I thought my questions were at an end. LOL
I am having no problem generating Responsive HTML for my large project, but the PDF output has not yet been successful. I've poured over my project to try to figure out why it keeps failing, but I'm at a loss. Here is what I've checked so far.
I did try to build a PDF of the About RoboHelp project sample, and that generated fine. I guess I will have to take a deeper look at that to see if I might be missing something, but so far things seem pretty straightforward.
I've checked and double-checked my project settings, css references, links, used and unused files. Given what I explain above, is there anything else I should look for?
given that the "failed to copy" message references the masterpages folder, have you checked the stylesheet attached to the master page used for the pdf output preset?
That was actually a good suggestion, but no reference in that stylesheet to that mystery _rhdefault.css. It's a real puzzle where it's coming from. LOL
Are you doing anything like importing Word files through the Link functionality? If so, perhaps the import settings are referencing it?
It was initially imported from a Word file, but then broken apart into individual HTML topics. The only external references in the project are external URLs.
There are a lot of topics, but other than that it's pretty straightforward HTML and CSS.
Use the Find and Replace to search for _rhdefault.css. In Settings, select Find in Source View.
Oh I do love that feature 8-)
0 occurrences in 0 files. And if I look at the Style Sheet report, there are no pages shown for using that particular style sheet.
I have to have some sort of mystery setting set somewhere that is preventing me from generating this PDF file, and I'm betting it has something to do with that little mystery css file. Wherever it is it has been really hard to find.
Long shot but close your project and go to File Explorer. Delete the files in the .rh folder. Open your project again.
Does the problem still occur?
Hi Peter, thanks.
I've tried that a couple of times and the problem persists afterward. But I had a thought. I am using the "online" project as a source for the PDF manual. Will I need to make a separate project for each of the two manuals to use for the "book" TOCs, or can the single online source be used for both PDF and online?
The other thought I had is if I create a subset TOC maybe I can figure out if there is a particular section that is giving me the grief. But because there are so many cross references between chapters I am afraid that any builds won't be successful anyway.
The online version builds perfectly.
I think in your other thread we have already established that you only need one project to achieve what you want.
You can use your online TOC for PDF purposes. To add front and back matter you would need to right click the TOC and change its properties to Book while you do that. This is explained in the About RoboHelp sample project.
Overall you need tags to identify different type of topics and use them with expressions to include/exclude them from a particular output.
You can use one TOC and keep changing its properties from Online to Book or vice versa but you might find it easier to create a duplicate of the Online, set its properties to Book and then edit that as required to get the front and back matter you want.
I may have stumbled onto something but I can't find how to fix it.
There are two index files in my documentation index folder. One named Default.ids, and the other named (myproject).idx.
I began observing that the PDF projects seemed to stop responding at the Index compilation. I opened each file in a text editor to see what was going on. Initially, they were both just about blank.
There were over 1000 index entries in the RoboHelp project index, but barely any in either of these idx files. Right now, only one of the two manuals is indexed (the Python manual). I don't want to build the index entries for the Reference manual until I sort this out.
Anyway, I used a text editor to add "Reference" conditional tags to the index that was there. And here is when I found something interesting.
When I add conditional tags to the Index tab in Robohelp, the changes are written in the default.idx file.
But the conditional tags themselves appear to be inherited from the (projectname).idx file.
In other words, there is some sort of mismatch going on, and this might be why the PDFs are failing.
I have used search and replace to find whatever is referencing this (projectname).idx file, but nothing comes up. I am not sure how it was created. I may have gotten over ambitious with a search and replace at one point and it might have happened that way.
There may be light at the end of the tunnel. Previous to this I hadn't created the index.html files for the PDF manuals, but they are there now. And that does indeed seem to be where things are being held up.
Correction. The "Python" condition was added to the Python-related topics. It's the "Reference" condition and index entries that haven't been added yet.
Well I deleted the (project).idx file and restarted RoboHelp. The index entries are now displaying fine while using the default.idx file.
What I've decided to do in regards to the TOC is create book TOC's for the two main PDF manuals (Reference and Python). Then I have a third online TOC that loads both of those TOCs in. That way I don't need to make changes to multiple TOCs when topics are added to either manual.
I used to be able to generate a PDF from the About RoboHelp project without issue. However, now even that is failing.
It seems when it's time to generate the PDF file, Robohelp hangs and goes unresponsive. Otherwise the status bar sticks on "merging _glossary.html." If I press Dismiss in the PDF generation status, nothing happens. If I then click the X in the PDF Output Presets tab, nothing happens. The only way to get out of this is to shut RoboHelp down and restart. No log generated.
I have a similar problem if I display a single topic and click "Save As PDF". Initially, the Save As dialog pops up and prompts for a save location. After that, the About Robohelp project screen goes completely white. No output, no log generated.
I've been doing a lot of wading around in the weeds while trying to sort out my PDF issues, and at one point I saw that the Microsoft PDF Output plugin might have been a culprit. So I deleted that, which may be what started this happening. I have since updated Windows, reinstalled Office, Creative Cloud (for Acrobat DC), and RoboHelp 2020. I'm still having the issue. Print to PDF seems to work in everything but RoboHelp.
The more I am trying to figure out why PDFs won't print, the worse it's getting! LOL
This really is looking like a machine issue so my next step would be to install RoboHelp on a second machine (trial mode will suffice for this) and put the project there.
RoboHelp must be installed with admin rights.
Then first open the sample projects and check they work. With that established make an offering to the gods and try with your project.
Let us know where that gets you.
I made an offering to the gods and wiped out my C drive, reinstalled Windows, and now I'm at least back to the point where I can print PDFs of the About RoboHelp project.
But I'm still at my original issue with the mysterious _rhdefault.css reference and being unable to output to PDFs.
A question for you, and I am not sure if this is related. Rather than having all of the index entries at the top of the page, I have put the index entries at the headings that the index entry points to, so that users can be taken directly to the topic instead of to the top of the page. Like so:
Could this be partly why my PDF output is failing?
It should be easy to test your hypothesis. Take a copy of that topic and create a TOC containing just that topic. Try generating the PDF using that TOC. Does it build? If it doesn't, move all the entries to the top of the topic. Does it build? If not, delete all the index entries. Does it build?
You have simply pointed your index entries to a bookmark and that's fine.
Can you share the project? If you can, then see the Contact page on my site and send the project as instructed there. Do make sure you include a link to this thread and please do not email the project direct.
Thank you so much, Peter. Email sent.
The problem was caused by an output topic having been imported into the source files. In Classic they were obvious as there would be red squares indicating the scripts within such topics.
Visually they are not obvious in the new UI. I have filed a ticket suggesting that the PDF generation process should check for such files and throw a warning if any are found.
If anyone thinks this is behind their PDF generation failure, the Find and Replace and search for CDATA. That will identify output files in your source.
Thank you again for your help with this, Peter. I don't think I would have learned that on my own. Now I know. 8-)
PDF successfully generating, now to fix its style sheets. I am a happy camper.