Finding templates was changed to be independent on document DPI. It is now possible to process documents from different sources (different document DPI) by one template.
The possibility of extracting data from line items was added (requires a license for line items). It is possible to set and edit attributes for line items in admin module for individual workspaces, tab “Item attributes”. It is possible to set modifications and default values in a similar way as for regular attributes. Type validation (number, date) is the same as for regular attributes, it is possible to add more validations to the stored procedure.
It is possible to add a SET attribute to line items. The new type is “fixed set” (fixed set of values is allowed in the attribute). In the admin module, you can set the allowed values for the set attribute in the tab “SET values”. Each value has a name (displayed in OCR client) and can have another identifier (for example id of the value from the different system; this identifier can be exported).
In case the template for the document wasn’t found (the document is waiting in the queue for some time) and the user clicks on the document, the system will try to find the template from the set of newly created templates (templates newer than the document). This is useful if the document was processed by OCR sooner than the new template was created and is now waiting in the document list.
In admin module, for workspace you can configure languages, which should be recognized on documents (tab “Workspace”). For workspaces where some set of particular languages is processed, setting just these languages for the workspace can increase the success of word extraction from the documents (limiting the number of languages to be recognized). If you are missing some language in the set of languages, please create the task for adding this language on http://support.socosit.cz.
If the OCR is installed on a server with limited internet access or other restrictions, offline activation can be done by generating the code for offline activation from admin module.
In admin module, it is now possible to set the string attribute as multiline (set the height of string attribute). You can edit this in the field “Edit height ration”. Standard height of the attribute will be multiplied by this value (if you set the value 2.0 attribute will be 2x higher).
In the validation, it was necessary to set the state of the validation by the variable @Result (set to 0 every time when the validation should be evaluated as invalid). Now the validation is invalid (document can’t be exported) if the state of at least one attribute is set as invalid or variable @Result is set to 0 (for backward compatibility). Validations should be easier now (no need to set the variable @Result if some attribute is set as invalid).
The modification "DeleteDashCharsFromLeft" was added for the attributes. This modification deletes all “-“ characters from the left.
For the Custom XML export, it is now possible to set the folder for the custom (XML) file. You can set the path in the ExportData5 parameter. By default, the file is saved to the output folder of the specific profile.
To the template date formats, the format YDM (year, day, month) was added.
For attribute (table AttributeDefaultValue) it is now possible to set the default value "ValueFromLastDocByTemplate" which will set the value from the last document exported by the same template.
Document processing was changed (system of processing more document simultaneously).
Attribute hint (information about attribute) in the OCR client is not now displayed for the user without the “administrator” permission.
In table OcrParams, value MaxConcurrentOcrProcesses, you can now set how many documents can be extracted simultaneously (i.e. how many CPU cores will be used for extracting. Default value is 2, for more powerful servers, you can try to increase this number to process the documents faster).
In admin module, you can now set a “Skip attribute (TAB key)” for the attribute. When checked, this attribute will be skipped in the OCR client if the user is switching between attributes by thy TAB key. It is useful for less used attributes in the attribute list.
Attribute modification BankAccount was adding a slash before the fourth character from the end of the string (the format used in some countries). This was not suitable in the case when the bank account extraction is divided into two attributes – bank account number and bank code. Adding of the slash in this modification was removed.
If the window for selecting a template for the document is opened in the client, it is now possible to move the window (it is not fixed) so the user can still see the text in the document.
In the DMS XML export, it is now possible to set the output file name from the attribute value. In this case, the parameter ExportData must be set to “Rename” and in the parameter ExportData1, the DevIdAttribute of the attribute with the file name must be set.
In the database, table Workspace, you can now set the attribute PageGroupSize for the workspace. The default value is 1000, minimal value 10. OCR Client will be displaying only the configured number of pages of the document at once (in one group of pages). This can be useful for Redact module or for workspaces where documents with more pages are processed (contracts, ..).
In the table Profile, attribute IdOcrTextType it is now possible to configure the type of extracted text (to increase the success rate of extracting special text types), one of this values can be set:
Codes for the VAT id from Netherland and Spain were added to VatId attribute modification.
File in input folder which couldn’t be deleted (invalid, used by another process) caused errors in further document processing. This state was fixed.
Error in displaying the document with different size/DPI was fixed.
Error in creating the license table for pages was fixed. Values for some months were not created and it was necessary to reactivate the license.
Fixed. Document list shows only documents from profiles with set permission for the user.
Fixed.
Automatic export was fixed – error due to changes in export definition.
Fixed, modification now deletes all occurrences of the configured string.