PDF editor
Add text, images, shapes and drawings directly on the page.
Your files stay in this browser. Nothing is uploaded.
About Editor
The editor opens your PDF as pages you can work on directly, so you can edit PDF text in your browser without sending the file anywhere. Click a line of existing text to retype it, add new text wherever you need it, drop in an image, draw a shape or a freehand mark, and cover anything you do not want showing with a Whiteout box.
Page work lives here too. Delete a page, rotate it, or insert a blank one after it, then click Apply changes and download the result. The file is opened and changed by code running in this tab, so it is never uploaded. Your original is left as it was, and the edited copy is a new file with "-edited" added to its name.
How it works
- Choose a PDF, or drop one anywhere on this page. To start something new instead, choose "Start with a blank page" for a single A4 page.
- Text is selected when the document opens. Click a line of existing text to edit it in place, or click empty space to start a new box. The toolbar above the box sets bold, italic, size, font and colour, and can add a link, move, duplicate or delete the box.
- Use Images for pictures and stamps, Whiteout for a box that hides part of the page, Annotate for highlights, underlines, strike-outs and freehand marks, and Shapes for lines, arrows, rectangles and ellipses. Drag on the page to place a box, mark or shape.
- Use the row above each page to rotate it, delete it, or insert a blank page after it. The Pages button opens thumbnails for jumping to any page, and Undo and Redo sit in the toolbar.
- Click Apply changes, then download the edited PDF.
How editing existing text works
A PDF does not store paragraphs. It stores separate runs of text at fixed coordinates, so the editor works on one run at a time, and a run is usually a line or part of one. Click a run and it becomes a box in the same place. Apply changes then paints a rectangle over the original words, in the background colour sampled next to them, and draws your text on top.
The size is estimated from the height of the line and the colour starts as black, so check both and change them from the box's toolbar if they differ. A longer replacement wraps inside the box on to the lines below without pushing anything down, so widen the box by dragging a handle, or shorten the text. On a photograph or gradient, the patch can show as a flat block.
Find & Replace, under More, edits every run containing your search text. It matches within one run, so a phrase split across two runs is not found, and its patches are always white.
Fonts: what is reused and what is substituted
When a run's font is embedded in the PDF, your edit reuses it, so new words match the old. Embedded fonts are usually subsets holding only the characters the document used. If you type one the subset lacks, the whole line switches to the closest font on the list, and it does the same when the font is not embedded at all.
The list has 45 families in five groups: Standard, Sans-serif, Serif, Monospace and Arabic. The stand-ins for Arial, Calibri, Cambria, Times New Roman, Courier New and Georgia have the same character widths as the originals, so line breaks stay put. Others, such as Verdana or Palatino, are matched to a different width, and a line can run longer or shorter.
Helvetica, Times New Roman and Courier are the PDF format's built-in fonts. They are referenced rather than embedded, and any character outside western European languages is written as a question mark. The list has no Chinese, Japanese or Korean font.
Every other family comes from Google Fonts. Picking one makes your browser fetch it for the preview, a request that names the family and nothing else. Applying changes fetches it again, asking this time for a subset cut to the characters in the text you added or edited: those characters reach fonts.googleapis.com, and the rest of the document does not. If the download fails, the export uses the nearest built-in font.
Covered text is not deleted text
Edited text, Whiteout and Delete existing image all work by painting over the page. What was underneath is still in the file, and selecting the page text or opening the file in another program can bring it back, so they suit tidying a layout and not redaction. Deleting a page is the same kind of limit: it comes out of the page list, so no viewer shows it, but nothing rewrites the file and the page is saved out with everything else.
If something must not be recoverable, apply your changes and run the result through Flatten PDF with Flatten everything. It builds a new file out of pictures of the pages, so nothing sits under the patches, and no text in it can be selected or searched afterwards.
What Apply changes builds, and what the editor cannot do
Apply changes rebuilds the PDF in this tab, showing progress such as "Applying page 3 of 12": form values first, then page rotations, deletions and blank pages, then each page's text, images, shapes, annotations and links, and finally attachments.
Drawn items become part of the page. Highlights, underlines and strike-outs are painted marks, not comment annotations, so viewers show them but do not list them as comments. Links are real links, and filled form fields stay live until you flatten the file.
Some jobs are outside the editor. It cannot reorder pages (Organize PDF does), merge in another file (Merge PDF does), create form fields (it fills the ones already there), or remove attachments already in the file. A file that needs a password must first go through Unlock PDF.
Frequently asked questions
Yes, one run of text at a time: click a line and retype it in place. The colour starts as black and the size is estimated from the line's height, so check both before you download. A longer replacement wraps inside its box instead of reflowing the page.
Because the file did not carry the font it needed. A PDF usually embeds a cut-down copy of each face, holding only the characters that document happened to use, so a letter you type that was not in the original has nowhere to come from. The line then falls back to the nearest of the 45 bundled families, and you can override that choice from the box's toolbar.
Not until it has been read. A scan is a picture of a page, so there are no characters to click on. Run it through OCR first to add a text layer, then open the result here.
No. Whiteout paints a box over the page, and the words underneath stay in the file, as they do under edited text. To remove them for good, run the applied file through Flatten PDF with Flatten everything.
Not here. You can delete pages, rotate them and insert blank ones. Organize PDF reorders pages and Merge PDF combines files, so use those first and bring the result back to the editor.
Not in the editor yet. Take the password off first with Unlock PDF, then open the file it gives you.
For files under 5 MB the editor keeps a recovery copy in this browser, saved about two seconds after your last change. Come back to this page and it offers to restore your edits or discard them. Only one document is kept at a time, and the copy is cleared when you click Apply changes unless you tick "Keep a recovery copy after download". More, then Local saving, switches it off.
Yes. There is no server here to receive the file. It is read from your disk into this tab, changed in memory and written back out as a download. Open your network tab while you work and none of the requests carries the file. What is kept is kept locally: a Recent files entry on this device, with the file itself when it is under 5 MB, deletable from the Recent files page.
Related tools
This tool ran entirely in your browser and nothing was uploaded. Your Recent files list keeps a copy of local files under 5 MB until you clear it. How your files are handled.