From 4edf3feb491245f4a837d9d1058ff18dece0bb76 Mon Sep 17 00:00:00 2001 From: Venkata Ramana Menda Date: Thu, 11 Dec 2025 16:33:10 +0530 Subject: [PATCH] Remove extra space before and after group items Signed-off-by: Venkata Ramana Menda --- docling_core/transforms/serializer/html.py | 2 +- .../transforms/serializer/markdown.py | 2 +- .../2408.09869v3_enriched_p2_p3_p5.gt.html | 2 +- .../doc/2408.09869v3_enriched_split.gt.html | 2 +- .../2408.09869v3_enriched_split_p2.gt.html | 2 +- test/data/doc/concatenated.html | 6 +- .../data/doc/constructed_doc.embedded.html.gt | 6 +- test/data/doc/constructed_doc.embedded.md.gt | 6 +- .../doc/constructed_doc.placeholder.html.gt | 6 +- .../doc/constructed_doc.placeholder.md.gt | 6 +- .../doc/constructed_doc.referenced.html.gt | 6 +- .../data/doc/constructed_doc.referenced.md.gt | 6 +- test/data/doc/constructed_document.yaml.html | 6 +- test/data/doc/constructed_document.yaml.md | 6 +- .../constructed_legacy_annot_mark_false.gt.md | 6 +- .../constructed_legacy_annot_mark_true.gt.md | 6 +- .../constructed_mode_always_valid_false.gt.md | 6 +- .../constructed_mode_always_valid_true.gt.md | 6 +- .../constructed_mode_auto_valid_false.gt.md | 6 +- .../constructed_mode_auto_valid_true.gt.md | 6 +- .../constructed_mode_never_valid_false.gt.md | 6 +- .../constructed_mode_never_valid_true.gt.md | 6 +- test/data/doc/constructed_orig_false.gt.html | 6 +- test/data/doc/constructed_orig_true.gt.html | 6 +- test/data/doc/inline_and_formatting.gt.html | 20 +-- test/data/doc/inline_and_formatting.gt.md | 20 +-- test/data/doc/polymers.gt.html | 30 ++-- test/data/doc/polymers.gt.md | 30 ++-- test/test_serialization.py | 149 ++++++++++++++++++ 29 files changed, 261 insertions(+), 112 deletions(-) diff --git a/docling_core/transforms/serializer/html.py b/docling_core/transforms/serializer/html.py index 4a92e303b..490d9041b 100644 --- a/docling_core/transforms/serializer/html.py +++ b/docling_core/transforms/serializer/html.py @@ -871,7 +871,7 @@ 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: diff --git a/docling_core/transforms/serializer/markdown.py b/docling_core/transforms/serializer/markdown.py index 2783d4ed4..aebce806c 100644 --- a/docling_core/transforms/serializer/markdown.py +++ b/docling_core/transforms/serializer/markdown.py @@ -803,7 +803,7 @@ 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) diff --git a/test/data/doc/2408.09869v3_enriched_p2_p3_p5.gt.html b/test/data/doc/2408.09869v3_enriched_p2_p3_p5.gt.html index f58ce03e8..e19f61a4a 100644 --- a/test/data/doc/2408.09869v3_enriched_p2_p3_p5.gt.html +++ b/test/data/doc/2408.09869v3_enriched_p2_p3_p5.gt.html @@ -105,7 +105,7 @@
  • Can leverage different accelerators (GPU, MPS, etc).

    • 2 Getting Started

    -To use Docling, you can simply install the docling package from PyPI. Documentation and examples are available in our GitHub repository at github.com/DS4SD/docling . 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. +To use Docling, you can simply install the docling package from PyPI. Documentation and examples are available in our GitHub repository at github.com/DS4SD/docling. 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.

    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.

    from docling.document_converter import DocumentConverter Large
    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 [...]"
    diff --git a/test/data/doc/2408.09869v3_enriched_split.gt.html b/test/data/doc/2408.09869v3_enriched_split.gt.html index 9ea787dd5..e305e3c3e 100644 --- a/test/data/doc/2408.09869v3_enriched_split.gt.html +++ b/test/data/doc/2408.09869v3_enriched_split.gt.html @@ -126,7 +126,7 @@

    1 Introduction

  • Can leverage different accelerators (GPU, MPS, etc).

    • 2 Getting Started

    -To use Docling, you can simply install the docling package from PyPI. Documentation and examples are available in our GitHub repository at github.com/DS4SD/docling . 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. +To use Docling, you can simply install the docling package from PyPI. Documentation and examples are available in our GitHub repository at github.com/DS4SD/docling. 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.

    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.

    from docling.document_converter import DocumentConverter Large
    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 [...]"
    diff --git a/test/data/doc/2408.09869v3_enriched_split_p2.gt.html b/test/data/doc/2408.09869v3_enriched_split_p2.gt.html index 5c43aecbc..e4b5f9441 100644 --- a/test/data/doc/2408.09869v3_enriched_split_p2.gt.html +++ b/test/data/doc/2408.09869v3_enriched_split_p2.gt.html @@ -106,7 +106,7 @@
  • Can leverage different accelerators (GPU, MPS, etc).

    • 2 Getting Started

    -To use Docling, you can simply install the docling package from PyPI. Documentation and examples are available in our GitHub repository at github.com/DS4SD/docling . 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. +To use Docling, you can simply install the docling package from PyPI. Documentation and examples are available in our GitHub repository at github.com/DS4SD/docling. 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.

    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.

    from docling.document_converter import DocumentConverter Large
    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 [...]"
    diff --git a/test/data/doc/concatenated.html b/test/data/doc/concatenated.html index b2894959a..c8127be7f 100644 --- a/test/data/doc/concatenated.html +++ b/test/data/doc/concatenated.html @@ -363,10 +363,10 @@

    1. Introduction

    @@ -387,7 +387,7 @@

    1. Introduction

    -Some formatting chops: bold italic underline strikethrough subscript superscript hyperlink & everything at the same time. +Some formatting chops:bolditalicunderlinestrikethroughsubscriptsuperscripthyperlink&everything at the same time.
    1. Item 1 in A
    2. Item 2 in A
      3. diff --git a/test/data/doc/constructed_doc.embedded.html.gt b/test/data/doc/constructed_doc.embedded.html.gt index 646da35ff..d39ea34bf 100644 --- a/test/data/doc/constructed_doc.embedded.html.gt +++ b/test/data/doc/constructed_doc.embedded.html.gt @@ -167,10 +167,10 @@ item 2 of neighboring list
      • item 1 of sub list
      • -Here a code snippet: print("Hello world") (to be displayed inline) +Here a code snippet:print("Hello world")(to be displayed inline)
      • -Here a formula: E=mc2E=mc^2 (to be displayed inline) +Here a formula:E=mc2E=mc^2(to be displayed inline)
      @@ -191,7 +191,7 @@ item 2 of neighboring list -Some formatting chops: bold italic underline strikethrough subscript superscript hyperlink & everything at the same time. +Some formatting chops:bolditalicunderlinestrikethroughsubscriptsuperscripthyperlink&everything at the same time.
      1. Item 1 in A
      2. Item 2 in A
        3. diff --git a/test/data/doc/constructed_doc.embedded.md.gt b/test/data/doc/constructed_doc.embedded.md.gt index 5a4f94404..6c8a34e27 100644 --- a/test/data/doc/constructed_doc.embedded.md.gt +++ b/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 @@ $$E=mc^2$$ -Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) & [~~***everything at the same time.***~~](https://github.com/DS4SD/docling) +Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&[~~***everything at the same time.***~~](https://github.com/DS4SD/docling) - (i) Item 1 in A - (ii) Item 2 in A diff --git a/test/data/doc/constructed_doc.placeholder.html.gt b/test/data/doc/constructed_doc.placeholder.html.gt index 1548ad1f2..f7bf7b4ce 100644 --- a/test/data/doc/constructed_doc.placeholder.html.gt +++ b/test/data/doc/constructed_doc.placeholder.html.gt @@ -167,10 +167,10 @@ item 2 of neighboring list
        • item 1 of sub list
        • -Here a code snippet: print("Hello world") (to be displayed inline) +Here a code snippet:print("Hello world")(to be displayed inline)
        • -Here a formula: E=mc2E=mc^2 (to be displayed inline) +Here a formula:E=mc2E=mc^2(to be displayed inline)
        @@ -191,7 +191,7 @@ item 2 of neighboring list -Some formatting chops: bold italic underline strikethrough subscript superscript hyperlink & everything at the same time. +Some formatting chops:bolditalicunderlinestrikethroughsubscriptsuperscripthyperlink&everything at the same time.
        1. Item 1 in A
        2. Item 2 in A
          3. diff --git a/test/data/doc/constructed_doc.placeholder.md.gt b/test/data/doc/constructed_doc.placeholder.md.gt index 9b8993271..bc4973a80 100644 --- a/test/data/doc/constructed_doc.placeholder.md.gt +++ b/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 @@ $$E=mc^2$$ -Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) & [~~***everything at the same time.***~~](https://github.com/DS4SD/docling) +Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&[~~***everything at the same time.***~~](https://github.com/DS4SD/docling) - (i) Item 1 in A - (ii) Item 2 in A diff --git a/test/data/doc/constructed_doc.referenced.html.gt b/test/data/doc/constructed_doc.referenced.html.gt index cdc58aa19..c642a973a 100644 --- a/test/data/doc/constructed_doc.referenced.html.gt +++ b/test/data/doc/constructed_doc.referenced.html.gt @@ -167,10 +167,10 @@ item 2 of neighboring list
          • item 1 of sub list
          • -Here a code snippet: print("Hello world") (to be displayed inline) +Here a code snippet:print("Hello world")(to be displayed inline)
          • -Here a formula: E=mc2E=mc^2 (to be displayed inline) +Here a formula:E=mc2E=mc^2(to be displayed inline)
          @@ -191,7 +191,7 @@ item 2 of neighboring list -Some formatting chops: bold italic underline strikethrough subscript superscript hyperlink & everything at the same time. +Some formatting chops:bolditalicunderlinestrikethroughsubscriptsuperscripthyperlink&everything at the same time.
          1. Item 1 in A
          2. Item 2 in A
            3. diff --git a/test/data/doc/constructed_doc.referenced.md.gt b/test/data/doc/constructed_doc.referenced.md.gt index 6a30a5822..d9806228e 100644 --- a/test/data/doc/constructed_doc.referenced.md.gt +++ b/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 @@ $$E=mc^2$$ -Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) & [~~***everything at the same time.***~~](https://github.com/DS4SD/docling) +Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&[~~***everything at the same time.***~~](https://github.com/DS4SD/docling) - (i) Item 1 in A - (ii) Item 2 in A diff --git a/test/data/doc/constructed_document.yaml.html b/test/data/doc/constructed_document.yaml.html index f2741aa26..31c0a9737 100644 --- a/test/data/doc/constructed_document.yaml.html +++ b/test/data/doc/constructed_document.yaml.html @@ -167,10 +167,10 @@

            1. Introduction

            • item 1 of sub list
            • -Here a code snippet: print("Hello world") (to be displayed inline) +Here a code snippet:print("Hello world")(to be displayed inline)
            • -Here a formula: E=mc2E=mc^2 (to be displayed inline) +Here a formula:E=mc2E=mc^2(to be displayed inline)
            @@ -191,7 +191,7 @@

            1. Introduction

            -Some formatting chops: bold italic underline strikethrough subscript superscript hyperlink & everything at the same time. +Some formatting chops:bolditalicunderlinestrikethroughsubscriptsuperscripthyperlink&everything at the same time.
            1. Item 1 in A
            2. Item 2 in A
              3. diff --git a/test/data/doc/constructed_document.yaml.md b/test/data/doc/constructed_document.yaml.md index ea562f91e..5d86bd2b9 100644 --- a/test/data/doc/constructed_document.yaml.md +++ b/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 @@ $$E=mc^2$$ -Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) & [~~***everything at the same time.***~~](https://github.com/DS4SD/docling) +Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&[~~***everything at the same time.***~~](https://github.com/DS4SD/docling) - (i) Item 1 in A - (ii) Item 2 in A diff --git a/test/data/doc/constructed_legacy_annot_mark_false.gt.md b/test/data/doc/constructed_legacy_annot_mark_false.gt.md index f0f3c4698..b8850b8a5 100644 --- a/test/data/doc/constructed_legacy_annot_mark_false.gt.md +++ b/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 @@ $$E=mc^2$$ -Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) & [~~***everything at the same time.***~~](https://github.com/DS4SD/docling) +Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&[~~***everything at the same time.***~~](https://github.com/DS4SD/docling) - (i) Item 1 in A - (ii) Item 2 in A diff --git a/test/data/doc/constructed_legacy_annot_mark_true.gt.md b/test/data/doc/constructed_legacy_annot_mark_true.gt.md index 0382ae172..f0a91e37d 100644 --- a/test/data/doc/constructed_legacy_annot_mark_true.gt.md +++ b/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 @@ $$E=mc^2$$ -Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) & [~~***everything at the same time.***~~](https://github.com/DS4SD/docling) +Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&[~~***everything at the same time.***~~](https://github.com/DS4SD/docling) - (i) Item 1 in A - (ii) Item 2 in A diff --git a/test/data/doc/constructed_mode_always_valid_false.gt.md b/test/data/doc/constructed_mode_always_valid_false.gt.md index 3121af342..cba2527df 100644 --- a/test/data/doc/constructed_mode_always_valid_false.gt.md +++ b/test/data/doc/constructed_mode_always_valid_false.gt.md @@ -44,8 +44,8 @@ item 2 of list after empty list ■ 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 @@ $$E=mc^2$$ -Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) & [~~***everything at the same time.***~~](https://github.com/DS4SD/docling) +Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&[~~***everything at the same time.***~~](https://github.com/DS4SD/docling) (i) Item 1 in A (ii) Item 2 in A diff --git a/test/data/doc/constructed_mode_always_valid_true.gt.md b/test/data/doc/constructed_mode_always_valid_true.gt.md index 630d5458d..b0fab96dd 100644 --- a/test/data/doc/constructed_mode_always_valid_true.gt.md +++ b/test/data/doc/constructed_mode_always_valid_true.gt.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 @@ $$E=mc^2$$ -Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) & [~~***everything at the same time.***~~](https://github.com/DS4SD/docling) +Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&[~~***everything at the same time.***~~](https://github.com/DS4SD/docling) 1. (i) Item 1 in A 2. (ii) Item 2 in A diff --git a/test/data/doc/constructed_mode_auto_valid_false.gt.md b/test/data/doc/constructed_mode_auto_valid_false.gt.md index 585e10b86..2e7d5278f 100644 --- a/test/data/doc/constructed_mode_auto_valid_false.gt.md +++ b/test/data/doc/constructed_mode_auto_valid_false.gt.md @@ -44,8 +44,8 @@ item 2 of list after empty list 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 @@ $$E=mc^2$$ -Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) & [~~***everything at the same time.***~~](https://github.com/DS4SD/docling) +Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&[~~***everything at the same time.***~~](https://github.com/DS4SD/docling) (i) Item 1 in A (ii) Item 2 in A diff --git a/test/data/doc/constructed_mode_auto_valid_true.gt.md b/test/data/doc/constructed_mode_auto_valid_true.gt.md index ea562f91e..5d86bd2b9 100644 --- a/test/data/doc/constructed_mode_auto_valid_true.gt.md +++ b/test/data/doc/constructed_mode_auto_valid_true.gt.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 @@ $$E=mc^2$$ -Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) & [~~***everything at the same time.***~~](https://github.com/DS4SD/docling) +Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&[~~***everything at the same time.***~~](https://github.com/DS4SD/docling) - (i) Item 1 in A - (ii) Item 2 in A diff --git a/test/data/doc/constructed_mode_never_valid_false.gt.md b/test/data/doc/constructed_mode_never_valid_false.gt.md index 1a065857b..1e69125bd 100644 --- a/test/data/doc/constructed_mode_never_valid_false.gt.md +++ b/test/data/doc/constructed_mode_never_valid_false.gt.md @@ -44,8 +44,8 @@ item 2 of list after empty list 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 @@ $$E=mc^2$$ -Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) & [~~***everything at the same time.***~~](https://github.com/DS4SD/docling) +Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&[~~***everything at the same time.***~~](https://github.com/DS4SD/docling) Item 1 in A Item 2 in A diff --git a/test/data/doc/constructed_mode_never_valid_true.gt.md b/test/data/doc/constructed_mode_never_valid_true.gt.md index c42f4e83b..c6d5ad49a 100644 --- a/test/data/doc/constructed_mode_never_valid_true.gt.md +++ b/test/data/doc/constructed_mode_never_valid_true.gt.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 @@ $$E=mc^2$$ -Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) & [~~***everything at the same time.***~~](https://github.com/DS4SD/docling) +Some formatting chops:**bold***italic*underline~~strikethrough~~subscriptsuperscript[hyperlink](.)&[~~***everything at the same time.***~~](https://github.com/DS4SD/docling) 1. Item 1 in A 2. Item 2 in A diff --git a/test/data/doc/constructed_orig_false.gt.html b/test/data/doc/constructed_orig_false.gt.html index e3f6f7a5b..1955c13ff 100644 --- a/test/data/doc/constructed_orig_false.gt.html +++ b/test/data/doc/constructed_orig_false.gt.html @@ -167,10 +167,10 @@

              1. Introduction

              • item 1 of sub list
              • -Here a code snippet: print("Hello world") (to be displayed inline) +Here a code snippet:print("Hello world")(to be displayed inline)
              • -Here a formula: E=mc2E=mc^2 (to be displayed inline) +Here a formula:E=mc2E=mc^2(to be displayed inline)
              @@ -191,7 +191,7 @@

              1. Introduction

              -Some formatting chops: bold italic underline strikethrough subscript superscript hyperlink & everything at the same time. +Some formatting chops:bolditalicunderlinestrikethroughsubscriptsuperscripthyperlink&everything at the same time.
              1. Item 1 in A
              2. Item 2 in A
                3. diff --git a/test/data/doc/constructed_orig_true.gt.html b/test/data/doc/constructed_orig_true.gt.html index f2741aa26..31c0a9737 100644 --- a/test/data/doc/constructed_orig_true.gt.html +++ b/test/data/doc/constructed_orig_true.gt.html @@ -167,10 +167,10 @@

                1. Introduction

                • item 1 of sub list
                • -Here a code snippet: print("Hello world") (to be displayed inline) +Here a code snippet:print("Hello world")(to be displayed inline)
                • -Here a formula: E=mc2E=mc^2 (to be displayed inline) +Here a formula:E=mc2E=mc^2(to be displayed inline)
                @@ -191,7 +191,7 @@

                1. Introduction

                -Some formatting chops: bold italic underline strikethrough subscript superscript hyperlink & everything at the same time. +Some formatting chops:bolditalicunderlinestrikethroughsubscriptsuperscripthyperlink&everything at the same time.
                1. Item 1 in A
                2. Item 2 in A
                  3. diff --git a/test/data/doc/inline_and_formatting.gt.html b/test/data/doc/inline_and_formatting.gt.html index 2165b1be7..c801af756 100644 --- a/test/data/doc/inline_and_formatting.gt.html +++ b/test/data/doc/inline_and_formatting.gt.html @@ -126,31 +126,31 @@

                  Contribution guideline example

                  This is simple.

                  -Foo emphasis strong emphasis both . -Create your feature branch: git checkout -b feature/AmazingFeature . +Fooemphasisstrong emphasisboth. +Create your feature branch:git checkout -b feature/AmazingFeature.
                  1. -Pull the repository . +Pull therepository.
                  2. -Create your feature branch ( git checkout -b feature/AmazingFeature ) +Create your feature branch (git checkout -b feature/AmazingFeature)
                  3. -Commit your changes ( git commit -m 'Add some AmazingFeature' ) +Commit your changes (git commit -m 'Add some AmazingFeature')
                  4. -Push to the branch ( git push origin feature/AmazingFeature ) +Push to the branch (git push origin feature/AmazingFeature)
                  5. Open a Pull Request
                  6. Whole list item has same formatting
                  7. -List item has mixed or partial formatting +List item hasmixed or partialformatting

                  Whole heading is italic

                   -Some formatted_code -

                  Partially formatted heading to_escape not_to_escape E=mc2E=mc^2 & ampersand

                  -A hyperlink on code in a line +Someformatted_code +

                  Partially formattedheading to_escapenot_to_escapeE=mc2E=mc^2& ampersand

                  +A hyperlink oncode in a line 
                  A hyperlink on code as paragraph

                  The end.

                  diff --git a/test/data/doc/inline_and_formatting.gt.md b/test/data/doc/inline_and_formatting.gt.md index 0455b064f..cac10221c 100644 --- a/test/data/doc/inline_and_formatting.gt.md +++ b/test/data/doc/inline_and_formatting.gt.md @@ -2,25 +2,25 @@ This is simple. -Foo *emphasis* **strong emphasis** ***both*** . +Foo*emphasis***strong emphasis*****both***. -Create your feature branch: `git checkout -b feature/AmazingFeature` . +Create your feature branch:`git checkout -b feature/AmazingFeature`. -1. Pull the [**repository**](https://github.com/docling-project/docling) . -2. Create your feature branch ( `git checkout -b feature/AmazingFeature` ) -3. Commit your changes ( `git commit -m 'Add some AmazingFeature'` ) -4. Push to the branch ( `git push origin feature/AmazingFeature` ) +1. Pull the[**repository**](https://github.com/docling-project/docling). +2. Create your feature branch (`git checkout -b feature/AmazingFeature`) +3. Commit your changes (`git commit -m 'Add some AmazingFeature'`) +4. Push to the branch (`git push origin feature/AmazingFeature`) 5. Open a Pull Request 6. **Whole list item has same formatting** -7. List item has *mixed or partial* formatting +7. List item has*mixed or partial*formatting # *Whole heading is italic* -Some *`formatted_code`* +Some*`formatted_code`* -## *Partially formatted* heading to\_escape `not_to_escape` [$E=mc^2$](https://en.wikipedia.org/wiki/Albert_Einstein) & ampersand +## *Partially formatted*heading to\_escape`not_to_escape`[$E=mc^2$](https://en.wikipedia.org/wiki/Albert_Einstein)& ampersand -A hyperlink on [`code in a line`](#link) +A hyperlink on[`code in a line`](#link) [`A hyperlink on code as paragraph`](#test) diff --git a/test/data/doc/polymers.gt.html b/test/data/doc/polymers.gt.html index d70699e9d..cf274a983 100644 --- a/test/data/doc/polymers.gt.html +++ b/test/data/doc/polymers.gt.html @@ -188,10 +188,10 @@

                  Safety and Regulatory Considerations

                  Extraction in food simulants
                  • -What it is : Samples of the packaging material are immersed in a liquid that mimics the chemical properties of a specific food type (e.g., aqueous, acidic, fatty). +What it is: Samples of the packaging material are immersed in a liquid that mimics the chemical properties of a specific food type (e.g., aqueous, acidic, fatty).
                  • -Typical simulants : +Typical simulants:
                    • 3% acetic acid (for acidic foods)
                    • 50% ethanol (for alcohol‑based foods)
                      • @@ -200,7 +200,7 @@

                      Safety and Regulatory Considerations

                  • -Procedure : +Procedure:
                    • Prepare a defined volume of simulant in a sealed vessel.
                    • Immerse the material for a set time at a controlled temperature (often 50 °C–70 °C).
                      • @@ -208,13 +208,13 @@

                      Safety and Regulatory Considerations

                  • -Analysis : GC‑MS, LC‑MS, or HPLC depending on the analyte class. +Analysis: GC‑MS, LC‑MS, or HPLC depending on the analyte class.
                  • -Advantages : Direct assessment of potential migration into a realistic medium; scalable for routine testing. +Advantages: Direct assessment of potential migration into a realistic medium; scalable for routine testing.
                  • -Limitations : Does not account for headspace gas migration; may underestimate migration of highly volatile substances. +Limitations: Does not account for headspace gas migration; may underestimate migration of highly volatile substances.
                  @@ -222,10 +222,10 @@

                  Safety and Regulatory Considerations

                  Headspace analysis
                  • -What it is : Measurement of volatile substances that migrate from the material into the surrounding gas phase. +What it is: Measurement of volatile substances that migrate from the material into the surrounding gas phase.
                  • -Procedure : +Procedure:
                    • Seal the material in a headspace vial or chamber.
                    • Equilibrate at a defined temperature (commonly 25 °C–60 °C).
                      • @@ -234,13 +234,13 @@

                      Safety and Regulatory Considerations

                  • -Applications : Assessment of aromas, flavor compounds, or volatile contaminants. +Applications: Assessment of aromas, flavor compounds, or volatile contaminants.
                  • -Advantages : Sensitive to low‑concentration volatiles; minimal sample preparation. +Advantages: Sensitive to low‑concentration volatiles; minimal sample preparation.
                  • -Limitations : Does not capture non‑volatile migration; results depend on equilibrium time and temperature. +Limitations: Does not capture non‑volatile migration; results depend on equilibrium time and temperature.
                  @@ -248,10 +248,10 @@

                  Safety and Regulatory Considerations

                  Direct contact tests
                  • -What it is : The packaging material is placed in direct contact with the food or food simulant, often using a defined food‑packaging configuration. +What it is: The packaging material is placed in direct contact with the food or food simulant, often using a defined food‑packaging configuration.
                  • -Procedure : +Procedure:
                    • Assemble the material and food (or simulant) in a mold or container that simulates real usage (e.g., sealed pouch, jar).
                    • Incubate for the intended storage time at the relevant temperature.
                      • @@ -260,10 +260,10 @@

                      Safety and Regulatory Considerations

                  • -Advantages : Mimics real consumer exposure; captures both liquid and vapor migration pathways. +Advantages: Mimics real consumer exposure; captures both liquid and vapor migration pathways.
                  • -Limitations : More labor‑intensive; requires careful control of contact area, thickness, and sealing integrity. +Limitations: More labor‑intensive; requires careful control of contact area, thickness, and sealing integrity.
                  diff --git a/test/data/doc/polymers.gt.md b/test/data/doc/polymers.gt.md index e44bcb8d1..3bc75356c 100644 --- a/test/data/doc/polymers.gt.md +++ b/test/data/doc/polymers.gt.md @@ -50,38 +50,38 @@ **Common migration testing methods** - **Extraction in food simulants** - - *What it is* : Samples of the packaging material are immersed in a liquid that mimics the chemical properties of a specific food type (e.g., aqueous, acidic, fatty). - - *Typical simulants* : + - *What it is*: Samples of the packaging material are immersed in a liquid that mimics the chemical properties of a specific food type (e.g., aqueous, acidic, fatty). + - *Typical simulants*: - 3% acetic acid (for acidic foods) - 50% ethanol (for alcohol‑based foods) - 95% ethanol (for high‑fat foods) - Distilled water (for aqueous foods) - - *Procedure* : + - *Procedure*: - Prepare a defined volume of simulant in a sealed vessel. - Immerse the material for a set time at a controlled temperature (often 50 °C–70 °C). - Remove, filter, and concentrate the extract for analysis. - - *Analysis* : GC‑MS, LC‑MS, or HPLC depending on the analyte class. - - *Advantages* : Direct assessment of potential migration into a realistic medium; scalable for routine testing. - - *Limitations* : Does not account for headspace gas migration; may underestimate migration of highly volatile substances. + - *Analysis*: GC‑MS, LC‑MS, or HPLC depending on the analyte class. + - *Advantages*: Direct assessment of potential migration into a realistic medium; scalable for routine testing. + - *Limitations*: Does not account for headspace gas migration; may underestimate migration of highly volatile substances. - **Headspace analysis** - - *What it is* : Measurement of volatile substances that migrate from the material into the surrounding gas phase. - - *Procedure* : + - *What it is*: Measurement of volatile substances that migrate from the material into the surrounding gas phase. + - *Procedure*: - Seal the material in a headspace vial or chamber. - Equilibrate at a defined temperature (commonly 25 °C–60 °C). - Sample the gas phase with a gas sampling needle or syringe. - Analyze via GC‑FID, GC‑MS, or PTR‑MS. - - *Applications* : Assessment of aromas, flavor compounds, or volatile contaminants. - - *Advantages* : Sensitive to low‑concentration volatiles; minimal sample preparation. - - *Limitations* : Does not capture non‑volatile migration; results depend on equilibrium time and temperature. + - *Applications*: Assessment of aromas, flavor compounds, or volatile contaminants. + - *Advantages*: Sensitive to low‑concentration volatiles; minimal sample preparation. + - *Limitations*: Does not capture non‑volatile migration; results depend on equilibrium time and temperature. - **Direct contact tests** - - *What it is* : The packaging material is placed in direct contact with the food or food simulant, often using a defined food‑packaging configuration. - - *Procedure* : + - *What it is*: The packaging material is placed in direct contact with the food or food simulant, often using a defined food‑packaging configuration. + - *Procedure*: - Assemble the material and food (or simulant) in a mold or container that simulates real usage (e.g., sealed pouch, jar). - Incubate for the intended storage time at the relevant temperature. - Extract or sample the food directly (e.g., through the material or by taking a portion of the food). - Analyze for migrated substances. - - *Advantages* : Mimics real consumer exposure; captures both liquid and vapor migration pathways. - - *Limitations* : More labor‑intensive; requires careful control of contact area, thickness, and sealing integrity. + - *Advantages*: Mimics real consumer exposure; captures both liquid and vapor migration pathways. + - *Limitations*: More labor‑intensive; requires careful control of contact area, thickness, and sealing integrity. These three approaches—extraction in food simulants, headspace analysis, and direct contact tests—complement each other to provide a comprehensive assessment of potential migration from packaging into food. diff --git a/test/test_serialization.py b/test/test_serialization.py index 78fb83d11..ae9ff5ea7 100644 --- a/test/test_serialization.py +++ b/test/test_serialization.py @@ -28,6 +28,7 @@ DoclingDocument, RefItem, RichTableCell, + Formatting, TableCell, TableData, TextItem, @@ -964,3 +965,151 @@ def test_webvtt_params(): actual_default = ser_default.serialize().text assert len(actual) <= len(actual_default) or actual != actual_default + verify(exp_file=src.with_suffix(".gt.idt.xml"), actual=actual) + + +# =============================== +# Tests for inline group join behavior without spaces +# =============================== + + +def test_md_inline_group_no_spaces(): + """Test that inline groups join text parts without spaces for continuous text.""" + doc = DoclingDocument(name="test") + + # Create an inline group with multiple text items that should be joined without spaces + # Simulating the case where "Docling" is split into "D" (bold) and "ocling" (normal) + group = doc.add_inline_group() + doc.add_text( + label="text", + parent=group, + text="D", + formatting=Formatting(bold=True, italic=False, underline=False, strikethrough=False, script="baseline") + ) + doc.add_text( + label="text", + parent=group, + text="ocling", + formatting=Formatting(bold=False, italic=False, underline=False, strikethrough=False, script="baseline") + ) + + # This should serialize as "**D**ocling" without space + ser = MarkdownDocSerializer(doc=doc) + actual = ser.serialize().text.strip() + + expected = "**D**ocling" + assert actual == expected + + +def test_html_inline_group_no_spaces(): + """Test that inline groups join text parts without spaces for continuous text.""" + doc = DoclingDocument(name="test") + + # Create an inline group with multiple text items that should be joined without spaces + group = doc.add_inline_group() + doc.add_text( + label="text", + parent=group, + text="Project", + formatting=Formatting(bold=True, italic=False, underline=False, strikethrough=False, script="baseline") + ) + doc.add_text( + label="text", + parent=group, + text="ing", + formatting=Formatting(bold=False, italic=False, underline=False, strikethrough=False, script="baseline") + ) + + # This should serialize as Projecting without space + ser = HTMLDocSerializer(doc=doc, params=HTMLParams(html_head="", prettify=False)) + actual = ser.serialize().text + + # Extract the body content between and + start = actual.find("") + 6 + end = actual.find("") + body_content = actual[start:end].strip() + + # Check that the span contains the expected content + assert 'Projecting' in body_content + + + + +def test_md_inline_group_mixed_formatting_mid_word(): + """Test inline group with different formatting mid-word.""" + doc = DoclingDocument(name="test") + + # Simulate "Parsing" with "Pars" normal and "ing" italic + group = doc.add_inline_group() + doc.add_text(label="text", parent=group, text="Pars") + doc.add_text( + label="text", + parent=group, + text="ing", + formatting=Formatting(bold=False, italic=True, underline=False, strikethrough=False, script="baseline") + ) + + ser = MarkdownDocSerializer(doc=doc) + actual = ser.serialize().text.strip() + + expected = "Pars*ing*" + assert actual == expected + + +def test_html_inline_group_mixed_formatting_mid_word(): + """Test inline group with different formatting mid-word.""" + doc = DoclingDocument(name="test") + + # Simulate "Parsing" with "Pars" normal and "ing" italic + group = doc.add_inline_group() + doc.add_text(label="text", parent=group, text="Pars") + doc.add_text( + label="text", + parent=group, + text="ing", + formatting=Formatting(bold=False, italic=True, underline=False, strikethrough=False, script="baseline") + ) + + ser = HTMLDocSerializer(doc=doc, params=HTMLParams(html_head="", prettify=False)) + actual = ser.serialize().text + + # Extract the body content between and + start = actual.find("") + 6 + end = actual.find("") + body_content = actual[start:end].strip() + + # Check that both parts are present without spaces between + assert 'Parsing' in body_content.replace('\n', ' ') + + +def test_md_inline_group_single_part(): + """Test inline group with single text part (no joining needed).""" + doc = DoclingDocument(name="test") + + group = doc.add_inline_group() + doc.add_text(label="text", parent=group, text="Single") + + ser = MarkdownDocSerializer(doc=doc) + actual = ser.serialize().text.strip() + + expected = "Single" + assert actual == expected + + +def test_html_inline_group_single_part(): + """Test inline group with single text part (no joining needed).""" + doc = DoclingDocument(name="test") + + group = doc.add_inline_group() + doc.add_text(label="text", parent=group, text="Single") + + ser = HTMLDocSerializer(doc=doc, params=HTMLParams(html_head="", prettify=False)) + actual = ser.serialize().text + + # Extract the body content between and + start = actual.find("") + 6 + end = actual.find("") + body_content = actual[start:end].strip() + + # Check that the single part content is present + assert 'Single' in body_content