Skip to content

Data Sources

A data source in the TableCurve 3D explorer must come from one of the supported file formats:

  • XYZ, and Multi-column ASCII files
  • Excel (XLS v3-XLS Office XP)
  • Lotus 123 (WK4, WK3, WK1, WKS, WRK)
  • Quattro Pro (WB2, WB1, WQ1, WKQ)
  • SigmaPlot (JNB, SPW, SP5, SPG)
  • XYZ, and Multi-column DIF files
  • dBase III+ and dBase IV (DBF)
  • SPSS Windows (SAV)
  • Systat Windows (SYS)

X,Y,Z ASCII Files

The is the default data format. These files optionally contain a main title, X title, Y title, and Z title on the first four lines followed by successive lines of data. On each data line, the X value must come first, followed by the Y value, and then the value of the dependent variable Z. There must be a delimiter between the values such as one or more spaces, a comma, or a separate line. Weights may follow the X,Y,Z values by having an uppercase W precede the weight value. XYZ ASCII files automatically place a data object in the TableCurve 3D notebook when imported.

Multi-Column ASCII Files

If the fifth line of an ASCII file contains more than three floating point values, it is treated as a multicolumn ASCII file. These files do not automatically place data objects in the TableCurve 3D notebook when imported. Data objects are added using the Autospecify All XYZ Data Items or Specify XYZ Data Item options.

When manually specifying an XYZ data object, you must specify an X column, a Y column, a Z column, and optionally a Weights column. Use the column number and the ASCII string information in the initial columns to determine the columns desired.

For TableCurve 3D to read a multi-column ASCII file successfully, it must be in one of two formats, both of which require all row entries for a given sequence of columns to be on a single line which concludes with a carriage return:

  • Comma Delimited Format - Here all values are separated by commas. You may have any number of spaces within the data, but each column entry, including empty positions, must be separated from other entries by a comma. With this format, ASCII character strings do not have to be delimited with quotes. You will have to delimit a string with quotes, however, if it contains a comma.
  • Space and Quote Delimited Format - In this case, there are no commas separating the different column entries. Rather, TableCurve 3D relies on a space or spaces between the entries of numeric values, and on double quotes beginning and concluding every character string. With this format, quoted character strings are essential. You must also use a pair of empty quotes to represent empty column positions.

The following two examples illustrate both formats and represent valid multi-column ASCII files:

Column 1,Column 2,Column 3,Column 4,Column 5 1.234,2.345,Data Missing,7.890,8.901 3.456,4.567,6.789,9.012,10.123 5.678,6.789,7.890,11.234,12.345 7.890,8.901,,13.456,14.567 "Column 1" "Column 2" "Column 3" "Column 4" "Column 5" 1.234 2.345 "Data Missing" 7.890 8.901 3.456 4.567 6.789 9.012 10.123 5.678 6.789 7.890 11.234 12.345 7.890 8.901 "" 13.456 14.567

It is always a good idea to carefully inspect the input data from multi-column files whose formats may not be fully compatible. The TableCurve 3D multi-column import procedure should read most multi-column ASCII files generated by commercial software.

Excel, Lotus, and Quattro Pro Files

Even when only three columns of data are present in each sheet of a spreadsheet file, data objects are not added to the TableCurve 3D notebook when imported. Data objects must be added using the Autospecify All XYZ Data Items or Specify XYZ Data Item options.

The first 100 rows of the spreadsheet are scanned for text strings. You must have at least one cell entry, data or string, in the first 100 rows in order for a column to be available for specification. When manually specifying an XYZ data object, use these text strings and the column designation to select the X, Y, and Z columns, and optionally a Weights column for the data table.

  • All data to be imported must be in numeric or formula format. Numbers saved as strings will not be read. Strings may appear anywhere within the columns.
  • The X, Y, Z, and Weights columns can come from any worksheet or page within these files, although all columns within an individual XYZ specification must come from the same worksheet. The column selection list box will show both the sheet and column in an Excel-like nomenclature. Across all sheets and pages, the maximum number of columns available for selection is 4096. You will not be able to select the same column for more than one variable.
  • A valid triplet of X,Y,Z values must exist within a spreadsheet row in order for the X,Y,Z data point to be added to the data table. An empty cell or a cell containing a string will prevent entry to the data table.
  • All instances of a zero value in an X or Y column will be included in the data imported. Such points, if invalid, can be toggled off in the Section Data option or via the TableCurve 3D Editor.
  • For Excel, TableCurve 3D will attempt to read a main title from the first row of the spreadsheet file for each data set in the sheet. It is recommended that the X column contain the main title when sequences of XYZ triplets are present, and in the Z column when a single X and Y column is followed by many Z columns.
  • For Lotus and Quattro files, TableCurve 3D will seek to read a main title from the first row of the sheet or page. This title will be available only if the first row contains a single string entry. This title must thus suffice for all sets on the page. It can be in any column. With the exception of the title column, TableCurve 3D will offer the first string in each selected column as the X, Y or Z variable title. In the title column, the second string is offered as the variable title.

SigmaPlot Files

TableCurve 3D can build its data table from the numeric values stored in SigmaPlot JNB, SPW, SP5, and SPG graph files. After adding a SigmaPlot data source, data objects must be added using the Autospecify All XYZ Data Items or Specify XYZ Data Item options.

The column specification process is very similar to that of the spreadsheets. There are a few differences:

  • Only the first available worksheet is available for import. This can be an embedded Excel worksheet.
  • The SigmaPlot column selection list consists of the column labels assigned within SigmaPlot rather than the strings in the initial columns.
  • Only numeric entries are read. Embedded strings and missing values are not imported.

SPSS Files

The SPSS SAV format used in v7.5 through v11 is supported for data import. After adding an SPSS data source, data objects must be added using the Autospecify All XYZ Data Items or Specify XYZ Data Item options. The process is again nearly identical to spreadsheet import. The differences are:

  • You must select columns based on the variable name and variable label assigned within SPSS.
  • Valid X,Y,Z triplets, those without system or variable-specific missing values, are added to the data table.

Systat Files

The Systat Windows SYS and SD formats used in v8.0 through v10.2 are supported for data import. After adding a Systat data source, data objects must be added using the Autospecify All XYZ Data Items or Specify XYZ Data Item options. The differences relative to spreadsheet import are:

  • You must select columns based on the variable name assigned within Systat.
  • Valid X,Y, Z triplets, those without missing values, are added to the data table.

Data Interchange Format (DIF) Files

When a DIF file contains three or more columns of numeric information, it is available as a data source. Even when only three columns of data are present in a DIF file, a data object is not added to the TableCurve 3D notebook when imported. Data objects must be added using the Autospecify All XYZ Data Items or Specify XYZ Data Item options.

You will be offered a column selection based upon a column ID and any title assigned to this column. Just as with the spreadsheet options, you must select a column for the X, Y, and Z-variables, and optionally for the Weights

dBase Files

TableCurve 3D can build its data table from the numeric information in dBase databases. Both dBase III+ and dBase IV type DBF files are supported. Even when only three columns of data are present in a dBase file, a data object is not added to the TableCurve 3D notebook when imported. Data objects must be added using the Autospecify All XYZ Data Items or Specify XYZ Data Item options.

The import process is similar to that of a spreadsheet except numeric fields are selected for the X, Y, Z, and optionally the Weights variables. The following rules apply:

  • Only numeric fields (type N in III+ and types N and F in IV) are offered for selection.
  • If a dBase record is marked for deletion, the data from the selected fields will still be added to the data table, but this XYZ triplet will be marked as inactive. The state of points can be toggled on and off in the Section Data and TableCurve 3D Editor options.
  • Zero values in dBase data files are sometimes legitimate and other times simply represent a field that has yet to receive a value. All instances of a zero value in one or more of the fields will be included in the data imported, but these data triplets will be marked as inactive. Again, the state of points can be toggled on and off in the Section Data and TableCurve 3D Editor options.
  • dBase database files do not use a binary format for storing numeric values. Numeric storage is generally in fixed decimal position ASCII strings. This limits the range and precision of floating point values. The import procedure will preserve whatever precision exists within the original dBase numeric field. Under certain conditions, versions III+ and IV can both store ASCII scientific notation. Type N and F fields containing scientific notation will be correctly read by TableCurve 3D.

Management of Zero Values

A single active 0,0,z data point will result in the exclusion of TableCurve 3D's logarithmic equations. In dBase files, a zero value often represents a numeric field that has yet to receive a value. Triplets containing a zero value in a dBase file are thus marked as inactive. For all other data formats, such points will be active and included in the fitting.

Titles

In all individual XYZ specification operations, you will be offered an X, Y, and Z title. You may accept or edit these titles. If you press Previous Titles in the titles entry dialog, the titles present prior to the specification, if any, are restored. Note that the modifications to these titles are stored in the TableCurve 3D notebook only, and not in the data source file.