fix: sync prosemirror-schema-basic to 1.2.4

- Add clearMark parse rules for em and strong marks (1.2.1)
- Add getAttrs on strong <b> tag rule for Google Docs workaround (1.2.1)
- Add getAttrs regex filter on strong font-weight style rule (1.2.1)
- Add code: true to code mark spec (1.2.4)
- Fix image parseDOM: use img[src] selector, extract alt attribute
- Fix link parseDOM: use a[href] selector, extract title attribute

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Author: sciyoshi

README.md

@@ -12,7 +12,7 @@ This package provides Python implementations of the following
 - [`prosemirror-model`](https://github.com/ProseMirror/prosemirror-model) version 1.25.4
 - [`prosemirror-transform`](https://github.com/ProseMirror/prosemirror-transform) version 1.8.0
 - [`prosemirror-test-builder`](https://github.com/ProseMirror/prosemirror-test-builder)
-- [`prosemirror-schema-basic`](https://github.com/ProseMirror/prosemirror-schema-basic) version 1.1.2
+- [`prosemirror-schema-basic`](https://github.com/ProseMirror/prosemirror-schema-basic) version 1.2.4
 - [`prosemirror-schema-list`](https://github.com/ProseMirror/prosemirror-schema-list) version 1.5.1 (node specs and `wrapRangeInList` only; command functions that depend on `prosemirror-state` are excluded)
 
 The original implementation has been followed as closely as possible during

prosemirror/schema/basic/schema_basic.py

@@ -1,3 +1,4 @@
+import re
 from typing import Any
 
 from prosemirror.model import Schema
@@ -65,10 +66,11 @@
         "draggable": True,
         "parseDOM": [
             {
-                "tag": "img",
+                "tag": "img[src]",
                 "getAttrs": lambda dom_: {
                     "src": dom_.get("src"),
                     "title": dom_.get("title"),
+                    "alt": dom_.get("alt"),
                 },
             },
         ],
@@ -101,22 +103,66 @@
             "title": {"default": None, "validate": "string|null"},
         },
         "inclusive": False,
-        "parseDOM": [{"tag": "a", "getAttrs": lambda d: {"href": d.get("href")}}],
+        "parseDOM": [
+            {
+                "tag": "a[href]",
+                "getAttrs": lambda d: {
+                    "href": d.get("href"),
+                    "title": d.get("title"),
+                },
+            },
+        ],
         "toDOM": lambda node, _: [
             "a",
             {"href": node.attrs["href"], "title": node.attrs["title"]},
             0,
         ],
     },
     "em": {
-        "parseDOM": [{"tag": "i"}, {"tag": "em"}, {"style": "font-style=italic"}],
+        "parseDOM": [
+            {"tag": "i"},
+            {"tag": "em"},
+            {"style": "font-style=italic"},
+            {
+                "style": "font-style=normal",
+                "clearMark": lambda m: m.type.name == "em",
+            },
+        ],
         "toDOM": lambda _, __: em_dom,
     },
     "strong": {
-        "parseDOM": [{"tag": "strong"}, {"tag": "b"}, {"style": "font-weight"}],
+        "parseDOM": [
+            {"tag": "strong"},
+            # This works around a Google Docs misbehavior where
+            # pasted content will be inexplicably wrapped in <b>
+            # tags with a font-weight normal.
+            {
+                "tag": "b",
+                "getAttrs": lambda node: (
+                    None
+                    if "font-weight:normal"
+                    not in (node.get("style") or "").replace(" ", "")
+                    else False
+                ),
+            },
+            {
+                "style": "font-weight=400",
+                "clearMark": lambda m: m.type.name == "strong",
+            },
+            {
+                "style": "font-weight",
+                "getAttrs": lambda value: (
+                    None if re.match(r"^(bold(er)?|[5-9]\d{2,})$", value) else False
+                ),
+            },
+        ],
         "toDOM": lambda _, __: strong_dom,
     },
-    "code": {"parseDOM": [{"tag": "code"}], "toDOM": lambda _, __: code_dom},
+    "code": {
+        "code": True,
+        "parseDOM": [{"tag": "code"}],
+        "toDOM": lambda _, __: code_dom,
+    },
 }