Computing desk
< December 14	<< Nov \| December \| Jan >>	December 16 >

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

December 15[edit]

Google search results difference in browsers[edit]

I've recently noticed that the look of Google search results differs between my Firefox and Opera. While Opera shows familiar older look, Firefox now shows a website icon, Google translate tool, etc. (screenshots). Is it just me or for everyone (an updated Firefox perhaps)? Brandmeister^talk 21:07, 15 December 2019 (UTC)[reply]

I have noticed this difference using the same browser (Opera) on two different devices. On my cell phone Opera shows the new layout, while on my Fire tablet it shows the old layout. Both devices are running the same version of Opera, and they both are sending the same User Agent. So I'm not sure what causes the difference, but it doesn't seem to be a difference in the browser. CodeTalker (talk) 19:35, 16 December 2019 (UTC)[reply]

SOLVED: Removing a watermark from a PDF[edit]

I downloaded the Casio fx-991EX user guide here:

https://support.casio.com/en/manual/manualfile.php?cid=004009138

When I look at it with Adobe Acrobat reader it looks fine, but when I try to print it it has a huge black diagonal "CASIO" across every page, making it unreadable.

Does anyone know how to either remove the watermark or where to download a printable version?— Preceding unsigned comment added by Guy Macon (talk • contribs) 22:46, 15 December 2019 (UTC)[reply]

Guy Macon, I don't know what you mean by "looks fine" - in my PDF viewer it shows a grey diagonal "CASIO" behind the other text. The black text on top sounds like some sort of bug. Have you tried a different PDF viewer application? Is your Acrobat up-to-date? What about your print drivers? Elizium23 (talk) 06:49, 16 December 2019 (UTC)[reply]

I used the latest version for Windows 10 (Adobe acrobat Reader DC 19.021.2058). No grey "CASIO" visible on my screen. I would be able to live with gray; when printed the CASIO is black and you can't read the parts of the page it covers. I tried googling "remove PDF watermark" and found a program that claims to be able to do it. It popped up a message "absolutely free for 7 days. Just enter your credit card and you can cancel any time." Yeah, right. :( That program also showed the black CASIO but wouldn't try to remove it without a CC#. Tried two different printers with different drivers. Still had the black CASIO. --Guy Macon (talk) 07:12, 16 December 2019 (UTC)[reply]

Suggest you try opening it in another app (e.g. word processor or image viewer), or another PDF viewer. There are plenty out there.--Shantavira|^{feed me} 09:14, 16 December 2019 (UTC)[reply]

There are 2 issues here which probably need to be discussed.

One: This PDF has permission restrictions. This means with any tool which respects these permissions you won't be able to edit it without knowing the owner/edit password unless these permissions are removed. Unsurprisingly this includes a lot of commercial tools, but also quite a few free ones. There are a lot of online only tools which can 'unlock' PDFs. I won't link to them because of DMCA concerns and also since I cannot guarantee that any of them don't add malicious data to the PDF even though I don't think this is a concern for your set-up. I will also say that our article QPDF says it can decrypt PDFs.

Two: This PDF seems to use a simple image watermark. Nothing fancy like some PDFs do to try and stop you removing it. If it is unlocked and you open it in a good PDF editor like Acrobat, you should be able to trivially remove the watermark.

A quick and dirty solution is to use the commandline/server PDFtk. (Note that the Windows version does respect security restrictions although I've heard this is removed in the Debian one.) Then do something like pdftk fx-570_991EX_EN.pdf output fx-570_991EX_EN.uncompressed.pdf uncompress verbose (well you don't need the verbose). You can then explore the PDF in a text editor.

You should find 5 objects (30, 115, 153, 211, 262) representing the watermark since for some reason some pages have their own copy. On Notepad++ with the newline option, you can use \d+? 0 obj(?:(?!\nendobj).)*/Private /Watermark.*?endobj \n to find these or make your own regex for your own editor. These all reference the same image, object 287.

You can use 287 0 obj.*?endobj \n to select this object. If you delete either all 5 of the watermark objects, or the image, it shouldn't show up anymore. Your PDF reader will likely complain that the file seems corrupt since each page still references the watermark but should work fine if decent AFAIK. (It looks fine on Adobe Reader although I barely looked.)

You may be able to use some corruption repair tool to repair the PDF after you've partly removed the watermark, although PDFtk didn't work with a brief try. Alternatively, you can probably find tools maybe even PDFtk, or if not GhostScript to more gracefully remove the image or whatever. Finally, you can probably properly remove all references to the watermark via a text editor. If you need to send it, you may want to recompress it beforehand.

Nil Einne (talk) 16:32, 16 December 2019 (UTC)[reply]

Someone on the HP forums solved this. See https://www.hpmuseum.org/forum/thread-14180.html

I am still going to try the suggestions above so I will be able to deal with similar situations in the future. --Guy Macon (talk) 16:49, 16 December 2019 (UTC)[reply]

For more generic stuff, you may be interested in [1] and maybe also [2] and [3] which discuss other possibilites, including text water marks. I think given your knowledge, decompressing the PDF will probably often be enough to be able to work it out, since even without much understanding of the format, you can maybe sort of see what is going on. Nil Einne (talk) 17:29, 16 December 2019 (UTC)[reply]