Everything that shipped, newest first. Anything that changes what happens to your data gets written down here.
11 October 2026
Make a scanned PDF searchable
New tool: OCR PDF. A scan is a picture of a page, so you cannot search it or copy a sentence out of it. OCR PDF reads the words in each page and lays them over the scan as invisible text, so the same file becomes searchable and you can select and copy from it.
The pages look exactly as they did. Nothing is redrawn or recompressed; the words go on top.
It runs in your browser, like everything else here. The reading engine and the language you pick download from this site the first time, a few megabytes, and your scan is never uploaded.
Sixteen languages, including Chinese, Japanese, Korean, Hindi and Nepali. Arabic is not offered yet.
Pages that already have text are left alone, and the page tells you which ones. You can also save the recognised text as a .txt file.
It reads printed text, not handwriting, and a faint or blurry scan comes back with mistakes. Pages it found hard to read are named when it finishes.
3 October 2026
PDF to Word keeps your paragraphs
PDF to Word was handing back each page as one unbroken block of text. A ten-page report arrived as ten paragraphs, each of them a wall. The count under the download said so, which made it look deliberate. It now breaks paragraphs where your document breaks them.
It works this out from the page itself: the gap between lines, a change of type size, a bullet or a number at the start of a line, an indented first line. Headings come back as their own paragraph and list items stay separate instead of running together.
Lines that were wrapped are joined back into sentences, so you no longer get a hard break every seventy characters when you try to edit.
One case it leaves alone deliberately. Where a document marks its paragraphs with nothing at all (no extra space, no indent, no bullet), there is nothing on the page to read, so those stay joined rather than being split in the wrong place. Pulling two paragraphs apart takes a second; finding the sentence that got cut in half does not.
What has not changed: it is still the words and not the design, and it still runs entirely in your browser.
24 August 2026
Install it, and use it on a plane
PDFsansar installs now. It sits on your home screen with its own icon and opens in its own window, and there is nothing extra to download: it is the same site.
It works with no connection. That was quietly true all along, because the tools never needed a server; it is just official now. Open a tool once with a connection and it keeps working offline, with your files still never going anywhere.
A tool you have never opened will say so rather than failing strangely. That first visit is when its code gets saved.
The install button is on Account Settings. On an iPhone, Safari does the installing: Share, then Add to Home Screen.
24 August 2026
Optional accounts, and Fill PDF forms
You can sign in with Google now, if you want to. It syncs three things (your language, your appearance choice and your starred tools) between your devices, and that is the whole of what an account is for.
It stays optional, and it always will. No tool has been put behind a sign-in, there is no page limit, and everything on this site works exactly as it did if you never make an account.
Your documents are not part of it. The tools still run entirely in your browser, and the sign-in code never sees a file, a filename or even a page count. We also do not keep a record of which tools you personally used. The only thing counted is a plain daily total per tool, with nobody attached to it.
Deleting an account is a button on Account Settings. It signs you out at once and removes the record after 30 days; signing in again before then cancels it.
New tool: Fill PDF forms. Type into a fillable PDF and download it filled in, with the option to lock the answers so they print the same in every reader.
It tells you when it cannot help rather than handing you a broken file: a scan of a form has nothing to type into, and an XFA form (Adobe's own format) needs Acrobat Reader. Both are detected and explained.
24 August 2026
Dark mode
PDFsansar follows your system theme now. If your phone or laptop is set to dark, so is this site: no setting to find, nothing to switch on.
You can also override it. Account settings has Light, Dark and System, and whichever you pick sticks on this device and applies before the page paints, so there is no white flash on the way in.
Page previews stay white in dark mode, because they are pictures of your document rather than part of the interface. Inverting one would show you something your PDF does not contain.
Made two of the greys darker while we were in there. The footer's small print and the "last updated" line on the legal pages were well below the contrast standard, and now every piece of text on the site clears it in both themes.
19 August 2026
Sign a PDF without printing it
New tool: Sign PDF. Draw your signature with a finger or a mouse, or use a photo of one you already have, then drag it where it belongs and resize it. Put it on as many pages as you need.
Your signature is trimmed to the ink, so the box you drag and the mark you see are the same thing, with no invisible margins pushing it out of place.
It stays on your device like everything else here. There is no account and nothing is stored, so the signature is gone when you close the tab.
The page is honest about what this is: the equivalent of signing a printout, not a cryptographic digital signature with a certificate. The tool page says so rather than letting you assume.
19 August 2026
A proper look, and a much better phone experience
Redesigned the whole site. Every tool now has its own colour and its own drawn icon, so you can find the one you want without reading each card.
The tools themselves got the most attention: a drop zone that lights up when you drag a file onto it, a download button that looks like the thing that finishes the job, and a confirmation at the end that actually says your file never left the device.
The header no longer eats half a phone screen. Links scroll sideways in one row instead of stacking three deep.
Fixed a layout fault that put a thousand pixels of empty space beside every Arabic page.
Made the small grey text darker so it meets the contrast standard rather than nearly meeting it.
Everything that moves now stops moving if your device asks for reduced motion.
19 August 2026
Seventeen languages, and things to read
The tools, the home page and the FAQ are now available in seventeen languages, picked from the menu in the header. Arabic reads right to left properly, rather than as English with the words moved.
Added guides: short walkthroughs for merging on a phone, splitting without Acrobat, and fixing scans that came out sideways.
Started the blog properly, with a piece on what actually happens to a file you upload to a converter site, and one on how browsers got good enough to make that unnecessary.
Published an accessibility statement that lists the gaps as well as the good parts, and added a skip link so keyboard users can get past the header in one keystroke.
Added a sitemap page for people, not just the XML one for crawlers.
18 August 2026
Every tool now explains itself
Each tool page ends with a plain explanation of what it is for, what it does, and what it deliberately does not do. That last part is the one people actually search for, like whether compressing loses quality or whether cropping deletes what it hides.
Added the questions people ask about compressing, cropping, redacting, removing and extracting pages, text extraction, page numbers and watermarks, with structured data so search engines and AI assistants can quote the answers.
Crop, Redact, Remove pages and Extract pages were showing their internal name instead of their title on the tools page. Fixed, and the build now refuses to ship a tool whose copy is missing.
Tool pages now point to QRSansar for sharing a document by link or QR code: the one job that cannot happen on your own device, and labelled as such.
17 August 2026
Ten new tools
Crop PDF: drag a box over the page and trim everything outside it. Nothing is re-rendered, so text stays text.
Redact PDF: cover private details and have them actually removed. Any page you mark is rebuilt as a flat image, so the hidden words are not sitting in the file waiting to be copied out.
Remove pages and Extract pages: click the ones you mean, or type 2, 5-7.
Organize PDF: reorder, turn and drop pages, then save the lot in one go.
Add page numbers: pick the corner, the style and the number to start from.
Add watermark: stamp text or your logo across every page, with size, angle and strength you control.
PDF to PNG, for when JPG softens the text too much.
PDF to text: pull the words out and copy them or save a .txt.
Compress PDF: recompresses the pictures inside the file and leaves everything else alone, so the text stays selectable and searchable. A file with no pictures is told there is nothing to squeeze rather than handed a pointless download.
The tool list is grouped now, because fifteen tools in one flat grid was getting hard to scan.
17 August 2026
Site pages, and a cookie choice that means something
Added a header and footer to every page, so the tools are no longer islands.
Wrote the pages a real site needs: about, how it works, security, FAQ, contact, and the legal set.
Analytics no longer loads until you agree to it. Decline and the script is never requested, which is a stronger promise than the usual banner.
Added a page where you can change that answer at any time, and see what it is set to now.
Two new tools: delete and reorder pages, and PDF to PNG.
Every tool page now ends with answers to the questions people actually ask about that tool, and links to the others.
You can write to us from the site itself now, instead of copying an address into your mail app. The page names the service that carries your message.
The site finally has a favicon, so the tab is identifiable.
Initial release
The first five tools
Merge PDF: combine files into one, in the order you choose.
Split PDF: cut a document into separate files, or pull out the pages you want.
Rotate PDF: turn sideways pages upright without turning them into images.
JPG to PDF: build a PDF out of JPG and PNG images.
PDF to JPG: render every page to an image and download the set as a zip.
All five run entirely in the browser, which was the whole point of starting over.
What is next
Roughly in order, though what people ask for tends to reshuffle it:
Fill in a PDF form, the interactive kind with real fields
Edit a PDF: add text, images and shapes anywhere on the page
Rotate several PDFs in one go, instead of one file at a time
Password protect a PDF, and remove a password you already know
Move the heavy rendering off the main thread, so big files stop freezing the tab
Want something moved up? Ask. A single email from someone who needs a tool counts for more than our own guessing.
Pick a tool and get on with it
Nothing to install, no account to make. Your file stays on your device.