Repository Issue Activity (beta)

#272 `search_items` returns one spurious match per item for an empty phrase

#271 `compress_wide_spaces` corrupts non-ASCII text (`byte as char`) → mojibake in sparse blocks

#270 `batch-parse` discards the output directory (writes to filesystem root) when the input dir has no trailing slash

#269 `--target-pages` eagerly expands large ranges → unbounded memory / OOM

#255 [Feature] Support export to HTML

#261 [Feature] Add e2e tests for Node.js package

#265 v2: every other text line is dropped on scanned pages with tight line spacing (OCR returns them, output omits them)

#256 [Bug] 提供 darwin-x64平台的原生依赖

#212 [Bug] --password is passed to LibreOffice as --infilter, breaking encrypted Office document conversion

#239 [Parsing] Incorrect order for text extraction (columns)

#253 [Bug] Tesseract errors

#264 [Bug] RECORD validation failure for liteparse-2.0.4

#263 [Feature] how about table parse comparing with camelot?

#250 Page-1 bounding boxes shifted ~one page-height off-page when a rotated text column (arXiv watermark) is present

#248 [Feature] Extract table of contents (outline) from PDF documents

#235 [Feature] Expose OCR text-length threshold as a config option

#243 TextItem.x and .y returned in grid-projection coordinates, not PDF points — contradicts docs

#238 Panic in projection.rs:80 on scanned PDFs with no native text

#230 [Bug] Office document conversion silently fails on Windows

#58 Feature request: Cloudflare Workers / edge runtime support

#196 [Feature] seperate python sdk from the js sdk

#214 [Feature] Add format conversion to screenshot() for non-PDF inputs (v1 parity)

#218 [Bug] PDF is loaded twice on every OCR-enabled parse (extract + ocr_merge each call load_document)

#210 [Bug] Buffer parse leaves staging temp directories on disk when conversion or parsing fails

#216 [Bug] lit extract and lit image-bounds cannot open password-protected PDFs (no --password flag)