PDF Search Term Highlighting in preview window

[DaVo]

Hi all,

I already posted my problem here: (http://forums.x1.com/viewtopic.php?f=48&t=5014) as it is exactly what I experience, but as it is the wrong board I would like to post it here again:

I'm currently searching for a supporting tool to find parts of text in my scanned PDFs as I'm trying to go for a paperless office. For this, I need a powerful search tool with PDF-indexing. I hoped to find such in X1 search and really like the GUI, the way of searching and the possibities like "Open folder". What I really don't like is the fact, that the file preview of X1 does not highlight the search terms that I was searching for. I learnt already, that this has something to do with the way the files are scanned. But honestly: The competitors that I'm trying out concurrently (Copernic Desktop search and Lookeen) indeed DO highlight the correct words in the same files. So this is definitively a matter of X1. Where I have to say that I generaty my PDFs with Kodak Capture Pro, which is a high quality professional tool and is not in discussion to be changed

So could anyone tell my whether there are plans to change the file preview of X1 to be able to highlight the search terms in the future? Otherwise I regret to be forced to take one of the (in many other cases worse) competitors.

[Kenward]

You will not get a different answer from the one posted in the other section.

What you are seeing is a "feature" of scanned PDF files. They consist of an image of the scan with a text overlay.

X1 seems to search the text overlay to find words but displays the image. It is not capable of presenting the two at the same time.

X1 uses Inso viewers to show the files it finds. These are also the viewers used in QuickView Plus. Unfortunately, these can't highlight words in scanned PDFs either. I'm not sure that it can even find words.

As I have said already, X1 is looking into using newer viewers. There is a new version of QuickView Plus, but it now requires an expensive annual licence, so I am not going to investigate that to see if it fixes this defect.

Out of interest, which index and search software have you found that can highlight text in scanned PDFs? The only software that seems to come near X1 is dtSearch, albeit much more expensive.

[DaVo]

First: Thank you for your support. Please don't fell offended, I just thought that probably anyone from X1 is looking into this board and can tell anything less-rumoric. For example it would be very interesting to know some timeline for "the new viewer". This year? 2020?

As I have said already, Copernic Desktop Search as well as Lookeen can both highlight the text in the same scanned PDFs, that X1 is obviously only showing as image. So it is working in some way and it would be very sad if X1 cannot solve this in the near future (especially as I'm not the only one looking for this feature).

[Kenward]

If the X1 team wants to add more to the information you have seen already, they will contribute to this discussion. They visit the forum regularly and will contribute when they have something that they want to add. For a personal response, send them a support request.

[tophee]

This is indeed a deal breaker. Are there any news about switching to a new pdf viewer?

[Kenward]

X1 uses the same same Stellent viewers as other software houses.

I have just looked at two alternative viewers that also use Stellent technology.

Both have the same defect and show PDF files with the wrong characters.

By coincidence, one of those viewers comes from QuickView, the people behind the viewer technology, Oracle.

In the past I have updated to the paid-for version of QuickView, QuickView Plus. That has sometimes mended these defects. This year, the turned QuickView into an expensive "subscription" product. So I passed up on the upgrade.

Looks like Oracle is shafting the software industry with its defective viewers.

Time for X1 to do something about this.

[Peter Pan]

I just thought that probably anyone from X1 is looking into this board and can tell anything less-rumoric. For example it would be very interesting to know some timeline for "the new viewer". This year? 2020?

Yeah! Maybe 2020 will be the right time to get this working! Because in june 2019, this "bug" is still there! I will give a try to the trial version of QuickView Plus 2017. Just by curiosity as it does not integrate in X1 and I can "simply" open files in the result panel in Adobe Reader or any other PDF reader and search again for the same patterns. AFAIK QuickView would not save any time in this process.

Another defect: The viewer cannot render fonts which are "Custom" encoded like NimbusSanL (embedded subset). Rendering is incorrect in the preview window with another police but wrong characters making it unreadable although indexing, searching and highlighting found patterns work fine! But found then highlighted patterns remain unreadable.

I find this defect more annoying because I can't read neither the preview nor the highlighted patterns found in it.

Cheers from France

[Simon C]

It's quite poor in this respect, isn't it? I've got well over 600,000 pdfs that consist of searchable images in this kind of way.

If I open that kind of PDF in a free PDF reader such as Acrobat or Foxit Reader and search for some text it has no problem highlighting the position of the found text even though the text itself is on an invisible layer.

I can understand that there might be an issue in displaying the text, because it's invisible, but why that should affect the ability of X1 to display and position its usual kind of highlight to show where the text is, remains something of a mystery when it's bread and butter to any number of PDF viewers. And so long as that highlight isn't fully opaque and you can see something of the background image version of the text through it job's done, isn't it.

[DaVo]

It's just ridiculous - especially the "support" here.

I'm really glad that I decided not to buy a licence after the test-period. Other products have developed quite good in the meantime...

[Peter Pan]

One year and a minor X1 update later, from 8.6.1 to 8.6.2 nothing has changed: the integrated PDF viewer cannot render a preview properly for some PDF files. I do know that this statement does not match exactly the issue encountered by the OP. Nevertheless, we both have the same result : some PDF files can't be viewed correctly in the X1 viewer. The "font" displayed is pure blank or changed to a one with wrong characters, making the matched file unreadable.

In my case, the searched text has been correctly indexed and it is found then highlighted in the viewer but what is highlighted is unreadable. Only for some PDF files.

And I do now know that this is related to the way X1 does wrongly handle the fonts in these PDF files, as suspected from the beginning, a few years ago. To confirm thos behavior, I did the following: I simply changed the wrongly displayed fonts with standard ones like Arial or Courier then the viewer displayed the file correctly. The wrongly displayed fonts may be standard or very specific ones. I believe that this issue (bug) is more related to the way the X1 viewer does handle the fonts to display them than to the fonts themselves.

It's surprising and disappointing that X1 has not been able or interested in resolving this over years. Especially when the viewers of the competitors do better job in viewing the files.