Picking pages by hand or by text
Every page of your PDF is a thumbnail with a checkbox. Click a page, or its checkbox, to pick it. Hold Shift and click a second page to pick the whole run between them. Select all, None, Odd pages, Even pages and Invert change many pages at once, and Invert is handy when it is easier to say which pages you do not want.
The box takes the way people write page lists: single pages, ranges and open ended ranges, separated by commas, spaces or semicolons. 3 is page 3, 2-5 is pages 2 to 5, 8- is page 8 to the end, and last is the last page. The words odd, even and all work too. A dash pasted from a word processor counts as a hyphen. If something is wrong the box says what it did not understand, for example that a range runs backwards or that page 40 does not exist in a 30 page file, and it keeps the previous pages in the grid until you fix the text.
The box and the grid are two views of the same choice. Tick a page and the box is rewritten, for example from 2, 5-6 to 2, 5-7. Type in the box and the ticks move. Nothing happens to your file until you press Extract.
The order of the pages
By default the new file follows the order of the original, whatever order you ticked or typed the pages in. If you choose Use the order I typed, the new file follows your list instead: 5, 2, 8-9, 1 gives page 5, page 2, pages 8 and 9, then page 1. Pages you pick by clicking count in the order you clicked them, so you can build an order without typing.
A page is used once. If you type 1, 1, 2, the new file has pages 1 and 2, and the first mention wins. The preview after extracting labels each page with its place in the new file and the page it came from, such as Page 1 · from 5, so you can see the order you got.
This tool picks and orders pages and nothing else: it does not rotate pages or merge files. To join the result with other files, use the PDF merger. To cut a document into several files at once, use the PDF splitter.
What the new file contains
A PDF is a list of pages, and extracting copies the pages you chose into a new file. Nothing is redrawn or recompressed, so text stays text you can select and search, photos keep their quality, and every page keeps its own size, rotation and crop box. A page that was landscape stays landscape.
Things that belong to the whole document cannot follow a page. Bookmarks are not copied. Links that jump to another page of the original are dropped, even when the page they point to is in your selection, because they would point into a file that no longer matches. Links to websites, highlights and comments stay. Form fields keep their look and their typed values, but the new file no longer contains a form. The title and author of the original are not copied.
Some PDF programs list every image and font of a document in one place that all pages share. A naive copy would drag all of them along, so one page of a big report would come out as large as the whole report. ClipLime copies only the pictures, fonts and patterns that the chosen pages draw, including the ones used inside the drawings those pages contain, as long as it can read the page's drawing instructions. If it cannot, the page keeps the whole shared list, the file can be bigger than you expect, and the result card says so. A font that several pages share is copied whole, so letters that only an unchosen page used can stay in the font file; it is the list of letters, never the words or their order. Layers keep the setting they had, and the private data some editing programs keep inside a page is left out.
Scans, big files and protected files
A scanned PDF stores each page as a picture. Extracting pages from it copies those pictures as they are, so the new file is about as big as those pages were. There is no text layer, and ClipLime does not recognise text in pictures, so a scan stays unsearchable.
The work happens in your browser, so the limit is your device's memory. Thumbnails are drawn only for the pages near the screen, which keeps long documents usable. ClipLime warns when a file is large for your device, roughly 60 MB on a phone and a few hundred megabytes on a computer, and refuses files above 1 GB. You can press Cancel while it works.
Password protected and encrypted PDFs cannot be used. Their pages stay scrambled until they are decrypted, and ClipLime does not decrypt files. A PDF that opens without a password but has permission settings, for example no copying, is encrypted too. If you are allowed to, open it in a PDF viewer, print or export it to a new PDF without protection, and add that copy.