Error -2113263849

PDF-X OCR SDK is a New product from us and intended to compliment our existing PDF and Imaging Tools to provide the Developer with an expanding set of professional tools for Optical Character Recognition tasks

Moderators: TrackerSupp-Daniel, Tracker Support, Vasyl-Tracker Dev Team, Chris - Tracker Supp, Sean - Tracker, Tracker Supp-Stefan

Post Reply
Dorwol
User
Posts: 275
Joined: Mon Aug 04, 2008 5:04 pm

Error -2113263849

Post by Dorwol »

Hello,

one of our clients get only this error code (with every document).

res = OCR_MakeSearchable(doc, Options, 0)
If IS_DS_FAILED(res) Then MsgBox "Error: " & str(res)


The messageBox shows "Error: -2113263849"

So please, where can I find the Error-Description to this error number!?
User avatar
Tracker Supp-Stefan
Site Admin
Posts: 17823
Joined: Mon Jan 12, 2009 8:07 am
Location: London
Contact:

Re: Error -2113263849

Post by Tracker Supp-Stefan »

Hi Dorwol,

You can use the following methods to format an error's description for the OCR Library:
OCRE_Err_FormatSeverity
OCRE_Err_FormatFacility
OCRE_Err_FormatErrorCode

And this particular error means:
"Lineart Cleanup Error"

Best,
Stefan
Dorwol
User
Posts: 275
Joined: Mon Aug 04, 2008 5:04 pm

Re: Error -2113263849

Post by Dorwol »

What exactly is a "Lineart Cleanup Error" ? :shock:
User avatar
Tracker Supp-Stefan
Site Admin
Posts: 17823
Joined: Mon Jan 12, 2009 8:07 am
Location: London
Contact:

Re: Error -2113263849

Post by Tracker Supp-Stefan »

Hi Dorwol,

Not 100% sure myself, so going to check that with a colleague.

We will post back here shortly.

Best,
Stefan
Walter-Tracker Supp
User
Posts: 381
Joined: Mon Jun 13, 2011 5:10 pm

Re: Error -2113263849

Post by Walter-Tracker Supp »

I'm afraid the description that Stefan gave you was incorrect; it is not a "line art" error.

It's an internal OCR error (OCR_ERR_INTERNAL). The most common cause of this is improper placement of the language files. We should have a special, and obvious, error code for that because it is so common. Next version will. You could implement a check in your own projects by looking for the expected language files and returning a nice error if they are missing.

If you are using English, and specify "C:\OCR\" as your language data file directory, you must have eng_pxvocr.dat placed in the directory "C:\OCR\ocrdats\". It is important that all OCR data files be placed in this "ocrdats" subdirectory which is hardcoded (won't be in the next major version, but it's a limitation now and we're too focused on new release development to change it in this version).

Please check that the language files are properly installed on the client's machine, and that your code properly specifies the directory.

If this does not help, perhaps you could send a sample file (or test it on your own development machine(s) to see if it is reproducible).

-Walter
Dorwol
User
Posts: 275
Joined: Mon Aug 04, 2008 5:04 pm

Re: Error -2113263849

Post by Dorwol »

Hi,

I have found now the problem.

When our application (with your DLL) is installed in a folder with special characters, the DLL does not work!

For exmaple:

This Installpath will work with your OCR-DLL >>> "E:\Programme\Mueller\<ApplicationName>"

This Installpath will NOT work with your OCR-DLL >>> "E:\Programme\Müller\<ApplicationName>"

Please can you fix this with next release? Thank you!
Walter-Tracker Supp
User
Posts: 381
Joined: Mon Jun 13, 2011 5:10 pm

Re: Error -2113263849

Post by Walter-Tracker Supp »

I can confirm that this is a bug in the handling of OCR data path names. Some workarounds while we investigate and prepare a fix would be:

1. Use relative pathnames (e.g. ".\") that do not include extended characters.
2. Ensure pathnames do not use extended ASCII or UTF-8/Unicode characters.

This is specifically a problem with language directory names only and does not impact PDF file names or directories.

With the holiday schedule I would not expected a fix until January.

-Walter
Walter-Tracker Supp
User
Posts: 381
Joined: Mon Jun 13, 2011 5:10 pm

Re: Error -2113263849

Post by Walter-Tracker Supp »

We will work on a temporary patch; please email support@pdf-xchange.com and request forwarding to Walter (reference this forum post if necessary), and I will return it to you via email. It changes the way unicode characters in pathnames are handled, but does not fully address the underlying issue (it will only fix issues where the unicode input maps to the current system's extended ASCII code table - so that, for example, chinese characters will still cause problems, but umlauts should be fine). On the surface it sounds like a trivial issue to completely fix, but there are some complexities to address which will take a bit more time. So this patch is just a workaround.


-Walter
Dorwol
User
Posts: 275
Joined: Mon Aug 04, 2008 5:04 pm

Re: Error -2113263849

Post by Dorwol »

Walter-Tracker Supp wrote:please email support@pdf-xchange.com and request forwarding to Walter
Done!

THX!!! :D
Walter-Tracker Supp
User
Posts: 381
Joined: Mon Jun 13, 2011 5:10 pm

Re: Error -2113263849

Post by Walter-Tracker Supp »

I have sent you the update directly by email. A fix will be available for download on the website (for everybody else) in the next day or so.

-Walter
Dorwol
User
Posts: 275
Joined: Mon Aug 04, 2008 5:04 pm

Re: Error -2113263849

Post by Dorwol »

I got another new error code "-1039531032" and the ocr process fails. :cry:

But DSErrorLookup tells me "Specified error code not found!"!?

What is the error description please? :(
User avatar
Tracker Supp-Stefan
Site Admin
Posts: 17823
Joined: Mon Jan 12, 2009 8:07 am
Location: London
Contact:

Re: Error -2113263849

Post by Tracker Supp-Stefan »

Hi Dorwool,

I could not locate the meaning of this error code either - so will ask Walter to take a look - but have you tried using the error code handling described in the OCR help?

Best,
Stefan
Walter-Tracker Supp
User
Posts: 381
Joined: Mon Jun 13, 2011 5:10 pm

Re: Error -2113263849

Post by Walter-Tracker Supp »

The error code name is "OCR_ERR_MEMALLOC" - an attempt to allocate memory for some object or data failed. You have most likely exhausted available memory with an OCR job. Most of the memory allocation occurs on a per page basis (freed up after the page has been processed), so it is likely that the page was particularly large or the resolution was set too high. You could try OCRing smaller regions of the page using fields (if you just want to extract the text), or at a lower resolution.
Post Reply