com.blockether.svar.internal.rlm.internal.pageindex.pdf — com.blockether/svar 0.1.2

PDF to images conversion and metadata extraction using Apache PDFBox.

Provides:

• pdf->images - Convert PDF file to vector of BufferedImage objects
• page-count - Get total page count of a PDF file
• pdf-metadata - Extract PDF metadata (author, title, dates, etc.)
• detect-text-rotation - Detect content rotation per page using text position heuristics

Uses PDFBox for reliable PDF rendering at configurable DPI. Handles error cases: encrypted PDFs, corrupted files, file not found.

detect-text-rotation clj

(detect-text-rotation pdf-path)

(detect-text-rotation pdf-path {:keys [page-set]})

Detects content rotation for each page of a PDF using text position heuristics.

Analyzes the direction of text characters on each page via PDFBox's TextPosition. If the majority of characters flow in a non-standard direction, the page content is rotated (e.g., landscape table on a portrait page).

TextPosition directions:

• 0°: Normal left-to-right text → no correction needed
• 90°: Text flows bottom-to-top → correct by rotating image 270° CW
• 180°: Upside-down text → correct by rotating image 180°
• 270°: Text flows top-to-bottom → correct by rotating image 90° CW

Params: pdf-path - String. Path to the PDF file. opts - Optional map with: :page-set - Set of 0-indexed page numbers to analyze, or nil for all pages.

Returns: Vector of integers, one per selected page. Each value is the clockwise rotation in degrees (0, 90, 180, or 270) needed to correct the rendered image.

Example: [0 0 90] means pages 0-1 are normal, page 2 needs 90° CW rotation.

page-count clj

(page-count pdf-path)

Returns the number of pages in a PDF file.

Params: pdf-path - String. Path to the PDF file.

Returns: Integer. Number of pages.

Throws: Same exceptions as pdf->images.

pdf->images clj

(pdf->images pdf-path)

(pdf->images pdf-path {:keys [dpi page-set] :or {dpi DEFAULT_DPI}})

Converts a PDF file to a vector of BufferedImage objects.

Params: pdf-path - String. Path to the PDF file. opts - Optional map with: :dpi - Integer. Rendering DPI (default 150). :page-set - Set of 0-indexed page numbers to render, or nil for all pages.

Returns: Vector of BufferedImage objects, one per selected page.

Throws: ex-info for not-found, corrupted, or encrypted PDFs.

pdf-metadata clj

(pdf-metadata pdf-path)

Extracts metadata from a PDF file.

Params: pdf-path - String. Path to the PDF file.

Returns: Map with: :author - String or nil. Document author. :title - String or nil. Document title. :subject - String or nil. Document subject. :creator - String or nil. Creating application. :producer - String or nil. PDF producer. :created-at - Instant or nil. Creation date. :updated-at - Instant or nil. Modification date. :keywords - String or nil. Document keywords.

Throws: Same exceptions as pdf->images.