Supprimer des Pages PDF

After you upload the PDF, the tool renders a visual thumbnail grid showing every page at a readable preview size with the page number displayed below each one. Clicking any thumbnail toggles it into the marked-for-removal state, indicated by a red overlay and a deletion icon; clicking the same thumbnail again deselects it. For contiguous ranges, Shift-clicking the first and last page in the range selects all pages in between in a single action. On desktop browsers, Ctrl-click (Windows and Linux) or Cmd-click (macOS) lets you add or remove individual non-contiguous pages from the selection without disturbing pages you have already marked. A persistent counter at the top of the grid shows how many pages are currently selected for deletion and how many will be retained in the output, giving you a constant sanity check before you commit. A text-field range input is also available for keyboard-driven workflows; entering a value such as 1-3, 7, 12-15 marks pages 1, 2, 3, 7, 12, 13, 14, and 15 simultaneously, which is faster than clicking individual thumbnails in long documents. The export button remains disabled until at least one page is marked for removal and at least one page remains unselected, preventing the accidental creation of an empty PDF. All selection state is held in browser memory and resets when you upload a new file or reload the page. No page is actually removed from any file until you click the export button, so you can revise your selection freely without risk. The original file you uploaded is never modified; the tool writes a new PDF containing only the kept pages.

The tool actively cleans the bookmark tree to reflect the new page structure of the output document. Any bookmark — also called a named destination or outline entry — whose target page number falls within the set of deleted pages is removed from the output's document outline entirely. Bookmarks that target retained pages are preserved with their labels and nesting hierarchy intact, and their destination page numbers are remapped to the correct positions in the renumbered output. For example, if a bookmark originally pointed to page 15 and you delete pages 1 through 5, the bookmark will point to page 10 in the output, which corresponds to the same content. Internal hyperlinks within the document body follow the same logic: links whose anchor target was on a deleted page become inactive dead links in the output, while links pointing to retained pages are remapped to updated page numbers. External hyperlinks pointing to URLs are untouched. Embedded form fields, annotations, digital signatures, and multimedia objects on retained pages are preserved without modification because the tool does a structural page extraction rather than a re-render; the kept page content streams are copied verbatim into the output file. Document-level metadata — title, author, creator, creation date, and subject — is carried forward to the output PDF unchanged. The output is always a new separate file; your original PDF is not written to at any stage. If the result does not match your expectations, you can close the download, reload the original, and adjust your selection without losing anything permanently.

There is no hard page-count limit enforced by the tool; the practical ceiling is determined by your device's available browser memory rather than an artificial restriction. On a desktop browser running on a modern laptop with 8 GB or more of RAM, the tool handles PDFs up to approximately 500 MB and several thousand pages for text-heavy documents reliably. Image-heavy PDFs — scanned archives, photo books, high-resolution technical drawings — consume significantly more memory per page and may reach memory pressure at a few hundred pages on the same hardware. Mobile browsers operate under tighter memory budgets set by the operating system; a mid-range phone can typically handle PDFs up to 100 MB before the browser tab risks being terminated. Processing time scales with the number of retained pages rather than the total page count of the source, because the engine copies only the kept content streams into the output structure. Deleting 800 pages from a 1,000-page document and keeping 200 is therefore faster than deleting 50 pages from a 250-page document where 200 are retained. As a rough benchmark, a 1,000-page text PDF with 200 pages removed completes in approximately 10 to 20 seconds on a modern desktop browser. For archival PDFs measured in gigabytes or tens of thousands of pages, a command-line tool such as pdftk or qpdf operating outside browser memory constraints is the more appropriate solution. All processing runs in your browser — no file is uploaded to any server at any point.

The tool is designed with a non-destructive workflow that makes accidental permanent deletion essentially impossible if you follow the standard procedure. The most important design principle is that the tool never modifies the file you uploaded; it always produces a new output file while leaving the source untouched in your browser session. Before you click the export button, you can freely click, shift-click, or Ctrl-click any selected page to deselect it, or use the Reset All button to clear the entire selection and start fresh from a clean state. The export button only becomes active once you have made a valid selection, and clicking it produces the output but does not automatically close or replace your original file in the upload area; you can immediately re-examine the retained pages in the output summary and, if anything looks wrong, click New File to reload your original and try again. Once you have downloaded the output, however, the undo path is simply to discard the downloaded file and re-upload your original. This is why keeping a backup copy of any important PDF before beginning deletion work is strongly recommended — store the original in a separate folder, cloud storage, or local drive before editing. For legal contracts, financial statements, medical records, and any other document where a permanent loss would be consequential, treat the backup step as mandatory rather than optional. The browser clears all in-memory state when you close the tab, so nothing persists beyond your session.