mirrors/docling
1
0
Fork 0
mirror of https://github.com/DS4SD/docling.git synced 2025-07-27 04:24:45 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge remote-tracking branch 'origin/main' into feat-figure-in-markdown

Browse Source
This commit is contained in:
Michele Dolfi 2024-09-24 15:41:22 +02:00
commit b1a3a7a56c
6 changed files with 120 additions and 83 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
CHANGELOG.md
View File

@ -1,3 +1,23 @@
## [v1.14.0](https://github.com/DS4SD/docling/releases/tag/v1.14.0) - 2024-09-24
### Feature
* Add URL support to CLI ([#99](https://github.com/DS4SD/docling/issues/99)) ([`3c46e42`](https://github.com/DS4SD/docling/commit/3c46e4266cf1ad8d3a99aa33636d84d34222b4fe))
### Fix
* Fix OCR setting for pypdfium, minor refactor ([#102](https://github.com/DS4SD/docling/issues/102)) ([`d96b96c`](https://github.com/DS4SD/docling/commit/d96b96c8481a8ae68545a34aaf9b8d5a6637a6be))
### Documentation
* Document CLI, minor README revamp ([#100](https://github.com/DS4SD/docling/issues/100)) ([`f8f2303`](https://github.com/DS4SD/docling/commit/f8f2303348c4bbcb7903ff172746a69607e53271))
## [v1.13.1](https://github.com/DS4SD/docling/releases/tag/v1.13.1) - 2024-09-23
### Fix
* Updated the render_as_doctags with the new arguments from docling-core ([#93](https://github.com/DS4SD/docling/issues/93)) ([`4794ce4`](https://github.com/DS4SD/docling/commit/4794ce460a542a730fd5a72a7be7f94a07ed5d12))
## [v1.13.0](https://github.com/DS4SD/docling/releases/tag/v1.13.0) - 2024-09-18 ## [v1.13.0](https://github.com/DS4SD/docling/releases/tag/v1.13.0) - 2024-09-18
### Feature ### Feature

94
README.md
View File

@ -22,8 +22,9 @@ Docling bundles PDF document conversion to JSON and Markdown in an easy, self-co
* ⚡ Converts any PDF document to JSON or Markdown format, stable and lightning fast * ⚡ Converts any PDF document to JSON or Markdown format, stable and lightning fast
* 📑 Understands detailed page layout, reading order and recovers table structures * 📑 Understands detailed page layout, reading order and recovers table structures
* 📝 Extracts metadata from the document, such as title, authors, references and language * 📝 Extracts metadata from the document, such as title, authors, references and language
* 🔍 Optionally applies OCR (use with scanned PDFs) * 🔍 Includes OCR support for scanned PDFs
* 🤖 Integrates easily with LLM app / RAG frameworks like 🦙 LlamaIndex and 🦜🔗 LangChain * 🤖 Integrates easily with LLM app / RAG frameworks like 🦙 LlamaIndex and 🦜🔗 LangChain
* 💻 Provides a simple and convenient CLI
## Installation ## Installation
@ -35,31 +36,33 @@ pip install docling
> [!NOTE] > [!NOTE]
> Works on macOS and Linux environments. Windows platforms are currently not tested. > Works on macOS and Linux environments. Windows platforms are currently not tested.
<details>
<summary><b>Alternative PyTorch distributions</b></summary>
### Use alternative PyTorch distributions The Docling models depend on the [PyTorch](https://pytorch.org/) library.
Depending on your architecture, you might want to use a different distribution of `torch`.
For example, you might want support for different accelerator or for a cpu-only version.
All the different ways for installing `torch` are listed on their website <https://pytorch.org/>.
The Docling models depend on the [PyTorch](https://pytorch.org/) library. One common situation is the installation on Linux systems with cpu-only support.
Depending on your architecture, you might want to use a different distribution of `torch`. In this case, we suggest the installation of Docling with the following options
For example, you might want support for different accelerator or for a cpu-only version.
All the different ways for installing `torch` are listed on their website <https://pytorch.org/>.
One common situation is the installation on Linux systems with cpu-only support. ```bash
In this case, we suggest the installation of Docling with the following options # Example for installing on the Linux cpu-only version
pip install docling --extra-index-url https://download.pytorch.org/whl/cpu
```
</details>
```bash <details>
# Example for installing on the Linux cpu-only version <summary><b>Docling development setup</b></summary>
pip install docling --extra-index-url https://download.pytorch.org/whl/cpu
```
To develop for Docling (features, bugfixes etc.), install as follows from your local clone's root dir:
```bash
poetry install --all-extras
```
</details>
### Development setup ## Getting started
To develop for Docling, you need Python 3.10 / 3.11 / 3.12 and Poetry. You can then install from your local clone's root dir:
```bash
poetry install --all-extras
```
## Usage
### Convert a single document ### Convert a single document
@ -70,7 +73,6 @@ from docling.document_converter import DocumentConverter
source = "https://arxiv.org/pdf/2408.09869" # PDF path or URL source = "https://arxiv.org/pdf/2408.09869" # PDF path or URL
converter = DocumentConverter() converter = DocumentConverter()
result = converter.convert_single(source) result = converter.convert_single(source)
print(result.render_as_markdown()) # output: "## Docling Technical Report[...]" print(result.render_as_markdown()) # output: "## Docling Technical Report[...]"
print(result.render_as_doctags()) # output: "<document><title><page_1><loc_20>..." print(result.render_as_doctags()) # output: "<document><title><page_1><loc_20>..."
``` ```
@ -86,6 +88,51 @@ python examples/batch_convert.py
``` ```
The output of the above command will be written to `./scratch`. The output of the above command will be written to `./scratch`.
### CLI
You can also use Docling directly from your command line to convert individual files —be it local or by URL— or whole directories.
A simple example would look like this:
```console
docling https://arxiv.org/pdf/2206.01062
```
To see all available options (export formats etc.) run `docling --help`.
<details>
<summary><b>CLI reference</b></summary>
Here are the available options as of this writing (for an up-to-date listing, run `docling --help`):
```console
$ docling --help
Usage: docling [OPTIONS] source
╭─ Arguments ──────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ * input_sources source PDF files to convert. Can be local file / directory paths or URL. [default: None] [required] │
╰──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
╭─ Options ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ --json --no-json If enabled the document is exported as JSON. [default: no-json] │
│ --md --no-md If enabled the document is exported as Markdown. [default: md] │
│ --txt --no-txt If enabled the document is exported as Text. [default: no-txt] │
│ --doctags --no-doctags If enabled the document is exported as Doc Tags. [default: no-doctags] │
│ --ocr --no-ocr If enabled, the bitmap content will be processed using OCR. [default: ocr] │
│ --backend [pypdfium2|docling] The PDF backend to use. [default: docling] │
│ --output PATH Output directory where results are saved. [default: .] │
│ --version Show version information. │
│ --help Show this message and exit. │
╰──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
```
</details>
### RAG
Check out the following examples showcasing RAG using Docling with standard LLM application frameworks:
- [Basic RAG pipeline with 🦙 LlamaIndex](https://github.com/DS4SD/docling/tree/main/examples/rag_llamaindex.ipynb)
- [Basic RAG pipeline with 🦜🔗 LangChain](https://github.com/DS4SD/docling/tree/main/examples/rag_langchain.ipynb)
## Advanced features
### Adjust pipeline features ### Adjust pipeline features
The example file [custom_convert.py](https://github.com/DS4SD/docling/blob/main/examples/custom_convert.py) contains multiple ways The example file [custom_convert.py](https://github.com/DS4SD/docling/blob/main/examples/custom_convert.py) contains multiple ways
@ -144,11 +191,6 @@ results = doc_converter.convert(conv_input)
You can limit the CPU threads used by Docling by setting the environment variable `OMP_NUM_THREADS` accordingly. The default setting is using 4 CPU threads. You can limit the CPU threads used by Docling by setting the environment variable `OMP_NUM_THREADS` accordingly. The default setting is using 4 CPU threads.
### RAG
Check out the following examples showcasing RAG using Docling with standard LLM application frameworks:
- [Basic RAG pipeline with 🦙 LlamaIndex](https://github.com/DS4SD/docling/tree/main/examples/rag_llamaindex.ipynb)
- [Basic RAG pipeline with 🦜🔗 LangChain](https://github.com/DS4SD/docling/tree/main/examples/rag_langchain.ipynb)
## Technical report ## Technical report
For more details on Docling's inner workings, check out the [Docling Technical Report](https://arxiv.org/abs/2408.09869). For more details on Docling's inner workings, check out the [Docling Technical Report](https://arxiv.org/abs/2408.09869).

78
docling/cli/main.py
View File

@ -8,7 +8,7 @@ from pathlib import Path
from typing import Annotated, Iterable, List, Optional from typing import Annotated, Iterable, List, Optional
import typer import typer
from pydantic import AnyUrl from docling_core.utils.file import resolve_file_source
from docling.backend.docling_parse_backend import DoclingParseDocumentBackend from docling.backend.docling_parse_backend import DoclingParseDocumentBackend
from docling.backend.pypdfium2_backend import PyPdfiumDocumentBackend from docling.backend.pypdfium2_backend import PyPdfiumDocumentBackend
@ -109,11 +109,11 @@ def export_documents(
@app.command(no_args_is_help=True) @app.command(no_args_is_help=True)
def convert( def convert(
input_sources: Annotated[ input_sources: Annotated[
List[Path], List[str],
typer.Argument( typer.Argument(
..., ...,
metavar="source", metavar="source",
help="PDF files to convert. Directories are also accepted.", help="PDF files to convert. Can be local file / directory paths or URL.",
), ),
], ],
export_json: Annotated[ export_json: Annotated[
@ -167,7 +167,8 @@ def convert(
logging.basicConfig(level=logging.INFO) logging.basicConfig(level=logging.INFO)
input_doc_paths: List[Path] = [] input_doc_paths: List[Path] = []
for source in input_sources: for src in input_sources:
source = resolve_file_source(source=src)
if not source.exists(): if not source.exists():
err_console.print( err_console.print(
f"[red]Error: The input file {source} does not exist.[/red]" f"[red]Error: The input file {source} does not exist.[/red]"
@ -179,58 +180,25 @@ def convert(
else: else:
input_doc_paths.append(source) input_doc_paths.append(source)
########################################################################### match backend:
case Backend.PYPDFIUM2:
do_cell_matching = ocr # only do cell matching when OCR enabled
pdf_backend = PyPdfiumDocumentBackend
case Backend.DOCLING:
do_cell_matching = True
pdf_backend = DoclingParseDocumentBackend
case _:
raise RuntimeError(f"Unexpected backend type {backend}")
# The following sections contain a combination of PipelineOptions pipeline_options = PipelineOptions(
# and PDF Backends for various configurations. do_ocr=ocr,
# Uncomment one section at the time to see the differences in the output. do_table_structure=True,
)
doc_converter = None pipeline_options.table_structure_options.do_cell_matching = do_cell_matching
if backend == Backend.PYPDFIUM2 and not ocr: # PyPdfium without OCR doc_converter = DocumentConverter(
pipeline_options = PipelineOptions() pipeline_options=pipeline_options,
pipeline_options.do_ocr = False pdf_backend=pdf_backend,
pipeline_options.do_table_structure = True )
pipeline_options.table_structure_options.do_cell_matching = False
doc_converter = DocumentConverter(
pipeline_options=pipeline_options,
pdf_backend=PyPdfiumDocumentBackend,
)
elif backend == Backend.PYPDFIUM2.value and ocr: # PyPdfium with OCR
pipeline_options = PipelineOptions()
pipeline_options.do_ocr = False
pipeline_options.do_table_structure = True
pipeline_options.table_structure_options.do_cell_matching = True
doc_converter = DocumentConverter(
pipeline_options=pipeline_options,
pdf_backend=PyPdfiumDocumentBackend,
)
elif backend == Backend.DOCLING.value and not ocr: # Docling Parse without OCR
pipeline_options = PipelineOptions()
pipeline_options.do_ocr = False
pipeline_options.do_table_structure = True
pipeline_options.table_structure_options.do_cell_matching = True
doc_converter = DocumentConverter(
pipeline_options=pipeline_options,
pdf_backend=DoclingParseDocumentBackend,
)
elif backend == Backend.DOCLING.value and ocr: # Docling Parse with OCR
pipeline_options = PipelineOptions()
pipeline_options.do_ocr = True
pipeline_options.do_table_structure = True
pipeline_options.table_structure_options.do_cell_matching = True
doc_converter = DocumentConverter(
pipeline_options=pipeline_options,
pdf_backend=DoclingParseDocumentBackend,
)
###########################################################################
# Define input files # Define input files
input = DocumentConversionInput.from_paths(input_doc_paths) input = DocumentConversionInput.from_paths(input_doc_paths)

7
examples/rag_langchain.ipynb
View File

@ -1,5 +1,12 @@
{ {
"cells": [ "cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# RAG with Docling and 🦜🔗 LangChain"
]
},
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 1, "execution_count": 1,

2
examples/rag_llamaindex.ipynb
View File

@ -4,7 +4,7 @@
"cell_type": "markdown", "cell_type": "markdown",
"metadata": {}, "metadata": {},
"source": [ "source": [
"# Quackling — Basic Pipeline" "# RAG with Docling and 🦙 LlamaIndex"
] ]
}, },
{ {

2
pyproject.toml
View File

@ -1,6 +1,6 @@
[tool.poetry] [tool.poetry]
name = "docling" name = "docling"
version = "1.13.0" # DO NOT EDIT, updated automatically version = "1.14.0" # DO NOT EDIT, updated automatically
description = "Docling PDF conversion package" description = "Docling PDF conversion package"
authors = ["Christoph Auer <cau@zurich.ibm.com>", "Michele Dolfi <dol@zurich.ibm.com>", "Maxim Lysak <mly@zurich.ibm.com>", "Nikos Livathinos <nli@zurich.ibm.com>", "Ahmed Nassar <ahn@zurich.ibm.com>", "Peter Staar <taa@zurich.ibm.com>"] authors = ["Christoph Auer <cau@zurich.ibm.com>", "Michele Dolfi <dol@zurich.ibm.com>", "Maxim Lysak <mly@zurich.ibm.com>", "Nikos Livathinos <nli@zurich.ibm.com>", "Ahmed Nassar <ahn@zurich.ibm.com>", "Peter Staar <taa@zurich.ibm.com>"]
license = "MIT" license = "MIT"