Grid Layout

Can be one of the following types:

The 'X Axis Extractor' property defines how column headers are detected at the top of the table. This extractor should match the entire header row and return sub-instances for each individual column.

• Use a regular expression with named groups or a Collation Provider such as Ordered Array to extract column headers.
• If 'Transpose' is set to false, the names of the sub-instances should match the names of the corresponding Data Column objects.

Configuration Details

• The extractor is typically a Pattern Match or List Match configured to match the full header row.
• Named groups in the pattern (e.g., (?<Revenue>...)) will be mapped to Data Column names.
• If using a Collation Provider, ensure the order and naming of the sub-instances aligns with your table columns.
• The extractor must be robust to variations in header formatting, such as extra whitespace or merged cells.

Practical Tips

• Test your extractor on multiple sample documents to ensure it reliably matches all expected header variations.
• If a column header is missing or misspelled in the document, the corresponding column may not be extracted.
• For tables with dynamic or optional columns, consider using a pattern that allows for missing groups.

Common Pitfalls

• If the extractor returns sub-instances with names that do not match any Data Column, those columns will be ignored.
• If the extractor matches too broadly, you may get extra columns or misaligned data.

Proper configuration of this extractor is critical for accurate column mapping and data extraction.

Can be one of the following types:

The 'Y Axis Extractor' property defines how row headers are detected on the left side of the table. This extractor should match the entire header column and return sub-instances for each individual row.

• Use a regular expression with named groups or a Collation Provider such as Ordered Array to extract row headers.
• If 'Transpose' is set to true, the names of the sub-instances should match the names of the corresponding Data Column objects.
• If no Y extractor is defined, table rows will be identified using line detection. In this case, enable snap to lines on the Y axis and set the 'Maximum Row Count' property to indicate the number of rows.

Configuration Details

• The extractor is typically a Pattern Match or List Match configured to match the full row header column.
• Named groups in the pattern (e.g., (?<January>...)) will be mapped to row labels.
• If using a Collation Provider, ensure the order and naming of the sub-instances aligns with your table rows.
• For tables with merged or multi-line row headers, use a pattern that can accommodate these variations.

Practical Tips

• Use the Y Axis Extractor for tables with explicit row labels (e.g., months, categories, account names).
• If your table does not have row headers, omit this property and use line detection instead.
• When using line detection, set 'Maximum Row Count' to the expected number of rows for best results.

Common Pitfalls

• If the extractor returns sub-instances with names that do not match any expected row, data may be misaligned.
• If the extractor is too broad or too narrow, you may get extra or missing rows.

This property is optional, but recommended for tables with explicit row headers.

The 'Maximum Row Count' property controls how many rows will be detected when using line detection (i.e., when no Y Axis Extractor is set).

• A value of 0 indicates an unlimited number of rows.
• Set this property to the expected number of rows for more predictable extraction.

This property is ignored if a Y Axis Extractor is configured.

The 'X Axis Snap Limits' property specifies how far the extractor will search for lines to snap column headers to, improving alignment and accuracy.

• Enter a value (e.g., 0.25in) to enable snapping.
• Leave empty to disable snap to lines for X axis headers.

Use this property to handle tables with imperfect or misaligned header lines.

The 'Y Axis Snap Limits' property specifies how far the extractor will search for lines to snap row headers to, improving alignment and accuracy.

• Enter a value (e.g., 0.25in) to enable snapping.
• Leave empty to disable snap to lines for Y axis headers.

Use this property to handle tables with imperfect or misaligned row header lines.

The 'Line Snap Margin' property allows you to shrink the snapping zone on each edge, providing finer control over which lines are considered for snapping.

• Enter a value (e.g., 0.05in) to reduce the snapping area.
• Use this property to avoid snapping to unintended lines or to fine-tune header alignment.

This property is only relevant when snap to lines is enabled.

The 'Minimum Fill Weight' property controls how Optical Mark Recognition (OMR) columns are read.

• If left blank, OMR reads are based on Layout Data generated by the Box Detection and Box Removal IP commands.
• If a value is specified (e.g., 2pt), OMR reads are based on the cell containing enough black pixels to fill a square region of the specified size.

Example: A value of 2pt means a cell is considered checked if it contains enough black pixels to fill a 2pt x 2pt area.

The 'Header Column' property determines which Data Column will be populated with the row header values detected by the Y Axis Extractor.

• This property is only relevant when both X and Y Axis Extractors are configured.
• Select the Data Column that should receive the row header text.

The 'Transpose' property allows you to handle tables where the data is transposed (i.e., rows and columns are swapped).

• Set to true to treat source rows as columns in the output table.
• Set to false (default) for standard table layouts.

This property is essential for extracting data from cross-tab or pivoted tables.