cuberite/TCLAP
1
0
Fork 0
mirror of https://github.com/cuberite/TCLAP.git synced 2025-09-07 19:29:07 -04:00
Code Issues Packages Projects Releases Wiki Activity

Updated the example

Browse Source
This commit is contained in:
macbishop 2006-04-18 10:59:16 +00:00
parent d45405908c
commit e7f9d30e5d
2 changed files with 297 additions and 267 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

527
docs/manual.html
View File

@ -1,26 +1,26 @@
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>Templatized C++ Command Line Parser Manual</title><link rel="stylesheet" href="style.css" type="text/css" /><meta name="generator" content="DocBook XSL Stylesheets V1.61.3" /></head><body><div class="book" lang="en" xml:lang="en"><div class="titlepage"><div><div><h1 class="title"><a id="id2868300"></a>Templatized C++ Command Line Parser Manual</h1></div><div><div class="author"><h3 class="author"><span class="firstname">Michael</span> <span class="othername">E</span> <span class="surname">Smoot</span></h3></div></div><div><p class="copyright">Copyright © 2003,2004,2005 Michael E. Smoot</p></div></div><div></div><hr /></div><div class="toc"><p><b>Table of Contents</b></p><dl><dt>1. <a href="#id2868084">Basic Usage</a></dt><dd><dl><dt><a href="#ARG_PROPERTIES">Argument Properties</a></dt><dt><a href="#ARGUMENT_TYPES">Types of Arguments</a></dt><dt><a href="#COMPILING">Compiling</a></dt></dl></dd><dt>2. <a href="#COMPLICATIONS">Complications</a></dt><dd><dl><dt><a href="#COMBINE_SWITCHES">I want to combine multiple switches into one argument...</a></dt><dt><a href="#MULTI_ARG">I tried passing multiple values on the command line with the
same flag and it didn't work...</a></dt><dt><a href="#UNLABELED_VALUE_ARG">I don't like labelling all of my arguments...</a></dt><dt><a href="#UNLABELED_MULTI_ARG">I want an arbitrary number of unlabeled arguments to be accepted...</a></dt><dt><a href="#XOR">I want one argument or the other, but not both...</a></dt><dt><a href="#NO_FLAG">I have more arguments than single flags make sense for...</a></dt><dt><a href="#CONSTRAINT">I want to constrain the values allowed for a particular
argument...</a></dt><dt><a href="#ARG_ADD_CMDLINE">I want the Args to add themselves to the CmdLine...</a></dt><dt><a href="#CHANGE_OUTPUT">I want different output than what is provided...</a></dt></dl></dd><dt>3. <a href="#EXCEPTIONS">Exceptions to the Rules</a></dt><dd><dl><dt><a href="#IGNORE_ARGS">Ignoring arguments</a></dt><dt><a href="#MULTIPLE_IDENTICAL_SWITCHES">Multiple Identical Switches</a></dt><dt><a href="#DESCRIPTION_EXCEPTIONS">Type Descriptions</a></dt></dl></dd><dt>4. <a href="#VISITORS">Visitors</a></dt><dt>5. <a href="#MORE_INFO">More Information</a></dt></dl></div><div class="chapter" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title"><a id="id2868084"></a>Chapter 1. Basic Usage</h2></div></div><div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl><dt><a href="#ARG_PROPERTIES">Argument Properties</a></dt><dt><a href="#ARGUMENT_TYPES">Types of Arguments</a></dt><dt><a href="#COMPILING">Compiling</a></dt></dl></div><p>
<html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>Templatized C++ Command Line Parser Manual</title><link rel="stylesheet" href="style.css" type="text/css" /><meta name="generator" content="DocBook XSL Stylesheets V1.68.1" /></head><body><div class="book" lang="en" xml:lang="en"><div class="titlepage"><div><div><h1 class="title"><a id="id765455"></a>Templatized C++ Command Line Parser Manual</h1></div><div><div class="author"><h3 class="author"><span class="firstname">Michael</span> <span class="othername">E</span> <span class="surname">Smoot</span></h3></div></div><div><p class="copyright">Copyright © 2003,2004,2005 Michael E. Smoot</p></div></div><hr /></div><div class="toc"><p><b>Table of Contents</b></p><dl><dt><span class="chapter"><a href="#id859483">1. Basic Usage</a></span></dt><dd><dl><dt><span class="sect1"><a href="#ARG_PROPERTIES">Argument Properties</a></span></dt><dt><span class="sect1"><a href="#ARGUMENT_TYPES">Types of Arguments</a></span></dt><dt><span class="sect1"><a href="#COMPILING">Compiling</a></span></dt></dl></dd><dt><span class="chapter"><a href="#COMPLICATIONS">2. Complications</a></span></dt><dd><dl><dt><span class="sect1"><a href="#COMBINE_SWITCHES">I want to combine multiple switches into one argument...</a></span></dt><dt><span class="sect1"><a href="#MULTI_ARG">I tried passing multiple values on the command line with the
same flag and it didn't work...</a></span></dt><dt><span class="sect1"><a href="#UNLABELED_VALUE_ARG">I don't like labelling all of my arguments...</a></span></dt><dt><span class="sect1"><a href="#UNLABELED_MULTI_ARG">I want an arbitrary number of unlabeled arguments to be accepted...</a></span></dt><dt><span class="sect1"><a href="#XOR">I want one argument or the other, but not both...</a></span></dt><dt><span class="sect1"><a href="#NO_FLAG">I have more arguments than single flags make sense for...</a></span></dt><dt><span class="sect1"><a href="#CONSTRAINT">I want to constrain the values allowed for a particular
argument...</a></span></dt><dt><span class="sect1"><a href="#ARG_ADD_CMDLINE">I want the Args to add themselves to the CmdLine...</a></span></dt><dt><span class="sect1"><a href="#CHANGE_OUTPUT">I want different output than what is provided...</a></span></dt></dl></dd><dt><span class="chapter"><a href="#EXCEPTIONS">3. Exceptions to the Rules</a></span></dt><dd><dl><dt><span class="sect1"><a href="#IGNORE_ARGS">Ignoring arguments</a></span></dt><dt><span class="sect1"><a href="#MULTIPLE_IDENTICAL_SWITCHES">Multiple Identical Switches</a></span></dt><dt><span class="sect1"><a href="#DESCRIPTION_EXCEPTIONS">Type Descriptions</a></span></dt></dl></dd><dt><span class="chapter"><a href="#VISITORS">4. Visitors</a></span></dt><dt><span class="chapter"><a href="#MORE_INFO">5. More Information</a></span></dt></dl></div><div class="chapter" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title"><a id="id859483"></a>Chapter 1. Basic Usage</h2></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl><dt><span class="sect1"><a href="#ARG_PROPERTIES">Argument Properties</a></span></dt><dt><span class="sect1"><a href="#ARGUMENT_TYPES">Types of Arguments</a></span></dt><dt><span class="sect1"><a href="#COMPILING">Compiling</a></span></dt></dl></div><p>
<span class="emphasis"><em>TCLAP</em></span> has a few key classes to be aware of.
The first is the
<tt class="classname">CmdLine</tt> (command line) class. This class parses
<code class="classname">CmdLine</code> (command line) class. This class parses
the command line passed to it according to the arguments that it
contains. Arguments are separate objects that are added to the
<tt class="classname">CmdLine</tt> object one at a time. The six
argument classes are: <tt class="classname">ValueArg</tt>,
<tt class="classname">UnlabeledValueArg</tt>,
<tt class="classname">SwitchArg</tt>, <tt class="classname">MultiSwitchArg</tt>,
<tt class="classname">MultiArg</tt> and
<tt class="classname">UnlabeledMultiArg</tt>.
<code class="classname">CmdLine</code> object one at a time. The six
argument classes are: <code class="classname">ValueArg</code>,
<code class="classname">UnlabeledValueArg</code>,
<code class="classname">SwitchArg</code>, <code class="classname">MultiSwitchArg</code>,
<code class="classname">MultiArg</code> and
<code class="classname">UnlabeledMultiArg</code>.
These classes are templatized, which means they can be defined to parse
a value of any <a href="#FOOTNOTES"> type**</a>. Once you add the
arguments to the <tt class="classname">CmdLine</tt> object, it parses the
arguments to the <code class="classname">CmdLine</code> object, it parses the
command line
and assigns the data it finds to the specific argument objects it
contains. Your program accesses the values parsed by
calls to the <tt class="methodname">getValue()</tt> methods of the
calls to the <code class="methodname">getValue()</code> methods of the
argument objects.
</p><p>
Here is a simple <a href="test1.cpp" target="_top"> example</a> ...
@ -31,25 +31,40 @@ Here is a simple <a href="test1.cpp" target="_top"> example</a> ...
#include &lt;algorithm&gt;
#include &lt;tclap/CmdLine.h&gt;
using namespace TCLAP;
using namespace std;
int main(int argc, char** argv)
{
using std::string;
using std::cout;
using std::cerr;
using std::endl;
// Wrap everything in a try block. Do this every time,
// because exceptions will be thrown for problems.
// because exceptions will be thrown for problems.
try {
// Define the command line object.
CmdLine cmd(&quot;Command description message&quot;, ' ', &quot;0.9&quot;);
// Define the command line object, and insert a messages
//that tells you what the program does etc.
//The "Command description message" is printed last in the help
//text. The second argument is the delimiter (usually space) and
//the last one is the version number. The CmdLine object is used
//for parsing.
TCLAP::CmdLine cmd("Command description message", ' ', "0.9");
// Define a value argument and add it to the command line.
ValueArg&lt;string&gt; nameArg(&quot;n&quot;,&quot;name&quot;,&quot;Name to print&quot;,true,&quot;homer&quot;,&quot;string&quot;);
// A value arg takes a switch and a value such as -n Bishop
TCLAP::ValueArg&lt;string&gt; nameArg("n","name","Name to print",true,"homer","string");
// Add the argument nameArg to the command line object. This
// makes it possible to match the argument on the command line
// during parsing.
cmd.add( nameArg );
// Define a switch and add it to the command line.
SwitchArg reverseSwitch(&quot;r&quot;,&quot;reverse&quot;,&quot;Print name backwards&quot;, false);
cmd.add( reverseSwitch );
// A switch arg is a binary argument and only takes a switch
// (true/false) such as -r. Also the command line object is added
// directly while creating the SwitchArg, eliminating the need
// for the call to cmd.add(), this can be used with any type of Arg.
TCLAP::SwitchArg reverseSwitch("r","reverse","Print name backwards", false, cmd);
// Parse the args.
cmd.parse( argc, argv );
@ -61,15 +76,15 @@ int main(int argc, char** argv)
// Do what you intend too...
if ( reverseName )
{
reverse(name.begin(),name.end());
cout &lt;&lt; &quot;My name (spelled backwards) is: &quot; &lt;&lt; name &lt;&lt; endl;
std::reverse(name.begin(),name.end());
cout &lt;&lt; "My name (spelled backwards) is: " &lt;&lt; name &lt;&lt; endl;
}
else
cout &lt;&lt; &quot;My name is: &quot; &lt;&lt; name &lt;&lt; endl;
cout &lt;&lt; "My name is: " &lt;&lt; name &lt;&lt; endl;
} catch (ArgException &amp;e) // catch any exceptions
{ cerr &lt;&lt; &quot;error: &quot; &lt;&lt; e.error() &lt;&lt; &quot; for arg &quot; &lt;&lt; e.argId() &lt;&lt; endl; }
} catch (TCLAP::ArgException &amp;e) // catch any exceptions
{ cerr &lt;&lt; "error: " &lt;&lt; e.error() &lt;&lt; " for arg " &lt;&lt; e.argId() &lt;&lt; endl; }
}
</pre><p>
@ -129,85 +144,85 @@ Where:
This example shows a number of different properties of the
library...
</p><div class="itemizedlist"><ul type="disc"><li>Arguments can appear in any order (...mostly,
<a href="#COMPLICATIONS" title="Chapter 2. Complications"> more</a> on this later).</li><li>The <i class="parameter"><tt>help</tt></i>, <i class="parameter"><tt>version</tt></i>
and <i class="parameter"><tt>--</tt></i><tt class="classname">SwitchArg</tt>s
are specified automatically. Using either the <i class="parameter"><tt>-h</tt></i> or
<i class="parameter"><tt>--help</tt></i> flag will cause the USAGE message to be displayed,
<i class="parameter"><tt>-v</tt></i> or <i class="parameter"><tt>--version</tt></i> will cause
<a href="#COMPLICATIONS" title="Chapter 2. Complications"> more</a> on this later).</li><li>The <em class="parameter"><code>help</code></em>, <em class="parameter"><code>version</code></em>
and <em class="parameter"><code>--</code></em><code class="classname">SwitchArg</code>s
are specified automatically. Using either the <em class="parameter"><code>-h</code></em> or
<em class="parameter"><code>--help</code></em> flag will cause the USAGE message to be displayed,
<em class="parameter"><code>-v</code></em> or <em class="parameter"><code>--version</code></em> will cause
any version information to
be displayed, and <i class="parameter"><tt>--</tt></i> or
<i class="parameter"><tt>--ignore_rest</tt></i> will cause the
be displayed, and <em class="parameter"><code>--</code></em> or
<em class="parameter"><code>--ignore_rest</code></em> will cause the
remaining labeled arguments to be ingored. These switches are
included automatically on every command line and there is no way to
turn this off (unless you change <tt class="filename">CmdLine.h</tt> yourself).
turn this off (unless you change <code class="filename">CmdLine.h</code> yourself).
More <a href="#VISITORS" title="Chapter 4. Visitors"> later</a> on how we get this to
work.</li><li>If there is an error parsing the command line (e.g. a required
argument isn't provided), the program exits and displays a brief
USAGE and an error message.</li><li>The program name is assumed to always be argv[0], so it isn't
specified directly.</li><li>A delimiter character can be specified. This means that if you
prefer arguments of the style <i class="parameter"><tt>-s=asdf</tt></i> instead of
<i class="parameter"><tt>-s asdf</tt></i>, you can do so.</li><li><span class="emphasis"><em>Always wrap everything in a try block that catches
prefer arguments of the style <em class="parameter"><code>-s=asdf</code></em> instead of
<em class="parameter"><code>-s asdf</code></em>, you can do so.</li><li><span class="emphasis"><em>Always wrap everything in a try block that catches
ArgExceptions!</em></span> Any problems found in constructing the
<tt class="classname">CmdLine</tt> or the <tt class="classname">Arg</tt>s will
throw an <tt class="classname">ArgException</tt>.</li></ul></div><p>
</p><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="ARG_PROPERTIES"></a>Argument Properties</h2></div></div><div></div></div><p>
<code class="classname">CmdLine</code> or the <code class="classname">Arg</code>s will
throw an <code class="classname">ArgException</code>.</li></ul></div><p>
</p><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="ARG_PROPERTIES"></a>Argument Properties</h2></div></div></div><p>
Arguments, whatever their type, have a few common basic properties.
These properties are set in the constructors of the arguments.
</p><div class="itemizedlist"><ul type="disc"><li>First is the flag or the character preceeded by a dash(-) that
signals the beginning of the argument on the command line.</li><li>Arguments also have names, which can, if desired also be used
as a flag on the command line, this time preceeded by two dashes
(--) [like the familiar <tt class="function">getopt_long()</tt>].</li><li>Next is the description of the argument. This is a short
(--) [like the familiar <code class="function">getopt_long()</code>].</li><li>Next is the description of the argument. This is a short
description of the argument displayed in the help/usage message
when needed.</li><li>The boolean value in <tt class="classname">ValueArg</tt>s
when needed.</li><li>The boolean value in <code class="classname">ValueArg</code>s
indicates whether the
argument is required to be present (<tt class="classname">SwitchArg</tt>s
argument is required to be present (<code class="classname">SwitchArg</code>s
can't be required, as that would defeat the purpose).</li><li>Next, the default value the arg should assume if the arg isn't
required or entered on the command line.</li><li>Last, for <tt class="classname">ValueArg</tt>s is a short
required or entered on the command line.</li><li>Last, for <code class="classname">ValueArg</code>s is a short
description of the type
that the argument expects (yes its an ugly
<a href="#DESCRIPTION_EXCEPTIONS" title="Type Descriptions"> hack</a>).
Note that the order of
arguments on the command line (so far) doesn't matter. Any argument
not matching an <tt class="classname">Arg</tt> added to the command
not matching an <code class="classname">Arg</code> added to the command
line will cause an
exception to be thrown (<a href="#COMPLICATIONS" title="Chapter 2. Complications"> for the
most part</a>, with some <a href="#EXCEPTIONS" title="Chapter 3. Exceptions to the Rules"> exceptions</a>).
</li></ul></div><p>
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="ARGUMENT_TYPES"></a>Types of Arguments</h2></div></div><div></div></div><p>
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="ARGUMENT_TYPES"></a>Types of Arguments</h2></div></div></div><p>
There are two primary types of arguments:
</p><p>
</p><div class="itemizedlist"><ul type="disc"><li><tt class="classname">SwitchArg</tt>s are what the name implies:
simple, on/off, boolean switches. Use <tt class="classname">SwitchArg</tt>s
</p><div class="itemizedlist"><ul type="disc"><li><code class="classname">SwitchArg</code>s are what the name implies:
simple, on/off, boolean switches. Use <code class="classname">SwitchArg</code>s
anytime you want to turn
some sort of system property on or off. <tt class="classname">SwitchArg</tt>s
don't parse a value. They return <tt class="constant">TRUE</tt> or
<tt class="constant">FALSE</tt>, depending on whether the switch has been found
on the command line and what the default value was defined as.</li><li><tt class="classname">ValueArg</tt>s are arguments that read a
some sort of system property on or off. <code class="classname">SwitchArg</code>s
don't parse a value. They return <code class="constant">TRUE</code> or
<code class="constant">FALSE</code>, depending on whether the switch has been found
on the command line and what the default value was defined as.</li><li><code class="classname">ValueArg</code>s are arguments that read a
value of some type
from the command line. Any time you need a file name, a number,
etc. use a <tt class="classname">ValueArg</tt> or one of its variants.
etc. use a <code class="classname">ValueArg</code> or one of its variants.
<a href="#UNLABELED_VALUE_ARG" title="I don't like labelling all of my arguments..."> UnlabedValueArg</a>,
<a href="#MULTI_ARG" title="I tried passing multiple values on the command line with the&#10;same flag and it didn't work..."> MultiArg</a>, and
<a href="#UNLABELED_MULTI_ARG" title="I want an arbitrary number of unlabeled arguments to be accepted..."> UnlabeledMultiArg</a> are
special cases of <tt class="classname">ValueArg</tt>s and are described below. All
<tt class="classname">ValueArg</tt>s are
special cases of <code class="classname">ValueArg</code>s and are described below. All
<code class="classname">ValueArg</code>s are
<a href="#FOOTNOTES"> templatized**</a> and will attempt to parse
the string its flag matches on the command line as the type it is
specified as. <tt class="classname">ValueArg&lt;int&gt;</tt>
specified as. <code class="classname">ValueArg&lt;int&gt;</code>
will attempt to parse an
int, <tt class="classname">ValueArg&lt;float&gt;</tt> will attempt to
parse a float, etc. If <tt class="methodname">operator&gt;&gt;</tt>
int, <code class="classname">ValueArg&lt;float&gt;</code> will attempt to
parse a float, etc. If <code class="methodname">operator&gt;&gt;</code>
for the specified type doesn't
recognize the string on the command line as its defined type, then
an exception will be thrown.</li></ul></div><p>
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="COMPILING"></a>Compiling</h2></div></div><div></div></div><p>
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="COMPILING"></a>Compiling</h2></div></div></div><p>
<span class="emphasis"><em>TCLAP</em></span> is implemented entirely in header files
which means you only need to include CmdLine.h to use the library.
</p><pre class="programlisting">
#include &lt;tclap/CmdLine.h&gt;
</pre><p>
You'll need to make sure that your compiler can see the header
files. If you do the usual &quot;make install&quot; then your compiler should
files. If you do the usual "make install" then your compiler should
see the files by default. Alternatively, you can use the -I
complier argument to specify the exact location of the libraries.
</p><pre class="programlisting">
@ -229,37 +244,37 @@ to work on Windows, Sun and Alpha platforms. We've made every
effort to keep the library compliant with the ANSI C++ standard so
if your compiler meets the standard, then this library should work
for you. Please let us know if this is not the case!
</p><div class="sect2" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="id2853441"></a>Windows Note</h3></div></div><div></div></div><p>
</p><div class="sect2" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="id859977"></a>Windows Note</h3></div></div></div><p>
As we understand things, Visual C++ does not have the file
<tt class="filename">config.h</tt> which is used to make platform
<code class="filename">config.h</code> which is used to make platform
specific definitions. In this situation, we assume that you
have access to <tt class="classname">sstream</tt>. Our understanding is that
have access to <code class="classname">sstream</code>. Our understanding is that
this should not be a problem for VC++ 7.x. However, if this
is not the case and you need to use <tt class="classname">strstream</tt>,
is not the case and you need to use <code class="classname">strstream</code>,
then simply tell your compiler to define the variable
<tt class="constant">HAVE_STRSTREAM</tt> and undefine
<tt class="constant">HAVE_SSTREAM</tt> That
<code class="constant">HAVE_STRSTREAM</code> and undefine
<code class="constant">HAVE_SSTREAM</code> That
<span class="emphasis"><em>should</em></span> work. We think. Alternatively, just edit
the files <tt class="filename">ValueArg.h</tt> and <tt class="filename">MultiArg.h</tt>.
the files <code class="filename">ValueArg.h</code> and <code class="filename">MultiArg.h</code>.
</p></div><p>
</p><div class="sect2" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="id2853501"></a>Random Note</h3></div></div><div></div></div><p>
If your compiler doesn't support the <tt class="methodname">using</tt> syntax used
in <tt class="classname">UnlabeledValueArg</tt> and
<tt class="classname">UnlabeledMultiArg</tt> to support two stage name lookup,
</p><div class="sect2" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="id860029"></a>Random Note</h3></div></div></div><p>
If your compiler doesn't support the <code class="methodname">using</code> syntax used
in <code class="classname">UnlabeledValueArg</code> and
<code class="classname">UnlabeledMultiArg</code> to support two stage name lookup,
then you have two options. Either comment out the statements if you don't
need two stage name lookup, or do a bunch of search and replace and use
the <tt class="methodname">this</tt> pointer syntax: e.g.
<tt class="methodname">this-&gt;_ignoreable</tt> instead
of just <tt class="methodname">_ignorable</tt> (do this for each variable
or method referenced by <tt class="methodname">using</tt>).
the <code class="methodname">this</code> pointer syntax: e.g.
<code class="methodname">this-&gt;_ignoreable</code> instead
of just <code class="methodname">_ignorable</code> (do this for each variable
or method referenced by <code class="methodname">using</code>).
</p></div><p>
</p></div></div><div class="chapter" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title"><a id="COMPLICATIONS"></a>Chapter 2. Complications</h2></div></div><div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl><dt><a href="#COMBINE_SWITCHES">I want to combine multiple switches into one argument...</a></dt><dt><a href="#MULTI_ARG">I tried passing multiple values on the command line with the
same flag and it didn't work...</a></dt><dt><a href="#UNLABELED_VALUE_ARG">I don't like labelling all of my arguments...</a></dt><dt><a href="#UNLABELED_MULTI_ARG">I want an arbitrary number of unlabeled arguments to be accepted...</a></dt><dt><a href="#XOR">I want one argument or the other, but not both...</a></dt><dt><a href="#NO_FLAG">I have more arguments than single flags make sense for...</a></dt><dt><a href="#CONSTRAINT">I want to constrain the values allowed for a particular
argument...</a></dt><dt><a href="#ARG_ADD_CMDLINE">I want the Args to add themselves to the CmdLine...</a></dt><dt><a href="#CHANGE_OUTPUT">I want different output than what is provided...</a></dt></dl></div><p>
</p></div></div><div class="chapter" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title"><a id="COMPLICATIONS"></a>Chapter 2. Complications</h2></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl><dt><span class="sect1"><a href="#COMBINE_SWITCHES">I want to combine multiple switches into one argument...</a></span></dt><dt><span class="sect1"><a href="#MULTI_ARG">I tried passing multiple values on the command line with the
same flag and it didn't work...</a></span></dt><dt><span class="sect1"><a href="#UNLABELED_VALUE_ARG">I don't like labelling all of my arguments...</a></span></dt><dt><span class="sect1"><a href="#UNLABELED_MULTI_ARG">I want an arbitrary number of unlabeled arguments to be accepted...</a></span></dt><dt><span class="sect1"><a href="#XOR">I want one argument or the other, but not both...</a></span></dt><dt><span class="sect1"><a href="#NO_FLAG">I have more arguments than single flags make sense for...</a></span></dt><dt><span class="sect1"><a href="#CONSTRAINT">I want to constrain the values allowed for a particular
argument...</a></span></dt><dt><span class="sect1"><a href="#ARG_ADD_CMDLINE">I want the Args to add themselves to the CmdLine...</a></span></dt><dt><span class="sect1"><a href="#CHANGE_OUTPUT">I want different output than what is provided...</a></span></dt></dl></div><p>
Naturally, what we have seen to this point doesn't satisfy all of
our needs.
</p><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="COMBINE_SWITCHES"></a>I want to combine multiple switches into one argument...</h2></div></div><div></div></div><p>
Multiple <tt class="classname">SwitchArg</tt>s can be combined into a
</p><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="COMBINE_SWITCHES"></a>I want to combine multiple switches into one argument...</h2></div></div></div><p>
Multiple <code class="classname">SwitchArg</code>s can be combined into a
single argument on the command line. If you have switches -a, -b and -c
it is valid to do either:
@ -282,11 +297,11 @@ it is valid to do either:
This is to make this library more in line with the POSIX and GNU
standards (as I understand them).
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="MULTI_ARG"></a>I tried passing multiple values on the command line with the
same flag and it didn't work...</h2></div></div><div></div></div><p>
Correct. You can neither specify mulitple <tt class="classname">ValueArg</tt>s
or <tt class="classname">SwitchArg</tt>s with the same flag in the code nor
same flag and it didn't work...</h2></div></div></div><p>
Correct. You can neither specify mulitple <code class="classname">ValueArg</code>s
or <code class="classname">SwitchArg</code>s with the same flag in the code nor
on the command line. Exceptions will occur in either case.
For <tt class="classname">SwitchArg</tt>s
For <code class="classname">SwitchArg</code>s
it simply doesn't make sense to allow a particular flag to be
turned on or off repeatedly on the command line. All you should
ever need is to set your state <span class="emphasis"><em>once</em></span> by specifying
@ -300,81 +315,81 @@ libraries...
% fooCompiler -L /dir/num1 -L /dir/num2 file.foo
</pre><p>
In situations like this, you will want to use a
<tt class="classname">MultiArg</tt>. A
<tt class="classname">MultiArg</tt> is essentially a
<tt class="classname">ValueArg</tt> that appends any
<code class="classname">MultiArg</code>. A
<code class="classname">MultiArg</code> is essentially a
<code class="classname">ValueArg</code> that appends any
value that it matches and parses onto a vector of values. When the
<tt class="methodname">getValue()</tt> method is called, a vector of
<code class="methodname">getValue()</code> method is called, a vector of
values, instead of a single value is returned. A
<tt class="classname">MultiArg</tt> is declared much like
a <tt class="classname">ValueArg</tt>:
<code class="classname">MultiArg</code> is declared much like
a <code class="classname">ValueArg</code>:
</p><pre class="programlisting">
MultiArg&lt;int&gt; itest(&quot;i&quot;, &quot;intTest&quot;, &quot;multi int test&quot;, false,&quot;int&quot; );
MultiArg&lt;int&gt; itest("i", "intTest", "multi int test", false,"int" );
cmd.add( itest );
</pre><p>
Note that <tt class="classname">MultiArg</tt>s can be added to the
<tt class="classname">CmdLine</tt> in any order (unlike
Note that <code class="classname">MultiArg</code>s can be added to the
<code class="classname">CmdLine</code> in any order (unlike
<a href="#UNLABELED_MULTI_ARG" title="I want an arbitrary number of unlabeled arguments to be accepted..."> UnlabeledMultiArg</a>).
</p><p>
<span class="emphasis"><em>New Feature!</em></span> <tt class="classname">MultiSwitchArg</tt> now
<span class="emphasis"><em>New Feature!</em></span> <code class="classname">MultiSwitchArg</code> now
allows you to set a switch multiple times on the command line. The call
to <tt class="methodname">getValue()</tt> returns the number (int) of times
to <code class="methodname">getValue()</code> returns the number (int) of times
the switch
has been found on the command line in addition to the default value.
Here is an example using the default initial value of 0:
</p><pre class="programlisting">
MultiSwitchArg quiet(&quot;q&quot;,&quot;quiet&quot;,&quot;Reduce the volume of output&quot;);
MultiSwitchArg quiet("q","quiet","Reduce the volume of output");
cmd.add( quiet );
</pre><p>
Alternatively, you can specify your own initial value:
</p><pre class="programlisting">
MultiSwitchArg quiet(&quot;q&quot;,&quot;quiet&quot;,&quot;Reduce the volume of output&quot;,5);
MultiSwitchArg quiet("q","quiet","Reduce the volume of output",5);
cmd.add( quiet );
</pre><p>
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="UNLABELED_VALUE_ARG"></a>I don't like labelling all of my arguments...</h2></div></div><div></div></div><p>
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="UNLABELED_VALUE_ARG"></a>I don't like labelling all of my arguments...</h2></div></div></div><p>
To this point all of our arguments have had labels (flags)
indentifying them on the command line, but there are some
situations where flags are burdensome and not worth the effort. One
example might be if you want to implement a magical command we'll
call <b class="command">copy</b>. All <b class="command">copy</b> does is
call <span><strong class="command">copy</strong></span>. All <span><strong class="command">copy</strong></span> does is
copy the file specified
in the first argument to the file specified in the second argument.
We can do this using <tt class="classname">UnlabeledValueArg</tt>s which are pretty
much just <tt class="classname">ValueArg</tt>s without the flag specified,
We can do this using <code class="classname">UnlabeledValueArg</code>s which are pretty
much just <code class="classname">ValueArg</code>s without the flag specified,
which tells
the <tt class="classname">CmdLine</tt> object to treat them accordingly.
the <code class="classname">CmdLine</code> object to treat them accordingly.
The code would look like this:
</p><pre class="programlisting">
UnlabeledValueArg&lt;float&gt; nolabel( &quot;name&quot;, &quot;unlabeled test&quot;, 3.14,
&quot;nameString&quot; );
UnlabeledValueArg&lt;float&gt; nolabel( "name", "unlabeled test", 3.14,
"nameString" );
cmd.add( nolabel );
</pre><p>
Everything else is handled identically to what is seen above. The
only difference to be aware of, and this is important: <span class="emphasis"><em>the order
that UnlabeledValueArgs are added to the <tt class="classname">CmdLine</tt>
that UnlabeledValueArgs are added to the <code class="classname">CmdLine</code>
is the order that they will be parsed!!!!</em></span>
This is <span class="emphasis"><em>not</em></span> the case for normal
<tt class="classname">SwitchArg</tt>s and <tt class="classname">ValueArg</tt>s.
<code class="classname">SwitchArg</code>s and <code class="classname">ValueArg</code>s.
What happens internally is the first argument that the
<tt class="classname">CmdLine</tt> doesn't recognize is assumed to be
the first <tt class="classname">UnlabeledValueArg</tt> and
<code class="classname">CmdLine</code> doesn't recognize is assumed to be
the first <code class="classname">UnlabeledValueArg</code> and
parses it as such. Note that you are allowed to intersperse labeled
args (SwitchArgs and ValueArgs) in between
<tt class="classname">UnlabeledValueArgs</tt> (either on the command line
or in the declaration), but the <tt class="classname">UnlabeledValueArgs</tt>
<code class="classname">UnlabeledValueArgs</code> (either on the command line
or in the declaration), but the <code class="classname">UnlabeledValueArgs</code>
will still be parsed in the order they are added. Just remember that order is
important for unlabeled arguments.
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="UNLABELED_MULTI_ARG"></a>I want an arbitrary number of unlabeled arguments to be accepted...</h2></div></div><div></div></div><p>
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="UNLABELED_MULTI_ARG"></a>I want an arbitrary number of unlabeled arguments to be accepted...</h2></div></div></div><p>
Don't worry, we've got you covered. Say you want a strange command
that searches each file specified for a given string (let's call it
<b class="command">grep</b>), but you don't want to have to type in all of the file
<span><strong class="command">grep</strong></span>), but you don't want to have to type in all of the file
names or write a script to do it for you. Say,
</p><pre class="programlisting">
@ -382,7 +397,7 @@ names or write a script to do it for you. Say,
</pre><p>
First remember that the <span class="emphasis"><em>*</em></span> is handled by the shell and
expanded accordingly, so what the program <b class="command">grep</b> sees is
expanded accordingly, so what the program <span><strong class="command">grep</strong></span> sees is
really something like:
</p><pre class="programlisting">
@ -390,13 +405,13 @@ really something like:
</pre><p>
To handle situations where multiple, unlabled arguments are needed,
we provide the <tt class="classname">UnlabeledMultiArg</tt>.
<tt class="classname">UnlabeledMultiArg</tt>s
we provide the <code class="classname">UnlabeledMultiArg</code>.
<code class="classname">UnlabeledMultiArg</code>s
are declared much like everything else, but with only a description
of the arguments. By default, if an <tt class="classname">UnlabeledMultiArg</tt>
of the arguments. By default, if an <code class="classname">UnlabeledMultiArg</code>
is specified, then at least one is required to be present or an
exception will be thrown. The most important thing to remember is,
that like <tt class="classname">UnlabeledValueArg</tt>s: order matters!
that like <code class="classname">UnlabeledValueArg</code>s: order matters!
In fact, <span class="emphasis"><em>an UnlabeledMultiArg must be the last argument added to the
CmdLine!</em></span>. Here is what a declaration looks like:
@ -405,7 +420,7 @@ CmdLine!</em></span>. Here is what a declaration looks like:
//
// UnlabeledMultiArg must be the LAST argument added!
//
UnlabeledMultiArg&lt;string&gt; multi(&quot;file names&quot;);
UnlabeledMultiArg&lt;string&gt; multi("file names");
cmd.add( multi );
cmd.parse(argc, argv);
@ -413,69 +428,69 @@ CmdLine!</em></span>. Here is what a declaration looks like:
</pre><p>
You must only ever specify one (1) <tt class="classname">UnlabeledMultiArg</tt>.
One <tt class="classname">UnlabeledMultiArg</tt> will read every unlabeled
You must only ever specify one (1) <code class="classname">UnlabeledMultiArg</code>.
One <code class="classname">UnlabeledMultiArg</code> will read every unlabeled
Arg that wasn't already processed by a
<tt class="classname">UnlabeledValueArg</tt> into a
<tt class="classname">vector</tt> of type T. Any
<tt class="classname">UnlabeledValueArg</tt> or other
<tt class="classname">UnlabeledMultiArg</tt> specified after the first
<tt class="classname">UnlabeledMultiArg</tt> will be ignored, and if
<code class="classname">UnlabeledValueArg</code> into a
<code class="classname">vector</code> of type T. Any
<code class="classname">UnlabeledValueArg</code> or other
<code class="classname">UnlabeledMultiArg</code> specified after the first
<code class="classname">UnlabeledMultiArg</code> will be ignored, and if
they are required,
exceptions will be thrown. When you call the
<tt class="methodname">getValue()</tt>
method of the <tt class="classname">UnlabeledValueArg</tt> argument,
a <tt class="classname">vector</tt>
<code class="methodname">getValue()</code>
method of the <code class="classname">UnlabeledValueArg</code> argument,
a <code class="classname">vector</code>
will be returned. If you can imagine a situation where there will
be multiple args of multiple types (stings, ints, floats, etc.)
then just declare the <tt class="classname">UnlabeledMultiArg</tt> as type
<tt class="classname">string</tt> and parse the different values yourself or use
several <tt class="classname">UnlabeledValueArg</tt>s.
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="XOR"></a>I want one argument or the other, but not both...</h2></div></div><div></div></div><p>
then just declare the <code class="classname">UnlabeledMultiArg</code> as type
<code class="classname">string</code> and parse the different values yourself or use
several <code class="classname">UnlabeledValueArg</code>s.
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="XOR"></a>I want one argument or the other, but not both...</h2></div></div></div><p>
Suppose you have a command that must read input from one of two
possible locations, either a local file or a URL. The command
<span class="emphasis"><em>must</em></span> read something, so <span class="emphasis"><em>one</em></span>
argument is required, but
not both, yet neither argument is strictly necessary by itself.
This is called &quot;exclusive or&quot; or &quot;XOR&quot;. To accomodate this
This is called "exclusive or" or "XOR". To accomodate this
situation, there is now an option to add two or more
<tt class="classname">Arg</tt>s to
a <tt class="classname">CmdLine</tt> that are exclusively or'd with one another:
<tt class="methodname">xorAdd()</tt>. This means that exactly one of the
<tt class="classname">Arg</tt>s must be set and no more.
<code class="classname">Arg</code>s to
a <code class="classname">CmdLine</code> that are exclusively or'd with one another:
<code class="methodname">xorAdd()</code>. This means that exactly one of the
<code class="classname">Arg</code>s must be set and no more.
</p><p>
<tt class="methodname">xorAdd()</tt> comes in two flavors, either
<tt class="methodname">xorAdd(Arg&amp; a, Arg&amp; b)</tt>
to add just two <tt class="classname">Arg</tt>s to be xor'd and
<tt class="methodname">xorAdd( vector&lt;Arg*&gt; xorList )</tt>
to add more than two <tt class="classname">Arg</tt>s.
<code class="methodname">xorAdd()</code> comes in two flavors, either
<code class="methodname">xorAdd(Arg&amp; a, Arg&amp; b)</code>
to add just two <code class="classname">Arg</code>s to be xor'd and
<code class="methodname">xorAdd( vector&lt;Arg*&gt; xorList )</code>
to add more than two <code class="classname">Arg</code>s.
</p><pre class="programlisting">
ValueArg&lt;string&gt; fileArg(&quot;f&quot;,&quot;file&quot;,&quot;File name to read&quot;,true,&quot;homer&quot;,
&quot;filename&quot;);
ValueArg&lt;string&gt; urlArg(&quot;u&quot;,&quot;url&quot;,&quot;URL to load&quot;,true,
&quot;http://example.com&quot;, &quot;URL&quot;);
ValueArg&lt;string&gt; fileArg("f","file","File name to read",true,"homer",
"filename");
ValueArg&lt;string&gt; urlArg("u","url","URL to load",true,
"http://example.com", "URL");
cmd.xorAdd( fileArg, urlArg );
cmd.parse(argc, argv);
</pre><p>
Once one <tt class="classname">Arg</tt> in the xor list is matched on the
<tt class="classname">CmdLine</tt> then the others in the xor list will be
Once one <code class="classname">Arg</code> in the xor list is matched on the
<code class="classname">CmdLine</code> then the others in the xor list will be
marked as set. The question then, is how to determine which of the
<tt class="classname">Arg</tt>s has been set? This is accomplished by calling the
isSet() method for each <tt class="classname">Arg</tt>. If the
<tt class="classname">Arg</tt> has been
matched on the command line, the <tt class="methodname">isSet()</tt> will return
<tt class="constant">TRUE</tt>, whereas if the <tt class="classname">Arg</tt>
has been set as a result of matching the other <tt class="classname">Arg</tt>
that was xor'd <tt class="methodname">isSet()</tt> will
return <tt class="constant">FALSE</tt>.
(Of course, if the <tt class="classname">Arg</tt> was not xor'd and
wasn't matched, it will also return <tt class="constant">FALSE</tt>.)
<code class="classname">Arg</code>s has been set? This is accomplished by calling the
isSet() method for each <code class="classname">Arg</code>. If the
<code class="classname">Arg</code> has been
matched on the command line, the <code class="methodname">isSet()</code> will return
<code class="constant">TRUE</code>, whereas if the <code class="classname">Arg</code>
has been set as a result of matching the other <code class="classname">Arg</code>
that was xor'd <code class="methodname">isSet()</code> will
return <code class="constant">FALSE</code>.
(Of course, if the <code class="classname">Arg</code> was not xor'd and
wasn't matched, it will also return <code class="constant">FALSE</code>.)
</p><pre class="programlisting">
@ -486,101 +501,101 @@ wasn't matched, it will also return <tt class="constant">FALSE</tt>.)
else
// Should never get here because TCLAP will note that one of the
// required args above has not been set.
throw(&quot;Very bad things...&quot;);
throw("Very bad things...");
</pre><p>
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="NO_FLAG"></a>I have more arguments than single flags make sense for...</h2></div></div><div></div></div><p>
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="NO_FLAG"></a>I have more arguments than single flags make sense for...</h2></div></div></div><p>
Some commands have so many options that single flags no longer map
sensibly to the available options. In this case, it is desirable to
specify <tt class="classname">Arg</tt>s using only long options. This one is easy to
accomplish, just make the flag value blank in the <tt class="classname">Arg</tt>
constructor. This will tell the <tt class="classname">Arg</tt> that only the long
specify <code class="classname">Arg</code>s using only long options. This one is easy to
accomplish, just make the flag value blank in the <code class="classname">Arg</code>
constructor. This will tell the <code class="classname">Arg</code> that only the long
option should be matched and will force users to specify the long
option on the command line. The help output is updated accordingly.
</p><pre class="programlisting">
ValueArg&lt;string&gt; fileArg(&quot;&quot;,&quot;file&quot;,&quot;File name&quot;,true,&quot;homer&quot;,&quot;filename&quot;);
ValueArg&lt;string&gt; fileArg("","file","File name",true,"homer","filename");
SwitchArg caseSwitch(&quot;&quot;,&quot;upperCase&quot;,&quot;Print in upper case&quot;,false);
SwitchArg caseSwitch("","upperCase","Print in upper case",false);
</pre><p>
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="CONSTRAINT"></a>I want to constrain the values allowed for a particular
argument...</h2></div></div><div></div></div><p>
argument...</h2></div></div></div><p>
<span class="emphasis"><em>Interface Change!!!</em></span> Sorry folks, but we've changed
the interface to constraining <tt class="classname">Arg</tt>s. Constraints are
now hidden behind the <tt class="classname">Constraint</tt> interface. To
constrain an <tt class="classname">Arg</tt> simply implement the interface
the interface to constraining <code class="classname">Arg</code>s. Constraints are
now hidden behind the <code class="classname">Constraint</code> interface. To
constrain an <code class="classname">Arg</code> simply implement the interface
and specify the new class in the constructor as before.
</p><p>
Fear not, you can still constrain <tt class="classname">Arg</tt>s based on
a list of values. Instead of adding a <tt class="classname">vector</tt> of
allowed values to the <tt class="classname">Arg</tt> directly,
create a <tt class="classname">ValuesConstraint</tt> object
with a <tt class="classname">vector</tt> of values and add that to the
<tt class="classname">Arg</tt>. The <tt class="classname">Arg</tt> constructors
Fear not, you can still constrain <code class="classname">Arg</code>s based on
a list of values. Instead of adding a <code class="classname">vector</code> of
allowed values to the <code class="classname">Arg</code> directly,
create a <code class="classname">ValuesConstraint</code> object
with a <code class="classname">vector</code> of values and add that to the
<code class="classname">Arg</code>. The <code class="classname">Arg</code> constructors
have been modified accordingly.
</p><p>
When the value for the
<tt class="classname">Arg</tt> is parsed,
<code class="classname">Arg</code> is parsed,
it is checked against the list of values specified in the
<tt class="classname">ValuesConstraint</tt>.
<code class="classname">ValuesConstraint</code>.
If the value is in the list then it is accepted. If
not, then an exception is thrown. Here is a simple example:
</p><pre class="programlisting">
vector&lt;string&gt; allowed;
allowed.push_back(&quot;homer&quot;);
allowed.push_back(&quot;marge&quot;);
allowed.push_back(&quot;bart&quot;);
allowed.push_back(&quot;lisa&quot;);
allowed.push_back(&quot;maggie&quot;);
allowed.push_back("homer");
allowed.push_back("marge");
allowed.push_back("bart");
allowed.push_back("lisa");
allowed.push_back("maggie");
ValuesConstraint&lt;string&gt; allowedVals( allowed );
ValueArg&lt;string&gt; nameArg(&quot;n&quot;,&quot;name&quot;,&quot;Name to print&quot;,true,&quot;homer&quot;,&amp;allowedVals);
ValueArg&lt;string&gt; nameArg("n","name","Name to print",true,"homer",&amp;allowedVals);
cmd.add( nameArg );
</pre><p>
When a <tt class="classname">ValuesConstraint</tt> is specified,
When a <code class="classname">ValuesConstraint</code> is specified,
instead of a type description being specified in the
<tt class="classname">Arg</tt>, a
<code class="classname">Arg</code>, a
type description is created by concatenating the values in the
allowed list using operator&lt;&lt; for the specified type. The
help/usage for the <tt class="classname">Arg</tt> therefore lists the
help/usage for the <code class="classname">Arg</code> therefore lists the
allowable values. Because of this, you might want to keep the list
relatively small, however there is no limit on this.
</p><p>
Obviously, a list of allowed values isn't always the best way to
constrain things. For instance, one might wish to allow only
integers greater than 0. In this case, simply create a class that
implements the <tt class="classname">Constraint&lt;int&gt;</tt> interface and
implements the <code class="classname">Constraint&lt;int&gt;</code> interface and
checks whether the value parsed is greater than 0 (done in the
<tt class="methodname">check()</tt> method) and create your
<tt class="classname">Arg</tt> with your new <tt class="classname">Constraint</tt>.
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="ARG_ADD_CMDLINE"></a>I want the Args to add themselves to the CmdLine...</h2></div></div><div></div></div><p>
New constructors have beed added for each <tt class="classname">Arg</tt>
that take a <tt class="classname">CmdLine</tt> object as an argument.
Each <tt class="classname">Arg</tt> then
<tt class="methodname">add</tt>s itself to the <tt class="classname">CmdLine</tt>
object. There is no difference in how the <tt class="classname">Arg</tt>
<code class="methodname">check()</code> method) and create your
<code class="classname">Arg</code> with your new <code class="classname">Constraint</code>.
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="ARG_ADD_CMDLINE"></a>I want the Args to add themselves to the CmdLine...</h2></div></div></div><p>
New constructors have beed added for each <code class="classname">Arg</code>
that take a <code class="classname">CmdLine</code> object as an argument.
Each <code class="classname">Arg</code> then
<code class="methodname">add</code>s itself to the <code class="classname">CmdLine</code>
object. There is no difference in how the <code class="classname">Arg</code>
is handled between this method and calling the
<tt class="methodname">add()</tt> method directly. At the moment, there is
no way to do an <tt class="methodname">xorAdd()</tt> from the constructor. Here
<code class="methodname">add()</code> method directly. At the moment, there is
no way to do an <code class="methodname">xorAdd()</code> from the constructor. Here
is an example:
</p><pre class="programlisting">
// Create the command line.
CmdLine cmd(&quot;this is a message&quot;, '=', &quot;0.99&quot; );
CmdLine cmd("this is a message", '=', "0.99" );
// Note that the following args take the &quot;cmd&quot; object as arguments.
SwitchArg btest(&quot;B&quot;,&quot;existTestB&quot;, &quot;exist Test B&quot;, false, cmd );
// Note that the following args take the "cmd" object as arguments.
SwitchArg btest("B","existTestB", "exist Test B", false, cmd );
ValueArg&lt;string&gt; stest(&quot;s&quot;, &quot;stringTest&quot;, &quot;string test&quot;, true, &quot;homer&quot;,
&quot;string&quot;, cmd );
ValueArg&lt;string&gt; stest("s", "stringTest", "string test", true, "homer",
"string", cmd );
UnlabeledValueArg&lt;string&gt; utest(&quot;unTest1&quot;,&quot;unlabeled test one&quot;,
&quot;default&quot;,&quot;string&quot;, cmd );
UnlabeledValueArg&lt;string&gt; utest("unTest1","unlabeled test one",
"default","string", cmd );
// NO add() calls!
@ -588,14 +603,14 @@ is an example:
cmd.parse(argc,argv);
</pre><p>
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="CHANGE_OUTPUT"></a>I want different output than what is provided...</h2></div></div><div></div></div><p>
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="CHANGE_OUTPUT"></a>I want different output than what is provided...</h2></div></div></div><p>
It is straightforward to change the output generated by
<span class="emphasis"><em>TCLAP</em></span>. Either subclass the
<tt class="classname">StdOutput</tt> class and re-implement the methods you choose,
<code class="classname">StdOutput</code> class and re-implement the methods you choose,
or write your own class that implements the
<tt class="classname">CmdLineOutput</tt> interface. Once you have done this,
then use the <tt class="classname">CmdLine</tt> <tt class="methodname">setOutput</tt>
method to tell the <tt class="classname">CmdLine</tt> to use your new output
<code class="classname">CmdLineOutput</code> interface. Once you have done this,
then use the <code class="classname">CmdLine</code> <code class="methodname">setOutput</code>
method to tell the <code class="classname">CmdLine</code> to use your new output
class. Here is a simple example:
</p><pre class="programlisting">
class MyOutput : public StdOutput
@ -603,28 +618,28 @@ class MyOutput : public StdOutput
public:
virtual void failure(CmdLineInterface&amp; c, ArgException&amp; e)
{
cerr &lt;&lt; &quot;My special failure message for: &quot; &lt;&lt; endl
cerr &lt;&lt; "My special failure message for: " &lt;&lt; endl
&lt;&lt; e.what() &lt;&lt; endl;
}
virtual void usage(CmdLineInterface&amp; c)
{
cout &lt;&lt; &quot;my usage message:&quot; &lt;&lt; endl;
cout &lt;&lt; "my usage message:" &lt;&lt; endl;
list&lt;Arg*&gt; args = c.getArgList();
for (ArgListIterator it = args.begin(); it != args.end(); it++)
cout &lt;&lt; (*it)-&gt;longID()
&lt;&lt; &quot; (&quot; &lt;&lt; (*it)-&gt;getDescription() &lt;&lt; &quot;)&quot; &lt;&lt; endl;
&lt;&lt; " (" &lt;&lt; (*it)-&gt;getDescription() &lt;&lt; ")" &lt;&lt; endl;
}
virtual void version(CmdLineInterface&amp; c)
{
cout &lt;&lt; &quot;my version message: 0.1&quot; &lt;&lt; endl;
cout &lt;&lt; "my version message: 0.1" &lt;&lt; endl;
}
};
int main(int argc, char** argv)
{
CmdLine cmd(&quot;this is a message&quot;, ' ', &quot;0.99&quot; );
CmdLine cmd("this is a message", ' ', "0.99" );
// set the output
MyOutput my;
@ -633,81 +648,81 @@ int main(int argc, char** argv)
// proceed normally ...
</pre><p>
See <tt class="filename">test4.cpp</tt> in the examples directory for the full
See <code class="filename">test4.cpp</code> in the examples directory for the full
example. <span class="emphasis"><em>NOTE</em></span>: if you supply your own Output object, we
will not delete it in the <tt class="classname">CmdLine</tt> destructor. This
will not delete it in the <code class="classname">CmdLine</code> destructor. This
could lead to a (very small) memory leak if you don't take care of the object
yourself.
</p></div></div><div class="chapter" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title"><a id="EXCEPTIONS"></a>Chapter 3. Exceptions to the Rules</h2></div></div><div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl><dt><a href="#IGNORE_ARGS">Ignoring arguments</a></dt><dt><a href="#MULTIPLE_IDENTICAL_SWITCHES">Multiple Identical Switches</a></dt><dt><a href="#DESCRIPTION_EXCEPTIONS">Type Descriptions</a></dt></dl></div><p>
</p></div></div><div class="chapter" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title"><a id="EXCEPTIONS"></a>Chapter 3. Exceptions to the Rules</h2></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl><dt><span class="sect1"><a href="#IGNORE_ARGS">Ignoring arguments</a></span></dt><dt><span class="sect1"><a href="#MULTIPLE_IDENTICAL_SWITCHES">Multiple Identical Switches</a></span></dt><dt><span class="sect1"><a href="#DESCRIPTION_EXCEPTIONS">Type Descriptions</a></span></dt></dl></div><p>
Like all good rules, there are many exceptions....
</p><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="IGNORE_ARGS"></a>Ignoring arguments</h2></div></div><div></div></div><p>
The <i class="parameter"><tt>--</tt></i> flag is automatically included in the
<tt class="classname">CmdLine</tt>.
</p><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="IGNORE_ARGS"></a>Ignoring arguments</h2></div></div></div><p>
The <em class="parameter"><code>--</code></em> flag is automatically included in the
<code class="classname">CmdLine</code>.
As (almost) per POSIX and GNU standards, any argument specified
after the <i class="parameter"><tt>--</tt></i> flag is ignored.
after the <em class="parameter"><code>--</code></em> flag is ignored.
<span class="emphasis"><em>Almost</em></span> because if an
<tt class="classname">UnlabeledValueArg</tt> that has not been set or an
<tt class="classname">UnlabeledMultiArg</tt> has been specified, by default
we will assign any arguments beyond the <i class="parameter"><tt>--</tt></i>
<code class="classname">UnlabeledValueArg</code> that has not been set or an
<code class="classname">UnlabeledMultiArg</code> has been specified, by default
we will assign any arguments beyond the <em class="parameter"><code>--</code></em>
to the those arguments as
per the rules above. This is primarily useful if you want to pass
in arguments with a dash as the first character of the argument. It
should be noted that even if the <i class="parameter"><tt>--</tt></i> flag is
passed on the command line, the <tt class="classname">CmdLine</tt> will
should be noted that even if the <em class="parameter"><code>--</code></em> flag is
passed on the command line, the <code class="classname">CmdLine</code> will
<span class="emphasis"><em>still</em></span> test to make sure all of the required
arguments are present.
</p><p>
Of course, this isn't how POSIX/GNU handle things, they explicitly
ignore arguments after the <i class="parameter"><tt>--</tt></i>. To accomodate this,
we can make both <tt class="classname">UnlabeledValueArg</tt>s and
<tt class="classname">UnlabeledMultiArg</tt>s ignoreable in their constructors.
ignore arguments after the <em class="parameter"><code>--</code></em>. To accomodate this,
we can make both <code class="classname">UnlabeledValueArg</code>s and
<code class="classname">UnlabeledMultiArg</code>s ignoreable in their constructors.
See the <a href="html/index.html" target="_top"> API Documentation</a> for details.
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="MULTIPLE_IDENTICAL_SWITCHES"></a>Multiple Identical Switches</h2></div></div><div></div></div><p>
No longer a problem! Just use <tt class="classname">MultiSwitchArg</tt>.
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="MULTIPLE_IDENTICAL_SWITCHES"></a>Multiple Identical Switches</h2></div></div></div><p>
No longer a problem! Just use <code class="classname">MultiSwitchArg</code>.
There is a description <a href="#MULTI_ARG" title="I tried passing multiple values on the command line with the&#10;same flag and it didn't work...">here</a>.
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="DESCRIPTION_EXCEPTIONS"></a>Type Descriptions</h2></div></div><div></div></div><p>
</p></div><div class="sect1" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="DESCRIPTION_EXCEPTIONS"></a>Type Descriptions</h2></div></div></div><p>
Ideally this library would use RTTI to return a human readable name
of the type declared for a particular argument. Unfortunately, at
least for <b class="command">g++</b>, the names returned aren't
least for <span><strong class="command">g++</strong></span>, the names returned aren't
particularly useful.
</p></div></div><div class="chapter" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title"><a id="VISITORS"></a>Chapter 4. Visitors</h2></div></div><div></div></div><p>
</p></div></div><div class="chapter" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title"><a id="VISITORS"></a>Chapter 4. Visitors</h2></div></div></div><p>
Disclaimer: Almost no one will have any use for
<tt class="classname">Visitor</tt>s, they were
<code class="classname">Visitor</code>s, they were
added to provide special handling for default arguments. Nothing
that <tt class="classname">Visitor</tt>s do couldn't be accomplished
that <code class="classname">Visitor</code>s do couldn't be accomplished
by the user after the
command line has been parsed. If you're still interested, keep
reading...
</p><p>
Some of you may be wondering how we get the <i class="parameter"><tt>--help</tt></i>,
<i class="parameter"><tt>--version</tt></i> and <i class="parameter"><tt>--</tt></i>
Some of you may be wondering how we get the <em class="parameter"><code>--help</code></em>,
<em class="parameter"><code>--version</code></em> and <em class="parameter"><code>--</code></em>
arguments to do their thing without mucking up the
<tt class="classname">CmdLine</tt> code with lots of <span class="emphasis"><em>if</em></span>
<code class="classname">CmdLine</code> code with lots of <span class="emphasis"><em>if</em></span>
statements and type checking. This is accomplished by using a
variation on the Visitor Pattern. Actually, it may not be a Visitor
Pattern at all, but that's what inspired me.
</p><p>
If we want some argument to do some sort of special handling,
besides simply parsing a value, then we add a <tt class="classname">Visitor</tt>
pointer to the <tt class="classname">Arg</tt>. More specifically, we add a
<span class="emphasis"><em>subclass</em></span> of the <tt class="classname">Visitor</tt>
besides simply parsing a value, then we add a <code class="classname">Visitor</code>
pointer to the <code class="classname">Arg</code>. More specifically, we add a
<span class="emphasis"><em>subclass</em></span> of the <code class="classname">Visitor</code>
class. Once the argument has been successfully parsed, the
<tt class="classname">Visitor</tt> for that argument is
<code class="classname">Visitor</code> for that argument is
called. Any data that needs to be operated on is declared in the
<tt class="classname">Visitor</tt> constructor and then operated on in the
<tt class="methodname">visit()</tt> method. A <tt class="classname">Visitor</tt>
is added to an <tt class="classname">Arg</tt> as the last argument in its
<code class="classname">Visitor</code> constructor and then operated on in the
<code class="methodname">visit()</code> method. A <code class="classname">Visitor</code>
is added to an <code class="classname">Arg</code> as the last argument in its
declaration. This may sound
complicated, but it is pretty straightforward. Let's see an
example.
</p><p>
Say you want to add an <i class="parameter"><tt>--authors</tt></i> flag to a program that
Say you want to add an <em class="parameter"><code>--authors</code></em> flag to a program that
prints the names of the authors when present. First subclass
<tt class="classname">Visitor</tt>:
<code class="classname">Visitor</code>:
</p><pre class="programlisting">
#include &quot;Visitor.h&quot;
#include "Visitor.h"
#include &lt;string&gt;
#include &lt;iostream&gt;
@ -717,29 +732,29 @@ class AuthorVisitor : public Visitor
string _author;
public:
AuthorVisitor(const string&amp; name ) : Visitor(), _author(name) {} ;
void visit() { cout &lt;&lt; &quot;AUTHOR: &quot; &lt;&lt; _author &lt;&lt; endl; exit(0); };
void visit() { cout &lt;&lt; "AUTHOR: " &lt;&lt; _author &lt;&lt; endl; exit(0); };
};
</pre><p>
Now include this class definition somewhere and go about creating
your command line. When you create the author switch, add the
<tt class="classname">AuthorVisitor</tt> pointer as follows:
<code class="classname">AuthorVisitor</code> pointer as follows:
</p><pre class="programlisting">
SwitchArg author(&quot;a&quot;,&quot;author&quot;,&quot;Prints author name&quot;, false,
new AuthorVisitor(&quot;Homer J. Simpson&quot;) );
SwitchArg author("a","author","Prints author name", false,
new AuthorVisitor("Homer J. Simpson") );
cmd.add( author );
</pre><p>
Now, any time the <i class="parameter"><tt>-a</tt></i> or
<i class="parameter"><tt>--author</tt></i> flag is specified,
Now, any time the <em class="parameter"><code>-a</code></em> or
<em class="parameter"><code>--author</code></em> flag is specified,
the program will print the author name, Homer J. Simpson and exit
without processing any further (as specified in the
<tt class="methodname">visit()</tt> method).
</p></div><div class="chapter" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title"><a id="MORE_INFO"></a>Chapter 5. More Information</h2></div></div><div></div></div><p>
<code class="methodname">visit()</code> method).
</p></div><div class="chapter" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title"><a id="MORE_INFO"></a>Chapter 5. More Information</h2></div></div></div><p>
For more information, look at the <a href="html/index.html" target="_top">
API Documentation</a> and the examples included with the
distribution.

37
docs/manual.xml
View File

@ -68,25 +68,40 @@ Here is a simple <ulink url="test1.cpp"> example</ulink> ...
#include &lt;algorithm&gt;
#include &lt;tclap/CmdLine.h&gt;
using namespace TCLAP;
using namespace std;
int main(int argc, char** argv)
{
using std::string;
using std::cout;
using std::cerr;
using std::endl;
// Wrap everything in a try block. Do this every time,
// because exceptions will be thrown for problems.
// because exceptions will be thrown for problems.
try {
// Define the command line object.
CmdLine cmd("Command description message", ' ', "0.9");
// Define the command line object, and insert a messages
//that tells you what the program does etc.
//The "Command description message" is printed last in the help
//text. The second argument is the delimiter (usually space) and
//the last one is the version number. The CmdLine object is used
//for parsing.
TCLAP::CmdLine cmd("Command description message", ' ', "0.9");
// Define a value argument and add it to the command line.
ValueArg&lt;string&gt; nameArg("n","name","Name to print",true,"homer","string");
// A value arg takes a switch and a value such as -n Bishop
TCLAP::ValueArg&lt;string&gt; nameArg("n","name","Name to print",true,"homer","string");
// Add the argument nameArg to the command line object. This
// makes it possible to match the argument on the command line
// during parsing.
cmd.add( nameArg );
// Define a switch and add it to the command line.
SwitchArg reverseSwitch("r","reverse","Print name backwards", false);
cmd.add( reverseSwitch );
// A switch arg is a binary argument and only takes a switch
// (true/false) such as -r. Also the command line object is added
// directly while creating the SwitchArg, eliminating the need
// for the call to cmd.add(), this can be used with any type of Arg.
TCLAP::SwitchArg reverseSwitch("r","reverse","Print name backwards", false, cmd);
// Parse the args.
cmd.parse( argc, argv );
@ -98,14 +113,14 @@ int main(int argc, char** argv)
// Do what you intend too...
if ( reverseName )
{
reverse(name.begin(),name.end());
std::reverse(name.begin(),name.end());
cout &lt;&lt; "My name (spelled backwards) is: " &lt;&lt; name &lt;&lt; endl;
}
else
cout &lt;&lt; "My name is: " &lt;&lt; name &lt;&lt; endl;
} catch (ArgException &amp;e) // catch any exceptions
} catch (TCLAP::ArgException &amp;e) // catch any exceptions
{ cerr &lt;&lt; "error: " &lt;&lt; e.error() &lt;&lt; " for arg " &lt;&lt; e.argId() &lt;&lt; endl; }
}
</programlisting>