fix: Remove Extra Space Before and After Group Items

docling_core/transforms/serializer/html.py

@@ -787,7 +787,7 @@ def serialize(
         )
 
         # Join all parts without separators
-        inline_html = " ".join([p.text for p in parts if p.text])
+        inline_html = "".join([p.text for p in parts if p.text])
 
         # Wrap in span if needed
         if inline_html:

docling_core/transforms/serializer/markdown.py

@@ -670,7 +670,7 @@ def serialize(
             visited=my_visited,
             **kwargs,
         )
-        text_res = " ".join([p.text for p in parts if p.text])
+        text_res = "".join([p.text for p in parts if p.text])
         return create_ser_result(text=text_res, span_source=parts)
 
 

test/data/doc/2408.09869v3_enriched_p2_p3_p5.gt.html

@@ -105,7 +105,7 @@
 <li style="list-style-type: '· ';">Can leverage different accelerators (GPU, MPS, etc).</li>
 </ul>
 <h2>2 Getting Started</h2>
-<span class='inline-group'>To use Docling, you can simply install the docling package from PyPI. Documentation and examples are available in our GitHub repository at  <a href="https://github.com/DS4SD/docling">github.com/DS4SD/docling</a> . All required model assets 1 are downloaded to a local huggingface datasets cache on first use, unless you choose to pre-install the model assets in advance.</span>
+<span class='inline-group'>To use Docling, you can simply install the docling package from PyPI. Documentation and examples are available in our GitHub repository at <a href="https://github.com/DS4SD/docling">github.com/DS4SD/docling</a>. All required model assets 1 are downloaded to a local huggingface datasets cache on first use, unless you choose to pre-install the model assets in advance.</span>
 <p>Docling provides an easy code interface to convert PDF documents from file system, URLs or binary streams, and retrieve the output in either JSON or Markdown format. For convenience, separate methods are offered to convert single documents or batches of documents. A basic usage example is illustrated below. Further examples are available in the Doclign code repository.</p>
 <pre><code>from docling.document_converter import DocumentConverter Large</code></pre>
 <pre><code>source = "https://arxiv.org/pdf/2206.01062" # PDF path or URL converter = DocumentConverter() result = converter.convert_single(source) print(result.render_as_markdown()) # output: "## DocLayNet: A Human -Annotated Dataset for Document -Layout Analysis [...]"</code></pre>

test/data/doc/2408.09869v3_enriched_split.gt.html

@@ -127,7 +127,7 @@ <h2>1 Introduction</h2>
 <li style="list-style-type: '· ';">Can leverage different accelerators (GPU, MPS, etc).</li>
 </ul>
 <h2>2 Getting Started</h2>
-<span class='inline-group'>To use Docling, you can simply install the docling package from PyPI. Documentation and examples are available in our GitHub repository at  <a href="https://github.com/DS4SD/docling">github.com/DS4SD/docling</a> . All required model assets 1 are downloaded to a local huggingface datasets cache on first use, unless you choose to pre-install the model assets in advance.</span>
+<span class='inline-group'>To use Docling, you can simply install the docling package from PyPI. Documentation and examples are available in our GitHub repository at <a href="https://github.com/DS4SD/docling">github.com/DS4SD/docling</a>. All required model assets 1 are downloaded to a local huggingface datasets cache on first use, unless you choose to pre-install the model assets in advance.</span>
 <p>Docling provides an easy code interface to convert PDF documents from file system, URLs or binary streams, and retrieve the output in either JSON or Markdown format. For convenience, separate methods are offered to convert single documents or batches of documents. A basic usage example is illustrated below. Further examples are available in the Doclign code repository.</p>
 <pre><code>from docling.document_converter import DocumentConverter Large</code></pre>
 <pre><code>source = "https://arxiv.org/pdf/2206.01062" # PDF path or URL converter = DocumentConverter() result = converter.convert_single(source) print(result.render_as_markdown()) # output: "## DocLayNet: A Human -Annotated Dataset for Document -Layout Analysis [...]"</code></pre>

test/data/doc/2408.09869v3_enriched_split_p2.gt.html

@@ -106,7 +106,7 @@
 <li style="list-style-type: '· ';">Can leverage different accelerators (GPU, MPS, etc).</li>
 </ul>
 <h2>2 Getting Started</h2>
-<span class='inline-group'>To use Docling, you can simply install the docling package from PyPI. Documentation and examples are available in our GitHub repository at  <a href="https://github.com/DS4SD/docling">github.com/DS4SD/docling</a> . All required model assets 1 are downloaded to a local huggingface datasets cache on first use, unless you choose to pre-install the model assets in advance.</span>
+<span class='inline-group'>To use Docling, you can simply install the docling package from PyPI. Documentation and examples are available in our GitHub repository at <a href="https://github.com/DS4SD/docling">github.com/DS4SD/docling</a>. All required model assets 1 are downloaded to a local huggingface datasets cache on first use, unless you choose to pre-install the model assets in advance.</span>
 <p>Docling provides an easy code interface to convert PDF documents from file system, URLs or binary streams, and retrieve the output in either JSON or Markdown format. For convenience, separate methods are offered to convert single documents or batches of documents. A basic usage example is illustrated below. Further examples are available in the Doclign code repository.</p>
 <pre><code>from docling.document_converter import DocumentConverter Large</code></pre>
 <pre><code>source = "https://arxiv.org/pdf/2206.01062" # PDF path or URL converter = DocumentConverter() result = converter.convert_single(source) print(result.render_as_markdown()) # output: "## DocLayNet: A Human -Annotated Dataset for Document -Layout Analysis [...]"</code></pre>

test/data/doc/concatenated.html

@@ -363,10 +363,10 @@ <h2>1. Introduction</h2>
 <ul>
 <li style="list-style-type: '□ ';">item 1 of sub list</li>
 <li style="list-style-type: '□ ';">
-<span class='inline-group'>Here a code snippet: <code>print("Hello world")</code> (to be displayed inline)</span>
+<span class='inline-group'>Here a code snippet:<code>print("Hello world")</code>(to be displayed inline)</span>
 </li>
 <li style="list-style-type: '□ ';">
-<span class='inline-group'>Here a formula: <math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mrow><mi>E</mi><mo>&#x0003D;</mo><mi>m</mi><msup><mi>c</mi><mn>2</mn></msup></mrow><annotation encoding="TeX">E=mc^2</annotation></math> (to be displayed inline)</span>
+<span class='inline-group'>Here a formula:<math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mrow><mi>E</mi><mo>&#x0003D;</mo><mi>m</mi><msup><mi>c</mi><mn>2</mn></msup></mrow><annotation encoding="TeX">E=mc^2</annotation></math>(to be displayed inline)</span>
 </li>
 </ul>
 </li>
@@ -387,7 +387,7 @@ <h2>1. Introduction</h2>
 
 </ul>
 </div>
-<span class='inline-group'>Some formatting chops: <strong>bold</strong> <em>italic</em> <u>underline</u> <del>strikethrough</del> <sub>subscript</sub> <sup>superscript</sup> <a href=".">hyperlink</a> &amp; <a href="https://github.com/DS4SD/docling"><del><u><em><strong>everything at the same time.</strong></em></u></del></a></span>
+<span class='inline-group'>Some formatting chops:<strong>bold</strong><em>italic</em><u>underline</u><del>strikethrough</del><sub>subscript</sub><sup>superscript</sup><a href=".">hyperlink</a>&amp;<a href="https://github.com/DS4SD/docling"><del><u><em><strong>everything at the same time.</strong></em></u></del></a></span>
 <ol>
 <li style="list-style-type: '(i) ';">Item 1 in A</li>
 <li style="list-style-type: '(ii) ';">Item 2 in A</li>

test/data/doc/constructed_doc.embedded.html.gt

@@ -167,10 +167,10 @@ item 2 of neighboring list
 <ul>
 <li style="list-style-type: '□ ';">item 1 of sub list</li>
 <li style="list-style-type: '□ ';">
-<span class='inline-group'>Here a code snippet: <code>print("Hello world")</code> (to be displayed inline)</span>
+<span class='inline-group'>Here a code snippet:<code>print("Hello world")</code>(to be displayed inline)</span>
 </li>
 <li style="list-style-type: '□ ';">
-<span class='inline-group'>Here a formula: <math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mrow><mi>E</mi><mo>&#x0003D;</mo><mi>m</mi><msup><mi>c</mi><mn>2</mn></msup></mrow><annotation encoding="TeX">E=mc^2</annotation></math> (to be displayed inline)</span>
+<span class='inline-group'>Here a formula:<math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mrow><mi>E</mi><mo>&#x0003D;</mo><mi>m</mi><msup><mi>c</mi><mn>2</mn></msup></mrow><annotation encoding="TeX">E=mc^2</annotation></math>(to be displayed inline)</span>
 </li>
 </ul>
 </li>
@@ -191,7 +191,7 @@ item 2 of neighboring list
 
 </ul>
 </div>
-<span class='inline-group'>Some formatting chops: <strong>bold</strong> <em>italic</em> <u>underline</u> <del>strikethrough</del> <sub>subscript</sub> <sup>superscript</sup> <a href=".">hyperlink</a> &amp; <a href="https://github.com/DS4SD/docling"><del><u><em><strong>everything at the same time.</strong></em></u></del></a></span>
+<span class='inline-group'>Some formatting chops:<strong>bold</strong><em>italic</em><u>underline</u><del>strikethrough</del><sub>subscript</sub><sup>superscript</sup><a href=".">hyperlink</a>&amp;<a href="https://github.com/DS4SD/docling"><del><u><em><strong>everything at the same time.</strong></em></u></del></a></span>
 <ol>
 <li style="list-style-type: '(i) ';">Item 1 in A</li>
 <li style="list-style-type: '(ii) ';">Item 2 in A</li>

test/data/doc/constructed_doc.embedded.md.gt

@@ -44,8 +44,8 @@ This is the caption of figure 2.
 - item 1 of neighboring list
 - item 2 of neighboring list
     - item 1 of sub list
-    - Here a code snippet: `print("Hello world")` (to be displayed inline)
-    - Here a formula: $E=mc^2$ (to be displayed inline)
+    - Here a code snippet:`print("Hello world")`(to be displayed inline)
+    - Here a formula:$E=mc^2$(to be displayed inline)
 
 Here a code block:
 
@@ -61,7 +61,7 @@ $$E=mc^2$$
 
 <!-- missing-form-item -->
 
-Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) &amp; [~~***everything at the same time.***~~](https://github.com/DS4SD/docling)
+Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&amp;[~~***everything at the same time.***~~](https://github.com/DS4SD/docling)
 
 - (i) Item 1 in A
 - (ii) Item 2 in A

test/data/doc/constructed_doc.placeholder.html.gt

@@ -167,10 +167,10 @@ item 2 of neighboring list
 <ul>
 <li style="list-style-type: '□ ';">item 1 of sub list</li>
 <li style="list-style-type: '□ ';">
-<span class='inline-group'>Here a code snippet: <code>print("Hello world")</code> (to be displayed inline)</span>
+<span class='inline-group'>Here a code snippet:<code>print("Hello world")</code>(to be displayed inline)</span>
 </li>
 <li style="list-style-type: '□ ';">
-<span class='inline-group'>Here a formula: <math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mrow><mi>E</mi><mo>&#x0003D;</mo><mi>m</mi><msup><mi>c</mi><mn>2</mn></msup></mrow><annotation encoding="TeX">E=mc^2</annotation></math> (to be displayed inline)</span>
+<span class='inline-group'>Here a formula:<math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mrow><mi>E</mi><mo>&#x0003D;</mo><mi>m</mi><msup><mi>c</mi><mn>2</mn></msup></mrow><annotation encoding="TeX">E=mc^2</annotation></math>(to be displayed inline)</span>
 </li>
 </ul>
 </li>
@@ -191,7 +191,7 @@ item 2 of neighboring list
 
 </ul>
 </div>
-<span class='inline-group'>Some formatting chops: <strong>bold</strong> <em>italic</em> <u>underline</u> <del>strikethrough</del> <sub>subscript</sub> <sup>superscript</sup> <a href=".">hyperlink</a> &amp; <a href="https://github.com/DS4SD/docling"><del><u><em><strong>everything at the same time.</strong></em></u></del></a></span>
+<span class='inline-group'>Some formatting chops:<strong>bold</strong><em>italic</em><u>underline</u><del>strikethrough</del><sub>subscript</sub><sup>superscript</sup><a href=".">hyperlink</a>&amp;<a href="https://github.com/DS4SD/docling"><del><u><em><strong>everything at the same time.</strong></em></u></del></a></span>
 <ol>
 <li style="list-style-type: '(i) ';">Item 1 in A</li>
 <li style="list-style-type: '(ii) ';">Item 2 in A</li>

test/data/doc/constructed_doc.placeholder.md.gt

@@ -44,8 +44,8 @@ This is the caption of figure 2.
 - item 1 of neighboring list
 - item 2 of neighboring list
     - item 1 of sub list
-    - Here a code snippet: `print("Hello world")` (to be displayed inline)
-    - Here a formula: $E=mc^2$ (to be displayed inline)
+    - Here a code snippet:`print("Hello world")`(to be displayed inline)
+    - Here a formula:$E=mc^2$(to be displayed inline)
 
 Here a code block:
 
@@ -61,7 +61,7 @@ $$E=mc^2$$
 
 <!-- missing-form-item -->
 
-Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) &amp; [~~***everything at the same time.***~~](https://github.com/DS4SD/docling)
+Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&amp;[~~***everything at the same time.***~~](https://github.com/DS4SD/docling)
 
 - (i) Item 1 in A
 - (ii) Item 2 in A

test/data/doc/constructed_doc.referenced.html.gt

@@ -167,10 +167,10 @@ item 2 of neighboring list
 <ul>
 <li style="list-style-type: '□ ';">item 1 of sub list</li>
 <li style="list-style-type: '□ ';">
-<span class='inline-group'>Here a code snippet: <code>print("Hello world")</code> (to be displayed inline)</span>
+<span class='inline-group'>Here a code snippet:<code>print("Hello world")</code>(to be displayed inline)</span>
 </li>
 <li style="list-style-type: '□ ';">
-<span class='inline-group'>Here a formula: <math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mrow><mi>E</mi><mo>&#x0003D;</mo><mi>m</mi><msup><mi>c</mi><mn>2</mn></msup></mrow><annotation encoding="TeX">E=mc^2</annotation></math> (to be displayed inline)</span>
+<span class='inline-group'>Here a formula:<math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mrow><mi>E</mi><mo>&#x0003D;</mo><mi>m</mi><msup><mi>c</mi><mn>2</mn></msup></mrow><annotation encoding="TeX">E=mc^2</annotation></math>(to be displayed inline)</span>
 </li>
 </ul>
 </li>
@@ -191,7 +191,7 @@ item 2 of neighboring list
 
 </ul>
 </div>
-<span class='inline-group'>Some formatting chops: <strong>bold</strong> <em>italic</em> <u>underline</u> <del>strikethrough</del> <sub>subscript</sub> <sup>superscript</sup> <a href=".">hyperlink</a> &amp; <a href="https://github.com/DS4SD/docling"><del><u><em><strong>everything at the same time.</strong></em></u></del></a></span>
+<span class='inline-group'>Some formatting chops:<strong>bold</strong><em>italic</em><u>underline</u><del>strikethrough</del><sub>subscript</sub><sup>superscript</sup><a href=".">hyperlink</a>&amp;<a href="https://github.com/DS4SD/docling"><del><u><em><strong>everything at the same time.</strong></em></u></del></a></span>
 <ol>
 <li style="list-style-type: '(i) ';">Item 1 in A</li>
 <li style="list-style-type: '(ii) ';">Item 2 in A</li>

test/data/doc/constructed_doc.referenced.md.gt

@@ -44,8 +44,8 @@ This is the caption of figure 2.
 - item 1 of neighboring list
 - item 2 of neighboring list
     - item 1 of sub list
-    - Here a code snippet: `print("Hello world")` (to be displayed inline)
-    - Here a formula: $E=mc^2$ (to be displayed inline)
+    - Here a code snippet:`print("Hello world")`(to be displayed inline)
+    - Here a formula:$E=mc^2$(to be displayed inline)
 
 Here a code block:
 
@@ -61,7 +61,7 @@ $$E=mc^2$$
 
 <!-- missing-form-item -->
 
-Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) &amp; [~~***everything at the same time.***~~](https://github.com/DS4SD/docling)
+Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&amp;[~~***everything at the same time.***~~](https://github.com/DS4SD/docling)
 
 - (i) Item 1 in A
 - (ii) Item 2 in A

test/data/doc/constructed_document.yaml.html

@@ -167,10 +167,10 @@ <h2>1. Introduction</h2>
 <ul>
 <li style="list-style-type: '□ ';">item 1 of sub list</li>
 <li style="list-style-type: '□ ';">
-<span class='inline-group'>Here a code snippet: <code>print("Hello world")</code> (to be displayed inline)</span>
+<span class='inline-group'>Here a code snippet:<code>print("Hello world")</code>(to be displayed inline)</span>
 </li>
 <li style="list-style-type: '□ ';">
-<span class='inline-group'>Here a formula: <math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mrow><mi>E</mi><mo>&#x0003D;</mo><mi>m</mi><msup><mi>c</mi><mn>2</mn></msup></mrow><annotation encoding="TeX">E=mc^2</annotation></math> (to be displayed inline)</span>
+<span class='inline-group'>Here a formula:<math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><mrow><mi>E</mi><mo>&#x0003D;</mo><mi>m</mi><msup><mi>c</mi><mn>2</mn></msup></mrow><annotation encoding="TeX">E=mc^2</annotation></math>(to be displayed inline)</span>
 </li>
 </ul>
 </li>
@@ -191,7 +191,7 @@ <h2>1. Introduction</h2>
 
 </ul>
 </div>
-<span class='inline-group'>Some formatting chops: <strong>bold</strong> <em>italic</em> <u>underline</u> <del>strikethrough</del> <sub>subscript</sub> <sup>superscript</sup> <a href=".">hyperlink</a> &amp; <a href="https://github.com/DS4SD/docling"><del><u><em><strong>everything at the same time.</strong></em></u></del></a></span>
+<span class='inline-group'>Some formatting chops:<strong>bold</strong><em>italic</em><u>underline</u><del>strikethrough</del><sub>subscript</sub><sup>superscript</sup><a href=".">hyperlink</a>&amp;<a href="https://github.com/DS4SD/docling"><del><u><em><strong>everything at the same time.</strong></em></u></del></a></span>
 <ol>
 <li style="list-style-type: '(i) ';">Item 1 in A</li>
 <li style="list-style-type: '(ii) ';">Item 2 in A</li>

test/data/doc/constructed_document.yaml.md

@@ -44,8 +44,8 @@ This is the caption of figure 2.
 - item 1 of neighboring list
 - item 2 of neighboring list
     - item 1 of sub list
-    - Here a code snippet: `print("Hello world")` (to be displayed inline)
-    - Here a formula: $E=mc^2$ (to be displayed inline)
+    - Here a code snippet:`print("Hello world")`(to be displayed inline)
+    - Here a formula:$E=mc^2$(to be displayed inline)
 
 Here a code block:
 
@@ -61,7 +61,7 @@ $$E=mc^2$$
 
 <!-- missing-form-item -->
 
-Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) &amp; [~~***everything at the same time.***~~](https://github.com/DS4SD/docling)
+Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&amp;[~~***everything at the same time.***~~](https://github.com/DS4SD/docling)
 
 - (i) Item 1 in A
 - (ii) Item 2 in A

test/data/doc/constructed_legacy_annot_mark_false.gt.md

@@ -46,8 +46,8 @@ This is the caption of figure 2.
 - item 1 of neighboring list
 - item 2 of neighboring list
     - item 1 of sub list
-    - Here a code snippet: `print("Hello world")` (to be displayed inline)
-    - Here a formula: $E=mc^2$ (to be displayed inline)
+    - Here a code snippet:`print("Hello world")`(to be displayed inline)
+    - Here a formula:$E=mc^2$(to be displayed inline)
 
 Here a code block:
 
@@ -63,7 +63,7 @@ $$E=mc^2$$
 
 <!-- missing-form-item -->
 
-Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) &amp; [~~***everything at the same time.***~~](https://github.com/DS4SD/docling)
+Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&amp;[~~***everything at the same time.***~~](https://github.com/DS4SD/docling)
 
 - (i) Item 1 in A
 - (ii) Item 2 in A

test/data/doc/constructed_legacy_annot_mark_true.gt.md

@@ -46,8 +46,8 @@ This is the caption of figure 2.
 - item 1 of neighboring list
 - item 2 of neighboring list
     - item 1 of sub list
-    - Here a code snippet: `print("Hello world")` (to be displayed inline)
-    - Here a formula: $E=mc^2$ (to be displayed inline)
+    - Here a code snippet:`print("Hello world")`(to be displayed inline)
+    - Here a formula:$E=mc^2$(to be displayed inline)
 
 Here a code block:
 
@@ -63,7 +63,7 @@ $$E=mc^2$$
 
 <!-- missing-form-item -->
 
-Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) &amp; [~~***everything at the same time.***~~](https://github.com/DS4SD/docling)
+Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&amp;[~~***everything at the same time.***~~](https://github.com/DS4SD/docling)
 
 - (i) Item 1 in A
 - (ii) Item 2 in A