Graph User Manual

Graph ► Tools ► Table Editor

The table component of a Task is used to enter and display the decimal and character representation of data. The cells of the table can represent a label (text), scalar (a single decimal number) or point (2 or 3 dimensional point) entries. A typical table is diagrammed below.

The following is an itemization of table features.

Columns

A column of a table is the vertical collection of cells of the table. Touching on the column header (the grey portion at the top of each column) selects cells in that column. Touch-hold and then touching the Edit menu item edits the description for the column which is useful for setting labels for a legend. See Legend for additional information.

Cell Type

Each cell in a table is comprised of indivisible pieces of data. There are two types of "indivisible pieces of data" as defined here:

• Atomic: For a bar graph the atomic element is a single number. For a line graph the atomic element is two numbers representing a x and y value at a 2D point. Thus, for a 2D point, each cell shows, and can can be used to edit, two numbers. The notion of an atomic cell is consistent with the idea that there is no way to specify a 2D point without specifying two numbers so the two numbers are considered indivisible (atomic) and must be located within one cell.

• Component: It is often the case that numbers come from other sources that do not represent a 2D point and those numbers need to be inserted into the table. Those numbers are often a sequence of scalars that represent only one dimension of a sequence of 2D points. As such, they incompletely specify an atomic element and are not usable by themselves. They need to be combined with another sequence of scalars to form a sequence of 2D points (and hence a curve). The component cell type is used to facilitate this need. In component mode, the table columns are arranged by interleaving dimension. Specifically, the x-dimension is specified in a column and then the y-dimension is specified in the adjacent column.

The type of cell used is specified in the table Options. By switching between types, dimensions can be resorted. For example, for 2D points the data is serialized in the format: x1 y1 ... xN yN. That is the atomic representation. By using the component representation the data is serialized in the format: x1 ... xN y1 ... yN. It is often the case that both serializations are useful and as a result, tables have two cell types.

The above discussion is in terms of 2D points but is equally applicable to 3D points in the obvious way. Notice that bar data is scalar and scalars only have one dimension so the atomic and component cell type are identical.

It is hard to overemphasize the fact that the atomic cell is the more consistent representation, but that the component cell is the more conventional representation. When you use the component cell type you will notice this fact. For example, deleting an interval of y-values will not remove the y-values but rather zero them. That is because those y-values are intrinsically (and implicitly) associated with the x-values. To remove the y-values you must also select the respective interval of x-values and then delete both of them simultaneously. Only then will the points be removed. By using the atomic cell, this is a non-issue because the x and y values are in the same cell and deleting a cell deletes both the x and y value pair of a point. This same issue applies to the many other operations applied to tables. By utilizing a sequence of operations that combine atomic and component representations the resulting composite operation can be fairly complex and useful. For example, in component element type paste in a column of x-values and then a column of y-values, switch to atomic element type and then copy the column of 2D points. In that way you have just transposed the matrix of data a.k.a.: changed serialization from x1 ... xN y1 ... yN to x1 y1 ... xN yN.

The figure below diagrams the difference between atomic and component cells for a 2D Points task.

Component Selection

• Touch on a column header to select a column. Touch-hold to delete, copy, paste, cut the entire column.

• Touch on a row header to select a row. Touch-hold to delete, copy, paste, cut the entire row.

• Touch to the left of the column headers, right above the row headers, to focus on the table. Touching on that area deselects any cells of the table. Then select all cells in the table by touch-hold and then touch Select All. Then touch-hold to delete all values in the table, paste to insert data from the pasteboard into the entire table or copy to copy all values to the pasteboard.

• Touch-drag on cells in a column to select an interval and also to scroll the column.

• Touch-hold a selection interval to delete, copy, paste, cut that selection.

• Use the sliders to scroll over the table. Note: sliders are much more efficient than a swipe to pan the data. Imagine what it would be like to swipe over a million data points (no fun!). Use a touch-drag to scroll more slowly. As a side effect, touch drag selects cell intervals so deselect afterwards by touching outside the table as needed.

• The data format for each import operation is define in the respective task section. In general, for scalar data the column import (paste) format is a list of numbers, for point values it is a list of point components (for example, for 2D points: x1 y1 x2 y2 ... xN yN) where numbers are separated by a blank (whitespace).

Operations On Selection

Once a component is selected then touch-hold to bring up a menu of options. Choose one of the following from the menu:

• Delete to delete the selected component.

• Copy to copy the selected component.

• Paste to paste the selected component.

• Cut to cut the selected component.

It should be noted that if you select only one column or row of data then that column or row is extended with a subsequent paste. However, if you touch-drag an interval of columns or rows then the subsequent paste wraps over the rectangular cell selection.

Cell Edit

When you touch on a cell the cell editor is brought forward. Once forward the following applies.

• Edit the cell text using the normal keypad edit facilities.

• Touch done to enter the data.

• Use the arrow buttons to enter the text and proceed to an adjacent cell in the direction of the arrow.

If you paste to the entire table then the data needs to be formatted as follows.

• Double-return delimited format defines column ends as two consecutive return characters in the string serialization of the data.