Skip to content

feat(ocr): Add parallel batch OCR processing via ThreadPoolExecutor#2197

Open
echojfree wants to merge 6 commits into
microsoft:mainfrom
echojfree:feat/ocr-parallel-batch-processing
Open

feat(ocr): Add parallel batch OCR processing via ThreadPoolExecutor#2197
echojfree wants to merge 6 commits into
microsoft:mainfrom
echojfree:feat/ocr-parallel-batch-processing

Conversation

@echojfree

Copy link
Copy Markdown

Replace serial image-by-image OCR with batched parallel processing across all four OCR-enhanced converters (PDF, DOCX, PPTX, XLSX).

Changes

OCR Service (_ocr_service.py)

  • Add max_workers parameter (default 5) to LLMVisionOCRService
  • Add extract_text_batch() method using ThreadPoolExecutor for parallel I/O-bound LLM API calls
  • Individual image failures captured in OCRResult.error field without interrupting the batch

PDF Converter (_pdf_converter_with_ocr.py)

  • Single-pass PDF processing: collect all images from all pages first, then batch OCR, then interleave results with text lines by Y position for correct document flow
  • _ocr_full_pages() now collects page renders first, then batch OCRs them (both pdfplumber and PyMuPDF fallback paths)
  • Fix scanned-PDF fallback detection: regex-strip page-number headers before checking for real content
  • Remove _extract_page_images() in favor of in-pass collection
  • Extract _extract_text_lines_from_page() helper

DOCX Converter (_docx_converter_with_ocr.py)

  • _extract_and_ocr_images(): collect all image rIds and streams first, then call extract_text_batch() once

PPTX Converter (_pptx_converter_with_ocr.py)

  • Two-phase parallel strategy: Round 1: all images -> LLM caption in parallel Round 2: failed images -> OCR in parallel via extract_text_batch
  • Pre-scan phase collects all image shapes across all slides (recursive _collect_image_shapes handles groups)
  • Rendering phase consumes pre-computed results via cursor, maintaining identical output structure

XLSX Converter (_xlsx_converter_with_ocr.py)

  • Cross-sheet collection: gather all images from all sheets before batch OCR, then distribute results by sheet
  • Rename _extract_and_ocr_sheet_images -> _extract_sheet_images (collection only; OCR moved to caller)

Plugin Registration (_plugin.py)

  • Accept max_workers kwarg (default 5), forward to OCR service

Tests (4 files)

  • Add extract_text_batch() to all MockOCRService classes
  • Update test_pdf_multipage expectation: batch-parallel path correctly extracts text+images in a single pass instead of artificially falling back to full-page OCR

Performance

Real-world test on 5-page scanned contract PDF (EMC):

  • Serial (max_workers=1): 74.4s
  • Parallel (max_workers=5): 24.1s
  • Speedup: 3.1x

All 36 existing tests pass.

Replace serial image-by-image OCR with batched parallel processing
across all four OCR-enhanced converters (PDF, DOCX, PPTX, XLSX).

## Changes

### OCR Service (_ocr_service.py)
- Add max_workers parameter (default 5) to LLMVisionOCRService
- Add extract_text_batch() method using ThreadPoolExecutor for
  parallel I/O-bound LLM API calls
- Individual image failures captured in OCRResult.error field
  without interrupting the batch

### PDF Converter (_pdf_converter_with_ocr.py)
- Single-pass PDF processing: collect all images from all pages
  first, then batch OCR, then interleave results with text lines
  by Y position for correct document flow
- _ocr_full_pages() now collects page renders first, then batch
  OCRs them (both pdfplumber and PyMuPDF fallback paths)
- Fix scanned-PDF fallback detection: regex-strip page-number
  headers before checking for real content
- Remove _extract_page_images() in favor of in-pass collection
- Extract _extract_text_lines_from_page() helper

### DOCX Converter (_docx_converter_with_ocr.py)
- _extract_and_ocr_images(): collect all image rIds and streams
  first, then call extract_text_batch() once

### PPTX Converter (_pptx_converter_with_ocr.py)
- Two-phase parallel strategy:
  Round 1: all images -> LLM caption in parallel
  Round 2: failed images -> OCR in parallel via extract_text_batch
- Pre-scan phase collects all image shapes across all slides
  (recursive _collect_image_shapes handles groups)
- Rendering phase consumes pre-computed results via cursor,
  maintaining identical output structure

### XLSX Converter (_xlsx_converter_with_ocr.py)
- Cross-sheet collection: gather all images from all sheets
  before batch OCR, then distribute results by sheet
- Rename _extract_and_ocr_sheet_images -> _extract_sheet_images
  (collection only; OCR moved to caller)

### Plugin Registration (_plugin.py)
- Accept max_workers kwarg (default 5), forward to OCR service

### Tests (4 files)
- Add extract_text_batch() to all MockOCRService classes
- Update test_pdf_multipage expectation: batch-parallel path
  correctly extracts text+images in a single pass instead of
  artificially falling back to full-page OCR

## Performance

Real-world test on 5-page scanned contract PDF (EMC):
- Serial (max_workers=1): 74.4s
- Parallel (max_workers=5): 24.1s
- Speedup: 3.1x

All 36 existing tests pass.

Co-Authored-By: Claude <noreply@anthropic.com>
@microsoft-github-policy-service

Copy link
Copy Markdown

@echojfree please read the following Contributor License Agreement(CLA). If you agree with the CLA, please reply with the following information.

@microsoft-github-policy-service agree [company="{your company}"]

Options:

  • (default - no company specified) I have sole ownership of intellectual property rights to my Submissions and I am not making Submissions in the course of work for my employer.
@microsoft-github-policy-service agree
  • (when company given) I am making Submissions in the course of work for my employer (or my employer has intellectual property rights in my Submissions by contract or applicable law). I have permission from my employer to make Submissions and enter into this Agreement on behalf of my employer. By signing below, the defined term “You” includes me and my employer.
@microsoft-github-policy-service agree company="Microsoft"
Contributor License Agreement

Contribution License Agreement

This Contribution License Agreement (“Agreement”) is agreed to by the party signing below (“You”),
and conveys certain license rights to Microsoft Corporation and its affiliates (“Microsoft”) for Your
contributions to Microsoft open source projects. This Agreement is effective as of the latest signature
date below.

  1. Definitions.
    “Code” means the computer software code, whether in human-readable or machine-executable form,
    that is delivered by You to Microsoft under this Agreement.
    “Project” means any of the projects owned or managed by Microsoft and offered under a license
    approved by the Open Source Initiative (www.opensource.org).
    “Submit” is the act of uploading, submitting, transmitting, or distributing code or other content to any
    Project, including but not limited to communication on electronic mailing lists, source code control
    systems, and issue tracking systems that are managed by, or on behalf of, the Project for the purpose of
    discussing and improving that Project, but excluding communication that is conspicuously marked or
    otherwise designated in writing by You as “Not a Submission.”
    “Submission” means the Code and any other copyrightable material Submitted by You, including any
    associated comments and documentation.
  2. Your Submission. You must agree to the terms of this Agreement before making a Submission to any
    Project. This Agreement covers any and all Submissions that You, now or in the future (except as
    described in Section 4 below), Submit to any Project.
  3. Originality of Work. You represent that each of Your Submissions is entirely Your original work.
    Should You wish to Submit materials that are not Your original work, You may Submit them separately
    to the Project if You (a) retain all copyright and license information that was in the materials as You
    received them, (b) in the description accompanying Your Submission, include the phrase “Submission
    containing materials of a third party:” followed by the names of the third party and any licenses or other
    restrictions of which You are aware, and (c) follow any other instructions in the Project’s written
    guidelines concerning Submissions.
  4. Your Employer. References to “employer” in this Agreement include Your employer or anyone else
    for whom You are acting in making Your Submission, e.g. as a contractor, vendor, or agent. If Your
    Submission is made in the course of Your work for an employer or Your employer has intellectual
    property rights in Your Submission by contract or applicable law, You must secure permission from Your
    employer to make the Submission before signing this Agreement. In that case, the term “You” in this
    Agreement will refer to You and the employer collectively. If You change employers in the future and
    desire to Submit additional Submissions for the new employer, then You agree to sign a new Agreement
    and secure permission from the new employer before Submitting those Submissions.
  5. Licenses.
  • Copyright License. You grant Microsoft, and those who receive the Submission directly or
    indirectly from Microsoft, a perpetual, worldwide, non-exclusive, royalty-free, irrevocable license in the
    Submission to reproduce, prepare derivative works of, publicly display, publicly perform, and distribute
    the Submission and such derivative works, and to sublicense any or all of the foregoing rights to third
    parties.
  • Patent License. You grant Microsoft, and those who receive the Submission directly or
    indirectly from Microsoft, a perpetual, worldwide, non-exclusive, royalty-free, irrevocable license under
    Your patent claims that are necessarily infringed by the Submission or the combination of the
    Submission with the Project to which it was Submitted to make, have made, use, offer to sell, sell and
    import or otherwise dispose of the Submission alone or with the Project.
  • Other Rights Reserved. Each party reserves all rights not expressly granted in this Agreement.
    No additional licenses or rights whatsoever (including, without limitation, any implied licenses) are
    granted by implication, exhaustion, estoppel or otherwise.
  1. Representations and Warranties. You represent that You are legally entitled to grant the above
    licenses. You represent that each of Your Submissions is entirely Your original work (except as You may
    have disclosed under Section 3). You represent that You have secured permission from Your employer to
    make the Submission in cases where Your Submission is made in the course of Your work for Your
    employer or Your employer has intellectual property rights in Your Submission by contract or applicable
    law. If You are signing this Agreement on behalf of Your employer, You represent and warrant that You
    have the necessary authority to bind the listed employer to the obligations contained in this Agreement.
    You are not expected to provide support for Your Submission, unless You choose to do so. UNLESS
    REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING, AND EXCEPT FOR THE WARRANTIES
    EXPRESSLY STATED IN SECTIONS 3, 4, AND 6, THE SUBMISSION PROVIDED UNDER THIS AGREEMENT IS
    PROVIDED WITHOUT WARRANTY OF ANY KIND, INCLUDING, BUT NOT LIMITED TO, ANY WARRANTY OF
    NONINFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE.
  2. Notice to Microsoft. You agree to notify Microsoft in writing of any facts or circumstances of which
    You later become aware that would make Your representations in this Agreement inaccurate in any
    respect.
  3. Information about Submissions. You agree that contributions to Projects and information about
    contributions may be maintained indefinitely and disclosed publicly, including Your name and other
    information that You submit with Your Submission.
  4. Governing Law/Jurisdiction. This Agreement is governed by the laws of the State of Washington, and
    the parties consent to exclusive jurisdiction and venue in the federal courts sitting in King County,
    Washington, unless no federal subject matter jurisdiction exists, in which case the parties consent to
    exclusive jurisdiction and venue in the Superior Court of King County, Washington. The parties waive all
    defenses of lack of personal jurisdiction and forum non-conveniens.
  5. Entire Agreement/Assignment. This Agreement is the entire agreement between the parties, and
    supersedes any and all prior agreements, understandings or communications, written or oral, between
    the parties relating to the subject matter hereof. This Agreement may be assigned by Microsoft.

echojfree and others added 5 commits July 7, 2026 16:02
The relative import 'from ._llm_caption' resolves to markitdown_ocr._llm_caption
which doesn't exist. The correct absolute import is from
markitdown.converters._llm_caption.

This was a pre-existing bug masked by try/except in the original code path.

Co-Authored-By: Claude <noreply@anthropic.com>
- Default ocr_dpi reduced from 300 to 150 (4x fewer pixels = faster
  rendering + smaller base64 payloads + faster LLM processing)
- Page rendering in _ocr_full_pages now uses ThreadPoolExecutor
  (both pdfplumber and PyMuPDF fallback paths)
- Configurable via ocr_dpi= kwarg (e.g. MarkItDown(..., ocr_dpi=200))

Co-Authored-By: Claude <noreply@anthropic.com>
Full-page OCR DPI stays at 300 by default. Users who want faster
processing can still pass ocr_dpi=150.

Co-Authored-By: Claude <noreply@anthropic.com>
Replace batch-then-OCR with a streaming pipeline: each page is
submitted for OCR the moment its render finishes, instead of
waiting for ALL pages to render before starting ANY OCR.

Rendering and OCR now run concurrently, so total time is
max(render_all, ocr_all) rather than render_all + ocr_all.

Single-page documents skip the thread pool entirely for zero
overhead on small inputs.

Co-Authored-By: Claude <noreply@anthropic.com>
## PPTX: fix literal backslash-n in output (HIGH)
The PPTX converter emitted '\n' (literal backslash-n) instead of
real newlines, producing malformed markdown. The upstream core
_pptx_converter.py uses real '\n' — this aligns the OCR variant.
Output is now clean markdown (leading newlines correctly stripped).
Test expectations updated to match verified-correct output.

## XLSX: single-pass sheet read (perf)
_convert_with_ocr re-parsed the entire workbook via pd.read_excel
once per sheet (O(N) full parses). Now reads all sheets in one call
with sheet_name=None, matching the standard-path approach.

All 36 tests pass.

Co-Authored-By: Claude <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant