PDF to Markdown

PDF to Markdown extracts text and writes one Markdown heading per page. A page with no extractable text gets an empty-page note. It does not rebuild the layout.

A heading, then the lines

The file starts with a level-one heading, PDF export. Each page is a level-two heading, Page, and the page number. Text on that page is split on blank lines into paragraphs. Inside a paragraph, single line breaks become a Markdown line break of two spaces. The button reads Run PDF to Markdown. The download name ends in export.md. The content type is Markdown. A page whose extracted text is empty becomes the heading plus the line _(No extractable text on this page)_. That underscore form is the empty marker in the file, not a warning banner in the form.

What the Markdown cannot keep

Columns are not detected. A two-column page can come out as interleaved lines. Images are not embedded. Fonts, colors, and positions are dropped. Form fields are not a list of values. A scan that is only a picture of text usually hits the empty-page line, because this path does not run optical character recognition. There is no language menu and no page range. Every page is included, including the empty ones, so the heading count matches the page count. Bookmarks are not a table of contents in the Markdown. Annotations are not exported as quotes. A footer number printed on the page is ordinary text if the extractor can see it, and it is not used as the heading index.

How to read a bad extract

If a heading is followed by the empty-page line, that page had no text layer. If the words are present but the order looks shuffled, the PDF stored the text in drawing order, not reading order. This page does not sort lines into columns. Extract PDF Text is the plain-text cousin. PDF to HTML is a different conversion. Neither one restores a table as Markdown pipes. If you need a table, copy it from a spreadsheet instead of expecting pipes here. A page number in the heading is the PDF page index, starting at 1, not a number printed in the footer of the source. The Markdown file is text, so you can open it in any editor that shows a .md file without a preview pane.

The PDF is uploaded

The PDF is uploaded for the extract, and the temporary file is removed after the download. The cap is 25 MB. A password-protected PDF is not unlocked by this page. Keep the PDF if the Markdown dropped a figure you still need. A three-page file produces three level-two headings even when the middle page is the empty-page line. The heading numbers follow page order, starting at 1. Line breaks that were single newlines inside a block become a Markdown hard break, two spaces before the newline, so a poem-like extract keeps its line ends. Blank-line gaps become paragraph gaps.