showtable - Show data in nicely formatted columns
showtable [-options] [file]
Showtable reads an input data stream and displays it in a nicely formatted listing, with exact formatting depending upon the options. The input stream, file or
STDIN by default should consist of data separated by tabs or the defined separator character (see -d).
The actual output formatting is peformed by the ShowTable module.
There are two general sets of options: those which help determine the format of the input, and those which determine the format of the output.
Set the inter-column break string to "str". The default is a tab ("
\t"). If -strip is also given, blanks surrounding the break string will also be ignored.
This is the same as
Do not ignore lines of separators, such as dashes, equal signs, or underlines. If -nodashes is given, and these lines do occur in the stream, they will be treated as normal data.
Treat the first NN rows of data as column titles; multiple words in the column titles may wrap vertically. If NN is omitted, it defaults to 1. No -titles option is the same as -titles=0.
Set the input type as type, which can be one of: box, list, table, or simple. A simple-type table is the same as a table-type, but no wrapping characters are recognized.
Strip blanks from around the column values.
Do not strip blanks from the input. Useful if there is formatted or aligned data within a boxed table.
Use a table format for output, with wrapping of column values longer than the given or determined column widths. See Data::ShowTable for more details.
Use a simple table format, without any wrapping of column values. See Data::ShowTable for more details.
Use a list style format. See Data::ShowTable for more details.
Use a "boxed" style table. See Data::ShowTable for more details.
Use HTML-formating. See Data::ShowTable for more details.
Define the column names explicitly. This is useful for naming columns of data from
STDIN, when showtable is being used as a filter. The first column name, name1, cannot begin with a digit. This option allows any column titles obtained from the input to be overridden.
Do not output any headers on the tables; -titles=0 implies this option.
- -fn1[,n2, ..., nN]
Select fields numbered n1, n2, etc., to display. Each nN is a field index, or a range of indexes in the form:
MThe default is to show all the fields in each row. Fields are numbered from 1. An example: to show the first, and three through five fields of the
showtable -d: -f1,2-5 /etc/passwd
- -fields=fname1[,fname2, ..., fnameN]
Select the named fields to display. The field names must be available, either through the data stream, or by using the -titles option. The field names given must match the existing field names exactly.
Using the file
/etc/passwdfor another example: to show the same first two fields, by name:
showtable -d: -titles=Login,UID -fields=Login,UID /etc/passwd
Set the maximum table width. This value is applied to the variable Data::Showtable::Max_Table_Width. When the total width of all columns to be displayed exceeds this value, all column widths are scaled uniformly.
If -width is not given, then for all output but -html, the default value is either "
COLUMNS", if defined, or 80, if not. Whith -html mode, there is no default value for -width; in other words, there is no limit to the width.
Set individual column widths to the specified values. Empty column widths imply no maximum width. If the -width option is also given, then the -cwidth column widths can also be given as fractions or percentages.
Example: To set the maximum width of the third column to 20 characters:
Do not perform HTML escape sequences on the data; this allows embedded HTML text in the data to be displayed properly with the -html option.
- -attributes='attr1 attr2 ...'
Declare the table attributes, which are inserted into the
TABLEtoken. For example, the option:
-attributes='BORDER=0 CELLSPACING=2 CELLPADDING=4'
would cause the following HTML:
<TABLE BORDER=0 CELLSPACING=2 CELLPADDING=4>
The default table attributes are:
<TABLE BORDER=1 CELLSPACING=1 CELLPADDING=1>
Set the HTML formats for the column titles. The -title_formats (or just -tf) can be given multiple times, for each column, or formats for multiple columns can be given on the same option separated by semi-colons "
Each fmtN can itself be multiple HTML items, separated by commas. Each HTML element can be given either as an HTML token (eg: "
\<BOLD\>"), or as a plain name (eg: "
For example, here is a title format specification for three columns, where the first column title should be bold italic, the second italic, and the third italic in a smaller font:
The same as -title_formats but applies to the column data.
Define a mapping from column names, or indexes, to URLs to be inserted as <A HREF's> around the values for the named columns. Each colN is a column name or index, and each urlN is a string representing the URL to be inserted for the given column.
The URL text may contain these substitution strings:
%K - will be substituted with the current column name (or key).
%V - will be substituted with the current column value.
Multiple -url options may be given, if desired, rather than creating one long argument for a single -url. For example:
showtable -d: -f1,6 -titles=Login,Homedir \ -url='Login=mailto:%V' \ -url='HomeDir=file:%V' \ /etc/passwd
If the input type is box, then vertical and horizontal box characters are removed from the input stream, and blanks surrounding the vertical box characters are removed. The vertical box characters (column separaters) are "
|" or "
:". The The horizontal box characters are "
+" and "
Morever, data wrapped within a column is recognized and parsed as one column value, by recognizing the presence of a wrapping prefix or wrapping suffix character. Currently, the wrapping prefix character is "<", and the wrapping suffix character is ">".
An example of data wrapped within a column is given here. The table below has just two logical rows of data; with both rows having data wrapped into multiple physical rows.
+---------+---------+---------+ | Col 1 | Col 2 | Col 3 | +---------+---------+---------+ | This is>| Another>| Row 1,3>| |< a cont>|< value. |<is also>| |<inued >| |<long. | |<value. | | | |This is >| Item2-2 | Item2-3 | +---------+---------+---------+
When using the -list or -input=list options, either, or both, the input and output may be in a "list" format, which is implemented using the following syntax:
r1c1_name: r1c1_value r1c2_name: r1c2_value ... r1cN_name: r1cN_value r2c1_name: r2c1_value r2c2_name: r2c2_value : r2c2_value_continued ... r2cN_name: r2cN_value rMc1_name: rMc1_value rMc2_name: rMc2_value ... rMcN_name: rMcN_value
Each row of data consists of one or more columns, and ends with a blank line.
Each column consists of a column name, followed by a colon ":", followed by an optional, single space or tab, followed by the column value, on the same line.
Continuation lines of the previous column value consist of one or more space or tab characters, a colon ":", one optional, single space or tab, followed by the continuation value. In the example above, The second column value of the second row was continued.
When using -html on data already containing HTML-formatted text, the -noescape option should be used. By default, all input text is assumed not to be HTML-formatted, and is escaped allowing embedded "<", ">" characters, if any, to be displayed correctly.
Alan K. Stebbens email@example.com
Currently, the box formatting characters are not configurable: '+' for the corners; '-' and '|' for the tops and sides, respectively. In an ideal world, these things would be configurable.
The continuation prefix and suffix characters, '<' and '>', respectively, are also not configurable:
When reading table input, any data ending with ">" will be considered to be continued by the next row of data. To avoid this, use -input=simple.
When selecting noncontiguous fields (ie: -f1,4>) without field names, the default field names will be consecutively numbered from 1, which is counter-intuitive to the original selection. To avoid this, name the fields using the -title=... option.