docling-project · maxmnemonic · Apr 15, 2025 · Apr 15, 2025 · Apr 15, 2025 · Apr 15, 2025
diff --git a/docling_core/experimental/serializer/html.py b/docling_core/experimental/serializer/html.py
@@ -57,6 +57,7 @@
     NodeItem,
     OrderedList,
     PictureItem,
+    PictureTabularChartData,
     SectionHeaderItem,
     TableCell,
     TableItem,
@@ -104,6 +105,9 @@ class HTMLParams(CommonParams):
     # Allow for different output styles
     output_style: HTMLOutputStyle = HTMLOutputStyle.SINGLE_COLUMN
 
+    # Enable charts to be printed into HTML as tables
+    enable_chart_tables: bool = True
+
 
 class HTMLTextSerializer(BaseModel, BaseTextSerializer):
     """HTML-specific text item serializer."""
@@ -402,9 +406,28 @@ def serialize(
                     and item.image.uri.scheme == "data"
                 ):
                     img_text = f'<img src="{quote(str(item.image.uri))}">'
+
         if img_text:
             res_parts.append(create_ser_result(text=img_text, span_source=item))
 
+        if params.enable_chart_tables:
+            # Check if picture has attached PictureTabularChartData
+            tabular_chart_annotations = [
+                ann
+                for ann in item.annotations
+                if isinstance(ann, PictureTabularChartData)
+            ]
+            if len(tabular_chart_annotations) > 0:
+                temp_doc = DoclingDocument(name="temp")
+                temp_table = temp_doc.add_table(
+                    data=tabular_chart_annotations[0].chart_data
+                )
+                html_table_content = temp_table.export_to_html(temp_doc)
+                if len(html_table_content) > 0:
+                    res_parts.append(
+                        create_ser_result(text=html_table_content, span_source=item)
+                    )
+
         text_res = "".join([r.text for r in res_parts])
         if text_res:
             text_res = f"<figure>{text_res}</figure>"

diff --git a/docling_core/types/doc/document.py b/docling_core/types/doc/document.py
@@ -3143,6 +3143,7 @@ def export_to_html(  # noqa: C901
         from_element: int = 0,
         to_element: int = sys.maxsize,
         labels: Optional[set[DocItemLabel]] = None,
+        enable_chart_tables: bool = True,
         image_mode: ImageRefMode = ImageRefMode.PLACEHOLDER,
         formula_to_mathml: bool = True,
         page_no: Optional[int] = None,
@@ -3176,6 +3177,7 @@ def export_to_html(  # noqa: C901
             start_idx=from_element,
             stop_idx=to_element,
             image_mode=image_mode,
+            enable_chart_tables=enable_chart_tables,
             formula_to_mathml=formula_to_mathml,
             html_head=html_head,
             html_lang=html_lang,

diff --git a/test/data/doc/barchart.dt b/test/data/doc/barchart.dt
@@ -0,0 +1,3 @@
+<doctag><page_header><loc_71><loc_14><loc_217><loc_20>Probability, Combinatorics and Control</page_header>
+<chart><loc_102><loc_37><loc_392><loc_148><bar_chart><ched>Number of impellers<ched>single-frequency<ched>multi-frequency<nl><fcel>1<fcel>0.06<fcel>0.16<nl><fcel>2<fcel>0.12<fcel>0.26<nl><fcel>3<fcel>0.16<fcel>0.27<nl><fcel>4<fcel>0.14<fcel>0.26<nl><fcel>5<fcel>0.16<fcel>0.25<nl><fcel>6<fcel>0.24<fcel>0.24<nl></chart>
+</doctag>
diff --git a/test/data/doc/barchart.gt.html b/test/data/doc/barchart.gt.html
@@ -0,0 +1,130 @@
+<!DOCTYPE html>
+<html>
+<head>
+<meta charset="UTF-8">
+<title>Document</title>
+<meta name="generator" content="Docling HTML Serializer">
+<style>
+    html {
+        background-color: #f5f5f5;
+        font-family: Arial, sans-serif;
+        line-height: 1.6;
+    }
+    body {
+        max-width: 800px;
+        margin: 0 auto;
+        padding: 2rem;
+        background-color: white;
+        box-shadow: 0 0 10px rgba(0,0,0,0.1);
+    }
+    h1, h2, h3, h4, h5, h6 {
+        color: #333;
+        margin-top: 1.5em;
+        margin-bottom: 0.5em;
+    }
+    h1 {
+        font-size: 2em;
+        border-bottom: 1px solid #eee;
+        padding-bottom: 0.3em;
+    }
+    table {
+        border-collapse: collapse;
+        margin: 1em 0;
+        width: 100%;
+    }
+    th, td {
+        border: 1px solid #ddd;
+        padding: 8px;
+        text-align: left;
+    }
+    th {
+        background-color: #f2f2f2;
+        font-weight: bold;
+    }
+    figure {
+        margin: 1.5em 0;
+        text-align: center;
+    }
+    figcaption {
+        color: #666;
+        font-style: italic;
+        margin-top: 0.5em;
+    }
+    img {
+        max-width: 100%;
+        height: auto;
+    }
+    pre {
+        background-color: #f6f8fa;
+        border-radius: 3px;
+        padding: 1em;
+        overflow: auto;
+    }
+    code {
+        font-family: monospace;
+        background-color: #f6f8fa;
+        padding: 0.2em 0.4em;
+        border-radius: 3px;
+    }
+    pre code {
+        background-color: transparent;
+        padding: 0;
+    }
+    .formula {
+        text-align: center;
+        padding: 0.5em;
+        margin: 1em 0;
+        background-color: #f9f9f9;
+    }
+    .formula-not-decoded {
+        text-align: center;
+        padding: 0.5em;
+        margin: 1em 0;
+        background: repeating-linear-gradient(
+            45deg,
+            #f0f0f0,
+            #f0f0f0 10px,
+            #f9f9f9 10px,
+            #f9f9f9 20px
+        );
+    }
+    .page-break {
+        page-break-after: always;
+        border-top: 1px dashed #ccc;
+        margin: 2em 0;
+    }
+    .key-value-region {
+        background-color: #f9f9f9;
+        padding: 1em;
+        border-radius: 4px;
+        margin: 1em 0;
+    }
+    .key-value-region dt {
+        font-weight: bold;
+    }
+    .key-value-region dd {
+        margin-left: 1em;
+        margin-bottom: 0.5em;
+    }
+    .form-container {
+        border: 1px solid #ddd;
+        padding: 1em;
+        border-radius: 4px;
+        margin: 1em 0;
+    }
+    .form-item {
+        margin-bottom: 0.5em;
+    }
+    .image-classification {
+        font-size: 0.9em;
+        color: #666;
+        margin-top: 0.5em;
+    }
+</style>
+</head>
+<body>
+<div class='page'>
+<figure><table><tbody><tr><td>Number of impellers</td><td>single-frequency</td><td>multi-frequency</td></tr><tr><td>1</td><td>0.06</td><td>0.16</td></tr><tr><td>2</td><td>0.12</td><td>0.26</td></tr><tr><td>3</td><td>0.16</td><td>0.27</td></tr><tr><td>4</td><td>0.14</td><td>0.26</td></tr><tr><td>5</td><td>0.16</td><td>0.25</td></tr><tr><td>6</td><td>0.24</td><td>0.24</td></tr></tbody></table></figure>
+</div>
+</body>
+</html>
diff --git a/test/data/doc/barchart.gt.md b/test/data/doc/barchart.gt.md
@@ -0,0 +1,10 @@
+<!-- image -->
+
+|   Number of impellers |   single-frequency |   multi-frequency |
+|-----------------------|--------------------|-------------------|
+|                     1 |               0.06 |              0.16 |
+|                     2 |               0.12 |              0.26 |
+|                     3 |               0.16 |              0.27 |
+|                     4 |               0.14 |              0.26 |
+|                     5 |               0.16 |              0.25 |
+|                     6 |               0.24 |              0.24 |
diff --git a/test/data/doc/barchart.json b/test/data/doc/barchart.json
diff --git a/test/data/doc/barchart.png b/test/data/doc/barchart.png
diff --git a/test/test_doctags_load.py b/test/test_doctags_load.py
@@ -3,7 +3,7 @@
 from PIL import Image as PILImage
 
 from docling_core.types.doc import DoclingDocument
-from docling_core.types.doc.document import DocTagsDocument
+from docling_core.types.doc.document import DocTagsDocument, PictureTabularChartData
 
 
 def test_doctags_load_from_files():
@@ -55,6 +55,20 @@ def test_multipage_doctags_load():
     # print(doc.export_to_html())
 
 
+def test_doctags_chart():
+    doc = DoclingDocument(name="Document")
+    doctags_doc = DocTagsDocument.from_doctags_and_image_pairs(
+        [Path("test/data/doc/barchart.dt")],
+        [Path("test/data/doc/barchart.png")],
+    )
+    doc.load_from_doctags(doctags_doc)
+    for pic in doc.pictures:
+        tabular_chart_annotations = [
+            ann for ann in pic.annotations if isinstance(ann, PictureTabularChartData)
+        ]
+        assert len(tabular_chart_annotations) > 0
+
+
 def test_doctags_table_provenances_and_captions():
     doc = DoclingDocument(name="Document")
     doctags_doc = DocTagsDocument.from_doctags_and_image_pairs(

diff --git a/test/test_serialization.py b/test/test_serialization.py
@@ -64,6 +64,34 @@ def test_md_cross_page_list_page_break_p2():
     verify(exp_file=src.parent / f"{src.stem}_p2.gt.md", actual=actual)
 
 
+def test_html_charts():
+    src = Path("./test/data/doc/barchart.json")
+    doc = DoclingDocument.load_from_json(src)
+
+    ser = HTMLDocSerializer(
+        doc=doc,
+        params=HTMLParams(
+            image_mode=ImageRefMode.PLACEHOLDER,
+        ),
+    )
+    actual = ser.serialize().text
+    verify(exp_file=src.parent / f"{src.stem}.gt.html", actual=actual)
+
+
+def test_md_charts():
+    src = Path("./test/data/doc/barchart.json")
+    doc = DoclingDocument.load_from_json(src)
+
+    ser = MarkdownDocSerializer(
+        doc=doc,
+        params=MarkdownParams(
+            image_mode=ImageRefMode.PLACEHOLDER,
+        ),
+    )
+    actual = ser.serialize().text
+    verify(exp_file=src.parent / f"{src.stem}.gt.md", actual=actual)
+
+
 def test_html_cross_page_list_page_break():
     src = Path("./test/data/doc/activities.json")
     doc = DoclingDocument.load_from_json(src)