Skip to content

RAGFlow is an agent-native database: a Retrieval-Augmented Generation (RAG) engine that grew into context infrastructure.

fix(parser): keep a one-row slide table instead of dropping it (#19705)

Lukas Buckf7498bae4928
diff --git a/deepdoc/parser/ppt_parser.py b/deepdoc/parser/ppt_parser.py
index 9d1537f7c3..f093a1688e 100644
--- a/deepdoc/parser/ppt_parser.py
+++ b/deepdoc/parser/ppt_parser.py
@@ -60,6 +60,11 @@ class RAGFlowPptParser:
# Handle table
if shape_type == 19:
tb = shape.table
+ if len(tb.rows) == 1:
+ # The loop below pairs every row with the first one, so a table
+ # that only has that row produces nothing at all. There is no
+ # header to pair against, so the row itself is the content.
+ return "; ".join(tb.cell(0, j).text for j in range(len(tb.columns)))
rows = []
for i in range(1, len(tb.rows)):
rows.append("; ".join([tb.cell(0, j).text + ": " + tb.cell(i, j).text for j in range(len(tb.columns)) if tb.cell(i, j)]))
diff --git a/test/unit_test/deepdoc/parser/test_ppt_parser.py b/test/unit_test/deepdoc/parser/test_ppt_parser.py
new file mode 100644
index 0000000000..2609ac9159
--- /dev/null
+++ b/test/unit_test/deepdoc/parser/test_ppt_parser.py
@@ -0,0 +1,94 @@
+#
+# Copyright 2025 The InfiniFlow 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.
+#
+
+"""Unit tests for RAGFlowPptParser table extraction.
+
+A slide table is turned into "header: value" pairs by pairing every row with the
+first one. A table that only has that one row therefore has nothing to pair with,
+and used to come back empty.
+"""
+
+import importlib.util
+import io
+import os
+import sys
+
+import pytest
+
+pptx = pytest.importorskip("pptx")
+from pptx import Presentation # noqa: E402
+from pptx.util import Inches # noqa: E402
+
+# Import by file path so deepdoc/parser/__init__.py, which pulls in heavy
+# dependencies, is not needed for this test.
+_PARSER_PATH = os.path.join(
+ os.path.dirname(os.path.abspath(__file__)),
+ "..",
+ "..",
+ "..",
+ "..",
+ "deepdoc",
+ "parser",
+ "ppt_parser.py",
+)
+_spec = importlib.util.spec_from_file_location("ppt_parser_under_test", os.path.normpath(_PARSER_PATH))
+_module = importlib.util.module_from_spec(_spec)
+sys.modules["ppt_parser_under_test"] = _module
+_spec.loader.exec_module(_module)
+RAGFlowPptParser = _module.RAGFlowPptParser
+
+
+def _deck_with_table(rows: int, cols: int = 2) -> bytes:
+ """A one-slide deck with a text box and a `rows` x `cols` table."""
+ presentation = Presentation()
+ slide = presentation.slides.add_slide(presentation.slide_layouts[6])
+
+ textbox = slide.shapes.add_textbox(Inches(0.5), Inches(0.2), Inches(4), Inches(0.5))
+ textbox.text_frame.text = "SLIDE TEXT"
+
+ table = slide.shapes.add_table(rows, cols, Inches(0.5), Inches(1.0), Inches(6), Inches(0.8 * rows)).table
+ for row in range(rows):
+ for col in range(cols):
+ table.cell(row, col).text = f"R{row}C{col}"
+
+ buffer = io.BytesIO()
+ presentation.save(buffer)
+ return buffer.getvalue()
+
+
+def _parse(payload: bytes) -> str:
+ return RAGFlowPptParser()(payload, 0, 100000)[0]
+
+
+def test_single_row_table_is_not_dropped():
+ text = _parse(_deck_with_table(rows=1))
+
+ assert "R0C0" in text
+ assert "R0C1" in text
+ assert "SLIDE TEXT" in text
+
+
+def test_two_row_table_still_pairs_the_header_with_the_value():
+ text = _parse(_deck_with_table(rows=2))
+
+ assert "R0C0: R1C0; R0C1: R1C1" in text
+
+
+def test_three_row_table_pairs_every_row_with_the_header():
+ text = _parse(_deck_with_table(rows=3))
+
+ assert "R0C0: R1C0; R0C1: R1C1" in text
+ assert "R0C0: R2C0; R0C1: R2C1" in text