refactor(search): one definition of what a search occurrence is (#6905)

Follow-up to #6901, closing two places where the workflow search index and the
Note card that mirrors it could drift apart. Neither is a live bug; both are the
shape that produced one — the card silently disagreeing with the panel about
which hit is which, counted in one place and painted in another.

THE SCAN. #6901 shared `foldSearchWhitespace` but left the scan around it
duplicated: normalize, then non-overlapping `indexOf` stepping by
`max(len, 1)`, written out once in the indexer and once in the renderer package.
They agree today. They would stop agreeing the moment either grew whole-word
matching, diacritic folding, or a regex mode, and the failure is silent. Both
now call one `forEachSearchOccurrence` in `@sim/utils/string` — the only place
either package can share, since the card renders from a package that cannot
import from `apps/*`.

THE DECLARATION. The indexer projects markdown escapes only for a field
declaring `searchTextFormat: 'markdown'`; the card projects unconditionally,
because it cannot read the block registry. Dropping that one line from the Note
config would leave them disagreeing with nothing to catch it, so a test now
pins it and explains why.

Net negative in lines: this deletes a duplicated loop rather than adding a
layer.

Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>

apps/sim/lib/workflows/search-replace/indexer.test.ts

@@ -12,6 +12,7 @@ import {
   SEARCH_REPLACE_BLOCK_CONFIGS,
 } from '@/lib/workflows/search-replace/search-replace.fixtures'
 import { WORKFLOW_SEARCH_SUBFLOW_FIELD_IDS } from '@/lib/workflows/search-replace/subflow-fields'
+import { NoteBlock } from '@/blocks/blocks/note'
 
 /**
  * Uses the real tool registry. Nothing here imports it directly — the dependency
@@ -167,6 +168,19 @@ describe('indexWorkflowSearchMatches', () => {
     expect(matches.some((match) => match.target.kind === 'block-name')).toBe(false)
   })
 
+  describe('the Note body declares the markdown format the card assumes', () => {
+    /*
+     * The canvas card projects markdown escapes unconditionally — it renders from a package that
+     * cannot read the block registry. The indexer projects only when the field says so. Dropping
+     * the declaration would leave the two disagreeing about what an occurrence is, and the failure
+     * is silent: the panel counts a hit the card marks somewhere else.
+     */
+    it('keeps searchTextFormat on the Note content field', () => {
+      const content = NoteBlock.subBlocks.find((subBlock) => subBlock.id === 'content')
+      expect(content?.searchTextFormat).toBe('markdown')
+    })
+  })
+
   describe('a markdown field is searched as it renders', () => {
     /*
      * The rich-text editor backslash-escapes every markdown-significant character in prose, so a

apps/sim/lib/workflows/search-replace/indexer.ts

@@ -1,5 +1,5 @@
 import { isRecordLike } from '@sim/utils/object'
-import { foldSearchWhitespace, projectEscapedMarkdownForSearch } from '@sim/utils/string'
+import { forEachSearchOccurrence, projectEscapedMarkdownForSearch } from '@sim/utils/string'
 import { DEFAULT_SUBBLOCK_TYPE } from '@sim/workflow-persistence/subblocks'
 import type { SubBlockType } from '@sim/workflow-types/blocks'
 import { isWorkflowBlockProtected } from '@sim/workflow-types/workflow'
@@ -57,16 +57,6 @@ import {
   type ToolParameterConfig,
 } from '@/tools/params'
 
-/**
- * Whitespace is folded before comparison (see {@link foldSearchWhitespace}):
- * the fold is one-to-one, so ranges found in the normalized string index the
- * original text correctly.
- */
-function normalizeForSearch(value: string, caseSensitive: boolean): string {
-  const folded = foldSearchWhitespace(value)
-  return caseSensitive ? folded : folded.toLowerCase()
-}
-
 /**
  * Ranges of `query` in `value`, always in `value`'s own coordinates.
  *
@@ -83,23 +73,21 @@ function findTextRanges(
   caseSensitive: boolean,
   searchTextFormat?: SubBlockConfig['searchTextFormat']
 ) {
-  if (!query) return []
-
   const projection = searchTextFormat === 'markdown' ? projectEscapedMarkdownForSearch(value) : null
-  const source = normalizeForSearch(projection ? projection.text : value, caseSensitive)
-  const target = normalizeForSearch(query, caseSensitive)
   const ranges: Array<{ start: number; end: number }> = []
 
-  let index = source.indexOf(target)
-  while (index !== -1) {
-    const end = index + target.length
-    ranges.push(
-      projection
-        ? { start: projection.starts[index], end: projection.starts[end] }
-        : { start: index, end }
-    )
-    index = source.indexOf(target, index + Math.max(target.length, 1))
-  }
+  forEachSearchOccurrence(
+    projection ? projection.text : value,
+    query,
+    (start, end) => {
+      ranges.push(
+        projection
+          ? { start: projection.starts[start], end: projection.starts[end] }
+          : { start, end }
+      )
+    },
+    caseSensitive
+  )
 
   return ranges
 }