Highlighting search terms within PDF files

[vallinem]

We're using the default search form; when I have a search where the results include PDF files, when I click on the PDF file different things happen depending on the PC.
1) PC 1: Adobe opens and takes me to the beginning of the file
2) PC 2: Adobe opens and takes me to the point in the file where the search term is; generally it highlights the actual search term, but in one case it highlights the line above the search term. Also, the highlight color is so dark that you can't really read the search term through it.

Is this something we can control at all through Webinator, or is it a client-specific problem?

[mark]

It's mostly under the control of adobe's pdf plugin in your web browser. Make sure you have a recent version of the plugin. You can get it for free from www.adobe.com.

There's a way to set the hilight color in the search script, but in our experience, adobe mostly ignores it. In the search script there's a call to "pdfxml" to generate the word postitions for hilighting. The 3rd argument "#00FF00" is the color. You can try changing it. A PC display with few colors (256 or less) may be beyond hope of getting a readable color.

[doran]

The feature Adobe calls "byteserving" may also affect some of the behavior you're seeing with pdfs.
http://www.adobe.com/support/techguides ... vmain.html

[josmani]

I would like to do hit-highlighting on PDF documents that I store in one of my database table. I am also storing the extracts of the PDFs in another column (using anytotxt).

Can I use "pdfxml" to do this? I do not use Webinator.

[John]

Yes you can. Looking over the Webinator scripts for pdfhi and pdfxml may help show how it can be used. Basically you add #xml=URL to the url where the URL points back to the script that outputs the result of pdfxml on the column you stored.

[jgdoke]

First: I have tested pdf highlighting and do not see any highlighting. PDF highlighting is turned on..

http://www.rockwellautomation.com/produ ... 45d7461b5b

Second: I turned off pdf highlighting and the pdfhi.txt part of the url did not go away. is it supposed to??

[mark]

The hostname for the pdf hilight seems wrong. It has "usmkeapln002.rockwellautomation.com" whereas your search is performed on "www.rockwellautomation.com". usmkeapln002 doesn't resolve and www doesn't understand the request. Are you republishing the results? If so you need to modify that hostname to resolve to a url that will get the highlight info from the search server.

http://usmkeapln002.rockwellautomation. ... 45d7461b5b

[jgdoke]

I was showing a result from our internal dev appliance.

Here is a search from our external box:
http://search2.rockwellautomation.com/t ... q=&jump=10

[mark]

Same problem more or less. The appliance doesn't know it's name so it's reverting to 127.0.0.1. Set the hostname on the appliance to agree with it's dns name.

#xml=http://127.0.0.1/
needs to be
#xml=http://search2.rockwellautomation.com/