jupyter-ydoc 3.3.2__tar.gz → 3.3.4__tar.gz

{jupyter_ydoc-3.3.2 → jupyter_ydoc-3.3.4}/.github/workflows/auto_author_assign.yml

@@ -12,4 +12,4 @@ jobs:
   assign-author:
     runs-on: ubuntu-latest
     steps:
-      - uses: toshimaru/auto-author-assign@v2.1.1
+      - uses: toshimaru/auto-author-assign@v3.0.1

{jupyter_ydoc-3.3.2 → jupyter_ydoc-3.3.4}/.github/workflows/check-release.yml

@@ -37,7 +37,7 @@ jobs:
           token: ${{ secrets.GITHUB_TOKEN }}
       - name: Upload Distributions
         if: ${{ matrix.group == 'check_release' }}
-        uses: actions/upload-artifact@v5
+        uses: actions/upload-artifact@v6
         with:
           name: jupyter-releaser-dist-${{ github.run_number }}
           path: .jupyter_releaser_checkout/dist

{jupyter_ydoc-3.3.2 → jupyter_ydoc-3.3.4}/.pre-commit-config.yaml

@@ -17,7 +17,7 @@ repos:
         exclude: ^\.yarn
 
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.14.6
+    rev: v0.14.10
     hooks:
       - id: ruff
         args: [--fix, --show-fixes]

{jupyter_ydoc-3.3.2 → jupyter_ydoc-3.3.4}/CHANGELOG.md

@@ -2,6 +2,47 @@
 
 <!-- <START NEW CHANGELOG ENTRY> -->
 
+## 3.3.4
+
+([Full Changelog](https://github.com/jupyter-server/jupyter_ydoc/compare/@jupyter/ydoc@3.3.3...654e8f80f619c7820dd82dfbe3c6d985b6a69e28))
+
+### Bugs fixed
+
+- Fix cell duplication when YStore diverges from disk [#376](https://github.com/jupyter-server/jupyter_ydoc/pull/376) ([@jordanhboxer](https://github.com/jordanhboxer), [@davidbrochart](https://github.com/davidbrochart), [@dmonad](https://github.com/dmonad), [@krassowski](https://github.com/krassowski))
+
+### Maintenance and upkeep improvements
+
+- Bump toshimaru/auto-author-assign from 3.0.0 to 3.0.1 [#375](https://github.com/jupyter-server/jupyter_ydoc/pull/375) ([@davidbrochart](https://github.com/davidbrochart))
+- Bump toshimaru/auto-author-assign from 2.1.1 to 3.0.0 [#373](https://github.com/jupyter-server/jupyter_ydoc/pull/373) ([@davidbrochart](https://github.com/davidbrochart))
+- Bump actions/upload-artifact from 5 to 6 [#371](https://github.com/jupyter-server/jupyter_ydoc/pull/371) ([@davidbrochart](https://github.com/davidbrochart))
+
+### Contributors to this release
+
+The following people contributed discussions, new ideas, code and documentation contributions, and review.
+See [our definition of contributors](https://github-activity.readthedocs.io/en/latest/#how-does-this-tool-define-contributions-in-the-reports).
+
+([GitHub contributors page for this release](https://github.com/jupyter-server/jupyter_ydoc/graphs/contributors?from=2025-12-10&to=2026-01-09&type=c))
+
+@davidbrochart ([activity](https://github.com/search?q=repo%3Ajupyter-server%2Fjupyter_ydoc+involves%3Adavidbrochart+updated%3A2025-12-10..2026-01-09&type=Issues)) | @dmonad ([activity](https://github.com/search?q=repo%3Ajupyter-server%2Fjupyter_ydoc+involves%3Admonad+updated%3A2025-12-10..2026-01-09&type=Issues)) | @jordanhboxer ([activity](https://github.com/search?q=repo%3Ajupyter-server%2Fjupyter_ydoc+involves%3Ajordanhboxer+updated%3A2025-12-10..2026-01-09&type=Issues)) | @krassowski ([activity](https://github.com/search?q=repo%3Ajupyter-server%2Fjupyter_ydoc+involves%3Akrassowski+updated%3A2025-12-10..2026-01-09&type=Issues))
+
+<!-- <END NEW CHANGELOG ENTRY> -->
+
+## 3.3.3
+
+([Full Changelog](https://github.com/jupyter-server/jupyter_ydoc/compare/@jupyter/ydoc@3.3.2...3900efdeefef21eb6e117fdd30827ee4c9149ff3))
+
+### Bugs fixed
+
+- Fix multi-byte Unicode handling in files [#370](https://github.com/jupyter-server/jupyter_ydoc/pull/370) ([@krassowski](https://github.com/krassowski))
+
+### Maintenance and upkeep improvements
+
+### Contributors to this release
+
+([GitHub contributors page for this release](https://github.com/jupyter-server/jupyter_ydoc/graphs/contributors?from=2025-12-01&to=2025-12-10&type=c))
+
+[@krassowski](https://github.com/search?q=repo%3Ajupyter-server%2Fjupyter_ydoc+involves%3Akrassowski+updated%3A2025-12-01..2025-12-10&type=Issues) | [@pre-commit-ci](https://github.com/search?q=repo%3Ajupyter-server%2Fjupyter_ydoc+involves%3Apre-commit-ci+updated%3A2025-12-01..2025-12-10&type=Issues)
+
 ## 3.3.2
 
 ([Full Changelog](https://github.com/jupyter-server/jupyter_ydoc/compare/@jupyter/ydoc@3.3.1...87e205209dbfed2d367424e6913c88916e77d98e))
@@ -20,8 +61,6 @@
 
 [@dependabot](https://github.com/search?q=repo%3Ajupyter-server%2Fjupyter_ydoc+involves%3Adependabot+updated%3A2025-11-21..2025-12-01&type=Issues) | [@krassowski](https://github.com/search?q=repo%3Ajupyter-server%2Fjupyter_ydoc+involves%3Akrassowski+updated%3A2025-11-21..2025-12-01&type=Issues) | [@pre-commit-ci](https://github.com/search?q=repo%3Ajupyter-server%2Fjupyter_ydoc+involves%3Apre-commit-ci+updated%3A2025-11-21..2025-12-01&type=Issues)
 
-<!-- <END NEW CHANGELOG ENTRY> -->
-
 ## 3.3.1
 
 ([Full Changelog](https://github.com/jupyter-server/jupyter_ydoc/compare/@jupyter/ydoc@3.3.0...0a0d0bc23b5894db0b00c5b073861025757c72db))

{jupyter_ydoc-3.3.2 → jupyter_ydoc-3.3.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: jupyter-ydoc
-Version: 3.3.2
+Version: 3.3.4
 Summary: Document structures for collaborative editing using Ypy
 Project-URL: Homepage, https://jupyter.org
 Project-URL: Source, https://github.com/jupyter-server/jupyter_ydoc

{jupyter_ydoc-3.3.2 → jupyter_ydoc-3.3.4}/javascript/package.json

@@ -1,6 +1,6 @@
 {
     "name": "@jupyter/ydoc",
-    "version": "3.3.2",
+    "version": "3.3.4",
     "type": "module",
     "description": "Jupyter document structures for collaborative editing using YJS",
     "homepage": "https://github.com/jupyter-server/jupyter_ydoc",

{jupyter_ydoc-3.3.2 → jupyter_ydoc-3.3.4}/jupyter_ydoc/_version.py

@@ -1,4 +1,4 @@
 # This file is auto-generated by Hatchling. As such, do not:
 #   - modify
 #   - track in version control e.g. be sure to add to .gitignore
-__version__ = VERSION = '3.3.2'
+__version__ = VERSION = '3.3.4'

{jupyter_ydoc-3.3.2 → jupyter_ydoc-3.3.4}/jupyter_ydoc/ynotebook.py

@@ -251,7 +251,13 @@ class YNotebook(YBaseDoc):
                 "id": str(uuid4()),
             }
         ]
-        old_ycells_by_id: dict[str, Map] = {ycell["id"]: ycell for ycell in self._ycells}
+        # Build dict of old cells by ID, keeping only the first occurrence of each ID
+        # to handle the case where the stored doc already has duplicate IDs.
+        old_ycells_by_id: dict[str, Map] = {}
+        for ycell in self._ycells:
+            cell_id = ycell.get("id")
+            if cell_id is not None and cell_id not in old_ycells_by_id:
+                old_ycells_by_id[cell_id] = ycell
 
         with self._ydoc.transaction():
             new_cell_list: list[dict] = []
@@ -267,34 +273,52 @@ class YNotebook(YBaseDoc):
                     )
 
                     if updated_granularly:
-                        new_cell_list.append(old_cell)
+                        new_cell_list.append(new_cell)
                         retained_cells.add(cell_id)
                         continue
                 # New or changed cell
                 new_cell_list.append(new_cell)
 
-            # First delete all non-retained cells
+            # First delete all non-retained cells and duplicates
             if not retained_cells:
                 # fast path if no cells were retained
                 self._ycells.clear()
             else:
                 index = 0
+                seen: set[str] = set()
                 for old_ycell in list(self._ycells):
-                    if old_ycell["id"] not in retained_cells:
+                    cell_id = old_ycell.get("id")
+                    if cell_id is None or cell_id not in retained_cells or cell_id in seen:
                         self._ycells.pop(index)
                     else:
+                        seen.add(cell_id)
                         index += 1
 
-            # Now add new cells
-            index = 0
-            for new_cell in new_cell_list:
-                if len(self._ycells) > index:
-                    if self._ycells[index]["id"] == new_cell.get("id"):
-                        # retained cell
-                        index += 1
-                        continue
+            # Now reorder/insert cells to match new_cell_list
+            for index, new_cell in enumerate(new_cell_list):
+                new_id = new_cell.get("id")
+
+                # Fast path: correct cell already at this position
+                if len(self._ycells) > index and self._ycells[index].get("id") == new_id:
+                    continue
+
+                # Retained cell: find and move it into position
+                if new_id is not None and new_id in retained_cells:
+                    # Linear scan to find the cell (O(n) per retained cell)
+                    for cur in range(index + 1, len(self._ycells)):
+                        if self._ycells[cur].get("id") == new_id:
+                            # Use delete+recreate instead of move() for yjs 13.x compatibility
+                            # (yjs 13.x doesn't support the move operation that pycrdt generates)
+                            del self._ycells[cur]
+                            self._ycells.insert(index, self.create_ycell(new_cell))
+                            break
+                    continue
+
+                # New cell: insert at position
                 self._ycells.insert(index, self.create_ycell(new_cell))
-                index += 1
+
+            # Remove any extra cells at the end
+            del self._ycells[len(new_cell_list) :]
 
             for key in [
                 k for k in self._ystate.keys() if k not in ("dirty", "path", "document_id")

jupyter_ydoc-3.3.4/jupyter_ydoc/yunicode.py

@@ -0,0 +1,181 @@
+# Copyright (c) Jupyter Development Team.
+# Distributed under the terms of the Modified BSD License.
+
+from collections.abc import Callable
+from difflib import SequenceMatcher
+from functools import partial
+from typing import Any
+
+from pycrdt import Awareness, Doc, Text
+
+from .ybasedoc import YBaseDoc
+
+# Heuristic threshold as recommended in difflib documentation
+SIMILARITY_THREESHOLD = 0.6
+
+
+class YUnicode(YBaseDoc):
+    """
+    Extends :class:`YBaseDoc`, and represents a plain text document, encoded as UTF-8.
+
+    Schema:
+
+    .. code-block:: json
+
+        {
+            "state": YMap,
+            "source": YText
+        }
+    """
+
+    def __init__(self, ydoc: Doc | None = None, awareness: Awareness | None = None):
+        """
+        Constructs a YUnicode.
+
+        :param ydoc: The :class:`pycrdt.Doc` that will hold the data of the document, if provided.
+        :type ydoc: :class:`pycrdt.Doc`, optional.
+        :param awareness: The :class:`pycrdt.Awareness` that shares non persistent data
+                          between clients.
+        :type awareness: :class:`pycrdt.Awareness`, optional.
+        """
+        super().__init__(ydoc, awareness)
+        self._ysource: Text = self._ydoc.get("source", type=Text)
+        self.undo_manager.expand_scope(self._ysource)
+
+    @property
+    def version(self) -> str:
+        """
+        Returns the version of the document.
+
+        :return: Document's version.
+        :rtype: str
+        """
+        return "1.0.0"
+
+    def get(self) -> str:
+        """
+        Returns the content of the document.
+
+        :return: Document's content.
+        :rtype: str
+        """
+        return str(self._ysource)
+
+    def set(self, value: str) -> None:
+        """
+        Sets the content of the document.
+
+        :param value: The content of the document.
+        :type value: str
+        """
+        old_value = self.get()
+        if old_value == value:
+            # no-op if the values are already the same,
+            # to avoid side-effects such as cursor jumping to the top
+            return
+
+        before_bytes = old_value.encode("utf-8")
+        after_bytes = value.encode("utf-8")
+
+        with self._ydoc.transaction():
+            matcher = SequenceMatcher(a=before_bytes, b=after_bytes)
+
+            if (
+                matcher.real_quick_ratio() >= SIMILARITY_THREESHOLD
+                and matcher.ratio() >= SIMILARITY_THREESHOLD
+            ):
+                operations = matcher.get_opcodes()
+
+                # Fix byte ranges and check for problematic overlaps
+                fixed_operations = []
+                prev_end = 0
+                prev_tag = None
+                has_overlap = False
+
+                for tag, i1, i2, j1, j2 in operations:
+                    # Fix byte ranges to proper UTF-8 character boundaries
+                    i1_fixed, i2_fixed = _fix_byte_range_to_char_boundary(before_bytes, i1, i2)
+                    j1_fixed, j2_fixed = _fix_byte_range_to_char_boundary(after_bytes, j1, j2)
+
+                    # Check if this operation overlaps with the previous one
+                    # which can happen with grapheme clusters (emoji + modifiers, etc.)
+                    if i1_fixed < prev_end and prev_tag != "equal":
+                        has_overlap = True
+                        break
+
+                    prev_end = i2_fixed
+                    prev_tag = tag
+                    fixed_operations.append((tag, i1_fixed, i2_fixed, j1_fixed, j2_fixed))
+
+                # If we detected overlapping operations, fall back to hard reload
+                if has_overlap:
+                    self._ysource.clear()
+                    if value:
+                        self._ysource += value
+                else:
+                    # Apply granular operations
+                    offset = 0
+                    for tag, i1, i2, j1, j2 in fixed_operations:
+                        match tag:
+                            case "replace":
+                                self._ysource[i1 + offset : i2 + offset] = after_bytes[
+                                    j1:j2
+                                ].decode("utf-8")
+                                offset += (j2 - j1) - (i2 - i1)
+                            case "delete":
+                                del self._ysource[i1 + offset : i2 + offset]
+                                offset -= i2 - i1
+                            case "insert":
+                                self._ysource.insert(
+                                    i1 + offset, after_bytes[j1:j2].decode("utf-8")
+                                )
+                                offset += j2 - j1
+                            case "equal":
+                                pass
+                            case _:
+                                raise ValueError(f"Unknown tag '{tag}' in sequence matcher")
+            else:
+                # for very different strings, just replace the whole content;
+                # this avoids generating a huge number of operations
+
+                # clear document
+                self._ysource.clear()
+                # initialize document
+                if value:
+                    self._ysource += value
+
+    def observe(self, callback: Callable[[str, Any], None]) -> None:
+        """
+        Subscribes to document changes.
+
+        :param callback: Callback that will be called when the document changes.
+        :type callback: Callable[[str, Any], None]
+        """
+        self.unobserve()
+        self._subscriptions[self._ystate] = self._ystate.observe(partial(callback, "state"))
+        self._subscriptions[self._ysource] = self._ysource.observe(partial(callback, "source"))
+
+
+def _is_utf8_continuation_byte(byte: int) -> bool:
+    """Check if a byte is a UTF-8 continuation byte (10xxxxxx)."""
+    return (byte & 0xC0) == 0x80
+
+
+def _fix_byte_range_to_char_boundary(data: bytes, start: int, end: int) -> tuple[int, int]:
+    """
+    Adjust byte indices to proper UTF-8 character boundaries.
+
+    :param data: The byte data.
+    :param start: The start byte index.
+    :param end: The end byte index.
+    :return: A tuple of (adjusted_start, adjusted_end).
+    """
+    # Move start backward to the beginning of a UTF-8 character
+    while start > 0 and start < len(data) and _is_utf8_continuation_byte(data[start]):
+        start -= 1
+
+    # Move end forward to the end of a UTF-8 character
+    while end < len(data) and _is_utf8_continuation_byte(data[end]):
+        end += 1
+
+    return start, end

jupyter_ydoc-3.3.2/jupyter_ydoc/yunicode.py

@@ -1,120 +0,0 @@
-# Copyright (c) Jupyter Development Team.
-# Distributed under the terms of the Modified BSD License.
-
-from collections.abc import Callable
-from difflib import SequenceMatcher
-from functools import partial
-from typing import Any
-
-from pycrdt import Awareness, Doc, Text
-
-from .ybasedoc import YBaseDoc
-
-# Heuristic threshold as recommended in difflib documentation
-SIMILARITY_THREESHOLD = 0.6
-
-
-class YUnicode(YBaseDoc):
-    """
-    Extends :class:`YBaseDoc`, and represents a plain text document, encoded as UTF-8.
-
-    Schema:
-
-    .. code-block:: json
-
-        {
-            "state": YMap,
-            "source": YText
-        }
-    """
-
-    def __init__(self, ydoc: Doc | None = None, awareness: Awareness | None = None):
-        """
-        Constructs a YUnicode.
-
-        :param ydoc: The :class:`pycrdt.Doc` that will hold the data of the document, if provided.
-        :type ydoc: :class:`pycrdt.Doc`, optional.
-        :param awareness: The :class:`pycrdt.Awareness` that shares non persistent data
-                          between clients.
-        :type awareness: :class:`pycrdt.Awareness`, optional.
-        """
-        super().__init__(ydoc, awareness)
-        self._ysource: Text = self._ydoc.get("source", type=Text)
-        self.undo_manager.expand_scope(self._ysource)
-
-    @property
-    def version(self) -> str:
-        """
-        Returns the version of the document.
-
-        :return: Document's version.
-        :rtype: str
-        """
-        return "1.0.0"
-
-    def get(self) -> str:
-        """
-        Returns the content of the document.
-
-        :return: Document's content.
-        :rtype: str
-        """
-        return str(self._ysource)
-
-    def set(self, value: str) -> None:
-        """
-        Sets the content of the document.
-
-        :param value: The content of the document.
-        :type value: str
-        """
-        old_value = self.get()
-        if old_value == value:
-            # no-op if the values are already the same,
-            # to avoid side-effects such as cursor jumping to the top
-            return
-
-        with self._ydoc.transaction():
-            matcher = SequenceMatcher(a=old_value, b=value)
-
-            if (
-                matcher.real_quick_ratio() >= SIMILARITY_THREESHOLD
-                and matcher.ratio() >= SIMILARITY_THREESHOLD
-            ):
-                operations = matcher.get_opcodes()
-                offset = 0
-                for tag, i1, i2, j1, j2 in operations:
-                    match tag:
-                        case "replace":
-                            self._ysource[i1 + offset : i2 + offset] = value[j1:j2]
-                            offset += (j2 - j1) - (i2 - i1)
-                        case "delete":
-                            del self._ysource[i1 + offset : i2 + offset]
-                            offset -= i2 - i1
-                        case "insert":
-                            self._ysource.insert(i1 + offset, value[j1:j2])
-                            offset += j2 - j1
-                        case "equal":
-                            pass
-                        case _:
-                            raise ValueError(f"Unknown tag '{tag}' in sequence matcher")
-            else:
-                # for very different strings, just replace the whole content;
-                # this avoids generating a huge number of operations
-
-                # clear document
-                self._ysource.clear()
-                # initialize document
-                if value:
-                    self._ysource += value
-
-    def observe(self, callback: Callable[[str, Any], None]) -> None:
-        """
-        Subscribes to document changes.
-
-        :param callback: Callback that will be called when the document changes.
-        :type callback: Callable[[str, Any], None]
-        """
-        self.unobserve()
-        self._subscriptions[self._ystate] = self._ystate.observe(partial(callback, "state"))
-        self._subscriptions[self._ysource] = self._ysource.observe(partial(callback, "source"))