Highlighting words/phrases in PDF XChange Viewer

bcraytor1 · Post by **bcraytor1** » Thu Dec 18, 2008 10:15 pm

dtSearch provides a function "MakePdfWebHighlightFile" that generates XML data used to highlight hits in Adobe Reader. The XML data file/string looks like this:

<XML>
<Body units=characters color=#ff00ff mode=active version=2>
<Highlight>
<loc pg=37 pos=911 len=9>
<loc pg=40 pos=1416 len=8>
<loc pg=42 pos=339 len=8>
<loc pg=70 pos=646 len=8>
...
</Highlight>
</Body>
</XML>

Note that dtSearch can be instructed to skip certain fields in calculating offsets with an instruction such as:

searchOptions.FieldFlags = FieldFlags.dtsoFfSkipFilenameField |
FieldFlags.dtsoFfSkipDocumentProperties;

Is there any way to pass this xml data or a translation to PDF XChange Viewer to have it do similar highlighting?

bcraytor1 · Post by **bcraytor1** » Thu Dec 18, 2008 10:20 pm

I guess another way to ask the question is: "Does PDF XChange Viewer support Adobe's PDF hit highlighting API."

- Also, we are developing for Vista 64.

Post by **Vasyl - PDF-XChange** » Fri Dec 19, 2008 6:06 pm

Is there any way to pass this xml data or a translation to PDF XChange Viewer to have it do similar highlighting?

This possibility can be added into the next build (NBuild>=41) as:

AX(32|64)->HighlightText(XMLFile, Flags);

bcraytor1 · Post by **bcraytor1** » Mon Dec 22, 2008 8:35 pm

That sounds great. What is the probability of that happening and do you have any idea of the time frame?

Post by **John - Tracker Supp** » Tue Dec 23, 2008 9:06 pm

It will be added into build 41 and this should be available in late January or possibly before,

Hope that helps.

bcraytor1 · Post by **bcraytor1** » Tue Dec 23, 2008 10:07 pm

Thanks. That will certainly work for us.

bcraytor1 · Post by **bcraytor1** » Wed Dec 24, 2008 12:39 am

By the way, just as a nice-to-have, but certainly not a requirement, is there any way that the XML file could be used to create the same kind of search window for multiple hits that you currently have? That is, as an alternative to highlighting multiple hits in the PDF?

Wed Dec 24, 2008 10:56 am

The full search and highlight functionality will be added into the next build (NBuild > 41).

..is there any way that the XML file could be used to create the same kind of search window for multiple hits that you currently have? That is, as an alternative to highlighting multiple hits in the PDF?

For now there is no window similar to the existing search window. In the future you will be able to use the search results (which will be contain complete info) for representation in your UI.

Thanks.

bcraytor1 · Post by **bcraytor1** » Wed Feb 18, 2009 12:29 am

I'm looking at build 41 and don't see any HighlightText() function. How is highlighting supposed to be done now?

Wed Feb 18, 2009 10:03 am

This feature is ready for usage.
We will to update AX help today, samples for this will be included also..
Please wait for new help.

Post by **Vasyl - PDF-XChange** » Wed Feb 18, 2009 7:09 pm

Look to new AX Help, bookmark:

Reference\Named Items\Named Objects\Documents\<Item>\Pages\<Item>\Text

Post by **John - Tracker Supp** » Wed Feb 18, 2009 9:13 pm

Unchecked help file sent by email.
This will be added to the SDK download on Thursday - once checked/edited.

bcraytor1 · Post by **bcraytor1** » Wed Feb 18, 2009 11:37 pm

I got the .chm file. It appears I should be using C# code such as:

try
{

AxCoPDFXCview1.DoVerb("Documents[#4095].Pages[0].Text",
"Highlight",
new x(1,
10,
0x800080,
0.65),
out dataOut,
0);

}
catch (Exception ex)
{
TraceErrorMessage(ex, out nRes);
if (zShowErrorMessage)
ShowErrorMessage(ex);

}

But this returns "Wrong Operation". Can you tell me what, if anything, is wrong with it?

bcraytor1 · Post by **bcraytor1** » Wed Feb 18, 2009 11:56 pm

BTW, just for test purposes, for the above code x is defined as:

class x {
private long first;
private long count;
private long color;
private double opacity;
public x(long zFirst,long zCount, long zColor, double zOpacity)
{
first = zFirst;
count = zCount;
color = zColor;
opacity = zOpacity;
}

Thu Feb 19, 2009 11:24 am

To pass more arguments through DoVerb you should pack them to an array of object's (that is to SAFEARRAY, as described in AX Help):

Code: Select all

object dataIn = null;
object dataOut = null;
object[] args = new object[4];
args[0] = 1;
args[1] = 10;
Color c = Color.FromArgb(0,0,255);
args[2] = ColorTranslator.ToWin32(c);
args[3] = 0.75;
dataIn = args;
AxCoPDFXCview1.DoVerb("Documents[#4095].Pages[0].Text", "Highlight", dataIn, out dataOut, 0);

bcraytor1 · Post by **bcraytor1** » Thu Feb 19, 2009 6:46 pm

OK, I got this to work. However, there is a MAJOR problem. Your offsets don't conform to the Adobe PDF Hit Higlighting Standard. So, the offsets returned by dtSearch don't work when they are passed to your routines. Is there any work-around?

Fri Feb 20, 2009 10:08 am

As I understand, here can be problem: for collect all text from document, we used own mechanism of text-collecting (this mechanism is undocumented, no Adobe documentation for it).
For test this case, please send me example of search results which received by you from dtSearch engine, and your code for highlights these results.

Also, in near future I planning to present AX-interface of own search engine..

Thanks.

bcraytor1 · Post by **bcraytor1** » Fri Feb 20, 2009 3:07 pm

Well you stated at the top of the thread that would provide an API that would support Adobe's PDF Hit Highlighting Standard. So, that was very disingenuous and misleading.

dtSearch is probably the most widely used search engine for PDF files and it is not likely that you will be able to compete with them any time soon. It would make you far more competitive to provide compatibility with dtsearch, by supporting the Adobe standard. It appears however, that you are taking the same path as Foxit in this regard - and I don't see that as an advantage.

We have large amounts of files to search and index and are dependent on dtSearch. Therefore, it looks like our only alternative here is to simply provide a 32 bit version of our program for Vista 64 until Adobe comes out with a native 64 bit version of Reader.

Fri Feb 20, 2009 5:06 pm

Let make clear what would you like to have: just a function which will take XML file (or memory buffer with XML) as argument and makes highlighting?
Ok, but in this case there is problem - we cannot guaranty than indexes of words/chars will be the same into our viewer and into Adobe. (if dtSearch uses adobe library for working with PDF files, for sure, indexes will be the same into Adobe and dtSearch. But we don't use adobe library).

bcraytor1 · Post by **bcraytor1** » Fri Feb 20, 2009 5:28 pm

I think the second post in this thread made it clear that dtSearch supports the Adobe Pdf Hit Highlighting Standard. That's why the dtSearch XML output file of offsets can be fed directly to Adobe.

In fact, your highlighting function is totally useless - since you don't provide the means for acquiring meaningful offsets. It currently has no purpose for anyone outside of your company.

So, you really need to read and follow the Adobe Pdf Hit Highlighting Standard if you want to be compatible with dtSearch and other tools which use the standard.

For test purposes, make sure to use Pdf files with image and other embedded data. It would seem you could get an eval version of dtSearch for test purposes.

Bert Craytor

Post by **Vasyl - PDF-XChange** » Fri Feb 20, 2009 6:19 pm

Ok. We understand you.
We will to support #xml parameter in input URL when document opening (as described in http://www.adobe.com/devnet/pdf/pdfs/Hi ... Format.pdf).
Please wait 1-2 days.

bcraytor1 · Post by **bcraytor1** » Fri Feb 20, 2009 7:15 pm

Well, per the standard the recommended offsets are in characters. If I called your existing function with offset = 0, it would often land on the 3 or 4th character. So, you would have to change your offset to reflect the actual character (page,character offset, character length) position, i.e. offset= 10 means the 10 character on the page. - Just to be clear.

Sun Feb 22, 2009 3:27 pm

While methods (algorithm) of reconstruction text from PDF page isn't standardized (yes, sequence into which text is shown of PDF page isn't the same as this text is written into content of PDF page), we cannot say that the character (or word) with our index, say, 10, will be the same as into Adobe.
In other words, it is impossible to say that Adobe's hit-highlight format is a standard.

bcraytor1 · Post by **bcraytor1** » Mon Feb 23, 2009 5:20 pm

dtSearch has been able to generate the XML files which work just fine with Adobe Reader (32Bit). So, it's doable. Maybe you could contact them and get some information to allow your product to be compatible with theirs.

Tue Feb 24, 2009 8:16 am

We will try. But if the are using Adobe PDF Library for extracting text from PDF files, it is "roadblock".

bcraytor1 · Post by **bcraytor1** » Tue Mar 17, 2009 10:33 pm

Has there been any update to highlighting?

PF4PDFX · Post by **PF4PDFX** » Sat Dec 03, 2011 10:21 pm

Hi to the developers and everybody else

Thanks for creating such a great software, I am loving it... The only problem so far is described by the links below
and after searching your website I decided to post in this thread instead of creating a new one. I hope that I am on topic

I am coming from
http://support.dtsearch.com/dts0229.htm
and
http://support.dtsearch.com/webhelp/dts ... files.html

I am trying to understand if this information is correct on their web site and your viewer is not compatible with this desktop search engine in this sense.
Have you made any progress with this feature?

thanks a lot
PF

Mon Dec 05, 2011 2:16 pm

Hello PF4PDFX,

You have posted in a developer topic, and the answer to the (older) developer question is that the function mentioned by Vasyl was implemented as HighlightTextByFile in our Viewer AX.

For the moment our end user Viewer will most likely not work with the dtSearch highlighter, but we are working on a new major version of our Viewer:
https://forum.pdf-xchange.com/ ... hp?t=10100
Which when released will support plug-ins, and I will now pass this to our devs, and we could contact dtSearch and ask them if they are interested in creating such a plug-in for our Viewer as they have done for Adobe X.

Best,
Stefan

PF4PDFX · Post by **PF4PDFX** » Mon Dec 05, 2011 2:28 pm

Tracker Supp-Stefan wrote:Hello PF4PDFX,

You have posted in a developer topic, and the answer to the (older) developer question is that the function mentioned by Vasyl was implemented as HighlightTextByFile in our Viewer AX.

For the moment our end user Viewer will most likely not work with the dtSearch highlighter, but we are working on a new major version of our Viewer:
https://forum.pdf-xchange.com/ ... hp?t=10100
Which when released will support plug-ins, and I will now pass this to our devs, and we could contact dtSearch and ask them if they are interested in creating such a plug-in for our Viewer as they have done for Adobe X.

Best,
Stefan

Hi Stefan

Thanks for replying. I am not sure if I understand what you mean by Viewer AX, I could not find this listed on your product list.
Is this some sort of developer library or API?

thanks
PF

Mon Dec 05, 2011 2:40 pm

Hello PF,

Yes this is a developer product. And as you posted in a developer topic - I did include that information in my reply for all developers that could read this topic.

As for when/if this can be implemented in our end user Viewer - we will need to check this with dtSearch directly and see if they are interested in developing a plug-in for our upcoming Viewer.

We will keep you posted here in this topic with any news on the subject.

Best,
Stefan

PF4PDFX · Post by **PF4PDFX** » Mon Dec 05, 2011 2:50 pm

Tracker Supp-Stefan wrote:Hello PF,

Yes this is a developer product. And as you posted in a developer topic - I did include that information in my reply for all developers that could read this topic.

As for when/if this can be implemented in our end user Viewer - we will need to check this with dtSearch directly and see if they are interested in developing a plug-in for our upcoming Viewer.

We will keep you posted here in this topic with any news on the subject.

Best,
Stefan

Now it is clear, I will patiently wait for the day when that feature will be available to mortals.
In the mean time I will just add here the fact that if you have Adobe Acrobat Reader installed on your system but Xchange PDF viewer configured as Default PDF viewer then you will be OK
dtSearch will use Acrobat Reader to show you where the keyword is (highlighting working) but if you double click on the doc it will be open in the default viewer (XCahve PDF). That is acceptable for now

thanks for the quick reply
PF

Mon Dec 05, 2011 3:20 pm

Hello PF,

Thanks for the hint about using both products

And thanks for the patience!

Cheers,
Stefan

Highlighting words/phrases in PDF XChange Viewer

Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer

Re: Highlighting words/phrases in PDF XChange Viewer