Welcome to the forum.
That fix is not available for RH7 as it is not needed. The bug was fixed.
I doubt you will "fry the processor". Try opening and compiling one of the supplied projects. That will identify whether or not the problem is your project.
Is your project on your local drive and are you compiling to that drive?
The 100% use is normal.
Thanks for the feedback. I went ahead and tested the default RoboHelp projects (Clownfish_School and CustomerSupport) that comes with the program and it generates without problems. Thus, it looks like it is a problem related specifically to the project I am working on. The strange thing is that I have reverted back to saved working copies of the same project dating all the way back to the beginning of January and they all are hanging up now. This makes no sense. What is even stranger is that I can produce technical documentation in RoboHelp via other Single Source Layouts....it is just the specific Single Source Layout that produces the .chm file that I am having a problem with.
I think I found the reason for the hangup. I removed the Conditional Build Expression from the Single Source Primary Layout (Microsoft HTML Help) and it compiled as it's supposed to. So:
1) Does the RH6 bug fix need to be adjusted for the RH7 because the conditional Build tags are causing the hangup?
2) Is there a limitation to the memory/speed of the computer I am working on to build a .chm file with a project that has extensive Coniditional Build Tags?
Please forget the RH6 fix. The bug that needed this fix was when a table had merged cells in the end column and the build used an expression. That is fixed. I am generating help with such tables and build expressions.
Your problem is not the combination of those two.
Create a new single source layout to generate a CHM. Don't copy this one. Sometimes the SSL can go corrupt and it is as simple as creating it afresh.
I have this same problem with cond'l build tags.
- The build tags in preview mode work fine.
- If I don't use the build tags in the project generation, it works fine
- If I set up the proj generation with any build tag, it hangs/crashes.
Another posting suggests first deleting hte CPD and XPJ files, then opening the project and compiling. This did not work for me.
Can anything be done about this??
Have you tried creating a new layout?
Maybe delete your SSL and create a new one?
Oops! Sorry Peter. Didn't see you standing there.
Nope. I got a bit further in the "updating files" stage, but it didn't complete. I'll research a bit more and then call Tech Support. I'll post if I get good info. Thanks, guys.
You have your project on your hard disk and are generating to your hard disk?
You are using RH7?
I too get crashes during help generation, always during the "Updating" phase of the process. Whenever I think I've found the culprit -- conditional build tags, broken link, multiple windows, single sourcing, DHTML weirdness -- I end up getting another crash later on. Some projects crash more often than others.
My current suspect: variables.
We ran into the same problem with RH7 people have described here the other day. And yes, it had to do with rowspans in tables in combination with conditional build tags. Maybe the problems was only partially fixed for RH7. It seems to definitely exist in certain situations. When compiling the project, the compiler searches for instances of conditional build tags, and during the process "stumbles" about code that has too many <tr> tags. I guess that the code that was created in this topic had been created with RH X5 and that RH7 would no longer create this type of faulty code. Anyway, here is how we located the problem, hopefully this may help anyone in the future. I'm sorry that the description is very short but I didn't do this myself, so I don't have any more details:
- Used Sysinternals' FileMonitor or ProcessMonitor to identify the HTML file that was processed by RoboHtml.exe when it stopped working
- Used an HTML editor to remove those rows of a table in the document that belonged to a rowspan. (We had td rowspan=5, and therefore removed five <tr>...</tr> sections): Then, it compiled.
- Removed rowspan, and replaced corresponding <td_null> by <td></td>. Then, it compiled.
The most important step is to locate the corrupt topic. There are probably multiple ways to "fix" the table/rowspan problem. It's possible that recreating the table with RH7 would be enough to fix it.