PDF won't update for end user

[Sky Captain]

I have a hyperlink on my webpage to a pdf. This pdf needs to be changed on a regular basis. Typically, I upload the new one to overwrite the old one.

But then my website client says visitors are still seeing the old version. I suspect this is because they've been to the site before and are seeing a cached version in their computer.

Is there any way in the html to force a refresh on load?

I suppose I could change the name of the pdf each time (and change the html link) but this might just move the refresh problem to the linking page. I don't know if that's an improvement.

[sspry]

I do the same thing for a client when they update their quarterly PDF order form... except last month, people were still seeing the old one!

Well, I HAD uploaded it - just to the wrong directory! Easily done especially when there are a mountain of updates to do all at the same time.

The other problem is to do with the file name - i.e. orderform.pdf not the same as orderform.PDF, order-form.pdf, or OrderForm.pdf... so I have to ensure the same name is used each time.

While these may not be the problem in your situation, it stresses the point about continually checking links to make sure you get the expected result

[weegillis]

Unfortunately there is no way to clear a user's local history cache. What you might do is add a NEW badge beside the PDF link each time you update the file. Remove at some later date once it's been well distributed. You could include the recommendation to refresh the PDF file once it opens in their browser. This will overwrite the locally cached version in one fell swoop. No need to delete history or clear cache.

In addition, there is probably a PHP approach to using HTTP Response Headers.

[Sky Captain]

What you might do is add a NEW badge beside the PDF link each time you update the file.

The problem with this is -- what if they're looking at a cached version of the HTML page, too? Then they won't see the "new" badge.

Here's my plan:

1) Datestamp the documents, so if a visitor sees an old one I can ask what date they see.

2) New PDF uploads get a new file name. File1.pdf gets replaced with File2.pdf etc. The corresponding link gets changed and the old PDF gets deleted.

3) Use a "pragma" meta tag on the linking page to prevent future visitors from mistakenly seeing their cache. This only helps with new visitors, not old ones who cached the old page, unfortunately. Details here: http://www.htmlgoodies.com/beyond/re...le.php/3472881

[weegillis]

Question: are you PDF's being indexed? If so, you may wish to keep the same name on each update. At least that way, the SERPs will always be pointing to correct file.

[Sky Captain]

Well, that's part of the problem, isn't it? I don't want old versions of the file showing up in searches.

[3DGrunge]

If cache is the issue couldnt you use something like a php header to stop the browser from cacheing? I am not certain if it would work, but I think it should.

<?php
// Date in the past
header("Expires: Mon, 26 Jul 1997 05:00:00 GMT");
header("Cache-Control: no-cache");
header("Pragma: no-cache");
?>

or force them to download the new .pdf file with something like
<?php
header("Content-type:application/pdf");

// It will be called downloaded.pdf
header("Content-Disposition:attachment;filename='downloaded.pdf'") ;

// The PDF source is in original.pdf
readfile("original.pdf");
?>

Again I am not certain if this will work, someone else might be able to confirm or refute this.

[Sky Captain]

Yes, I just started using the Pragma meta tag. (See above.)

That second bit of code is new to me. I'd like to know if it will work before trying it out.

Also is there an equivalent for straight HTML? Not all my websites are in PHP.

[wige]

What you are describing actually sounds like a symptom of a server misconfiguration - for some reason, the server is not properly communicating with clients about when the file was created. This could be because the modification date of the file is not changing, or it could be because the web server is not sending the proper headers.

First, a couple of questions:
Is the server Apache on Linux?
How is the PDF file being put into its download location, through FTP or some other method?

[Sky Captain]

PDF files are always uploaded through FTP.

The server is running Linux CentOS. I believe it's Apache, because I'm successfully using 301 redirects for other areas.

(That gives me an idea... is there a way to solve this in an HTACCESS file?)