v0.12.0

What's Changed

• Add otel observability by @avvertix in #24
• Drivers command now check for installed dependencies by @avvertix in #28
• Pdf elaboration commands by @avvertix in #27
• Parallel execution option by @avvertix in #29
• Llamaparse job details by @avvertix in #35
• Manipulate PDF attachments by @avvertix in #36
• Batch mode by @avvertix in #42
• Improve Llama Parse configuration by @avvertix in #43
• Bring Pdf manipulation service to core by @avvertix in #44
• Empty pages handling by @avvertix in #45
• Document structure roles by @avvertix in #46
• Improve rate limit errors by @avvertix in #47
• Guidance for using with coding agents by @avvertix in #49
• Improve unexpected error checking by @avvertix in #52
• CI: Execute PDFAct tests by @avvertix in #48
• Introduce Table and Image blocks by @avvertix in #53
• Improve batches by @avvertix in #54
• Improve responsiveness of parse command with multiple files and drivers by @avvertix in #56
• Align markdown command to parse command by @avvertix in #58
• Check if bounding box available before read values by @avvertix in #63
• Handle empty parse response by @avvertix in #64
• Split a page range to a single document by @avvertix in #66
• Contentmd by @avvertix in #57
• Remove unused preview command by @avvertix in #67

Full Changelog: v0.11.0...v0.12.0

v0.11.0

What's Changed

• Collect pymupdf warnings and make them available on the document model by @avvertix in #15
• Update doc for cli usage by @avvertix in #16
• Use Pydantic model for configuration values in driver initialization by @avvertix in #17
• Refactor commands in own files by @avvertix in #18
• Add Landing.ai driver by @avvertix in #19
• Add coding agents guidelines by @avvertix in #20
• Tracing cleanup by @avvertix in #21
• Add styled console output by @avvertix in #22
• Preview and Parse command improvements by @avvertix in #23

Full Changelog: v0.10.0...v0.11.0

v0.10.0

Parxy 0.10 is a complete rewrite from the ground up. It focuses on library and CLI usage, bringing a totally rewritten unified document model.

We decided to jump from version 0.6 to 0.10 to give a strong cut between the two paradigms.

Parxy 0.10 supports:

• PyMuPDF
• PDFAct
• LlamaParse
• Unstruct LLMWhisperer
• Unstructured.io locally installed using the open source distribution

What's Changed

• Test and release pipeline by @avvertix in #13
• Improve logging isolation and exception tracing by @avvertix in #14

Full Changelog: https://github.com/OneOffTech/parxy/commits/v0.10.0

v0.6.0

What's Changed

• Add Sections and improve Pymupdf handling by @andreaponti5 in #10
• Add bounding box support to unstructured parser by @AnnaMarika01 in #11
• Update LlamaParser by @andreaponti5 in #12

New Contributors

• @AnnaMarika01 made their first contribution in #11

Full Changelog: v0.5.0...v0.6.0

v0.5.0

What's Changed

• Add llamaparse and unstructured by @andreaponti5 in #9

Full Changelog: v0.4.0...v0.5.0

v0.4.0

What's Changed

• Removed the duplicate document download in #8
• Added checks for authentication required to download file in #8
• PDFAct processing errors are now forwarded in #8
• Fix errors in determine the heading level in #8

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

v0.3.1

• Rename from PDF Text Extract to Parxy

Full Changelog: v0.3.0...v0.3.1

v0.3.0

What's Changed

• Add heading level by @avvertix in #7

Full Changelog: v0.2.2...v0.3.0

v0.2.2

What's Changed

• Fix aggregate paragraphs by @AnnaMarika01 in #6

Full Changelog: v0.2.1...v0.2.2

v0.2.1

What's Changed

• Feature formatter by @avvertix in #3
• Fix optional env var by @avvertix in #4
• Prevent heading process if document is empty by @avvertix in #5

Full Changelog: v0.2.0...v0.2.1