Import Data

Upon importing, JMP converts file types such as text files and Microsoft Excel files to the data table format. On Windows, JMP relies on three-letter filename extensions to identify the type of file and how to interpret its contents. Here are some examples of filename extensions identified by JMP:

Open( "$SAMPLE_IMPORT_DATA/Bigclass.xlsx" ); // Microsoft Excel file

Open( "$SAMPLE_IMPORT_DATA/Bigclass.txt" ); // text file

Open( "$SAMPLE_IMPORT_DATA/Carpoll.xpt" ); // SAS transport file

On Macintosh, JMP relies on the Macintosh Type and Creator codes (if present) and secondarily on three-letter filename extensions. Be sure to add the file extension before importing the file on Macintosh.

•	Type and Creator codes are invisible data that enable the Finder to display a file with the correct icon corresponding to the application that created it.

•	Files with generic icons should have the filename extensions (as with files created on other operating system).

Open( "$SAMPLE_IMPORT_DATA/Bigclass.txt", text );

Additional supported formats include .csv, .jsl, .dat, .tsv, and .jrn.

For more detailed information about import options, refer to File Functions in the JSL Syntax Reference.

Import Data from a Text File

The Import Settings in the Text Data Files preferences determine how text files are imported. For example, column names begin on line one and data begin on line two by default. To use different settings, specify the import settings as Open() options in your script.

The default Import Settings and your custom import settings are saved in the data table Source script, so you can reimport the data using the same settings. However, the default Import Settings are optional in the script.

The following Open() options are available:

CharSet("option")

	// "Best Guess", "utf-8", "utf-16", "us-ascii", "windows-1252", "x-max-roman", "x-mac-japanese", "shift-jis", "euc-jp", "utf-16be", "gb2312"

Number of Columns(Number)

Columns(colName=colType(colWidth),... )

	// colType is Character|Numeric

	// colWidth is an integer specifying the width of the column

Treat Empty Columns as Numeric(Boolean)

Scan Whole File(Boolean)

End Of Field(Tab|Space|Comma|Semicolon|Other|None)

EOF Other("char")

End Of Line(CRLF|CR|LF|Semicolon|Other)

EOL Other("Char")

Strip Quotes|Strip Enclosing Quotes(Boolean)

Labels | Table Contains Column Headers(Boolean)

Year Rule | Two digit year rule ("Decade Start")

Column Names Start | Column Names are on line(Number)

Data Starts | Data Starts on Line(Number)

Lines to Read(Number)

Use Apostrophe as Quotation Mark

CompressNumericColumns(Boolean)

CompressCharacterColumns(Boolean)

CompressAllowListCheck(Boolean)

The following script opens a text file of comma-delimited text, which includes no column names. The script defines the column names and the column widths.

Open(

	"$SAMPLE_IMPORT_DATA/EOF_comma.txt",

	End of Field( comma ),

	Labels( 0 ),

	Columns(

		name = Character( 12 ),

		age = Numeric( 5 ),

		sex = Character( 5 ),

		height = Numeric( 3 ),

		weight = Numeric( 3 )

);

Here is an example of opening a text file in which the field separator is a space:

Open(

	"$SAMPLE_IMPORT_DATA/EOF_space.txt",

	Labels( 0 ),

	End of Field( Space )

);

To set the import options interactively, include the Text Wizard argument. A preview of the text file opens in the text import window.

Open( "$SAMPLE_IMPORT_DATA/EOF_space.txt", "Text Wizard" );

The following sections describe each argument in more detail. For more detailed information about import options, refer to File Functions in the JSL Syntax Reference.

Number of Columns

Specifies the total number of columns in the source file. This option is important if data is not clearly delimited.

Columns

Identifies column names, column types, and column widths with a Columns argument as shown in the preceding examples.

If you specify settings for a column other than the first column in the file, you must also specify settings for all the columns that precede it. Suppose that you want to open a text file that has four columns (name, sex, and age, and ID, in that order). age is a numeric column, and the width should be 5. You must also set the name and sex column types and widths, and list them in the same order:

Columns(

	name = Character( 15 ),

	sex = Character( 5 ),

	age = Numeric( 5 )

);

You are not required to provide settings for any columns that follow the one that you want to set (in this example, ID).

After the data is imported, you use the modeling type for a column. See Set or Get Data and Modeling Types.

Note: Most of the following arguments are defined in the JMP preferences. To override the preference, include the corresponding argument described below in your import scripts.

Treat Empty Columns as Numeric

Imports columns of missing data as numeric rather than character data. A period, Unicode dot, NaN, or a blank string are possible missing value indicators. This is a Boolean value. The default value is false.

Scan Whole File

Specifies how long JMP scans the file to determine data types for the columns. This is a Boolean value. The default value is true; the entire file is scanned until the data type is determined. To import large files, consider setting the value to false, which scans the file for five seconds.

Strip Quotes | Strip Enclosing Quotes

Specifies whether to include or remove the double quotation marks ( " ) that surround string values. This is a Boolean value. The default value is true.

For example, suppose that the field delimiter is a space:

•	John Doe is interpreted as two separate strings (John and Doe).

•	"John Doe" is interpreted as a single string. Most programs (including JMP) read a quotation mark and ignore other field delimiters until the second quotation occurs.

•	If you include Strip Quotes(1), "John Doe" is interpreted as John Doe (one string without quotation marks).

Note that many word processors have a “smart quotation marks” feature that automatically converts double quotation marks ( " ) into left and right curled quotation marks ( “ ” ). Smart quotation marks are interpreted literally as characters when the text file is imported, even when JMP strips double quotation marks.

End of Line(CR | LF | CRLF | Semicolon | Other)

Specifies the character or characters that separate rows. The choices are as follows:

•	CR for carriage returns (typical for text files created on Macintosh OS up to version 9)

•	LF for linefeeds (typical for UNIX and Macintosh OS X text files)

•	CRLF for both a carriage return followed by a linefeed (typical for Windows text files).

All three characters are line delimiters by default.

Use the Other option to use an additional character for the row separator, which you must specify in the EOLOther argument. JMP interprets either this character or the default character as a row separator.

Specifies the character or characters used to separate fields. Note the following:

•	The default field delimiter is Tab.

•	Use the Other option to use a different character for the field separator, which you must specify in the EOFOther argument.

•	The Space option uses a single space as a delimiter.

•	The Spaces option uses two or more spaces.

EOFOther, EOLOther

Specifies the character or characters used to separate fields or rows. For example, EOLOther("*") indicates that an asterisk separates rows in the text file.

Labels | Table Contains Column Headers

Indicates whether the first line of the text file contains column names. This is a Boolean value. The default value is true.

Year Rule | Two Digit Year Rule

Specifies how to import two-digit year values. If the earliest date is 1979, specify "1970". If the earliest date is 2001, specify "20xx".

Column Names Start | Column Names Are on Line

Specifies the starting line for column names. The following example specifies that the column names in the text file start on line three.

Open(

    "$SAMPLE_IMPORT_DATA/Animals_line3.txt",

    Columns(

        Column( "species", Character, "Nominal" ),

        Column( "subject", Numeric, "Continuous", Format( "Best", 10 ) ),

        Column( "miles", Numeric, "Continuous", Format( "Best", 10 ) ),

        Column( "season", Character, "Nominal" )

),

	Column Names Start( 3 )

);

Data Starts | Data Starts on Line

Specifies the starting line for data.

The following example specifies that the data in the text file start on line five.

Open(

	"$SAMPLE_IMPORT_DATA/Bigclass_L.txt",

	Columns(

		Column( "name", Character, "Nominal" ),

		Column( "age", Numeric, "Continuous", Format( "Best", 10 ) ),

		Column( "sex", Character, "Nominal" ),

		Column( "height", Numeric, "Continuous", Format( "Best", 10 ) ),

		Column( "weight", Numeric, "Continuous", Format( "Best", 10 ) )

),

	Data Starts( 5 )

);

Lines to Read

Specifies the number of lines to include in the data table. JMP starts counting after column names are read.

The following example includes only the first 10 lines in the data table.

Open(

    "$SAMPLE_IMPORT_DATA/Bigclass_L.txt",

    Columns(

        Column( "name", Character, "Nominal"),

        Column( "age", Numeric, "Continuous", Format( "Best", 10 ) ),

        Column( "sex", Character, "Nominal" ),

        Column( "height", Numeric, "Continuous", Format( "Best", 10 ) ),

        Column( "weight", Numeric, "Continuous", Format( "Best", 10 ) )

),

	Lines To Read( 10 )

);

Use Apostrophe as Quotation Mark

For data that are enclosed in apostrophes, this option treats apostrophes as quotation marks and omits them. For example, ’2010’ is imported as 2. This is a Boolean value. The default value is false.

This option is not recommended unless your data comes from a nonstandard source that places apostrophes around data fields rather than quotation marks.

Set Text Import Preferences

If you want to set preferences for importing text, it can be helpful to first see a list of all preferences. To do so, use the Show Preferences (All) function.

Then, copy and paste only the section you want into a Preferences function. For example, to specify import settings, write the following expression:

Preferences(

	Import Settings(

		End Of Field( Tab, Spaces, Comma )

);

Import Data from a Microsoft Excel File

When you open a Microsoft Excel workbook in JMP, the file is converted to a data table. JMP supports .xls, .xlsm, and .xlsx formats. See 导入 Microsoft Excel 文件 in the Using JMP book for details about Microsoft Excel support.

In the JMP preferences, settings in the General group can help determine how worksheets are imported:

•	Excel Open Method specifies how a Microsoft Excel file should be opened by default, when using a non-specific open statement.

–	Use Excel Wizard opens the Excel Import Wizard to import the file. This is the default setting.

–	Open All Sheets opens all worksheets in the Microsoft Excel file.

–	Select Individual Worksheets prompts users to select one or more worksheets when they open the file.

•	Use Excel Labels as Headings determines whether text in the first row of the worksheet is converted to column headings in the data table.

By default, JMP takes the best guess. If names have been defined for all cells in the first row, the text in those cells is converted to column heading. Otherwise, columns are named Column 1, Column 2, and so on.

To override a preference, include the corresponding argument described later in this section.

Note: See Create Excel Workbooks for information about exporting data tables to an Excel workbook.

Open a Workbook in the Excel Import Wizard

The Excel Import Wizard shows a preview of the data and lets you modify the settings before importing the data. Specify "Excel Wizard" as the argument.

Open( "$SAMPLE_IMPORT_DATA/Team Results.xlsx", "Excel Wizard" );

You can also customize settings such as the number of rows in the column headers and the last row of the data.

dt = Open(

	"C:\Excel Wizard Demo\PotatoProduction.xls",

	Worksheets( "Worldprod" ),

	Use for all sheets( 1 ),

	Concatenate Worksheets( 0 ),

	Create Concatenation Column( 0 ),

	Worksheet Settings(

1,

		Has Column Headers( 1 ),

		Number of Rows in Headers( 1 ),

		Headers Start on Row( 2 ),

		Data Starts on Row( 5 ),

		Data Starts on Column( 1 ),

		Data Ends on Row( 40 ),

		Data Ends on Column( 0 ),

		Replicated Spanned Rows( 1 ),

		Replicated Spanned Headers( 0 ),

		Suppress Hidden Rows( 1 ),

		Suppress Hidden Columns( 1 ),

		Suppress Empty Columns( 1 ),

		Treat as Hierarchy( 0 ),

		Multiple Series Stack( 0 ),

		Import Cell Colors( 0 ),

		Limit Column Detect( 0 ),

		Column Separator String( "-" )

);

Notes:

•	The Excel Open Method preference applies when you run the Open() expression with no argument and select another open method using the down arrow on the Open button.

•	When you click a custom button to open a workbook, the Excel Open Method preference applies. To force the worksheet to open directly, specify the Excel Open Method in your script:

Preference( Excel Open Method( "Open All Sheets" ) );

Import Specific Worksheets

Suppose that you want to import data from specific worksheets in your workbook. Specify those worksheets using the Worksheets argument. In the following example, the worksheet named small is imported into JMP.

Open( "C:\My Data\cars.xlsx", Worksheets( "small" ) );

Or specify the number of the worksheet, the third worksheet in the following example:

Open( "C:\My Data\cars.xlsx", Worksheets( "3" ) );

Import multiple or all worksheets by including the worksheet names in a list:

Open( "C:\My Data\cars.xlsx", Worksheets( {"small", "medium", "large"} ) );

Import SAS Data Sets

Open a SAS file as a data table without connecting to a SAS server.

sasxpt = Open( "$SAMPLE_IMPORT_DATA/carpoll.xpt" );

To convert the labels to column headings, include the Use Labels for Var Names argument.

sasdbf = Open( "$SAMPLE_IMPORT_DATA/Bigclass.sas7bdat", Use Labels for Var Names( 1 ));

.xpt and .stx file formats are also supported.

On Windows, you can also open SAS data sets from a SAS server. See Connect to a SAS Metadata Server for details.

Import Web Pages and Remote Files

You can import data from websites or from other computers. The data might be a JMP data table, a table defined in a web page, or another file type that JMP supports.

Open or Import Data from a Website

In the Open() command, specify the quoted URL to open a file from a website. You can open JMP data tables or other supported file types this way.

Open( "http://company1.com/Repairs.jmp" );

Open( "http://company1.com/My Data.txt", text); // specify text on Macintosh

Import a Web Page

A web page can include data in tabular format. Import the table as a JMP data table as follows:

Open( "http://company1.com", HTML Table( n ) );

n identifies which table you want to import. For example, to import the fourth table on the page, specify HTML Table(4). If you omit the value, only the first table on the page is imported.

JMP attempts to preserve the table header defined in a <th> HTML tag. The table header is converted to column headings in the data table. If the <th> tag is wrong or missing, use ColumnNames(n) to specify the nth row. By default, DataStarts(n) will be the next row, or you can specify the DataStarts row.

Import Images from a Web Page

You can import images on the web into a data table, using the Expression data type. The following example creates a new data table and adds two images to it.

dt = New Table( "test", New Column( "Image", Expression ) );

dt << Add Rows( 1 );

dt:Image = New Image( Open( "http://www.jmp.com/support/help/13/images/MosaicPlot.png" ) );

dt << Add Rows( 1 );

dt:Image = New Image( Open( "http://www.jmp.com/support/help/13/images/tTest_NoLoc.png" ) );

Import a File from a Shared Computer

JMP can import files stored on a shared computer, such as another computer or a network drive. The file path can be absolute or relative. The following examples show how to open files from a shared computer named Data. If you plan to share the script, it’s safer to use a relative path to the computer, not a path to the mapped drive.

Open( "\\Data\Repairs.jmp" );

Open( "\\Data\My Data.txt" );

Import HDF5 Files

Hierarchical Data Format, Version 5 (HDF5) is a portable file format for storing data. An HDF5 file consists of groups and datasets. When you import the file, JMP opens a group to present the names of the inner datasets.

JMP handles only tables with numeric (integer, float, double) and string types, and compound files with three or fewer dimensions that contain only simple types.

You can import up to 1,000,000 columns and an unbounded number of rows.

The syntax is as follows:

Open( "filename.h5", {"list_of", "dataset_names"});

Errors are written to the log when invalid data set names are passed.

Import ESRI Shapefiles

An ESRI shapefile is a geospatial vector data format used to create maps. JMP imports shapefiles as data tables. A .shp shapefile consists of coordinates for each shape. A .dbf shapefile includes values that refer to regions. To create maps in JMP, you modify the structure of the data and save the files with specific suffixes.

The following example imports a .shp file and saves it with the -XY suffix.

dt = Open( "$SAMPLE_IMPORT_DATA/Parishes.shp",

:X << Format( "Longitude DDD", 14, 4 );

:Y << Format( "Latitude DDD", 14, 4 ) );

dt << Save( "c:/Parishes-XY.jmp" );

Save the .dbf file with the -Name suffix.

dt = Open( "$SAMPLE_IMPORT_DATA/Parishes.dbf" );

dt << Save( "c:/Parishes-Name.jmp" );

Restructuring the data requires several steps, including adding a Map Role column property to names in the -Name.jmp file. For details, see 地图角色 in the Essential Graphing book.

Import a Password-Protected Microsoft Excel 2007 File

JMP does not support importing password-protected .xlsx files. Import password-protected Excel 2007 .xls files by including the Password argument.

Open( "Housing.xls", Password( "helloworld" ) );

Note: Password-protected Excel version 2010 and 2013 .xlsx files are not supported even if you save the file as .xls.

Import a Database

Open Database() opens a database using Open Database Connectivity (ODBC) and extracts data into a JMP data table. See the Database Access for more information.

JMP also converts DataBase Files (.dbf) files to data table format.

sasdbf = Open( "$SAMPLE_IMPORT_DATA/Bigclass.dbf",

	Use Labels for Var Names( 1 )

);