Odd Behavior with "Control Online Help" Function

[smercurio_fc]

So I'm doing a little playing around with opening compiled help files (.chm) files with the Control Online Help function and I've seen a strange behavior when trying to acces multi-level index keywords. According to the documentation, if you set the operation to "Key", you can enter a value for "string to search for" to jump to that topic. The help further states that:

 If this string contains an index keyword, the string must match the index keyword as it appears in the help file index. To jump to a topic with a multi-level index keyword, enter the first-level index keyword, a colon, and the second-level index keyword. Do not separate the elements with spaces. For example, a error codes:GPIB string jumps to a topic with a first-level index keyword of error codes and a second-level index keyword of GPIB.

 All right, seems straightforward enough. Except that when I tried it, I couldn't get it to work, even with the example provided. I tried the following simple test:

 

The help file opens, but it opens to this non-descript page, not to the "error codes:GPIB" page. I've seen this in 8.2 and 8.6. Anybody have any success with accessing a multi-level keyword?

Side-note

Interestingly enough, the Index for the lvconcepts.chm help file looks like this:

 

In other words, the index entry text is just "GPIB", but the text that shows up  in the keyword field is "error codes, GPIB". I'm curious to know how this was accomplished. I decompiled the lvconcepts.chm file and looked at the .hhk file, and it's not immediately obvious to me. (By the way, if you try to use "error codes, GPIB" it doesn't work either). 

Message Edited by smercurio_fc on 10-15-2008 03:19 PM

[smercurio_fc]

EDIT: For the above post, wherever you see "lvconcepts", it should be "lverror". Doesn't make a difference, though, as the behavior is still the same.

[Root_Canal]

hmmm, I see what you mean. using the html link address as the keyword worked for me:

Message Edited by Root Canal on 10-15-2008 05:37 PM

[smercurio_fc]

Well, yes, that will work, but you may not know what the internal structure of the .chm file is unless you created the .chm file in the first place or you decompile it, as I did. That's why the keywords are there.

Still think there's something funny about that function. NI?

[NI-Bob · Accepted solution]

Hello,

Thank you for contacting National Instruments!  This is very strange behavior.  I have tested it out and seen the exact same results that you are experiencing.  This was reported to R&D (CAR ID# 130246) for further investigation.  From what I can tell the one possible workaround we have is documented in the post by Root Canal.  I can continue to look into this in order to see if there are any other workarounds or a reason why it is executing in this manner.  Thank you very much for this feedback!  If you have any further questions on this issue please post them here.  Have a great day!

[smercurio_fc]

Thanks for confirming the behavior and opening a CAR. I don't really consider the mechanism of specifying the HTML file that's contained within the .chm as a "workaround" since as I said, one doesn't know the names of the HTML files unless one decompiles the .chm file. That may be applicable if you're accessing your own .chm file which you created, though.

[Root_Canal]

I agree. While it is technically possible to dig around and find the HTML file from a decompiled .chm, I think it is a bit impertinent to label this as a solution. I think that this unintended behavior should be regarded as having no 'acceptable' workaround, so that a suitably high priority may be placed on corrective action.
 
-root

[smercurio_fc]

I wasn't marking the workaround as the solution. I was marking the confirmation that it's a bug and that a CAR has been created for it. Unfortunately, the "Accepted Solution" mechanism applies to the whole post, and I felt it was better to mark it as a solution because the bug was confirmed, rather than not mark it at all.

[Root_Canal]

Ah, I see. I had the wrong impression about how to use the accepted solution marker.

[NI-Bob]

Hello,

I would agree that the "workaround" is not an ideal solution because we will not always have access to the structure of the HTML files.  At this point I have not discovered another workaround for this issue and I am not sure if there will be one that will be sufficient.  I will continue to look into this and update you as soon as I find any new information.  Thank you very much for your patience on this issue.  If you have any further questions or suggestions please post them here.  Have a great day!