feat: add 50% threshold for diffing - show full content when changes are large

When showDiffSinceLastCall detects >50% of lines changed, returns full
new content instead of a diff patch. This makes large changes more readable.

playwriter/src/clean-html.ts

@@ -1,6 +1,6 @@
 import { Page, Locator } from 'playwright-core'
-import { createPatch } from 'diff'
 import { formatHtmlForPrompt } from './htmlrewrite.js'
+import { createSmartDiff } from './diff-utils.js'
 
 export interface GetCleanHTMLOptions {
   locator: Locator | Page
@@ -81,16 +81,15 @@ export async function getCleanHTML(options: GetCleanHTMLOptions): Promise<string
       return 'No previous snapshot available. This is the first call for this locator. Full snapshot stored for next diff.'
     }
 
-    const patch = createPatch('html', previousSnapshot, htmlStr, 'previous', 'current', {
-      context: 3,
+    const diffResult = createSmartDiff({
+      oldContent: previousSnapshot,
+      newContent: htmlStr,
+      label: 'html',
     })
 
     pageSnapshots.set(snapshotKey, htmlStr)
 
-    if (patch.split('\n').length <= 4) {
-      return 'No changes detected since last snapshot'
-    }
-    return patch
+    return diffResult.content
   }
 
   // Store snapshot for future diffs

playwriter/src/diff-utils.test.ts

@@ -0,0 +1,119 @@
+import { describe, it, expect } from 'vitest'
+import { createSmartDiff } from './diff-utils.js'
+
+describe('createSmartDiff', () => {
+  it('returns no-change when content is identical', () => {
+    const result = createSmartDiff({
+      oldContent: 'hello\nworld',
+      newContent: 'hello\nworld',
+    })
+    expect(result.type).toBe('no-change')
+    expect(result.content).toBe('No changes detected since last snapshot')
+  })
+
+  it('returns diff for small changes', () => {
+    const result = createSmartDiff({
+      oldContent: 'line1\nline2\nline3\nline4\nline5',
+      newContent: 'line1\nline2-modified\nline3\nline4\nline5',
+    })
+    expect(result.type).toBe('diff')
+    expect(result.content).toContain('-line2')
+    expect(result.content).toContain('+line2-modified')
+  })
+
+  it('returns full content when changes exceed 50% threshold', () => {
+    const result = createSmartDiff({
+      oldContent: 'a\nb\nc\nd',
+      newContent: 'x\ny\nz\nw',
+    })
+    expect(result.type).toBe('full')
+    expect(result.content).toContain('Content changed significantly')
+    expect(result.content).toContain('x\ny\nz\nw')
+  })
+
+  it('respects custom threshold', () => {
+    // 2 out of 4 lines changed = 50%
+    const oldContent = 'a\nb\nc\nd'
+    const newContent = 'a\nX\nY\nd'
+
+    // With 40% threshold, should return full (50% >= 40%)
+    const resultLow = createSmartDiff({
+      oldContent,
+      newContent,
+      threshold: 0.4,
+    })
+    expect(resultLow.type).toBe('full')
+
+    // With 60% threshold, should return diff (50% < 60%)
+    const resultHigh = createSmartDiff({
+      oldContent,
+      newContent,
+      threshold: 0.6,
+    })
+    expect(resultHigh.type).toBe('diff')
+  })
+
+  it('uses custom label in diff output', () => {
+    const result = createSmartDiff({
+      oldContent: 'line1\nline2\nline3\nline4\nline5',
+      newContent: 'line1\nmodified\nline3\nline4\nline5',
+      label: 'snapshot',
+      threshold: 0.5, // 20% change, will return diff
+    })
+    expect(result.type).toBe('diff')
+    expect(result.content).toContain('--- snapshot')
+    expect(result.content).toContain('+++ snapshot')
+  })
+
+  it('handles empty old content (all additions)', () => {
+    const result = createSmartDiff({
+      oldContent: '',
+      newContent: 'new\ncontent\nhere',
+    })
+    // 100% new content should trigger full
+    expect(result.type).toBe('full')
+    expect(result.content).toContain('new\ncontent\nhere')
+  })
+
+  it('handles empty new content (all deletions)', () => {
+    const result = createSmartDiff({
+      oldContent: 'old\ncontent\nhere',
+      newContent: '',
+    })
+    expect(result.type).toBe('full')
+    expect(result.content).toContain('100%')
+  })
+
+  it('handles single line changes', () => {
+    const result = createSmartDiff({
+      oldContent: 'single line',
+      newContent: 'modified line',
+    })
+    // 100% change on single line
+    expect(result.type).toBe('full')
+  })
+
+  it('calculates percentage correctly in output', () => {
+    // 4 lines changed out of 5 = 80%
+    const result = createSmartDiff({
+      oldContent: 'a\nb\nc\nd\ne',
+      newContent: 'a\nX\nY\nZ\nW',
+    })
+    expect(result.type).toBe('full')
+    expect(result.content).toContain('(80% of lines)')
+  })
+
+  it('includes context lines in diff output', () => {
+    const result = createSmartDiff({
+      oldContent: 'line1\nline2\nline3\nline4\nline5\nline6\nline7\nline8\nline9\nline10',
+      newContent: 'line1\nline2\nline3\nline4\nMODIFIED\nline6\nline7\nline8\nline9\nline10',
+      threshold: 0.9, // Force diff
+    })
+    expect(result.type).toBe('diff')
+    // Should have context around the change
+    expect(result.content).toContain('line4')
+    expect(result.content).toContain('-line5')
+    expect(result.content).toContain('+MODIFIED')
+    expect(result.content).toContain('line6')
+  })
+})

playwriter/src/diff-utils.ts

@@ -0,0 +1,75 @@
+import { structuredPatch } from 'diff'
+
+export interface SmartDiffResult {
+  type: 'diff' | 'full' | 'no-change'
+  content: string
+}
+
+export interface CreateSmartDiffOptions {
+  oldContent: string
+  newContent: string
+  /** Threshold ratio (0-1) above which full content is returned instead of diff. Default 0.5 (50%) */
+  threshold?: number
+  /** Label for the diff output */
+  label?: string
+}
+
+/**
+ * Creates a smart diff that returns full content when changes exceed threshold.
+ * 
+ * When more than `threshold` (default 50%) of lines have changed, showing a diff
+ * is not useful - we return the full new content instead.
+ */
+export function createSmartDiff(options: CreateSmartDiffOptions): SmartDiffResult {
+  const { oldContent, newContent, threshold = 0.5, label = 'content' } = options
+
+  const patch = structuredPatch(label, label, oldContent, newContent, 'previous', 'current', {
+    context: 3,
+  })
+
+  // Count added and removed lines
+  let addedLines = 0
+  let removedLines = 0
+  for (const hunk of patch.hunks) {
+    for (const line of hunk.lines) {
+      if (line.startsWith('+')) {
+        addedLines++
+      } else if (line.startsWith('-')) {
+        removedLines++
+      }
+    }
+  }
+
+  // No changes
+  if (addedLines === 0 && removedLines === 0) {
+    return { type: 'no-change', content: 'No changes detected since last snapshot' }
+  }
+
+  // Calculate change ratio: use max(added, removed) since a replacement counts as both
+  // This ensures the ratio stays in 0-100% range
+  const oldLineCount = oldContent.split('\n').length
+  const newLineCount = newContent.split('\n').length
+  const maxLines = Math.max(oldLineCount, newLineCount, 1) // Avoid division by zero
+  const changedLines = Math.max(addedLines, removedLines)
+  const changeRatio = Math.min(changedLines / maxLines, 1) // Cap at 100%
+
+  if (changeRatio >= threshold) {
+    const percentChanged = Math.round(changeRatio * 100)
+    return {
+      type: 'full',
+      content: `Content changed significantly (${percentChanged}% of lines). Full new content:\n\n${newContent}`,
+    }
+  }
+
+  // Build unified diff string from structured patch
+  const diffLines: string[] = [
+    `--- ${label} (previous)`,
+    `+++ ${label} (current)`,
+  ]
+  for (const hunk of patch.hunks) {
+    diffLines.push(`@@ -${hunk.oldStart},${hunk.oldLines} +${hunk.newStart},${hunk.newLines} @@`)
+    diffLines.push(...hunk.lines)
+  }
+
+  return { type: 'diff', content: diffLines.join('\n') }
+}

playwriter/src/executor.ts

@@ -12,7 +12,7 @@ import util from 'node:util'
 import { createRequire } from 'node:module'
 import { fileURLToPath } from 'node:url'
 import vm from 'node:vm'
-import { createPatch } from 'diff'
+import { createSmartDiff } from './diff-utils.js'
 import { getCdpUrl } from './utils.js'
 import { waitForPageLoad, WaitForPageLoadOptions, WaitForPageLoadResult } from './wait-for-page-load.js'
 import { getCDPSessionForPage, CDPSession, ICDPSession } from './cdp-session.js'
@@ -427,11 +427,13 @@ export class PlaywrightExecutor {
             this.lastSnapshots.set(targetPage, snapshotStr)
             return 'No previous snapshot available. This is the first call for this page. Full snapshot stored for next diff.'
           }
-          const patch = createPatch('snapshot', previousSnapshot, snapshotStr, 'previous', 'current', { context: 3 })
-          if (patch.split('\n').length <= 4) {
-            return 'No changes detected since last snapshot'
-          }
-          return patch
+          const diffResult = createSmartDiff({
+            oldContent: previousSnapshot,
+            newContent: snapshotStr,
+            label: 'snapshot',
+          })
+          this.lastSnapshots.set(targetPage, snapshotStr)
+          return diffResult.content
         }
 
         this.lastSnapshots.set(targetPage, snapshotStr)