Hi RoboHelp Community!
My name is Scottie and our team is working with RoboHelp 15 generating Responsive HTML5 help. Our software's installer picks up our compiled help files nightly and we test our files using different web browsers (Chrome, Firefox, and IE). We noticed help updates were not always showing up until we reloaded or refreshed the the browser page.
We cannot force users to manually refresh/reload and our developer recommended adding the following to the html:
<meta http-equiv="cache-control" content="max-age=0" />
<meta http-equiv="cache-control" content="no-cache" />
<meta http-equiv="expires" content="0" />
<meta http-equiv="expires" content="Tue, 01 Jan 1980 1:00:00 GMT" />
This works, but I'm not sure if:
A. Do or can we create a master page with this code and then apply it to each page in our projects?
B. Anyone else has had a similar issue? If so, how did you resolve it?!
Thanks for any time and help!
My own question would be to ask where your users are. Are you planning on having the help hosted in a central location and updating it perhaps daily? If so, then I'd say the issue is a valid concern and you need to find a way to ensure they always see the latest and greatest.
But if your help is shipped with an application and the end user installs the app and the help doesn't really change until a new release is made available and shipped to them, it would seem the concerns are a bit overinflated.
Just some thoughts from the edge as we wait for others to chime in.
Thanks for getting back to me, Rick!
Currently, end users install updated files quarterly, but we are migrating the help to a local server for continuous updates by the end of summer. Our documentation team installs daily and is constantly clearing our cache.
I am curious how other users handle this issue, but I don't see any related topics! Is this a less common issue? Do you think the meta tags are the best solution?
I just did a quick search for "cache" in some of the RH forums & it seems that sticking the meta tags in you Master page and then applying it to all topics (either at output via the SSL recipe or within the project itself) will get the code into the resulting HTML topics. Some threads reported issues with the TOC not getting refreshed (in fact some talked about seeing a scrambled version after content updates), but I didn't see a clear programmatic method being promoted to hack the .js code that generates the TOC.
I've added similar code into my screen layout because at the time I researched it, there was some question about whether meta tags in the body of an html page would be honoured. Further research suggests that it should be fine for HTML5 (based on the HTML5 specs).
However, we still experience content stickiness, especially with the table of contents. *Usually* waiting an hour after getting an update ensures the correct content displays, but even that isn't guaranteed.
(I suspect server settings and the gXMLBuffer.js file are involved, but have not gotten any further in investigating due to time and staffing constraints. I just included a Troubleshooting topic in help which explains how to change a cache setting in IE and direct people to that when bugs are submitted.)
Thanks for this recommendation! We're trying it and thus far it seems to be working' although, we are seeing some issues with the TOCs. At least we have a start!