Tables#

On the right workspace panel, all template tables that have been previously recognized are displayed. There can be an unlimited number of them. Each table is automatically assigned a number, starting from 0. The number of columns in each recognized table is also displayed. This value can be used to determine which table(s) to continue working with.

The main tools for working with tables:

No.	Tool	Description
1.	button	Allows you to add a new table.
2.	button	Allows you to copy the selected table.
3.	button	Allows you to delete the selected table.
4.	button	Allows you to bind an Anchor to the selected property.
5.	button	Allows you to remove an Anchor from the selected property.
6.	button	Allows you to add a table column.
7.	button	Allows you to delete a table column.
8.	button	Import all tables from the document.

Table Properties#

The workspace panel opens a list of properties and tools for working with the table.

Properties and tools for working with tables:

No.	Property/Tool	Description
1.	Index	OCR index of the table. To improve binding, it is recommended to specify the minimum number of columns. If the Template Extractor extracts data from only one table of the scanned document, the index must be set to `0`.
2.	Skip Rows	If necessary, you can specify the number of initial rows to skip so that the data from these rows do not appear in the robot's results. When selecting a value of `1`, the first row will be skipped. The column names will correspond to the values of the first row. If column names are specified in the template, these values will not be changed.
3.	Header Rows	When constructing the table, the specified number of rows will be used to create the column headers. If a value of `0` is specified, the names will be taken from the column template; otherwise, the name will be summed from the cells of the column based on the specified number of rows. For example, if information from the top two rows of the table should be included in the header, you need to specify the number of header rows as `2`.
4.	Binding Anchor	When working with a document that contains multiple tables, you can specify a binding Anchor of type Optional. The Anchor will point to a word/header/item after which the desired table begins. When such an Anchor is found, the table following it will be recognized.
5.	Finish Anchor	The Finish Anchor limits the extraction of the table to the Anchor, which can be of type Optional. For example, a row in the table "Total Amount Due" can serve as such an Anchor. The table will be extracted from the document up to the finish Anchor.
6.	Upper Boundary Anchor	The Anchor after which the table will begin to be constructed.
7.	Lower Boundary Anchor	The Anchor up to which the table will be constructed.
8.	Minimum Number of Columns	The minimum number of columns in the OCR table. All tables with fewer columns than specified will be skipped.
9.	Maximum Number of Columns	The maximum number of columns in the OCR table. All tables with more columns than specified will be skipped.
10.	Exact Column Equality	When this setting is enabled, only those tables will be extracted where the number of columns in the document is less than or equal to that specified in the template.
11.	Columns on the First Page	When this setting is enabled, columns will be taken only from the first page. This setting indicates that there are no headers on subsequent pages for the table.
12.	Merge Tables	When this setting is enabled, tables with the same number of columns will be merged. If the setting is not enabled, such tables will be added as new.
13.	Error if Table is Empty	When processing the template, if the table is empty, a recognition error will be generated.
14.	Columns	Table columns. The corresponding number of columns is displayed for each table. Also, the following will be indicated here: Index (automatically assigned); Name (assigned by the user); Data Type: Text – text; Int – integer; Float – floating-point number; Date Time – date and time; Chars – letters only (digits and other symbols are excluded). If several columns are missing in the results of the full scan, we can add them.