v0.5.6 (2024-05-01)

What's Changed

• Llama Index Integration by @Filimoa in #41

Full Changelog: v0.5.5...v0.5.6

v0.5.5

What's Changed

• [Memory Leak Fix] Create Fitz Pdf From Bytestream by @Filimoa in #39

Full Changelog: v0.5.4...v0.5.5

v0.5.4

What's Changed

• #29 [minor tweak to mashihua's branch] by @Filimoa in #32
• Fix bug with parse.py by @mashihua in #29

New Contributors

• @mashihua made their first contribution in #29

Full Changelog: v0.5.3...v0.5.4

v0.5.3 (2024-04-21)

Minor bug fixes

What's Changed

• Update pymupdf.md by @ada-lovecraft in #20
• update the cookbooks link by @brianjking in #24
• fix: Fix sequence item 2: expected str instance, NoneType found exception when table output is set to markdown. by @ic-xu in #27

New Contributors

• @ada-lovecraft made their first contribution in #20
• @brianjking made their first contribution in #24
• @ic-xu made their first contribution in #27

Full Changelog: v0.5.2...v0.5.3

v0.5.2 (2024-04-11)

Features

• Better version display
• Fixed pytorch device bug. Thanks @jinmang2
• Add global config to set pytorch device

v0.5.1 (2024-04-05)

Bug Fixes

• Fixed type hinting bug for python < 3.10

Full Changelog: v0.5.0...v0.5.1

v0.5.0 (2024-04-07)

0.5.0 (2024-04-01)

What's Changed

• SemanticProcessing! This is the recommended processing pipeline.
• Add optional annotations to the pdf draw functions
• Fixed reading order bug

Breaking Changes

1. Renaming

• Node.aggregate_position renamed to Node.reading_order.
• RemoveStubs to RemoveNodesBelowNTokens

2. Refactored processing pipelines to use a class to promote ease of reuse

Previously

from openparse import ProcessingStep, default_pipeline, Node
from typing import List


class CustomCombineTables(ProcessingStep):
    def process(self, nodes: List[Node]) -> List[Node]:
        return nodes


# copy the default pipeline (or create a new one)
custom_pipeline = default_pipeline.copy()
custom_pipeline.append(CustomCombineTables())

parser = openparse.DocumentParser(
    table_args={"parsing_algorithm": "pymupdf"}, processing_pipeline=custom_pipeline
)
custom_10k = parser.parse(meta10k_path)

Now becomes

from openparse import processing, Node
from typing import List


class CustomCombineTables(processing.ProcessingStep):
    def process(self, nodes: List[Node]) -> List[Node]:
        return nodes


# copy the default pipeline (or create a new one)
custom_pipeline = processing.BasicIngestionPipeline()
custom_pipeline.append_transform(CustomCombineTables())

parser = openparse.DocumentParser(
    table_args={"parsing_algorithm": "pymupdf"}, processing_pipeline=custom_pipeline
)
custom_10k = parser.parse(meta10k_path)

3. openai and numpy as now required dependencies, will likely split this out in the future.

Full Changelog: v0.4.1...v0.5.0

v0.4.1 (2024-04-05)

What's Changed

• Better error messages for missing weights
• Type hinting bug with python 3.8 fixed

Full Changelog: v0.4.0...v0.4.1

0.4.0 (2024-04-04)

What's Changed

• ✨ Unitable support for table content extraction!
• 🐛 Fixed bug with table transformers failing on multiple pages.
• ✨ Improved table docs

What's Changed

• Unitable by @Filimoa in #6

New Contributors

• @Filimoa made their first contribution in #6

Full Changelog: v0.3.1...v0.4.0

0.3.1 (2024-04-01)

What's Changed

• Fixed #4