How do I find and suppress broken hypertext marker links?

Report · Jan 25, 2012

Hi,

I have a .book with 27 .fm chapters. Typically, our books go online as a single PDF plus 27 chapter PDFs. Various text in each chapter uses hypertext message URL markers to link to other chapters. Marketing has decided to send the single PDF to early-access customers well before the chapters will be online to link to. So, clicking the links in the single PDF give the 404 not found browser error. 😞

What we are looking for is a way to post-process the PDF file and suppress all the dead links. Can Acrobat Pro v9.4 do this? Is there another tool available? Thx.

Also, if there is a super-easy way to change the 100 hypertext markers to internal cross-references, we'd consider that, but I can't see how and it would only be a temporary solution because we'll have to put them back when the files go online. What would be even better is to have a straightforward way to make a link internal in the single PDF and external from each individual chapter. But again, the pressing issue is stopping the 404 errors.

Thanks!

Dave

Report · Jan 25, 2012

Dave,

It's not clear to me if the "single" PDF is the sum of all of the individual chapter PDFs or something different that links to 27 separate chapter files. If it is the sum, why do you need separate manual links to the chapter files? FM can maintain the links for either route (single or chapter file PDFs). What do your hypertext markers look like?

Report · Jan 25, 2012

Single- PDF is the sum of all chapters plus some front and back matter. We need separate links because typically our handbooks are available online as individual chapters and as the entire handbook. So when in an individual-chapter PDF, the link needs to jump to the appropriate other individual-chapter PDF. For those we use markers of type Hypertext with the message URL http... format for the markertext.

Yes, I understand that FM handles my single and chapter cases separately. We've always just "bitten the bullet" on having the link in the single PDF jump to the individual-chapter PDFs, rather than stay within the single-PDF file, because it was too much of a hassle to try to set up conditional text on every (hundreds) of links (to have a cond text marker link for the indv chapters and an internal cross-reference for the single). If there is a way to do this in one shot, I'd love to know how! 🙂

Update for the 404 pressing issue: we found a tool called Aerialist which seems to supress the dead links. We are testing it now. Again, this is just a short term solution. Once the chapter PDFs go online, we will want the links back.

Report · Jan 25, 2012

You don't need to use separate links or any conditionals at all. If you tell FM to create the PDFs as separate chapters, then it will automagically adjust the link to point to the appropriate chapter PDF file.Using cross-refs, this is a snap. If you use hypertext links, then as long as you link to the appropriate .fm file, FM will also take care of things. However, if you've hardwired to PDF files in your hypertext links, then you're creating a lot of work for yourself.

As I asked earlier, what do your URL links look like? Could you post one sample of your jump from the book to a single chapter.

Report · Jan 25, 2012

Sorry about misnderstanding your link example question. I thought you meant the markertext format. A typical link looks like http://www.mycompany.com/pdfs/myhandbook.pdf. and the markertext looks like messageURL

http://www.mycompany.com/pdfs/myhandbook.pdf.

What you say is VERY interesting to me. So are you saying that when we make the chapter PDFs we should highlight the chapter in the bookfile, then use File > Print Selected Files? And in the files, we should use the cross-reference (Source Type: Paragraphs is what we usually use for internal xrefs) format?

I imagine this means that on the web the files must be in a specific realtive position to each other for the links to work, correct? In the multi-file case, do the chapter-PDFs store relative or absolute paths?

Report · Jan 25, 2012

The paths are created relative to each other. However, going from one PDF to another via a weblink is a bit dodgy, IMHO. What if the user prefers to download the PDFs before opening? The links will take them back up to a web page and would probably PO the user.

Anyway, when creating multi-chapter PDF output you have two routes - SaveAsPDF (Check the "Generate Separate PDF Document for Each Selected File" option) or from the print book dialogue enter only an asterisk for the file name (not *.ps).

Cross-refs will render correctly and be created as relative links.

In the PDF files, you can then set the Base URL (Document Properties) to lock into your site PDF library, if required.

You might also want to have a look at Shlomo Perets' TimeSavers tool. This will let you maintain PDFs from FM in a more robust manner. See: http://www.microtype.com/timesavers.html and have a look at the TimeSavers Shortcuts PDF on that page.

Report · Jan 25, 2012

Thanks a bunch! I'll see if I can get this to work and report back. 🙂

Report · Jan 31, 2012

Hi Arnis,

I'm getting a little resistence from co-workers. Your wisdom can help! 🙂 What happens with your methods when the user downloads just one chapter to his/her local machine, opens it, and clicks the xrefs to other chapters? Will setting the Base URL force it to try to go to the web when there is no local copy of the destination chapter?

Report · Feb 01, 2012

Dave,

The Base URL only affects links that are defined to jump to URI's not to other (local) Files. If you create a "message URL myfile.pdf" in a FM file and supply a base URL for the generated PDF, then the file will look for the PDF file at the base URL address both when it is on the web server and when it has been locally downloaded. However, if you don't supply the base URL and use just the filename, then while the file is on the web server the link will continue to work (as it is still relative to the other chapter PDFs - hopefully). However, the downloaded chapter PDF will try to find the link locally at file:///<drive>|<path>/myfile.pdf

I don't completely understand your requirement for having both a single PDF book file and the individual PDF files on the web. If the "book" PDF file has been optimized for web serving, then links (as a user browses the file) take them to the correct destination and only download the appropriate page as required. If you want to provide multiple points of entry to your PDF book, then instead of having a separate chapter PDF to start in, you only need to provide a link to the specific page in the Book PDF from your web page. In the Book PDF case, you can then build the internal cross-chapter links using cross-refs.

If your single chapter PDFs serve some other purpose from your Book PDF, then suppressing the dead links using a third-party PDF tool, like Aerliast, is probably the best way to go (if you don't want to invest in some scripting solutions in FrameMaker).