diff --git a/paddleocr/_doc2md/math/latex_dict.py b/paddleocr/_doc2md/math/latex_dict.py
index cd65132a7d0..050b75e29e4 100644
--- a/paddleocr/_doc2md/math/latex_dict.py
+++ b/paddleocr/_doc2md/math/latex_dict.py
@@ -320,6 +320,9 @@
"lim": "\\lim_{{{lim}}}",
"max": "\\max_{{{lim}}}",
"min": "\\min_{{{lim}}}",
+ "inf": "\\inf_{{{lim}}}",
+ "sup": "\\sup_{{{lim}}}",
+ "gcd": "\\gcd_{{{lim}}}",
}
LIM_TO = ("\\rightarrow", "\\to")
diff --git a/paddleocr/_doc2md/math/omml.py b/paddleocr/_doc2md/math/omml.py
index 8fadc1cef8c..adca74d637b 100644
--- a/paddleocr/_doc2md/math/omml.py
+++ b/paddleocr/_doc2md/math/omml.py
@@ -141,7 +141,7 @@ def process_unknow(self, elm, stag):
class Pr(Tag2Method):
text = ""
- __val_tags = ("chr", "pos", "begChr", "endChr", "type")
+ __val_tags = ("chr", "pos", "begChr", "endChr", "sepChr", "type")
__innerdict = None # can't use the __dict__
@@ -174,6 +174,7 @@ def do_common(self, elm):
"pos": do_common,
"begChr": do_common,
"endChr": do_common,
+ "sepChr": do_common,
"type": do_common,
}
@@ -227,33 +228,51 @@ def do_acc(self, elm):
the accent function
"""
c_dict = self.process_children_dict(elm)
+ pr = c_dict.get("accPr")
latex_s = get_val(
- c_dict["accPr"].chr, default=CHR_DEFAULT.get("ACC_VAL"), store=CHR
+ pr.chr if pr is not None else None,
+ default=CHR_DEFAULT.get("ACC_VAL"),
+ store=CHR,
)
- return latex_s.format(c_dict["e"])
+ return (pr.text if pr is not None else "") + latex_s.format(c_dict["e"])
def do_bar(self, elm):
"""
the bar function
"""
c_dict = self.process_children_dict(elm)
- pr = c_dict["barPr"]
- latex_s = get_val(pr.pos, default=POS_DEFAULT.get("BAR_VAL"), store=POS)
- return pr.text + latex_s.format(c_dict["e"])
+ pr = c_dict.get("barPr")
+ latex_s = get_val(
+ pr.pos if pr is not None else None,
+ default=POS_DEFAULT.get("BAR_VAL"),
+ store=POS,
+ )
+ return (pr.text if pr is not None else "") + latex_s.format(c_dict["e"])
def do_d(self, elm):
"""
the delimiter object
"""
- c_dict = self.process_children_dict(elm)
- pr = c_dict["dPr"]
+ children = list(self.process_children_list(elm))
+ pr = next((text for stag, text, _ in children if stag == "dPr"), None)
null = D_DEFAULT.get("null")
-
- s_val = get_val(pr.begChr, default=D_DEFAULT.get("left"), store=T)
- e_val = get_val(pr.endChr, default=D_DEFAULT.get("right"), store=T)
- delim = pr.text + D.format(
+ expressions = [text for stag, text, _ in children if stag == "e"]
+ separator = get_val(pr.sepChr if pr is not None else None, default=",")
+ separator = r"\middle|" if separator == "|" else escape_latex(separator)
+
+ s_val = get_val(
+ pr.begChr if pr is not None else None,
+ default=D_DEFAULT.get("left"),
+ store=T,
+ )
+ e_val = get_val(
+ pr.endChr if pr is not None else None,
+ default=D_DEFAULT.get("right"),
+ store=T,
+ )
+ delim = (pr.text if pr is not None else "") + D.format(
left=null if not s_val else escape_latex(s_val),
- text=c_dict["e"],
+ text=separator.join(expressions),
right=null if not e_val else escape_latex(e_val),
)
return delim
@@ -288,6 +307,8 @@ def do_func(self, elm):
"""
c_dict = self.process_children_dict(elm)
func_name = c_dict.get("fName")
+ if func_name is None:
+ return c_dict.get("e", "")
return func_name.replace(FUNC_PLACE, c_dict.get("e"))
def do_fname(self, elm):
@@ -315,9 +336,12 @@ def do_groupchr(self, elm):
the Group-Character object
"""
c_dict = self.process_children_dict(elm)
- pr = c_dict["groupChrPr"]
- latex_s = get_val(pr.chr)
- return pr.text + latex_s.format(c_dict["e"])
+ pr = c_dict.get("groupChrPr")
+ latex_s = get_val(
+ pr.chr if pr is not None else None,
+ default=CHR["\u23df"],
+ )
+ return (pr.text if pr is not None else "") + latex_s.format(c_dict["e"])
def do_rad(self, elm):
"""
@@ -355,9 +379,10 @@ def do_limlow(self, elm):
t_dict = self.process_children_dict(elm, include=("e", "lim"))
latex_s = LIM_FUNC.get(t_dict["e"])
if not latex_s:
- raise RuntimeError("Not support lim {}".format(t_dict["e"]))
- else:
- return latex_s.format(lim=t_dict.get("lim"))
+ return r"\underset{{{lim}}}{{{text}}}".format(
+ lim=t_dict.get("lim"), text=t_dict["e"]
+ )
+ return latex_s.format(lim=t_dict.get("lim"))
def do_limupp(self, elm):
"""
diff --git a/tests/unit/test_doc2md_omml.py b/tests/unit/test_doc2md_omml.py
new file mode 100644
index 00000000000..b4cd62e7448
--- /dev/null
+++ b/tests/unit/test_doc2md_omml.py
@@ -0,0 +1,94 @@
+# Copyright (c) 2026 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Regression tests for standalone OMML conversion."""
+
+import importlib.util
+import sys
+import types
+from pathlib import Path
+
+import pytest
+
+etree = pytest.importorskip("lxml.etree")
+pytest.importorskip("pylatexenc")
+
+MATH_DIR = Path(__file__).resolve().parents[2] / "paddleocr" / "_doc2md" / "math"
+OMML_NS = "http://schemas.openxmlformats.org/officeDocument/2006/math"
+
+
+def _load_omml():
+ """Load the converter without importing the full PaddleOCR model stack."""
+ package = types.ModuleType("test_doc2md_math")
+ package.__path__ = [str(MATH_DIR)]
+ sys.modules[package.__name__] = package
+ spec = importlib.util.spec_from_file_location(
+ "test_doc2md_math.omml", MATH_DIR / "omml.py"
+ )
+ module = importlib.util.module_from_spec(spec)
+ sys.modules[spec.name] = module
+ spec.loader.exec_module(module)
+ return module
+
+
+@pytest.mark.parametrize(
+ ("tag", "expected"),
+ [
+ ("acc", r"\hat{x}"),
+ ("bar", r"\overline{x}"),
+ ("d", r"\left(x\right)"),
+ ("groupChr", r"\underbrace{x}"),
+ ("func", "x"),
+ ],
+)
+def test_optional_property_nodes_do_not_drop_formula(tag, expected):
+ omml = _load_omml()
+ xml = (
+ f''
+ "x"
+ f""
+ )
+ assert str(omml.oMath2Latex(etree.fromstring(xml))) == expected
+
+
+def test_explicit_bar_position_is_preserved():
+ omml = _load_omml()
+ xml = (
+ f''
+ ''
+ "x"
+ ""
+ )
+ assert str(omml.oMath2Latex(etree.fromstring(xml))) == r"\underline{x}"
+
+
+@pytest.mark.parametrize(
+ ("base", "expected"),
+ [
+ ("min", r"\min_{n}"),
+ ("inf", r"\inf_{n}"),
+ ("sup", r"\sup_{n}"),
+ ("gcd", r"\gcd_{n}"),
+ ("foo", r"\underset{n}{foo}"),
+ ],
+)
+def test_limit_below_preserves_unsupported_base(base, expected):
+ omml = _load_omml()
+ xml = (
+ f''
+ f"{base}"
+ "n"
+ ""
+ )
+ assert str(omml.oMath2Latex(etree.fromstring(xml))) == expected
diff --git a/tests/unit/test_doc2md_omml_delimiter.py b/tests/unit/test_doc2md_omml_delimiter.py
new file mode 100644
index 00000000000..5e4c6f01f9b
--- /dev/null
+++ b/tests/unit/test_doc2md_omml_delimiter.py
@@ -0,0 +1,64 @@
+# Copyright (c) 2026 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Regression tests for delimiters with multiple OMML expressions."""
+
+import importlib.util
+import sys
+import types
+from pathlib import Path
+
+import pytest
+
+etree = pytest.importorskip("lxml.etree")
+pytest.importorskip("pylatexenc")
+
+MATH_DIR = Path(__file__).resolve().parents[2] / "paddleocr" / "_doc2md" / "math"
+OMML_NS = "http://schemas.openxmlformats.org/officeDocument/2006/math"
+
+
+def _load_omml():
+ package = types.ModuleType("test_doc2md_delimiter_math")
+ package.__path__ = [str(MATH_DIR)]
+ sys.modules[package.__name__] = package
+ spec = importlib.util.spec_from_file_location(
+ "test_doc2md_delimiter_math.omml", MATH_DIR / "omml.py"
+ )
+ module = importlib.util.module_from_spec(spec)
+ sys.modules[spec.name] = module
+ spec.loader.exec_module(module)
+ return module
+
+
+@pytest.mark.parametrize(
+ ("values", "separator", "expected"),
+ [
+ (["n", "k"], "|", r"\left(n\middle|k\right)"),
+ (["a", "b", "c"], ";", r"\left(a;b;c\right)"),
+ (["n", "k"], None, r"\left(n,k\right)"),
+ (["x"], "|", r"\left(x\right)"),
+ ],
+)
+def test_delimiter_keeps_all_expressions_and_separator(values, separator, expected):
+ omml = _load_omml()
+ prop = (
+ f''
+ if separator is not None
+ else ""
+ )
+ expressions = "".join(
+ f"{value}" for value in values
+ )
+ xml = f'{prop}{expressions}'
+ assert str(omml.oMath2Latex(etree.fromstring(xml))) == expected