Inspect data - Data360_Analyze - Latest

View data in the data viewer

1. To expand the data viewer, click an input pin or an output pin that contains data (pins with data are green), or click the record count to the right of a node. You can hover over a pin to see its name.
   • For each data set that you inspect, a new tab is created in the data viewer.
   • From the tab menu, select the Search Tabs option. This will bring up the Search Tabs popover, which lists all the open tabs. You can filter tabs using the search field. Selecting a tab will select it in the data viewer.
   • The data viewer shows up to a maximum of the first 1000 records in any data set. For nodes which have been executed but produced no records, you can still view the corresponding metadata in the data viewer (field names and data types).
   • If you re-run a node, the corresponding tab remains open During the execution, a spinner icon replaces the pop-out button to indicate that the node is currently running. If the node or any of its upstream nodes fail during the re-execution, the tab is marked as stale. In this case, a warning icon will replace the pop-out button.

     Note: When a tab is marked as stale, the options to export to CSV and apply filters are disabled.
   • To help you to organize and work with your data in the data viewer, row numbers are displayed on the far left of the rows, and in the bottom left corner of the screen, you will see a count of how many records are displayed, against the total that exist in the entire data set. A selection summary is displayed in the footer to the right, showing Min, Max, Average, Sum, Rows selected, and Columns selected.
   • You can reorder columns in the data viewer to make it easier to inspect the fields of interest in a data set. You can drag and drop columns to a new location, or you can choose one of the following options from the column menu: Move to Beginning, Move to End, Move Left or Move Right. The column order will be remembered for as long as a selected tab is open. However, if you close the tab and then re-open the data set, the original column order will be reset.
   • Any applied filters, sorts, column moves, and column resizes will be preserved as long as the data metadata remains unchanged. This ensures that your customized views are maintained across node executions unless there are changes in the underlying data structure.

   • Select Search Columns from the tab menu to open the Search Columns popover, which lists all columns. Use the search bar to filter columns. The popover can be moved and resized. Selecting a row highlights the column in the data viewer and scrolls it into view. Unchecking a column hides it, and unchecking all hides all columns.

   • You can hide columns by choosing the Hide option from the column header menu. If columns are hidden, use the Unhide All option available in any column header or tab menu.

   • You can resize the data viewer columns by hovering over the right border of the column until your cursor changes shape, then clicking and dragging the border to the required width. To make the width of a specific column fit the content, double-click the right column border, or select Resize to Fit from the column menu. The Resize to Fit option now resizes the selected column. To resize all columns, use the Resize All Sub-menu with the following options:
     • Fit - resizes all columns to fit their data and column header
     • Stretch - stretches columns to fit the data viewer container
     • Fixed - resizes the columns to a fixed width
     • This - resizes columns to the width of the selected column
2. To resize the data viewer, drag the top edge to the required size.
3. To collapse the data viewer, click the top edge.

Freeze rows and columns in data viewer

To freeze rows in the data viewer:

• Freeze Rows: Right-click on a data viewer cell and select Freeze Rows Top. This action will freeze the currently selected row and all rows above it.

• Frozen rows remain stationary as you scroll vertically through the data. These rows are visually distinguished by a blue background in the row header (where the row number is displayed).

• Unfreeze Rows: To unfreeze the rows, right-click on any cell and select the Unfreeze Rows option.

To freeze a column in the data viewer:

• Freeze a Single Column: Select Freeze from the column header menu to position the column between any already frozen columns (with a blue header) and unfrozen ones..

• Freeze Multiple Columns: Use Freeze Left to freeze the current column and all columns to the left of it, maintaining their order. Frozen columns remain fixed during horizontal scrolling.

• Unfreeze a Column: To unfreeze a specific column, select Unfreeze from its header menu.

• Unfreeze All Columns: To unfreeze all columns, choose Unfreeze All from any header menu.

Compare data sets side by side

You can select a data set to view in a pop-out window, allowing you to compare two or more data sets side by side. For example, you might want to compare the input to a node with the output. Or, you might want to compare the output of a node after an initial run, then reconfigure the node with different property values to generate a second data set for comparison.

1. To view a data set in a pop-out window, click the View in pop-out window button on the data viewer tab:

   

   The data set is displayed in a pop-out window which you can move around the screen independently of the browser window. The tab is removed from the data viewer.

2. To close the pop-out window and return the data to the data viewer, click the Dock back into main window button in the top right corner. Column moves and resizes are preserved on pop out/in.

   

   Note: If you are viewing data in a pop-out window, you can add a Filter or Split node to your data flow provided that the source data remains available. If your source data is no longer available, you will see a warning at the bottom of the pop-out window. Any changes that you make to the source data while the pop-out window is open will not be reflected in the data viewer.

Data quality indicators

When you view a data set in the data viewer, colored bars indicating the quality of the data are displayed for each column.

The color and the length of the bar indicates the quality of the data, ranging from a dark green bar that runs the full width of the column to indicate the highest quality data, to a short, red bar to indicate the lowest quality data.

For Boolean, Integer / Long, Date / DateTime and Float / Double fields, quality is a measure of how many NULL values are found in the data set.

For String / Unicode fields, quality is a measure of how many NULL values, empty strings or values with leading or trailing white space are found in the data set.

When you hover over a data quality bar, a tooltip shows the quality as a percentage value.

You can click the data quality bar to display a histogram of the data values in the field and a number of statistics on the (sample) data shown in the data viewer, such as min, max, average, Std Dev. You can also display this dialog by choosing Statistics from the column menu.

You can change your view of the histogram by:

• Moving the ends of the gray bar below the histogram to zoom in on a section of the data values.
• Dragging the gray bar below the histogram to the left or right to view a different section of the histogram.

Copy data from the data viewer

You can copy content from the data viewer by pressing Ctrl+C, or by right-clicking and selecting Copy from the menu. You can then paste the selected data into an external text editor by pressing Ctrl+V.

For example:

• To copy the content of single cell, click the cell and press Ctrl+C. Or, to select multiple cells, click and drag over the cells that you want to select.
• To copy the content of an entire row of data, click the row number on the far left of the data viewer, then press Ctrl+C. To copy the content of an entire column, click the column header and press Ctrl+C.
• You can select all data in multiple adjacent columns by pressing Shift and clicking the column header of the first and last adjacent column that you want to select. You can select all data in multiple columns that are not adjacent by pressing Ctrl and clicking the column headers of the columns that you want to select. To select all data in the data viewer, press Ctrl+A, right-click and choose Select All, or click the empty cell in the top left corner of the data viewer.