wrap

Wrap text for titles and labels.

Installing via net install

To install, user can use the net install command to download from the project's Github page:

net install wrap, from("https://aarondwolf.github.io/wrap")

Syntax

wrap {varname | anything} [if] , [ local(name, local_options) at(integer)]

Options

• local(name, options) Store wrapped string or variable label in local name
• at(integer) Wrap text at every Nth character. wrap will split the string only after completion of a full word. The default value is 40.

Local Options

• label Store wrapped string or variable label in local name
• relabel Wrap text at every Nth character. wrap will split the string only after completion of a full word. The default value is 40.

Description

wrap takes a single string argument (anything) or a variable name (varname) as and outputs the full text (if anything), or variable label and value labels (if varname) wrapped into multiple lines of text using compound double quotes.

If local(name) is specified, the command will store the string in local name.

When if is specified (only valid when using wrap varname), wrap will only wrap value labels for levels of varname where if holds.

Examples

Load NLSW data:

    .  sysuse nlsw88, clear

Example 1: Wrap a Simple String

We have a title that is too long, and want to wrap it after 40 characters (the default):

    .  wrap "The quick brown fox jumps over the lazy dog."

We can now use this in a graph:

    .  twoway scatter wage grade, title(`r(name)')

We can store the string in local title, and specify when it will wrap:

    .  wrap "The quick brown fox jumps over the laxy dog.", local(title) at(20)
    .  twoway scatter wage grade, title(`title')

Example 2: Wrap Variable Label

Suppose we want the same graph using the variable label of the y-variable as the title. We could use the wrap varname command:

    .  wrap wage, at(10)
    .  twoway scatter wage grade, title(`r(varlabel)')

Example 3: Wrap Value Labels

We may wish to wrap the text of all value labels attached to a variable. To do this, we could use the wrap varname command on a variable with attached value labels:

    .  wrap collgrad, at(15)
    .  return list

​ This is particularly useful when using the graph bar command with the relabel() option.

    .  wrap collgrad, at(15)
    .  graph bar wage, over(collgrad, relabel(`r(relabel)'))

We can omit subgroups in wrap to mirror what we specify in graph bar:

    .  wrap industry if industry != 4, at(15)
    .  graph hbar wage if industry != 4, over(industry, relabel(`r(relabel)'))

Example 4: Specify a local macro to store results

In all cases, we can optionally store the results in a local macro using the local() option:

    .  wrap "The quick brown fox jumps over the laxy dog.", local(title) at(20)
    .  wrap collgrad, local(title) at(10)

By default, wrap will store the variable label for varname in local(). If you wish to store the results in r(labels) (the value-label pairs), specify local(name, labels):

    .  wrap industry, local(ind_labs, labels) at(10)

If you wish to use the results in in r(relabel) (the value-label pairs re-numbered, starting at 1), specify local(name, relabel):

    .  wrap industry, local(ind_labs, relabel) at(10)    

You can use this to create wrapped labels for use with graph bar:

    .  wrap industry if industry != 4, local(ind_labs, relabel) at(15)
    .  graph hbar wage if industry != 4, over(industry, relabel(`ind_labs'))

Stored results

wrap stores the following in r():

If specifying wrap anything:

Scalars

• r(length): The length of the original string.
• r(pieces): The number of pieces (lines) the string was wrapped to.

Macros

• r(name): New (wrapped) string.

If specifying wrap varname:

Scalars

• r(length) : The length of the variable label.
• r(pieces): The number of pieces (lines) the variable label was wrapped to.

Macros

• r(varlab): New (wrapped) variable label.
• r(vallab): Value label for varname.
• r(label): New (wrapped) value labels.
• r(relabel): New (wrapped) value labels, re-numbered to comply with the relabel() option in graph bar.

Author

Aaron Wolf, Northwestern University [email protected]

Acknowledgements

wrap was inspired by an answer on Statlist by Scott Merryman to a similar question: https://www.stata.com/statalist/archive/2007-03/msg00778.html