diff --git a/docling_core/transforms/serializer/base.py b/docling_core/transforms/serializer/base.py index 67d0a727..8d29aa0b 100644 --- a/docling_core/transforms/serializer/base.py +++ b/docling_core/transforms/serializer/base.py @@ -317,7 +317,7 @@ def _humanize_text(self, text: str, title: bool = False) -> str: return tmp.title() if title else tmp.capitalize() -@deprecated("Use BaseMetaSerializer() instead.") +# deprecated: use BaseMetaSerializer instead class BaseAnnotationSerializer(ABC): """Base class for annotation serializers.""" diff --git a/docling_core/transforms/serializer/common.py b/docling_core/transforms/serializer/common.py index 4720ada0..33f0a10e 100644 --- a/docling_core/transforms/serializer/common.py +++ b/docling_core/transforms/serializer/common.py @@ -7,6 +7,7 @@ import logging import re import sys +import warnings from abc import abstractmethod from functools import cached_property from pathlib import Path @@ -206,7 +207,9 @@ class CommonParams(BaseModel): include_hyperlinks: bool = True caption_delim: str = " " use_legacy_annotations: bool = Field( - default=False, description="Use legacy annotation serialization." + default=False, + description="Use legacy annotation serialization.", + deprecated="Legacy annotations considered only when meta not present.", ) allowed_meta_names: Optional[set[str]] = Field( default=None, @@ -336,7 +339,7 @@ def serialize( my_item = item or self.doc.body if my_item == self.doc.body: - if my_item.meta and not my_params.use_legacy_annotations: + if my_item.meta: meta_part = self.serialize_meta(item=my_item, **my_kwargs) if meta_part.text: parts.append(meta_part) @@ -355,7 +358,7 @@ def serialize( my_visited.add(my_item.self_ref) - if my_item.meta and not my_params.use_legacy_annotations: + if my_item.meta: meta_part = self.serialize_meta(item=my_item, **my_kwargs) if meta_part.text: parts.append(meta_part) @@ -655,3 +658,30 @@ def _get_page_breaks(self, text: str) -> Iterable[Tuple[str, int, int]]: prev_page_nr = int(match.group(1)) next_page_nr = int(match.group(2)) yield (full_match, prev_page_nr, next_page_nr) + + +def _should_use_legacy_annotations( + *, + params: CommonParams, + item: Union[PictureItem, TableItem], + kind: Optional[str] = None, +) -> bool: + if item.meta: + return False + with warnings.catch_warnings(record=True) as caught_warnings: + warnings.simplefilter("ignore", DeprecationWarning) + if ( + incl_attr := getattr(params, "include_annotations", None) + ) is not None and not incl_attr: + return False + use_legacy = bool( + [ + ann + for ann in item.annotations + if ((ann.kind == kind) if kind is not None else True) + ] + ) + if use_legacy: + for w in caught_warnings: + warnings.warn(w.message, w.category) + return use_legacy diff --git a/docling_core/transforms/serializer/doctags.py b/docling_core/transforms/serializer/doctags.py index 0195cd8e..0deaa991 100644 --- a/docling_core/transforms/serializer/doctags.py +++ b/docling_core/transforms/serializer/doctags.py @@ -23,6 +23,7 @@ from docling_core.transforms.serializer.common import ( CommonParams, DocSerializer, + _should_use_legacy_annotations, create_ser_result, ) from docling_core.types.doc.base import BoundingBox @@ -236,18 +237,25 @@ def serialize( # handle classification data predicted_class: Optional[str] = None - if item.meta and item.meta.classification: - predicted_class = ( - item.meta.classification.get_main_prediction().class_name - ) - elif ( - classifications := [ + if item.meta: + if item.meta.classification: + predicted_class = ( + item.meta.classification.get_main_prediction().class_name + ) + elif _should_use_legacy_annotations( + params=params, + item=item, + kind=PictureClassificationData.model_fields["kind"].default, + ): + if classifications := [ ann for ann in item.annotations if isinstance(ann, PictureClassificationData) - ] - ) and classifications[0].predicted_classes: - predicted_class = classifications[0].predicted_classes[0].class_name + ]: + if classifications[0].predicted_classes: + predicted_class = ( + classifications[0].predicted_classes[0].class_name + ) if predicted_class: body += DocumentToken.get_picture_classification_token(predicted_class) if predicted_class in [ @@ -263,25 +271,39 @@ def serialize( # handle molecule data smi: Optional[str] = None - if item.meta and item.meta.molecule: - smi = item.meta.molecule.smi - elif smiles_annotations := [ - ann for ann in item.annotations if isinstance(ann, PictureMoleculeData) - ]: - smi = smiles_annotations[0].smi + if item.meta: + if item.meta.molecule: + smi = item.meta.molecule.smi + elif _should_use_legacy_annotations( + params=params, + item=item, + kind=PictureMoleculeData.model_fields["kind"].default, + ): + if smiles_annotations := [ + ann + for ann in item.annotations + if isinstance(ann, PictureMoleculeData) + ]: + smi = smiles_annotations[0].smi if smi: body += _wrap(text=smi, wrap_tag=DocumentToken.SMILES.value) # handle tabular chart data chart_data: Optional[TableData] = None - if item.meta and item.meta.tabular_chart: - chart_data = item.meta.tabular_chart.chart_data - elif tabular_chart_annotations := [ - ann - for ann in item.annotations - if isinstance(ann, PictureTabularChartData) - ]: - chart_data = tabular_chart_annotations[0].chart_data + if item.meta: + if item.meta.tabular_chart: + chart_data = item.meta.tabular_chart.chart_data + elif _should_use_legacy_annotations( + params=params, + item=item, + kind=PictureTabularChartData.model_fields["kind"].default, + ): + if tabular_chart_annotations := [ + ann + for ann in item.annotations + if isinstance(ann, PictureTabularChartData) + ]: + chart_data = tabular_chart_annotations[0].chart_data if chart_data and chart_data.table_cells: temp_doc = DoclingDocument(name="temp") temp_table = temp_doc.add_table(data=chart_data) diff --git a/docling_core/transforms/serializer/html.py b/docling_core/transforms/serializer/html.py index 71bd798e..1a02d48c 100644 --- a/docling_core/transforms/serializer/html.py +++ b/docling_core/transforms/serializer/html.py @@ -38,6 +38,7 @@ CommonParams, DocSerializer, _get_annotation_text, + _should_use_legacy_annotations, create_ser_result, ) from docling_core.transforms.serializer.html_styles import ( @@ -496,7 +497,11 @@ def get_img_row(imgb64: str, ind: int) -> str: if img_text: res_parts.append(create_ser_result(text=img_text, span_source=item)) - if params.enable_chart_tables: + if params.enable_chart_tables and _should_use_legacy_annotations( + params=params, + item=item, + kind=PictureTabularChartData.model_fields["kind"].default, + ): # Check if picture has attached PictureTabularChartData tabular_chart_annotations = [ ann @@ -867,8 +872,13 @@ def _serialize_meta_field(self, meta: BaseMeta, name: str) -> Optional[str]: elif isinstance(field_val, MoleculeMetaField): txt = field_val.smi elif isinstance(field_val, TabularChartMetaField): - # suppressing tabular chart serialization - return None + temp_doc = DoclingDocument(name="temp") + temp_table = temp_doc.add_table(data=field_val.chart_data) + table_content = temp_table.export_to_html(temp_doc).strip() + if table_content: + txt = table_content + else: + return None elif tmp := str(field_val or ""): txt = tmp else: @@ -1119,17 +1129,16 @@ def serialize_captions( results.append(cap_ser_res) if ( - params.use_legacy_annotations - and params.include_annotations - and item.self_ref not in excluded_refs + item.self_ref not in excluded_refs + and isinstance(item, (PictureItem, TableItem)) + and _should_use_legacy_annotations(params=params, item=item) ): - if isinstance(item, (PictureItem, TableItem)): - ann_res = self.serialize_annotations( - item=item, - **kwargs, - ) - if ann_res.text: - results.append(ann_res) + ann_res = self.serialize_annotations( + item=item, + **kwargs, + ) + if ann_res.text: + results.append(ann_res) text_res = params.caption_delim.join([r.text for r in results]) if text_res: diff --git a/docling_core/transforms/serializer/markdown.py b/docling_core/transforms/serializer/markdown.py index 6292761d..c9660366 100644 --- a/docling_core/transforms/serializer/markdown.py +++ b/docling_core/transforms/serializer/markdown.py @@ -33,6 +33,7 @@ CommonParams, DocSerializer, _get_annotation_text, + _should_use_legacy_annotations, create_ser_result, ) from docling_core.types.doc.base import ImageRefMode @@ -72,23 +73,6 @@ ) -def _get_annotation_ser_result( - ann_kind: str, ann_text: str, mark_annotation: bool, doc_item: DocItem -): - return create_ser_result( - text=( - ( - f'' - f"{ann_text}" - f"" - ) - if mark_annotation - else ann_text - ), - span_source=doc_item, - ) - - class OrigListItemMarkerMode(str, Enum): """Display mode for original list item marker.""" @@ -315,8 +299,13 @@ def _serialize_meta_field( elif isinstance(field_val, MoleculeMetaField): txt = field_val.smi elif isinstance(field_val, TabularChartMetaField): - # suppressing tabular chart serialization - return None + temp_doc = DoclingDocument(name="temp") + temp_table = temp_doc.add_table(data=field_val.chart_data) + table_content = temp_table.export_to_markdown(temp_doc).strip() + if table_content: + txt = table_content + else: + return None elif tmp := str(field_val or ""): txt = tmp else: @@ -397,7 +386,7 @@ def serialize( if item.self_ref not in doc_serializer.get_excluded_refs(**kwargs): - if params.use_legacy_annotations and params.include_annotations: + if _should_use_legacy_annotations(params=params, item=item): ann_res = doc_serializer.serialize_annotations( item=item, @@ -466,7 +455,7 @@ def serialize( res_parts.append(cap_res) if item.self_ref not in doc_serializer.get_excluded_refs(**kwargs): - if params.use_legacy_annotations and params.include_annotations: + if _should_use_legacy_annotations(params=params, item=item): ann_res = doc_serializer.serialize_annotations( item=item, **kwargs, @@ -483,7 +472,11 @@ def serialize( if img_res.text: res_parts.append(img_res) - if params.enable_chart_tables: + if params.enable_chart_tables and _should_use_legacy_annotations( + params=params, + item=item, + kind=PictureTabularChartData.model_fields["kind"].default, + ): # Check if picture has attached PictureTabularChartData tabular_chart_annotations = [ ann diff --git a/docling_core/types/doc/document.py b/docling_core/types/doc/document.py index 626a9734..f2ea223c 100644 --- a/docling_core/types/doc/document.py +++ b/docling_core/types/doc/document.py @@ -4715,7 +4715,7 @@ def save_as_markdown( include_annotations: bool = True, *, mark_meta: bool = False, - use_legacy_annotations: bool = False, + use_legacy_annotations: Optional[bool] = None, # deprecated ): """Save to markdown.""" if isinstance(filename, str): @@ -4772,7 +4772,7 @@ def export_to_markdown( # noqa: C901 include_annotations: bool = True, mark_annotations: bool = False, *, - use_legacy_annotations: bool = False, + use_legacy_annotations: Optional[bool] = None, # deprecated allowed_meta_names: Optional[set[str]] = None, blocked_meta_names: Optional[set[str]] = None, mark_meta: bool = False, @@ -4815,17 +4815,15 @@ def export_to_markdown( # noqa: C901 :param page_break_placeholder: The placeholder to include for marking page breaks. None means no page break placeholder will be used. :type page_break_placeholder: Optional[str] = None - :param include_annotations: bool: Whether to include annotations in the export. - (Default value = True). + :param include_annotations: bool: Whether to include annotations in the export; only considered if item does not + have meta. (Default value = True). :type include_annotations: bool = True - :param mark_annotations: bool: Whether to mark annotations in the export; only - relevant if include_annotations is True. (Default value = False). + :param mark_annotations: bool: Whether to mark annotations in the export; only considered if item does not have + meta. (Default value = False). :type mark_annotations: bool = False - :param use_legacy_annotations: bool: Whether to use legacy annotation serialization. - (Default value = False). - :type use_legacy_annotations: bool = False - :param mark_meta: bool: Whether to mark meta in the export; only - relevant if use_legacy_annotations is False. (Default value = False). + :param use_legacy_annotations: bool: Deprecated; legacy annotations considered only when meta not present. + :type use_legacy_annotations: Optional[bool] = None + :param mark_meta: bool: Whether to mark meta in the export :type mark_meta: bool = False :returns: The exported Markdown representation. :rtype: str @@ -4845,6 +4843,13 @@ def export_to_markdown( # noqa: C901 if included_content_layers is not None else DEFAULT_CONTENT_LAYERS ) + + if use_legacy_annotations is not None: + warnings.warn( + "Parameter `use_legacy_annotations` has been deprecated and will be ignored.", + DeprecationWarning, + ) + serializer = MarkdownDocSerializer( doc=self, params=MarkdownParams( @@ -4863,7 +4868,6 @@ def export_to_markdown( # noqa: C901 page_break_placeholder=page_break_placeholder, mark_meta=mark_meta, include_annotations=include_annotations, - use_legacy_annotations=use_legacy_annotations, allowed_meta_names=allowed_meta_names, blocked_meta_names=blocked_meta_names or set(), mark_annotations=mark_annotations, @@ -5421,25 +5425,51 @@ def _add_text( ) pic.captions.append(caption.get_ref()) pic_title = "picture" + + pic_classification = None if chart_type is not None: - pic.annotations.append( - PictureClassificationData( - provenance="load_from_doctags", - predicted_classes=[ - # chart_type - PictureClassificationClass( - class_name=chart_type, confidence=1.0 - ) - ], - ) + # pic.annotations.append( + # PictureClassificationData( + # provenance="load_from_doctags", + # predicted_classes=[ + # # chart_type + # PictureClassificationClass( + # class_name=chart_type, confidence=1.0 + # ) + # ], + # ) + # ) + pic_classification = PictureClassificationMetaField( + predictions=[ + PictureClassificationPrediction( + class_name=chart_type, + confidence=1.0, + created_by="load_from_doctags", + ) + ] ) pic_title = chart_type + + pic_tabular_chart = None if table_data is not None: - # Add chart data as PictureTabularChartData - pd = PictureTabularChartData( - chart_data=table_data, title=pic_title + # # Add chart data as PictureTabularChartData + # pd = PictureTabularChartData( + # chart_data=table_data, title=pic_title + # ) + # pic.annotations.append(pd) + pic_tabular_chart = TabularChartMetaField( + title=pic_title, + chart_data=table_data, + ) + + if ( + pic_classification is not None + or pic_tabular_chart is not None + ): + pic.meta = PictureMeta( + classification=pic_classification, + tabular_chart=pic_tabular_chart, ) - pic.annotations.append(pd) else: if bbox: # In case we don't have access to an binary of an image @@ -6090,7 +6120,10 @@ def concatenate(cls, docs: Sequence["DoclingDocument"]) -> "DoclingDocument": def _validate_rules(self): def validate_furniture(doc: DoclingDocument): - if doc.furniture.children: + with warnings.catch_warnings(): + warnings.simplefilter("ignore", category=DeprecationWarning) + has_furniture_children = len(doc.furniture.children) > 0 + if has_furniture_children: raise ValueError( f"Deprecated furniture node {doc.furniture.self_ref} has children" ) diff --git a/test/data/doc/2408.09869v3_enriched_p1_include_annotations_false.gt.md b/test/data/doc/2408.09869v3_enriched_p1_include_annotations_false.gt.md deleted file mode 100644 index dd345623..00000000 --- a/test/data/doc/2408.09869v3_enriched_p1_include_annotations_false.gt.md +++ /dev/null @@ -1,44 +0,0 @@ -# Docling Technical Report - - - -Version 1.0 - -Christoph Auer Maksym Lysak Ahmed Nassar Michele Dolfi Nikolaos Livathinos Panos Vagenas Cesar Berrospi Ramis Matteo Omenetti Fabian Lindlbauer Kasper Dinkla Lokesh Mishra Yusik Kim Shubham Gupta Rafael Teixeira de Lima Valery Weber Lucas Morin Ingmar Meijer Viktor Kuropiatnyk Peter W. J. Staar - -AI4K Group, IBM Research R¨ uschlikon, Switzerland - -## Abstract - -This technical report introduces Docling , an easy to use, self-contained, MITlicensed open-source package for PDF document conversion. It is powered by state-of-the-art specialized AI models for layout analysis (DocLayNet) and table structure recognition (TableFormer), and runs efficiently on commodity hardware in a small resource budget. The code interface allows for easy extensibility and addition of new features and models. - -## 1 Introduction - -Converting PDF documents back into a machine-processable format has been a major challenge for decades due to their huge variability in formats, weak standardization and printing-optimized characteristic, which discards most structural features and metadata. With the advent of LLMs and popular application patterns such as retrieval-augmented generation (RAG), leveraging the rich content embedded in PDFs has become ever more relevant. In the past decade, several powerful document understanding solutions have emerged on the market, most of which are commercial software, cloud offerings [3] and most recently, multi-modal vision-language models. As of today, only a handful of open-source tools cover PDF conversion, leaving a significant feature and quality gap to proprietary solutions. - -With Docling , we open-source a very capable and efficient document conversion tool which builds on the powerful, specialized AI models and datasets for layout analysis and table structure recognition we developed and presented in the recent past [12, 13, 9]. Docling is designed as a simple, self-contained python library with permissive license, running entirely locally on commodity hardware. Its code architecture allows for easy extensibility and addition of new features and models. - -torch runtimes backing the Docling pipeline. We will deliver updates on this topic at in a future version of this report. - -Table 1: Runtime characteristics of Docling with the standard model pipeline and settings, on our test dataset of 225 pages, on two different systems. OCR is disabled. We show the time-to-solution (TTS), computed throughput in pages per second, and the peak memory used (resident set size) for both the Docling-native PDF backend and for the pypdfium backend, using 4 and 16 threads. - -| CPU | Thread budget | native backend | native backend | native backend | pypdfium backend | pypdfium backend | pypdfium backend | -|----------------------------------|-----------------|------------------|------------------|------------------|--------------------|--------------------|--------------------| -| | | TTS | Pages/s | Mem | TTS | Pages/s | Mem | -| Apple M3 Max | 4 | 177 s 167 s | 1.27 1.34 | 6.20 GB | 103 s 92 s | 2.18 2.45 | 2.56 GB | -| (16 cores) Intel(R) Xeon E5-2690 | 16 4 16 | 375 s 244 s | 0.60 0.92 | 6.16 GB | 239 s 143 s | 0.94 1.57 | 2.42 GB | - -## 5 Applications - -Thanks to the high-quality, richly structured document conversion achieved by Docling, its output qualifies for numerous downstream applications. For example, Docling can provide a base for detailed enterprise document search, passage retrieval or classification use-cases, or support knowledge extraction pipelines, allowing specific treatment of different structures in the document, such as tables, figures, section structure or references. For popular generative AI application patterns, such as retrieval-augmented generation (RAG), we provide quackling , an open-source package which capitalizes on Docling's feature-rich document output to enable document-native optimized vector embedding and chunking. It plugs in seamlessly with LLM frameworks such as LlamaIndex [8]. Since Docling is fast, stable and cheap to run, it also makes for an excellent choice to build document-derived datasets. With its powerful table structure recognition, it provides significant benefit to automated knowledge-base construction [11, 10]. Docling is also integrated within the open IBM data prep kit [6], which implements scalable data transforms to build large-scale multi-modal training datasets. - -## 6 Future work and contributions - -Docling is designed to allow easy extension of the model library and pipelines. In the future, we plan to extend Docling with several more models, such as a figure-classifier model, an equationrecognition model, a code-recognition model and more. This will help improve the quality of conversion for specific types of content, as well as augment extracted document metadata with additional information. Further investment into testing and optimizing GPU acceleration as well as improving the Docling-native PDF backend are on our roadmap, too. - -We encourage everyone to propose or implement additional features and models, and will gladly take your inputs and contributions under review . The codebase of Docling is open for use and contribution, under the MIT license agreement and in alignment with our contributing guidelines included in the Docling repository. If you use Docling in your projects, please consider citing this technical report. - -## References - -- [1] J. AI. Easyocr: Ready-to-use ocr with 80+ supported languages. https://github.com/ JaidedAI/EasyOCR , 2024. Version: 1.7.0. -- [2] J. Ansel, E. Yang, H. He, N. Gimelshein, A. Jain, M. Voznesensky, B. Bao, P. Bell, D. Berard, E. Burovski, G. Chauhan, A. Chourdia, W. Constable, A. Desmaison, Z. DeVito, E. Ellison, W. Feng, J. Gong, M. Gschwind, B. Hirsh, S. Huang, K. Kalambarkar, L. Kirsch, M. Lazos, M. Lezcano, Y. Liang, J. Liang, Y. Lu, C. Luk, B. Maher, Y. Pan, C. Puhrsch, M. Reso, M. Saroufim, M. Y. Siraichi, H. Suk, M. Suo, P. Tillet, E. Wang, X. Wang, W. Wen, S. Zhang, X. Zhao, K. Zhou, R. Zou, A. Mathews, G. Chanan, P. Wu, and S. Chintala. Pytorch 2: Faster diff --git a/test/data/doc/2408.09869v3_enriched_p1_mark_annotations_false.gt.md b/test/data/doc/2408.09869v3_enriched_p1_mark_meta_false.gt.md similarity index 98% rename from test/data/doc/2408.09869v3_enriched_p1_mark_annotations_false.gt.md rename to test/data/doc/2408.09869v3_enriched_p1_mark_meta_false.gt.md index 61f88f35..fe55c055 100644 --- a/test/data/doc/2408.09869v3_enriched_p1_mark_annotations_false.gt.md +++ b/test/data/doc/2408.09869v3_enriched_p1_mark_meta_false.gt.md @@ -22,8 +22,7 @@ With Docling , we open-source a very capable and efficient document conversion t torch runtimes backing the Docling pipeline. We will deliver updates on this topic at in a future version of this report. -summary: Typical Docling setup runtime characterization. -type: performance data +{'summary': 'Typical Docling setup runtime characterization.', 'type': 'performance data'} Table 1: Runtime characteristics of Docling with the standard model pipeline and settings, on our test dataset of 225 pages, on two different systems. OCR is disabled. We show the time-to-solution (TTS), computed throughput in pages per second, and the peak memory used (resident set size) for both the Docling-native PDF backend and for the pypdfium backend, using 4 and 16 threads. diff --git a/test/data/doc/2408.09869v3_enriched_p1_mark_meta_true.gt.md b/test/data/doc/2408.09869v3_enriched_p1_mark_meta_true.gt.md index 3f8a9266..14a091c6 100644 --- a/test/data/doc/2408.09869v3_enriched_p1_mark_meta_true.gt.md +++ b/test/data/doc/2408.09869v3_enriched_p1_mark_meta_true.gt.md @@ -24,9 +24,6 @@ torch runtimes backing the Docling pipeline. We will deliver updates on this top [Docling Legacy Misc] {'summary': 'Typical Docling setup runtime characterization.', 'type': 'performance data'} -summary: Typical Docling setup runtime characterization. -type: performance data - Table 1: Runtime characteristics of Docling with the standard model pipeline and settings, on our test dataset of 225 pages, on two different systems. OCR is disabled. We show the time-to-solution (TTS), computed throughput in pages per second, and the peak memory used (resident set size) for both the Docling-native PDF backend and for the pypdfium backend, using 4 and 16 threads. | CPU | Thread budget | native backend | native backend | native backend | pypdfium backend | pypdfium backend | pypdfium backend | diff --git a/test/data/doc/2408.09869v3_enriched_p1_use_legacy_annotations_true_mark_annotations_true.gt.md b/test/data/doc/2408.09869v3_enriched_p1_use_legacy_annotations_true_mark_annotations_true.gt.md deleted file mode 100644 index c08732f2..00000000 --- a/test/data/doc/2408.09869v3_enriched_p1_use_legacy_annotations_true_mark_annotations_true.gt.md +++ /dev/null @@ -1,49 +0,0 @@ -# Docling Technical Report - -In this image we can see a cartoon image of a duck holding a paper. - - - -Version 1.0 - -Christoph Auer Maksym Lysak Ahmed Nassar Michele Dolfi Nikolaos Livathinos Panos Vagenas Cesar Berrospi Ramis Matteo Omenetti Fabian Lindlbauer Kasper Dinkla Lokesh Mishra Yusik Kim Shubham Gupta Rafael Teixeira de Lima Valery Weber Lucas Morin Ingmar Meijer Viktor Kuropiatnyk Peter W. J. Staar - -AI4K Group, IBM Research R¨ uschlikon, Switzerland - -## Abstract - -This technical report introduces Docling , an easy to use, self-contained, MITlicensed open-source package for PDF document conversion. It is powered by state-of-the-art specialized AI models for layout analysis (DocLayNet) and table structure recognition (TableFormer), and runs efficiently on commodity hardware in a small resource budget. The code interface allows for easy extensibility and addition of new features and models. - -## 1 Introduction - -Converting PDF documents back into a machine-processable format has been a major challenge for decades due to their huge variability in formats, weak standardization and printing-optimized characteristic, which discards most structural features and metadata. With the advent of LLMs and popular application patterns such as retrieval-augmented generation (RAG), leveraging the rich content embedded in PDFs has become ever more relevant. In the past decade, several powerful document understanding solutions have emerged on the market, most of which are commercial software, cloud offerings [3] and most recently, multi-modal vision-language models. As of today, only a handful of open-source tools cover PDF conversion, leaving a significant feature and quality gap to proprietary solutions. - -With Docling , we open-source a very capable and efficient document conversion tool which builds on the powerful, specialized AI models and datasets for layout analysis and table structure recognition we developed and presented in the recent past [12, 13, 9]. Docling is designed as a simple, self-contained python library with permissive license, running entirely locally on commodity hardware. Its code architecture allows for easy extensibility and addition of new features and models. - -torch runtimes backing the Docling pipeline. We will deliver updates on this topic at in a future version of this report. - -summary: Typical Docling setup runtime characterization. -type: performance data - -Table 1: Runtime characteristics of Docling with the standard model pipeline and settings, on our test dataset of 225 pages, on two different systems. OCR is disabled. We show the time-to-solution (TTS), computed throughput in pages per second, and the peak memory used (resident set size) for both the Docling-native PDF backend and for the pypdfium backend, using 4 and 16 threads. - -| CPU | Thread budget | native backend | native backend | native backend | pypdfium backend | pypdfium backend | pypdfium backend | -|----------------------------------|-----------------|------------------|------------------|------------------|--------------------|--------------------|--------------------| -| | | TTS | Pages/s | Mem | TTS | Pages/s | Mem | -| Apple M3 Max | 4 | 177 s 167 s | 1.27 1.34 | 6.20 GB | 103 s 92 s | 2.18 2.45 | 2.56 GB | -| (16 cores) Intel(R) Xeon E5-2690 | 16 4 16 | 375 s 244 s | 0.60 0.92 | 6.16 GB | 239 s 143 s | 0.94 1.57 | 2.42 GB | - -## 5 Applications - -Thanks to the high-quality, richly structured document conversion achieved by Docling, its output qualifies for numerous downstream applications. For example, Docling can provide a base for detailed enterprise document search, passage retrieval or classification use-cases, or support knowledge extraction pipelines, allowing specific treatment of different structures in the document, such as tables, figures, section structure or references. For popular generative AI application patterns, such as retrieval-augmented generation (RAG), we provide quackling , an open-source package which capitalizes on Docling's feature-rich document output to enable document-native optimized vector embedding and chunking. It plugs in seamlessly with LLM frameworks such as LlamaIndex [8]. Since Docling is fast, stable and cheap to run, it also makes for an excellent choice to build document-derived datasets. With its powerful table structure recognition, it provides significant benefit to automated knowledge-base construction [11, 10]. Docling is also integrated within the open IBM data prep kit [6], which implements scalable data transforms to build large-scale multi-modal training datasets. - -## 6 Future work and contributions - -Docling is designed to allow easy extension of the model library and pipelines. In the future, we plan to extend Docling with several more models, such as a figure-classifier model, an equationrecognition model, a code-recognition model and more. This will help improve the quality of conversion for specific types of content, as well as augment extracted document metadata with additional information. Further investment into testing and optimizing GPU acceleration as well as improving the Docling-native PDF backend are on our roadmap, too. - -We encourage everyone to propose or implement additional features and models, and will gladly take your inputs and contributions under review . The codebase of Docling is open for use and contribution, under the MIT license agreement and in alignment with our contributing guidelines included in the Docling repository. If you use Docling in your projects, please consider citing this technical report. - -## References - -- [1] J. AI. Easyocr: Ready-to-use ocr with 80+ supported languages. https://github.com/ JaidedAI/EasyOCR , 2024. Version: 1.7.0. -- [2] J. Ansel, E. Yang, H. He, N. Gimelshein, A. Jain, M. Voznesensky, B. Bao, P. Bell, D. Berard, E. Burovski, G. Chauhan, A. Chourdia, W. Constable, A. Desmaison, Z. DeVito, E. Ellison, W. Feng, J. Gong, M. Gschwind, B. Hirsh, S. Huang, K. Kalambarkar, L. Kirsch, M. Lazos, M. Lezcano, Y. Liang, J. Liang, Y. Lu, C. Luk, B. Maher, Y. Pan, C. Puhrsch, M. Reso, M. Saroufim, M. Y. Siraichi, H. Suk, M. Suo, P. Tillet, E. Wang, X. Wang, W. Wen, S. Zhang, X. Zhao, K. Zhou, R. Zou, A. Mathews, G. Chanan, P. Wu, and S. Chintala. Pytorch 2: Faster diff --git a/test/data/doc/barchart.dt.out.json b/test/data/doc/barchart.dt.out.json new file mode 100644 index 00000000..a761f2d1 --- /dev/null +++ b/test/data/doc/barchart.dt.out.json @@ -0,0 +1,699 @@ +{ + "schema_name": "DoclingDocument", + "version": "1.8.0", + "name": "Document", + "furniture": { + "self_ref": "#/furniture", + "children": [], + "content_layer": "furniture", + "name": "_root_", + "label": "unspecified" + }, + "body": { + "self_ref": "#/body", + "children": [ + { + "$ref": "#/texts/0" + }, + { + "$ref": "#/pictures/0" + } + ], + "content_layer": "body", + "name": "_root_", + "label": "unspecified" + }, + "groups": [], + "texts": [ + { + "self_ref": "#/texts/0", + "parent": { + "$ref": "#/body" + }, + "children": [], + "content_layer": "furniture", + "label": "page_header", + "prov": [ + { + "page_no": 1, + "bbox": { + "l": 150.80399999999997, + "t": 42.980000000000004, + "r": 460.908, + "b": 61.4, + "coord_origin": "TOPLEFT" + }, + "charspan": [ + 0, + 38 + ] + } + ], + "orig": "Probability, Combinatorics and Control", + "text": "Probability, Combinatorics and Control" + } + ], + "pictures": [ + { + "self_ref": "#/pictures/0", + "parent": { + "$ref": "#/body" + }, + "children": [], + "content_layer": "body", + "meta": { + "classification": { + "predictions": [ + { + "confidence": 1.0, + "created_by": "load_from_doctags", + "class_name": "bar_chart" + } + ] + }, + "tabular_chart": { + "title": "bar_chart", + "chart_data": { + "table_cells": [ + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 0, + "end_row_offset_idx": 1, + "start_col_offset_idx": 0, + "end_col_offset_idx": 1, + "text": "Number of impellers", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 0, + "end_row_offset_idx": 1, + "start_col_offset_idx": 1, + "end_col_offset_idx": 2, + "text": "single-frequency", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 0, + "end_row_offset_idx": 1, + "start_col_offset_idx": 2, + "end_col_offset_idx": 3, + "text": "multi-frequency", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 1, + "end_row_offset_idx": 2, + "start_col_offset_idx": 0, + "end_col_offset_idx": 1, + "text": "1", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 1, + "end_row_offset_idx": 2, + "start_col_offset_idx": 1, + "end_col_offset_idx": 2, + "text": "0.06", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 1, + "end_row_offset_idx": 2, + "start_col_offset_idx": 2, + "end_col_offset_idx": 3, + "text": "0.16", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 2, + "end_row_offset_idx": 3, + "start_col_offset_idx": 0, + "end_col_offset_idx": 1, + "text": "2", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 2, + "end_row_offset_idx": 3, + "start_col_offset_idx": 1, + "end_col_offset_idx": 2, + "text": "0.12", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 2, + "end_row_offset_idx": 3, + "start_col_offset_idx": 2, + "end_col_offset_idx": 3, + "text": "0.26", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 3, + "end_row_offset_idx": 4, + "start_col_offset_idx": 0, + "end_col_offset_idx": 1, + "text": "3", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 3, + "end_row_offset_idx": 4, + "start_col_offset_idx": 1, + "end_col_offset_idx": 2, + "text": "0.16", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 3, + "end_row_offset_idx": 4, + "start_col_offset_idx": 2, + "end_col_offset_idx": 3, + "text": "0.27", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 4, + "end_row_offset_idx": 5, + "start_col_offset_idx": 0, + "end_col_offset_idx": 1, + "text": "4", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 4, + "end_row_offset_idx": 5, + "start_col_offset_idx": 1, + "end_col_offset_idx": 2, + "text": "0.14", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 4, + "end_row_offset_idx": 5, + "start_col_offset_idx": 2, + "end_col_offset_idx": 3, + "text": "0.26", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 5, + "end_row_offset_idx": 6, + "start_col_offset_idx": 0, + "end_col_offset_idx": 1, + "text": "5", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 5, + "end_row_offset_idx": 6, + "start_col_offset_idx": 1, + "end_col_offset_idx": 2, + "text": "0.16", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 5, + "end_row_offset_idx": 6, + "start_col_offset_idx": 2, + "end_col_offset_idx": 3, + "text": "0.25", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 6, + "end_row_offset_idx": 7, + "start_col_offset_idx": 0, + "end_col_offset_idx": 1, + "text": "6", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 6, + "end_row_offset_idx": 7, + "start_col_offset_idx": 1, + "end_col_offset_idx": 2, + "text": "0.24", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 6, + "end_row_offset_idx": 7, + "start_col_offset_idx": 2, + "end_col_offset_idx": 3, + "text": "0.24", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + } + ], + "num_rows": 7, + "num_cols": 3, + "grid": [ + [ + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 0, + "end_row_offset_idx": 1, + "start_col_offset_idx": 0, + "end_col_offset_idx": 1, + "text": "Number of impellers", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 0, + "end_row_offset_idx": 1, + "start_col_offset_idx": 1, + "end_col_offset_idx": 2, + "text": "single-frequency", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 0, + "end_row_offset_idx": 1, + "start_col_offset_idx": 2, + "end_col_offset_idx": 3, + "text": "multi-frequency", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + } + ], + [ + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 1, + "end_row_offset_idx": 2, + "start_col_offset_idx": 0, + "end_col_offset_idx": 1, + "text": "1", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 1, + "end_row_offset_idx": 2, + "start_col_offset_idx": 1, + "end_col_offset_idx": 2, + "text": "0.06", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 1, + "end_row_offset_idx": 2, + "start_col_offset_idx": 2, + "end_col_offset_idx": 3, + "text": "0.16", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + } + ], + [ + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 2, + "end_row_offset_idx": 3, + "start_col_offset_idx": 0, + "end_col_offset_idx": 1, + "text": "2", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 2, + "end_row_offset_idx": 3, + "start_col_offset_idx": 1, + "end_col_offset_idx": 2, + "text": "0.12", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 2, + "end_row_offset_idx": 3, + "start_col_offset_idx": 2, + "end_col_offset_idx": 3, + "text": "0.26", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + } + ], + [ + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 3, + "end_row_offset_idx": 4, + "start_col_offset_idx": 0, + "end_col_offset_idx": 1, + "text": "3", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 3, + "end_row_offset_idx": 4, + "start_col_offset_idx": 1, + "end_col_offset_idx": 2, + "text": "0.16", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 3, + "end_row_offset_idx": 4, + "start_col_offset_idx": 2, + "end_col_offset_idx": 3, + "text": "0.27", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + } + ], + [ + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 4, + "end_row_offset_idx": 5, + "start_col_offset_idx": 0, + "end_col_offset_idx": 1, + "text": "4", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 4, + "end_row_offset_idx": 5, + "start_col_offset_idx": 1, + "end_col_offset_idx": 2, + "text": "0.14", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 4, + "end_row_offset_idx": 5, + "start_col_offset_idx": 2, + "end_col_offset_idx": 3, + "text": "0.26", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + } + ], + [ + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 5, + "end_row_offset_idx": 6, + "start_col_offset_idx": 0, + "end_col_offset_idx": 1, + "text": "5", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 5, + "end_row_offset_idx": 6, + "start_col_offset_idx": 1, + "end_col_offset_idx": 2, + "text": "0.16", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 5, + "end_row_offset_idx": 6, + "start_col_offset_idx": 2, + "end_col_offset_idx": 3, + "text": "0.25", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + } + ], + [ + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 6, + "end_row_offset_idx": 7, + "start_col_offset_idx": 0, + "end_col_offset_idx": 1, + "text": "6", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 6, + "end_row_offset_idx": 7, + "start_col_offset_idx": 1, + "end_col_offset_idx": 2, + "text": "0.24", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + }, + { + "row_span": 1, + "col_span": 1, + "start_row_offset_idx": 6, + "end_row_offset_idx": 7, + "start_col_offset_idx": 2, + "end_col_offset_idx": 3, + "text": "0.24", + "column_header": false, + "row_header": false, + "row_section": false, + "fillable": false + } + ] + ] + } + } + }, + "label": "picture", + "prov": [ + { + "page_no": 1, + "bbox": { + "l": 216.648, + "t": 113.58999999999999, + "r": 832.6080000000001, + "b": 454.35999999999996, + "coord_origin": "TOPLEFT" + }, + "charspan": [ + 0, + 0 + ] + } + ], + "captions": [], + "references": [], + "footnotes": [], + "image": { + "mimetype": "image/png", + "dpi": 72, + "size": { + "width": 616.0, + "height": 341.0 + }, + "uri": "data:image/png;base64,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" + }, + "annotations": [] + } + ], + "tables": [], + "key_value_items": [], + "form_items": [], + "pages": { + "1": { + "size": { + "width": 1062.0, + "height": 1535.0 + }, + "image": { + "mimetype": "image/png", + "dpi": 72, + "size": { + "width": 1062.0, + "height": 1535.0 + }, + "uri": "data:image/png;base64,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" + }, + "page_no": 1 + } + } +} diff --git a/test/data/doc/barchart.gt.html b/test/data/doc/barchart.gt.html index 05cbb81f..75166c7b 100644 --- a/test/data/doc/barchart.gt.html +++ b/test/data/doc/barchart.gt.html @@ -125,7 +125,7 @@
Bar chart
-
Number of impellerssingle-frequencymulti-frequency
10.060.16
20.120.26
30.160.27
40.140.26
50.160.25
60.240.24
+
Number of impellerssingle-frequencymulti-frequency
10.060.16
20.120.26
30.160.27
40.140.26
50.160.25
60.240.24
diff --git a/test/data/doc/barchart.gt.md b/test/data/doc/barchart.gt.md index 0adc5569..8f7d5025 100644 --- a/test/data/doc/barchart.gt.md +++ b/test/data/doc/barchart.gt.md @@ -1,7 +1,5 @@ Bar chart - - | Number of impellers | single-frequency | multi-frequency | |-----------------------|--------------------|-------------------| | 1 | 0.06 | 0.16 | @@ -10,3 +8,5 @@ Bar chart | 4 | 0.14 | 0.26 | | 5 | 0.16 | 0.25 | | 6 | 0.24 | 0.24 | + + diff --git a/test/data/doc/constructed_legacy_annot_mark_false.gt.md b/test/data/doc/constructed_legacy_annot_mark_false.gt.md new file mode 100644 index 00000000..f0f3c469 --- /dev/null +++ b/test/data/doc/constructed_legacy_annot_mark_false.gt.md @@ -0,0 +1,80 @@ +- item of leading list + +# Title of the Document + +Author 1 +Affiliation 1 + +Author 2 +Affiliation 2 + +## 1. Introduction + +This paper introduces the biggest invention ever made. ... + +- list item 1 +- list item 2 +- list item 3 + 1. list item 3.a + 2. list item 3.b + 3. list item 3.c + 1. list item 3.c.i +- list item 4 + +This is the caption of table 1. + +This is a description of table 1. + +| Product | Years | Years | +|-----------|---------|---------| +| Product | 2016 | 2017 | +| Apple | 49823 | 695944 | + +This is the caption of figure 1. + + + +This is the caption of figure 2. + + + +- item 1 of list + +* item 1 of list after empty list +- 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 block: + +``` +print("Hello world") +``` + +Here a formula block: + +$$E=mc^2$$ + + + + + +Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) & [~~***everything at the same time.***~~](https://github.com/DS4SD/docling) + +- (i) Item 1 in A +- (ii) Item 2 in A +- (iii) Item 3 in A + 1. Item 1 in B + 42. Item 2 in B + 1. Item 1 in C + 2. Item 2 in C + 3. Item 3 in B +- (iv) Item 4 in A + +- List item without parent list group + +The end. diff --git a/test/data/doc/constructed_legacy_annot_mark_true.gt.md b/test/data/doc/constructed_legacy_annot_mark_true.gt.md new file mode 100644 index 00000000..0382ae17 --- /dev/null +++ b/test/data/doc/constructed_legacy_annot_mark_true.gt.md @@ -0,0 +1,80 @@ +- item of leading list + +# Title of the Document + +Author 1 +Affiliation 1 + +Author 2 +Affiliation 2 + +## 1. Introduction + +This paper introduces the biggest invention ever made. ... + +- list item 1 +- list item 2 +- list item 3 + 1. list item 3.a + 2. list item 3.b + 3. list item 3.c + 1. list item 3.c.i +- list item 4 + +This is the caption of table 1. + +This is a description of table 1. + +| Product | Years | Years | +|-----------|---------|---------| +| Product | 2016 | 2017 | +| Apple | 49823 | 695944 | + +This is the caption of figure 1. + + + +This is the caption of figure 2. + + + +- item 1 of list + +* item 1 of list after empty list +- 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 block: + +``` +print("Hello world") +``` + +Here a formula block: + +$$E=mc^2$$ + + + + + +Some formatting chops: **bold** *italic* underline ~~strikethrough~~ subscript superscript [hyperlink](.) & [~~***everything at the same time.***~~](https://github.com/DS4SD/docling) + +- (i) Item 1 in A +- (ii) Item 2 in A +- (iii) Item 3 in A + 1. Item 1 in B + 42. Item 2 in B + 1. Item 1 in C + 2. Item 2 in C + 3. Item 3 in B +- (iv) Item 4 in A + +- List item without parent list group + +The end. diff --git a/test/test_doctags_load.py b/test/test_doctags_load.py index a6bf5bea..5355c2d1 100644 --- a/test/test_doctags_load.py +++ b/test/test_doctags_load.py @@ -4,7 +4,7 @@ from PIL import Image as PILImage from docling_core.types.doc import DoclingDocument -from docling_core.types.doc.document import DocTagsDocument, PictureTabularChartData +from docling_core.types.doc.document import DocTagsDocument from .test_data_gen_flag import GEN_TEST_DATA @@ -118,11 +118,11 @@ def test_doctags_chart(): [Path("test/data/doc/barchart.png")], ) doc = DoclingDocument.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 + exp = "test/data/doc/barchart.dt.out.json" + verify( + exp_file=exp, + actual=doc.export_to_dict(), + ) def test_doctags_table_provenances_and_captions(): diff --git a/test/test_serialization.py b/test/test_serialization.py index a8ebcdaa..c0de487d 100644 --- a/test/test_serialization.py +++ b/test/test_serialization.py @@ -1,15 +1,10 @@ """Test serialization.""" from pathlib import Path -from typing import Any -from typing_extensions import override +import pytest -from docling_core.transforms.serializer.base import ( - BaseDocSerializer, - SerializationResult, -) -from docling_core.transforms.serializer.common import _DEFAULT_LABELS, create_ser_result +from docling_core.transforms.serializer.common import _DEFAULT_LABELS from docling_core.transforms.serializer.doctags import DocTagsDocSerializer from docling_core.transforms.serializer.html import ( HTMLDocSerializer, @@ -19,18 +14,15 @@ from docling_core.transforms.serializer.markdown import ( MarkdownDocSerializer, MarkdownParams, - MarkdownTableSerializer, OrigListItemMarkerMode, - _get_annotation_ser_result, ) from docling_core.transforms.visualizer.layout_visualizer import LayoutVisualizer from docling_core.types.doc.base import ImageRefMode from docling_core.types.doc.document import ( + DescriptionAnnotation, DoclingDocument, - MiscAnnotation, TableCell, TableData, - TableItem, ) from docling_core.types.doc.labels import DocItemLabel @@ -38,48 +30,6 @@ from .test_docling_doc import _construct_doc, _construct_rich_table_doc -class CustomAnnotationTableSerializer(MarkdownTableSerializer): - @override - def serialize( - self, - *, - item: TableItem, - doc_serializer: BaseDocSerializer, - doc: DoclingDocument, - **kwargs: Any, - ) -> SerializationResult: - params = MarkdownParams(**kwargs) - - res_parts: list[SerializationResult] = [] - - if params.include_annotations: - for ann in item.annotations: - if isinstance(ann, MiscAnnotation): - - # custom serialization logic: - ann_txt = "\n".join([f"{k}: {ann.content[k]}" for k in ann.content]) - - ann_ser_res = _get_annotation_ser_result( - ann_kind=ann.kind, - ann_text=ann_txt, - mark_annotation=params.mark_annotations, - doc_item=item, - ) - res_parts.append(ann_ser_res) - - # reusing the existing result (excluding the annotations): - parent_res = super().serialize( - item=item, - doc_serializer=doc_serializer, - doc=doc, - **{**kwargs, **{"include_annotations": False}}, - ) - res_parts.append(parent_res) - - text_res = "\n\n".join([part.text for part in res_parts]) - return create_ser_result(text=text_res, span_source=res_parts) - - def verify(exp_file: Path, actual: str): if GEN_TEST_DATA: with open(exp_file, "w", encoding="utf-8") as f: @@ -263,84 +213,82 @@ def test_md_list_item_markers(): ) -def test_md_legacy_include_annotations_false(): +def test_md_mark_meta_true(): src = Path("./test/data/doc/2408.09869v3_enriched.json") doc = DoclingDocument.load_from_json(src) ser = MarkdownDocSerializer( doc=doc, - table_serializer=CustomAnnotationTableSerializer(), params=MarkdownParams( - use_legacy_annotations=True, - include_annotations=False, + mark_meta=True, pages={1, 5}, ), ) actual = ser.serialize().text verify( - exp_file=src.parent / f"{src.stem}_p1_include_annotations_false.gt.md", + exp_file=src.parent / f"{src.stem}_p1_mark_meta_true.gt.md", actual=actual, ) -def test_md_legacy_mark_annotations_false(): +def test_md_mark_meta_false(): src = Path("./test/data/doc/2408.09869v3_enriched.json") doc = DoclingDocument.load_from_json(src) ser = MarkdownDocSerializer( doc=doc, - table_serializer=CustomAnnotationTableSerializer(), params=MarkdownParams( - use_legacy_annotations=True, - include_annotations=True, - mark_annotations=False, + mark_meta=False, pages={1, 5}, ), ) actual = ser.serialize().text verify( - exp_file=src.parent / f"{src.stem}_p1_mark_annotations_false.gt.md", + exp_file=src.parent / f"{src.stem}_p1_mark_meta_false.gt.md", actual=actual, ) -def test_md_mark_meta_true(): - src = Path("./test/data/doc/2408.09869v3_enriched.json") - doc = DoclingDocument.load_from_json(src) - - ser = MarkdownDocSerializer( - doc=doc, - table_serializer=CustomAnnotationTableSerializer(), - params=MarkdownParams( - mark_meta=True, - pages={1, 5}, - ), - ) - actual = ser.serialize().text +def test_md_legacy_annotations_mark_true(): + exp_file = Path("./test/data/doc/constructed_legacy_annot_mark_true.gt.md") + doc = _construct_doc() + with pytest.warns(DeprecationWarning): + doc.tables[0].annotations.append( + DescriptionAnnotation( + text="This is a description of table 1.", provenance="foo" + ) + ) + ser = MarkdownDocSerializer( + doc=doc, + params=MarkdownParams( + mark_annotations=True, + ), + ) + actual = ser.serialize().text verify( - exp_file=src.parent / f"{src.stem}_p1_mark_meta_true.gt.md", + exp_file=exp_file, actual=actual, ) -def test_md_legacy_mark_annotations_true(): - src = Path("./test/data/doc/2408.09869v3_enriched.json") - doc = DoclingDocument.load_from_json(src) - - ser = MarkdownDocSerializer( - doc=doc, - table_serializer=CustomAnnotationTableSerializer(), - params=MarkdownParams( - use_legacy_annotations=True, - include_annotations=True, - mark_annotations=True, - pages={1, 5}, - ), - ) - actual = ser.serialize().text +def test_md_legacy_annotations_mark_false(): + exp_file = Path("./test/data/doc/constructed_legacy_annot_mark_false.gt.md") + doc = _construct_doc() + with pytest.warns(DeprecationWarning): + doc.tables[0].annotations.append( + DescriptionAnnotation( + text="This is a description of table 1.", provenance="foo" + ) + ) + ser = MarkdownDocSerializer( + doc=doc, + params=MarkdownParams( + mark_annotations=False, + ), + ) + actual = ser.serialize().text verify( - exp_file=src.parent - / f"{src.stem}_p1_use_legacy_annotations_true_mark_annotations_true.gt.md", + exp_file=exp_file, actual=actual, )