Copy link to clipboard
Copied
I have RH 2017 (latest), but I'm re-trialing RH-2020 to see if the printed manual generation is better or at least the same in RH 2020, Update 7.
We have a large help project that in RH 2017 we generate by sections into .doc files and then post-process the doc files prior to public consumption. We generate by sections because our project is too large to generate everything in the TOC in a single go as MS Word typically hangs at some point, possibly from memory issues.
Anyway, in RH 2017, you can choose a part of the TOC as your "Chapter layout". This is what we've done.
However, once converted into RH 2020, these SSLs (single source layouts -- ie "Output Presets" in RH 2020) don't seem to behave the same. All them appear to reference the entire TOC.
In fact, does RH 2020 even have the ability to export just a part of the TOC in an "output preset" like 2017 can? If not, is there a good workaround that works for people?
Copy link to clipboard
Copied
So, from some additional research and messing around with this, it looks like I'll have to create a new TOC for each section of my TOC and use those section-based TOCs to generate printable outputs by section.
Then for my HTML5 output I'd create a holder TOC with all these other section TOCs inside of it. Does that sound correct? Just making sure I understand the process/workaround to do this.
Copy link to clipboard
Copied
As an experiment, did you try generating output using the 1 TOC method? Maybe RH2020 can handle it?
Copy link to clipboard
Copied
Jeff's suggestion to try generating the whole thing in one go isn't as crazy as it sounds. Adobe changed things so Word isn't required to generate a doc file any more. Possibly whatever utility used to produce the Word output might cope with the large output when Word didn't.
"Adobe RoboHelp does not require Microsoft Word installed on your machine to generate Word Document output."
https://helpx.adobe.com/robohelp/using/generate-word-output.html
Copy link to clipboard
Copied
Hi Jeff and Amebr. Thank you for responding.
Just to clarify, I'm generating a Word doc, not PDF files--but if I don't need to do any post processing would love to just do PDF files. (In RH 2017 we have to use a bunch of Word VBA macros to clean everything up.)
Yes, I did try generating everything with our one big TOC. That was the first thing I tried. It still hangs for me somewhere along the lines though, so I assume there's still a Word memory issue going on where Word balloons out of bounds--hence the desire for generated builds at at each chapter/TOC section or to verify which chapters do build fine.
Anyway, it's possible there's an issue with something in my topics or topics that RH doesn't like or can't handle, but for me, the log files aren't very helpful in finding out this info.
I wish I could watch it being built so I could determine what the problem is.
Copy link to clipboard
Copied
Okay. So now that I figured out where my RH2020 temporary files and log files are stored, I'm attempting to build everything as a PDF output in one go (not in sections) to see if I can resolve why it's hanging.
But the problem is this:
The generation is stuck somewhere doing...something (or maybe doing nothing). However, there is no active log to see what's causing it to hang. This is because, for some backwards reason, the .ehtml log file only appears at the end of generation rather than being actively and continually updated as a visible file somewhere that we can open in Np++ or something.
I can only assume that the log file details are being held in memory and then are dumped to the .ehtml log file at the very end.
If it's in a hung state like this, this is when we most need the log file to see why and where.
But all I see is this:
Which does not give any indication that there's even a problem. (I suspect there's a problem because there hasn't been any status change on this message for 45 minutes now.)
Anyway, in my temp location, inside the "Test - Everything (PDF)-copy" folder there's nothing else happening. The PDF file isn't being built or anything. The folder of temp files seems to just be sitting there waiting for something.
The only clue I have is the small status message from RoboHelp that says, "Merging HTML Files" and then it lists the first part of a topic filename: "Viewing_Hits_in..."
But looking at that topic file provides no clues as to why it's hung. I don't even know if that's the cause of why it's hung.
Eventually, I click Dismiss on that message and click the X button on the OUTPUT PRESET but still no .ehtml log file appears, and I'm no closer determining what's causing it to hang.
Copy link to clipboard
Copied
Time to contact support using the email address here. 🙂 https://helpx.adobe.com/contact/enterprise-support.other.html#robohelp
Copy link to clipboard
Copied
Thanks Amebr. I'm going to break out the TOC into multiple section-based TOCs and build by section and see if I can figure out where it hangs that way. But yes, they should fix this issue. I'll file a bug report with them.
Copy link to clipboard
Copied
Okay update time.
So, from my main TOC, I created individual TOCs for each main section. I then created Output Presets for each of those individual section-based TOCs for both Word and PDF outputs.
After cleaning up some issues in the HTML code, I can now build and open the .docx or .pdf file each section without any error. I had hoped if everything could build separately, I would then be able to build everything together in a single output preset.
Alas, but RoboHelp still hangs when I try to build a single Word or PDF file from my main TOC.
My next idea is to conditionally comment out most of my sections in my main TOC and see if that builds, and if so, then remove the comment condition one section at a time until the build fails to find out where the problem is.
Copy link to clipboard
Copied
I filed a bug on this here: https://tracker.adobe.com/#/view/RH-10597
Copy link to clipboard
Copied
Voted and wrote a vote essay 😛
Copy link to clipboard
Copied
Tagging out say five sections could make it work but that doesn't mean there's a problem with those five. Tagging out a different five could get it working again. The fact that one section at a time works suggests there's nothing wrong with the content. It could just be volume or the memory on the machine.
Each time you generate removing one section at a time, note the size of the document when it starts working. Then log a bug with that information.
________________________________________________________
My site www.grainge.org includes many free Authoring and RoboHelp resources that may be of help.
Copy link to clipboard
Copied
Thanks Peter. I filed https://tracker.adobe.com/#/view/RH-10617 on this issue. For me, it hangs sometime after 27,669 MB or 1219 pages. Those aren't exact limits, but when I add the next section of a few topics on top of all that, it hangs.
Copy link to clipboard
Copied
Whilst Word is not used to generate the document, maybe the old Word limits still apply? They were not an absolute size or number of pages, rather it was the mix of text and images and some other factors.
________________________________________________________
My site www.grainge.org includes many free Authoring and RoboHelp resources that may be of help.
Copy link to clipboard
Copied
There very well could be limits with the new way of doing things though I hope not. I am impressed by RH 2020's generation speed for printable outputs and would love to switch to it for this reason alone.
But generating only by sections in RH 2020 makes it so that links that point to other sections don't work. That's a bad user experience I'm hoping to avoid.
Copy link to clipboard
Copied
I was talking about limits that have always been there. This relates to Word 2000!
Before you created a partial output by one method, now you create two copies of the full TOC and create the same partial outputs.
________________________________________________________
My site www.grainge.org includes many free Authoring and RoboHelp resources that may be of help.
Copy link to clipboard
Copied
Hi Peter,
Sorry, I'm not sure what you're getting at exactly.
I generated a PDF file for my bug report to Adobe, not a Word document.
Perhaps I'm missing something, but my assumption is that if Word is not used to generate a PDF file then Word limitations shouldn't be taken into consideration, right?
Copy link to clipboard
Copied
The start of the thread referred to printables which could be DOC or PDF, then you mention DOC so I thought that's what we were talking about. I didn't go back through the whole thread.
The new process is based on Word's open source XML. You don't have to have Word installed on your machine for that but I was just wondering if that when the XML generated behind the scenes is saved as a DOCX file, the old limits kick in.
________________________________________________________
My site www.grainge.org includes many free Authoring and RoboHelp resources that may be of help.
Copy link to clipboard
Copied
Understood and sorry for the confusion, as I did start out intitially with .docx printables. I transitioned to testing PDF as that seemed a route with less steps. If you're right about the .docx size limits, it's too bad that Microsoft hasn't come up with a solution for that over the years with their latest Word offerings. Perhaps this underscores all the more reason to shift to PDF, assuming I can get my complete output preset to generate the pdf without hanging.
Copy link to clipboard
Copied
No worries. Adobe should be able to generate PDF better as they created the format. 🙂
The problem with PDF is any after generation editing, as you will know.
________________________________________________________
My site www.grainge.org includes many free Authoring and RoboHelp resources that may be of help.
Copy link to clipboard
Copied
True. However, I found that you can do quite a bit in Acrobat DC to edit a PDF, at least at the sentence, word, and paragraph level, such as modifying text, images, borders, etc.
Also, given that RH 2020 can generate PDF printables more quickly than RH 2017, I'm hoping we won't need to do much post processing anyway and can just fix any issues in the source and regenerate.