@ansonlai/docx-redline-js 0.5.0 → 0.5.1

package/AGENTS.md

@@ -57,13 +57,16 @@ reverted to the pre-revision baseline and re-diffed to the new text, cleanly
 merging the edits without accumulating intermediate revisions or nesting markup.
 If the paragraph contains revisions from a different reviewer, the edit fails
 with `EXISTING_REVISIONS` to safeguard third-party marks. Pass
-`existingRevisions: 'accept-all-first'` (or `--existing-revisions accept-all-first`
+`existingRevisions: 'slice-cross-author'` (or `--existing-revisions slice-cross-author`)
+to preserve the other reviewer's attribution while applying Word-native
+insertions and deletions inside their pending insertion. Pass
+`existingRevisions: 'accept-all-first'` (or `--existing-revisions accept-all-first`
 via CLI) to normalize all prior revisions first, or `'reject-input'` to refuse any
 paragraph with open revisions. Use `'accept-all-first-keep-normalized'` only when
-accepted revisions should be returned as a real change even on a no-op edit.
-Same-author merging also fails with `COMMENTED_CONTENT_MERGE` when the revised
-paragraph contains comment anchors, because reverting the prior revision could
-remove or orphan those comments. Resolve the comments before re-editing.
+accepted revisions should be returned as a real change even on a no-op edit.
+Same-author merging also fails with `COMMENTED_CONTENT_MERGE` when the revised
+paragraph contains comment anchors, because reverting the prior revision could
+remove or orphan those comments. Resolve the comments before re-editing.
 
 ### Apply a text edit without tracked changes (Direct Edits)
 
@@ -251,7 +254,7 @@ explicitly supplied.
 
 #### Standard Workflow (Fast & Direct)
 
-`apply` is fast, progressive, and self-validating by default. It supports inline one-liners as well as batch operations files:
+Use this for everything by default. `apply` is fast, progressive, and self-validating by default—it validates the resulting package and revision markup internally before writing. **Do not insert a `preflight` or baseline `validate` step on top of it "to be safe"**; `apply` already covers that internally. It supports inline one-liners as well as batch operations files:
 
 ```bash
 # 1. Inline one-liner edit (fastest for 1–2 edits; no JSON file needed)
@@ -275,9 +278,11 @@ For multi-clause or multi-page reviews, apply edits **section-by-section** or cl
 
 #### High-Assurance / Staged Verification Workflow (Optional)
 
-For high-stakes legal contracts, large automated batch migrations, or workflows
-requiring explicit non-mutating pre-checks and an independent baseline audit
-report, use the extended verification cycle:
+This is an opt-in, higher-latency path for cases like large automated batch
+migrations or workflows where the user specifically requests a non-mutating dry run
+and an independent baseline audit report. **Never switch into it on your own initiative**
+(not even for "high-stakes" contracts); unless the user explicitly requests it, stick with the
+Standard workflow above. Use the extended verification cycle:
 
 ```bash
 docx-redline inspect contract.docx --non-empty
@@ -351,6 +356,57 @@ compatibility fallback is `DOCX_REDLINE_AUTHOR` and then `Agent`. Consumers
 must use the JSON status and process exit code; failed atomic work has
 `written: false`, `outputPath: null`, and does not modify the output path.
 
+#### Safe Operations File Creation (JSON vs. Shell Heredocs)
+
+When composing batch operations files (`operations.json`):
+
+- **Use structured file-writing tools or JSON serializers**: Write operations files via your environment's file-creation tools or a language JSON serializer (`JSON.stringify`).
+- **Never compose operations in raw shell heredocs** (e.g., `cat << 'EOF'` in bash or PowerShell `@" ... "@`): Legal clauses routinely contain curly quotes (`“ ”`), smart apostrophes (`’`), em-dashes (`—`), section symbols (`§`), non-breaking spaces, and backslashes. Shell heredocs frequently mangle Unicode character encodings, quote escaping, and whitespace formatting, causing immediate `TARGET_NOT_FOUND` failures.
+
+#### Walking Progressive Batch Results (Status & Partial Execution)
+
+In default progressive mode (`atomic: false`), operations execute independently: valid operations commit to the document while failing operations report errors without aborting the batch:
+
+- **Do not rely solely on top-level `written: true` or `status !== "error"`**: A progressive batch can return `status: "partial"` with `written: true` when some operations succeed and others fail.
+- **Walk every entry in `results`**: Check `results[i].status` and `results[i].error`. Any `status: "error"` entry in `results` represents an unapplied change that must be investigated and resolved.
+- **`written: false`**: Indicates that zero operations were committed (or an atomic rollback occurred). Never treat or present an unwritten or partial output file as complete.
+
+#### Human-Readable References vs. Internal Machine Handles
+
+Target handles such as `ref` (`P<index>`), `targetRef`, and bare paragraph `index` numbers are **strictly internal machine handles** for the CLI and engine. They do not correspond to any visual or followable marker in Microsoft Word:
+
+- **Never surface `P11`, `P42`, or bare paragraph numbers** in user-facing prose, comments, redline summaries, or negotiation notes.
+- Instead, cite locations using the human-readable fields provided by `inspect` / `extract`:
+  - **`provision`**: Lead with section/clause numbers when present (e.g., `§14.1 Entire Agreement`).
+  - **`nearestHeading` + ordinal offset**: When `provision` is absent, describe position relative to the nearest heading (e.g., `under "Limitation of Liability", 2nd paragraph`).
+  - **Structural context**: For unnumbered clauses prior to the first heading, use plain language (e.g., `opening recital, before Section 1`).
+  - **`humanReference`**: Use the pre-joined citation string provided directly on inspected paragraph objects.
+
+#### Actionable Error Recovery Matrix
+
+When the CLI or runner returns an error code, follow these specific recovery actions:
+
+| Error Code | Meaning | Actionable Recovery |
+|---|---|---|
+| `TARGET_NOT_FOUND` | Target text did not match any paragraph. | **Do NOT retry with paraphrased text.** Re-run `extract`/`inspect`, copy `exactText` verbatim (including exact whitespace/punctuation), and add a discriminator (`paragraphId`, `fingerprint`, or `occurrence`). |
+| `AMBIGUOUS_TARGET` | Multiple paragraphs match identical text. | Disambiguate by supplying `paragraphId`, `fingerprint`, `occurrence`, or `index` in the target descriptor. |
+| `ANCHOR_NOT_FOUND` / `AMBIGUOUS_ANCHOR` | Comment anchor text was not uniquely matched in paragraph. | Narrow `textToComment` to a unique exact substring, or omit `textToComment` to anchor the comment to the entire paragraph. |
+| `OVERLAPPING_TEXT_EDITS` | Multiple operations target the same paragraph concurrently. | Consolidate all changes to the same paragraph into a single `redline` or `replace` operation. |
+| `EXISTING_REVISIONS` | Target paragraph contains tracked changes from another author. | Fails closed to protect third-party review marks. Report the other reviewer's name to the user. Do not pass `accept-all-first` without explicit authorization. |
+| `COMMENTED_CONTENT_MERGE` / `COMMENTED_CONTENT_DELETE` | Operation would overwrite, revert, or delete content with comments. | Fails closed to prevent orphaned comment threads. Report the comment author and text to the user; resolve the comment before re-editing. |
+| `INVALID_OPERATION` | Operation object violates schema or has incompatible fields. | Validate the JSON structure against [`document-operations.schema.json`](file:///c:/Users/Phara/Desktop/Projects/Docx%20Redline%20JS/docs/schemas/document-operations.schema.json) before targeting is attempted. |
+| `STRUCTURED_CONTENT_INVALID` | Malformed Markdown table or structure in replacement text. | Ensure tables include a separator row (`\| --- \| --- \|`) and consistent column counts; do not downgrade to raw text. |
+
+**Important Rule:** Never repeat the exact same failing command without correcting the reported cause. If an error persists after one correction attempt, stop and report the diagnostic code to the user.
+
+#### Document Scope & Boundary Invariants
+
+The `docx-redline` engine and CLI operate specifically on the **main document body**:
+
+- **Supported Content**: Body paragraphs, numbered/bulleted lists, tables and table cells, comments, and comment replies.
+- **Unsupported Content**: Headers, footers, footnotes, endnotes, floating text boxes, shape drawings, watermarks, and embedded macros.
+- Do not attempt to target, edit, or comment on header/footer text or footnote citations using `docx-redline`. Use specialized document manipulation tools or manual editing for layout frames outside the body text.
+
 ### Convert paragraph text into a Word list
 
 ```js
@@ -475,7 +531,7 @@ orchestration/
 | `structuredContent` | `boolean` | `true` | Auto-detects Markdown tables, headings (`#`), and lists in replacement text and renders them as native Word elements (`w:tbl`, `w:pStyle`, `w:numPr`). Pass `false` to treat replacement text strictly as plain text. |
 | `pairReplacements` | `boolean` | `true` | Links adjacent `<w:del>` and `<w:ins>` revisions with matching timestamps so Word groups them as a single replacement in the Reviewing Pane. |
 | `strictTargets` | `boolean` | `true` (CLI/facade) | Requires exact target descriptors (`exactText`, `paragraphId`, `index`, `occurrence`, `fingerprint`) and forbids ambiguous matching. Defaults to `false` in low-level runner for backwards compatibility. |
-| `existingRevisions` | `string` | `'merge-same-author'` | How to handle paragraphs with existing tracked changes. `'merge-same-author'` automatically merges subsequent edits from the same author against the pre-revision baseline while protecting different authors' revisions with `EXISTING_REVISIONS`. Pass `'accept-all-first'` to normalize prior revisions or `'reject-input'` to refuse editing revised paragraphs. |
+| `existingRevisions` | `string` | `'merge-same-author'` | How to handle paragraphs with existing tracked changes. `'merge-same-author'` merges the same author's work and protects other authors with `EXISTING_REVISIONS`. `'slice-cross-author'` retains same-author merging while allowing Word-native edits inside another author's pending insertion. Pass `'accept-all-first'` to normalize prior revisions or `'reject-input'` to refuse editing revised paragraphs. |
 | `removeFormatting` | `boolean` | `false` | When `true` and the text is unchanged with no Markdown hints, strips existing bold/italic/underline/strikethrough formatting. |
 | `sanitizeInput` | `boolean` | `false` | Opt-in removal of standalone leading assistant-preface lines. Literal dollar signs and `\n` sequences are always preserved. |
 
@@ -508,8 +564,8 @@ orchestration/
 }
 ```
 
-Known error codes include `PARSE_ERROR`, `TARGET_NOT_FOUND`, `PARTIAL_TARGET`,
-`EXISTING_REVISIONS`, `COMMENTED_CONTENT_MERGE`, `UNSAFE_REVISION_NESTING`, `UNSUPPORTED_REVISION_VIEW_MUTATION`,
+Known error codes include `PARSE_ERROR`, `TARGET_NOT_FOUND`, `PARTIAL_TARGET`,
+`EXISTING_REVISIONS`, `COMMENTED_CONTENT_MERGE`, `UNSAFE_REVISION_NESTING`, `UNSUPPORTED_REVISION_VIEW_MUTATION`,
 `UNSAFE_PARAGRAPH_BOUNDARY`, `DIFF_TOKEN_LIMIT`, and `BATCH_OPERATION_FAILED`.
 
 For ingestion that must distinguish an empty document from malformed OOXML,
@@ -561,7 +617,7 @@ directly into `word/document.xml`.
 5. `useNativeApi: true` means standalone mode cannot fully handle that operation path.
 6. `deleteCommentsByAuthorInOoxml` removes definitions and linked anchors only when they are present in the same OOXML payload. In a real `.docx`, `word/comments.xml` and `word/document.xml` are separate parts and must both be updated by the package integration layer.
 7. If output begins with `<pkg:package`, treat it as package-level OOXML and normalize it before writing anything back to `word/document.xml`.
-8. Existing revisions from the same author are merged by default against the pre-revision baseline (`merge-same-author`), while third-party revisions fail closed with `EXISTING_REVISIONS`. Pass `existingRevisions: 'accept-all-first'` to normalize all prior revisions first, or `'reject-input'` to refuse any revised paragraph.
+8. Existing revisions from the same author are merged by default against the pre-revision baseline (`merge-same-author`), while third-party revisions fail closed with `EXISTING_REVISIONS`. Pass `existingRevisions: 'slice-cross-author'` to preserve third-party attribution while editing inside pending insertions, `'accept-all-first'` to normalize all prior revisions first, or `'reject-input'` to refuse any revised paragraph.
 9. Caller content is not sanitized by default. Pass `sanitizeInput: true` only for raw assistant output; literal dollar delimiters and `\\n` sequences are never rewritten.
 10. Hyperlinks, bookmarks, comment markers, tabs/breaks, and footnote/endnote references are structural OOXML and should survive adjacent redline edits.
 11. Internally, create Word elements through `createWordElement` and tracked-change metadata through `createRevisionMetadata`.

package/CHANGELOG.md

@@ -1,6 +1,6 @@
 # Changelog
 
-## Unreleased
+## 0.5.0
 
 ### ⚠️ Breaking changes
 

package/README.md

@@ -183,7 +183,7 @@ See [the agent workflow in AGENTS.md](./AGENTS.md#agent-document-workflow-cli) a
 | `structuredContent` | `boolean` | `true` | Auto-detects Markdown tables, headings (`#`), and lists in replacement text and renders them as native Word elements (`w:tbl`, `w:pStyle`, `w:numPr`). Pass `false` to treat replacement text strictly as plain text. |
 | `pairReplacements` | `boolean` | `true` | Links adjacent `<w:del>` and `<w:ins>` revisions with matching timestamps so Word groups them as a single replacement in the Reviewing Pane. |
 | `strictTargets` | `boolean` | `true` (CLI/facade) | Requires exact target descriptors (`exactText`, `paragraphId`, `index`, `occurrence`, `fingerprint`) and forbids ambiguous matching. Defaults to `false` in low-level runner for backwards compatibility. |
-| `existingRevisions` | `string` | `'merge-same-author'` | How to handle paragraphs with existing tracked changes. `'merge-same-author'` automatically merges subsequent edits from the same author against the pre-revision baseline while protecting different authors' revisions with `EXISTING_REVISIONS`. Pass `'accept-all-first'` to normalize prior revisions or `'reject-input'` to refuse editing revised paragraphs. |
+| `existingRevisions` | `string` | `'merge-same-author'` | How to handle paragraphs with existing tracked changes. `'merge-same-author'` merges revisions from the same author and protects other authors with `EXISTING_REVISIONS`. `'slice-cross-author'` keeps that same-author merge behavior while allowing Word-native edits inside another author's pending insertion. Pass `'accept-all-first'` to normalize prior revisions or `'reject-input'` to refuse editing revised paragraphs. |
 | `removeFormatting` | `boolean` | `false` | When `true` and the text is unchanged with no Markdown hints, strips existing bold/italic/underline/strikethrough formatting. |
 | `sanitizeInput` | `boolean` | `false` | Opt-in removal of standalone leading assistant-preface lines. Literal dollar signs and `\n` sequences are always preserved. |
 

package/core/redline-validation.js

@@ -3,8 +3,9 @@
  *
  * Mirrors the invariants enforced by the test-suite round-trip harness so
  * downstream consumers can verify output before writing it into a package:
- * no nested revisions, deleted text uses w:delText, revision metadata is
- * complete, revision ids are unique, and boundary whitespace is preserved.
+ * only schema-permitted nested revisions, deleted text uses w:delText,
+ * revision metadata is complete, revision ids are unique, and boundary
+ * whitespace is preserved.
  */
 
 import { parseXml } from '../adapters/xml-adapter.js';
@@ -107,13 +108,18 @@ export function validateRedlineOoxml(oxml) {
         }
     }
 
-    // No w:ins/w:del nested inside another w:ins/w:del.
+    // A w:del may be a direct revision child of w:ins. All other insertion /
+    // deletion nesting is rejected, including deeper revisions inside that del.
     for (const revision of revisions) {
         const nested = Array.from(revision.getElementsByTagName('*'))
             .filter(el => el !== revision && ['ins', 'del'].includes(localNameOf(el)));
-        if (nested.length > 0) {
+        const invalidNested = nested.find(candidate => {
+            if (localNameOf(revision) !== 'ins' || localNameOf(candidate) !== 'del') return true;
+            return candidate.parentNode !== revision;
+        });
+        if (invalidNested) {
             addIssue('NESTED_REVISION', 'error',
-                `<${revision.nodeName}> (w:id="${wordAttribute(revision, 'id')}") contains nested <${nested[0].nodeName}>.`);
+                `<${revision.nodeName}> (w:id="${wordAttribute(revision, 'id')}") contains invalid nested <${invalidNested.nodeName}>.`);
         }
     }