You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
PptxConverterWithOCR uses "\\n" instead of "\n" in multiple output paths.
As a result, converted PPTX files contain literal backslash-n sequences rather
than Markdown line breaks.
The OCR PPTX tests currently encode this malformed output as expected behavior.
Additionally, the optional LLM caption path imports ._llm_caption, but the
module exists in markitdown.converters, not in markitdown_ocr.
Thank you for reporting both the newline problem and the caption-import issue.
#2506, included in v0.1.8, replaced the duplicated PPTX OCR conversion code with the core presentation pipeline. This removes the literal backslash-n formatting paths and uses the core caption implementation. The relevant formatting and caption regressions pass against v0.1.8.
Please upgrade both markitdown to 0.1.8 and markitdown-ocr to 0.1.1. Closing as fixed. Please reopen this issue if the problem still occurs with those versions, and include a reproduction so we can investigate.
Activity
afourney commentedon Sep 30, 2026
Thank you for reporting both the newline problem and the caption-import issue.
#2506, included in v0.1.8, replaced the duplicated PPTX OCR conversion code with the core presentation pipeline. This removes the literal backslash-n formatting paths and uses the core caption implementation. The relevant formatting and caption regressions pass against v0.1.8.
Please upgrade both
markitdownto0.1.8andmarkitdown-ocrto0.1.1. Closing as fixed. Please reopen this issue if the problem still occurs with those versions, and include a reproduction so we can investigate.