Writer Feature Type Properties: Format Parameters
To access feature type format parameters, open the Feature Type Properties dialog by clicking the properties button on a feature type in the workspace.
These parameters apply only to a selected feature type, not to the entire writer.
Tip: If a feature type parameter listed here conflicts with a writer-level parameter, then the writer parameter will be ignored and this feature type parameter will be used.
- Drop sheet: Worksheets are deleted and recreated after all existing sheets. (The sheet order may change.)
- Drop named range: Formatting and data in the extents of the existing named range are lost, but formatting in cells that exist outside the named range is unaffected. Original extents of the named range are lost. A new named range is created using the numbers specified in the Feature Type parameters Named Range Start Column and Named Range Start Row.
If a template file or template sheet is specified as a writer parameter, the option to truncate the existing sheet or named range is applied to the template sheet before it is copied to the destination file. Inserted items such as charts and pictures are not deleted.
- Truncate sheet: Formatting such as cell color and column widths that exist in cells outside the data extents remain.
- Truncate named range: Formatting in the extents of the existing named range is lost, but formatting in cells outside this existing named range is unaffected.
The Feature Type parameters Named Range Start Column and Named Range Start Row are ignored.
Data written to the named range use the original start positions and begin writing at the specified offset.
Note: Worksheet names must be less than 31 characters in length.
This parameter specifies a relative order of the feature type. Valid values are all positive integers (starting from 1).
The sheet with the lowest relative order will be the first sheet in the workbook, with sheets having a higher order placed after it. If a value is not specified, then the sheet will be placed after all sheets with a specified sheet order. If two sheets have identical sheet order values, then the two will be ordered according to their incoming order.
If the sheet feature type has no specified order, but it contains a named range feature type which does, then the order on the named range will be used. If the sheet with no order contains two named ranges with different sheet orders, then one of those values will be selected arbitrarily. Therefore, to avoid ambiguity, it is always best to specify the order directly on the sheet feature type.
Example
If you write 7 sheets in the form of [sheetName,sheetPosition]
{(A, ),(B,9),(C, ),(D,4),(E,4),(F, ),(G,2)}
the sheets are written in the order
G,D/E,B,A/C/F
where the order of (A/C/F) and (D/E) are dependent on their incoming order to the writer.
Using Feature Type Fanout will pass the sheet order of the template feature type to all its fanned out feature types. (See the topic Setting Feature Type Fanout in FME Workbench help.)
Set to Yes to write the headers of the attributes.
This parameter allows the users to freeze the first n rows, where n is the last row to be frozen. For example,
- 1: the first row is frozen
- 2: the first two rows are frozen
- 5: the first five rows are frozen
In Excel this feature is called Freeze Panes, and it allows you to freeze a portion of the sheet so that certain rows like column headers remain visible while scrolling.
This option has no effect on named ranges.
Set to Yes to treat attribute names, such as A, B, or C as the column name in which to write the data. If set to No (the default), the columns will be written in the order on the schema.
If writing a Named Range, as determined by the feature type name, this parameter specifies the starting column in the sheet where the Named Range will be defined if it does not exist.
This parameter is ignored if the Named Range name supplied already exists and has a defined range.
If writing a Named Range, as determined by the feature type name, this parameter specifies the starting row in the sheet where the Named Range will be defined if it does not exist. This parameter is ignored if the Named Range name supplied already exists and has a defined range.
This parameter specifies a starting column location for writing data in the sheet identified by the feature type name. This index applies to cell-based data but does not apply to raster or other non-cell-based data.
Row indexes are valid values between 1 and the maximum number of columns allowable in the sheet.
This parameter specifies a starting row location for writing data in the sheet identified by the feature type name. This index applies to cell-based data but does not apply to raster or other non-cell-based data.
Row indexes are valid values between 1 and the maximum number of rows allowable in the sheet.
Specifies whether to enable sheet protection on the current sheet corresponding to this feature type. Sheet protection is useful to prevent someone from accidentally or deliberately changing, moving, or deleting important data. In Excel, both password and non-password protection are supported, but only non-password protection is currently supported in FME. A similar setting exists for cell protection per column; however, there currently is no way to protect the entire workbook to prevent addition, removal, or rearranging of sheets. This setting does not apply to named ranges.
You can embed a map or raster image from another source directly into a spreadsheet. This parameter determines which raster format will be used for rasters written to the workbook. There are three supported options
- PNG
- JPEG
- BMP
When sending a raster feature to the writer, there are a few attributes you can add to modify the result.
Attribute Name | Description |
---|---|
xlsx_row_id |
Determines the row on which the top of the picture will be placed. The picture will align with the top of the row. Only positive integer values are valid. If not specified, this value will default to 1, which corresponds to row 1. |
xlsx_col_id |
Determines the column on which the left of the picture will be placed. The picture will align on the left side of the column. Only positive integer values or the Microsoft Excel Column letter designations (A,B,..,XFD) are valid. If not specified, this value will default to 1, which corresponds to column A. |
xlsx_raster_scale_factor |
Allows the raster to be scaled by a factor. This can be any positive numeric value greater than 0.0. If not specified, the raster will be written out using the xlsx_raster_height and xlsx_raster_width. If specified, xlsx_raster_height and xlsx_raster_width will be ignored. For example, a scale factor of 2.0 will double the width and height of the raster. If the image is larger than the height or width of a cell, the image will cover multiple cells. The cell height will not change. |
xlsx_raster_height |
Sets the raster height to the specified number of pixels. Only positive integer value greater than 0 are valid. Note: If xlsx_raster_scale_factor is specified, this value will be ignored. If this is used, you should also specify xlsx_raster_width. If neither is specified, the raster will default to its actual size. If the image is larger than the height or width of a cell, the image will cover multiple cells. The cell height will not change. |
xlsx_raster_width |
Sets the raster width to the specified number of pixels. Only positive integer value greater than 0 are valid. Note: If xlsx_raster_scale_factor is specified, this value will be ignored. If this is used, you should also specify xlsx_raster_height. If neither is specified, the raster will default to its actual size. If the image is larger than the height or width of a cell, the image will cover multiple cells. The cell height will not change. |
The writer mode can be specified on the feature type or on individual features.
At the feature type level, the writer mode is specified by this parameter. It may be superseded on individual features by the fme_db_operation format attribute.
Valid feature type writer modes are listed below:
Mode | Description |
---|---|
Insert |
Appends features to the end of the sheet, or writes them to an exact row and/or column. Any data that already exists in that location will be overwritten. If there is a row value specified for the Row ID Attribute, that row will be used instead. |
Update |
Causes the row corresponding to the feature’s value specified in the Row ID Attribute to have its data changed to the values on the incoming feature. If an attribute is missing on the incoming feature, the existing cell value will not be overwritten. |
Delete |
Deletes the values in the row corresponding to the feature’s value specified in the Row ID Attribute. Only a number of columns equal to the amount of columns on the incoming feature will have their data and formatting removed. |
The writer allows you to specify an input attribute to contain the value of the row identifier to be used for per-feature Update and Delete operations.