14
Replies
Hi there
I'm personally unaware of any setting that controls this aspect. Of course, I could be wrong, but I don't recall ever seeing any setting one would use that says: Create this link as a secure link.
Mind you I'm not a "security expert". To me, the fact something is using HTTPS vs standard HTTP seems that it should be simply that the basic HTML pages and coding are somehow behind other pages or a firewall or something that makes the HTTPS magick happen.
Obviously I have something to learn about it all and likely will via this thread.
Off to Google now to see what I can glean about this aspect.
Cheers... Rick 
Yeah, there's apparently not a setting on the RH interface. I suspect I am either doing something entirely wrong or I need to change a .js file. Since I'm not a programmer of any sort, much less a javascript one, I'm not sure which
So, after replying to you I googled and found a page that discussed how HTTPS works. Basically, boiled down, it seems that what you do is to place pages that should be encrypted into a special folder of the web server. Because the content is inside that folder, things are encrypted when communicating with the browser. And that makes total sense for a page here and there that may deal with something like Credit Card information or other kinds of information.
What baffles me a bit here is why anyone would see the need for help information to be encrypted via HTTPS.
What I read seemed to imply that there is no way to construct a hyperlink in a "secure" fashion. All that rot happens via the location of the page, the security certificate and the server. There is nothing special you have to do with the linking.
As you say pages "go missing", I'm somewhat concerned in how exactly your links are being constructed. Are you familiar with the concept of Relative VS Absolute links? If so, I'm wondering if the pages that perform the disappearing act have possibly been created using Absolute links. Is that a possibility?
Cheers... Rick
It's not a relative v absolute link issue. The problem exists in the compiled project's TOC, which isn't actually a constructed file -- it uses the whtdhtml.htm file, which calls like half a dozen .js files (which is where I assume the TOC is built).
When I change the address of the help project to use https, the TOC doesn't list everything unless you tell it to show unprotected content -- I guess that happens when you click an unsecured link on a secured page.
As for why they're doing this -- they're really doing it for the whole shebang. (It's not that big a company.) It's not that they're worried about help text, it's that it's a global change for everything.
It fell on me to figure this out because I forgot to take a sick day during the last team meeting
Hmmm, well, if I were in your shoes, here is how I would approach the troubleshooting.
From what you have reported, it sounds as if you are seeing a partial TOC. Right?
Assuming that's true, then it means that you have some links that work, and some that don't? And in typing that, it occurs to me that maybe you don't even see the links in the TOC until you click that option to show all content?
If that's the case, what's the difference between a link in the TOC that's visible VS one that isn't? Are both links configured the same way in the TOC? Are both links at the same TOC level? Or is one perhaps at the immediately visible level while the other is inside a book? Or what about the topics that each link point to. Same level in the folder structure or different levels in the folder structure?
My guess is that you are going to find some obscure issue where maybe a folder needs some rights changed. But that's just a hunch.
I'm assuming that if you choose to view the output immediately after you generate, all is okay there? The issue only arises once you have copied the files to the web server?
Cheers... Rick
Getting back to this...
It doesn't appear to be a folder permissions thing. But the TOC links that are missing are actually links to other projects that get published into subfolders beneath the main project. When you click into them, you actually get another tab and you're browsing through a different project, and they're all maintained and published separately.
The javascripting that builds the TOCs in both the main project and the sub-projects seems to scan through the folders to build the TOC. (I don't know if that's literally what the js is doing, but that's how it looks). But now in our new https world, it isn't finding those separately published sub-projects any more.
Does that make sense?
(In my defense, that weirdo structure was here before I was )
AdChoices