') {
-        throw new Error('Failed to get codepoint')
+        // Reached the end of the base string - the indicies won't be correct,
+        // but we also cannot advance any further to find a closer index.
+        return idx
       }
@@ -122,2 +72,4 @@
       start2: advanceTo(patch.start2),
+      utf8Start1: patch.utf8Start1,
+      utf8Start2: patch.utf8Start2,
       length1: patch.length1,
@@ -124,0 +76,0 @@ length2: patch.length2,

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Worsened metrics

Total package byte prevSize

526222

decreased by-0.22%

Lines of code

6037

decreased by-0.54%

No dependency changes