Output
The output determines what happens to the scanned document at the end of a workflow — which folder it is saved in, whether it is sent by email or which file format is used, for example. You can define several outputs and tie them to conditions, so that invoices end up in a different folder from delivery notes automatically.
Permission required
To access this function you need the following permission:
- Show outputs (
AdminUI_WorkflowManagement_ListOutputs)
You grant permissions through groups — see permissions – overview.
The output is the last step in the processing of scanned documents. You can tie the output to conditions that draw on the document content recognised (text, barcodes or the form type) or on metadata (who did the scanning? which device was it scanned on?).
You can create any number of outputs, with some of them carried out only where their conditions are met. You can equally set up several outputs in order to file copies of a document in different places — even in different colour depths and file formats.
Page stacks and batch processing
Section titled “Page stacks and batch processing”When one or more pages are passed to Docuflair’s output processing, the processing begins with the first entry in the list of outputs. Depending on the check of the conditions, an output is actually carried out (by saving a document, for example) or — where the condition is not met — the further processing of that entry is skipped. In that case the output processing moves on to the next line and there checks any conditions and a possible output anew.
Note the following basics:
- Outputs do not have to be tied to conditions (an output can take place always, if that is what you want).
- Within an output you can remove pages processed from the stack once the processing step has finished (after saving, for example), so that they are not processed further in the steps that follow.
- The order of the outputs matters. Where pages are removed from the stack in a line further up (an earlier working step), they can no longer be reached in the steps that follow.
- When the output processing reaches the end of the list, every page still in the stack is discarded and not processed further. Every page that was not saved in one of the processing steps is therefore lost.
For that reason we recommend adding a working step at the end of the output list that files every remaining page at a storage location, without any condition at all.
You can change the order of the entries at any time: by drag and drop — by clicking the number in the Order column — you move an entry to any position you like.
How it works
Section titled “How it works”Docuflair processes incoming scan jobs with one or more pages as a stack, which is passed from top to bottom to each output in turn for processing. Where a condition has been set, every page of the stack is run through and checked against that condition. In the next working step, every page with a “true” result is marked and combined with the pages that follow into blocks, which are passed on together to the final processing (file at a storage location, send an email and so on).
Creating a new output
Section titled “Creating a new output”You add a new output with a click on Create output.
Enter a name for the output in the Name field. We recommend a name that says as much as possible (“PDF output”, for example).
In the delivery type list you determine what is to happen to the pages, or where they are to be delivered. The following options are available:
- Discard the document
The pages in this stack are not saved. - File at a storage location
The pages are delivered to a storage location in the file format you want — to a local drive or a network drive as well, for example. - Use contact information
Delivers the pages scanned to a contact from the address book. - Pass to a workflow
Forwards the pages to another workflow. - Send an email
Delivers the document to an email address. The document can optionally be sent as a link, to keep the size of the email down. The recipient then has to download the document (which requires network access to the Docuflair server). - Pass to an external application
Starts an external application and passes the name of the file as a parameter. The external application should delete the temporary file by itself once the processing has finished.
Various placeholders are available for generating folder and file names; they are filled with content when the file is saved. A click on Placeholders shows a list of the possibilities available.
| Placeholder | Description |
|---|---|
| {variablename} | Inserts the value of a variable that was filled in the course of the recognition |
| %q | A random 5-character string |
| %b | Inserts, at this point, the value of the variable a condition was checked against |
| %e | The creator’s email address |
| %u | The creator’s name |
| %o! | The original file name without its extension |
| %o | The original file name |
| %r | The directory from a recursive search |
| %T | The creation time (the time of the scan) |
| %d | The current date in the format day.month.year (dd.MM.yyyy) |
| %t | The current time |
| %hh | The current hour |
| %mm | The current minute |
| %ss | The current second |
| %DD | The current day |
| %MM | The current month |
| %YY | The current year (2 digits) |
| %YYYY | The current year (4 digits) |
| %W | The weekday in German |
| %w | The weekday in English |
| %p | The number of output pages (pagecount) |
| %g | A sequential 5-digit number starting at 00000 (checks the file names that exist) |
| %S | A sequential 7-digit number starting at 0000000 (a global counter for scan jobs) |
| %6S | A sequential 6-digit number starting at 000000 (a global counter for scan jobs) |
| %5S | A sequential 5-digit number starting at 00000 (a global counter for scan jobs) |
| %4S | A sequential 4-digit number starting at 0000 (a global counter for scan jobs) |
| %3S | A sequential 3-digit number starting at 000 (a global counter for scan jobs) |
| %X | The number of seconds since 1.1.1970 |
The table shows the most common placeholders. The full list, with the substitution order, examples and limitations, is under filename placeholders.
The email notification lets you have users informed by email when this output has created a document.
The Release the document by hand option is only available where the scan release option has been licensed. With this option switched on, the document is indeed processed but held back until an operator with the right permission reviews the documents through the Documents/Scan release menu entry and releases them. Where needed, the operator also has functions available for reworking the scanned document.
With Remove blank pages you exclude pages with nothing printed on them when saving in this output. This function analyses every image in order to establish whether a page contains information or is entirely empty. The detection works essentially by counting the pixels that are not white. Where a given proportion of white, as a percentage, is exceeded, the page counts as blank and is removed.
Conditions
Section titled “Conditions”Configuring conditions is not compulsory and is only available in the more complex types of scan workflow.
Where no condition has been set, or where the simple types of scan workflow are in use, an output is carried out in every case and, once the processing has finished, the next entry in the processing list is moved on to.
Where a condition has been set for an output, Docuflair checks before creating the output whether that condition is met. If it is, the output continues and the document is saved or sent by email, say, depending on the delivery type chosen. If the condition is not met, the processing of that output is abandoned and continues with the next entry.
Conditions are normally checked against variables that were filled in the course of the recognition.
Various kinds of condition are predefined in Docuflair. Most conditions require the variable to be checked against to be given. As described under recognition, you can define several searches for different content in different areas of a page. The variable name is how the output processing learns which variable is of particular interest and is to be drawn on for splitting a job into individual documents, for instance.
The examples that follow relate to a recognition configured to search for any 1D barcode, which has found a barcode on a page with the following content: “ABC12345”.
| Condition | Description |
|---|---|
| The search failed | This condition only becomes true where the variable given was not filled on a page in the course of the recognition — because no barcode could be found, for example. |
| The search succeeded | This condition becomes true where the variable given was filled with any content at all, so where some barcode with some content could be recognised, for example. |
| The search result starts with | Becomes true where the string given under content is identical to the first characters of the variable to be checked. “ABC”, for example |
| The search result ends with | Becomes true where the string given under content is identical to the last characters of the variable to be checked. “45”, for example |
| The search result contains | Becomes true where the string given under content is contained in the variable to be checked. |
| Ignore the search result | Regardless of every recognition, this condition is always true, so the output is always carried out. |
| The search result matches a regular expression | Checks whether a regular expression can be found in the content of a variable. |
| The search result matches exactly | Checks whether the content matches the value of a variable exactly. In the example above it therefore only becomes true where the value “ABC12345” is checked for. |
This holds for every check on text content: upper and lower case are distinguished. A check with the condition the search result starts with and the value “AbC” would not become true in the example above.
Options
Section titled “Options”The options are how you control various parameters of the processing — to determine how a workflow is to be worked through, for instance.
| Option | Description |
|---|---|
| Do not process the pages again in the outputs that follow | With this option switched on, the pages processed by this output are removed from the stack and not processed again in later output steps. Requires any condition to deliver the result true. |
| Every recognition separates | An option for processing identical recognitions on consecutive pages. Where it is not switched on, Docuflair ignores identical recognitions (barcodes, for example) on consecutive pages; as long as identical results are recognised, all the pages are combined into one block. |
| Every page contains data for the recognition | With this option switched on, the separation into part-jobs is forced after every page. |
| Remove the page with the data recognised | Removes the page the recognition (of a barcode, for example) took place on. Required where separator sheets are to be filtered out that contain only the information for the separation and no other data. |
| Remove the back of a separator sheet | Removes the back of the page the recognition (of a barcode, for example) took place on. Useful where scanning is double-sided and separator sheets are used whose back is blank. |
| Use only the first recognition in a document | Normally a page stack is split at every recognition. With this option switched on, no further check for recognitions on the pages that follow is made after the first recognition. Where a barcode is printed on pages 1 and 3 of a five-page scan job, for example, the barcode on page 3 is ignored. |
Redaction
Section titled “Redaction”Redaction is how you make sensitive information in scanned documents illegible. Text recognition (OCR) is used to detect sensitive information. Dictionaries and directory services can serve as the data sources. The redaction takes place at pixel level. That makes sure sensitive information is filtered out in every output file format. Using the redaction requires the corresponding licence.
Where you use a dictionary as the data source, you can create or edit it with any text editor. Enter every word that is to be redacted before the output on a line of its own. For words that do not match the entry exactly (because of typing errors or a slightly different spelling, for instance), you can set a tolerance: give a number x at the end of the line, after a comma. When the text recognition finds a word in the image data, it is compared with the dictionary. The number x states whether the word found could be turned into the word in the dictionary with x or fewer substitutions, deletions or additions. If you add the entry mouse,1, for example, house would be found too, because the distance between the words is exactly 1. Where you enter two or more words on one line, those words have to occur together, in relative proximity, for a redaction to take place.
A sample dictionary
Mona Muster <br />7000 Eisenstadt<br />Hauptstraße 9<br />Deutsch,2<br />Englisch,2<br />Französisch,2<br />Italienisch,2<br />Where you use a directory service as the data source, every identity is read out (the list of internal staff, for example) along with their personal data (address, telephone number and so on). LDAP, Active Directory or AD LDS can serve as the directory service. The reading out of the identities can be restricted to individual OUs by entering a corresponding ConnectionString in the Active Directory connection. From the information extracted, Docuflair builds a dictionary and sets up error tolerances by itself, in order to filter out slightly wrong spellings or common abbreviations.
Translation
Section titled “Translation”Where the translation function is licensed, scanned documents can be translated into another language automatically. The translation runs through an external translation service.
| Setting | Description |
|---|---|
| Enable translation | Switches the automatic translation on for this output. |
| Translator | The choice of translation service (DeepL or Pepl/Tilde, for example). |
| API key | The API key for access to the translation service. |
| Input language | The language of the original document. |
| Output language | The target language the document is to be translated into. |
Embedded metadata
Section titled “Embedded metadata”This option is how you embed metadata directly in the output file (in the PDF properties, for example). That is useful where downstream systems are to read the metadata straight out of the file.
| Setting | Description |
|---|---|
| Embed metadata | Switches on the embedding of metadata in the output file. |
| Content | The freely configurable content of the metadata embedded. Variables from the recognition can be inserted through curly brackets ({myBarcode}, for example). |
The metadata file
Section titled “The metadata file”On the Metadata page you can have text and XML files created at the same time, when a document scanned and processed is delivered (transferred to a storage location, for example).
Where the metadata file is to be saved alongside the scan job, choose Deliver the same way as the document. Otherwise you can give a local directory and a file name.
The content of the metadata file is freely configurable. Where you want to draw on data from a recognition, insert the variable name inside curly brackets. Inserting {myBarcode}, for example, inserts the content of the variable myBarcode into the file at that point.