|
Ne.app NeANT
The Ne.app NeANT User module of the NeANT distribution.
|
The RWTest library supports all formatted output directives specified by the C99 standard as well as many other directive extensions. Only deviations, extensions, and other implmementation-defined behavior are described below. Consult your system and/or compiler documentation for specifications of the standard formatting output directives.
Unless specified otherwise, the following caveats apply to all directives and conversions. Directives that accept a pointer argument will convert a null pointer into the string literal (null).
Specifiers within formatting directives are denoted by angle brackets. The specifiers are placeholders whose range of possible values are listed in the description of the directive.
The argument is a pointer to void. The value of the pointer is converted to a sequence of hexadecimal digits. The number of converted digits depends on the size of the void* type. The "alternate form" expressed by %#p immediately precedes converted digits with a 0x prefix.
No known deviations or implementation-defined behavior.
The argument is a value convertible to bool type. If the value of the argument is nonzero, the value is converted to the string literal true. Otherwise, the value converts to false.
The argument is a pointer to a function. The conversion and formatting is performed as specified by the %p directive.
The argument is a pointer to a composite member. The conversion and formatting is performed as specified by the %p directive.
The argument is an integer value converted to unsigned char type. For non-printable characters, the resulting output is formatted as the corresponding escape sequence. (The value for a horizontal tab character for example is formatted as the string literal \t and a zero value may be formatted as the string literal \x00. In the alternate form, the resulting output is immediately surrounded with a single quotation mark '.
Similar to the escaped character directive %{c} and quoted escaped character directive %{#c} respectively, except that the argument is a value of wchar_t type.
Similar to the character string directive s except that the resulting output is surrounded with a double quotation mark "</tt>.
@subsection rwtest-fmtspec-ext-ls Quoted Wide Character String \%{\#ls}
Similar to the quoted character string directive <tt>%{\#s}</tt>
except that the argument is a pointer to a null-terminate sequence of
<tt>wchar_t</tt> type.
@subsection rwtest-fmtspec-ext-err Error Code \%m \%{\#m}
The argument is an integer value representing a system-dependent error
code. In the first form, the value is interpreted as an argument of
the \c strerror() function. The formatted result is the return value
of this function using this parameter value. In the second form, the
formatted result is a character sequence representing the
corresponding error name (e.g. \c ENOMEM) if applicable or equivalent
numeric sequence in decimal notation otherwise.
@subsection rwtest-fmtspec-ext-sig Signal \%K
The argument is an integer value representing a system-dependent
signal. The resulting output is a character sequence representing
the name of corresponding signal (e.g. \c SIGABRT) if
applicable or equivalent numeric sequence in decimal notation
otherwise.
@subsection rwtest-fmtspec-ext-bits Bitset \%{b}
The argument is a pointer to an array of \c char. Each bit in the
array, ordered from MSB of the first element to LSB of the last
element is converted into a sequence of '0' or '1' characters.
@subsection rwtest-fmtspec-ext-Ai Integer Array \%{Ac} \%{Ao} \%{Ad} \%{Ax}
The argument is a pointer to an array of integer type. The resulting
output is formatted as a string of comma-separated integer values.
The notation of each integer value is specified by <tt>c</tt> for
characters, <<tt>o</tt> for octal, <tt>d</tt> for decimal, or
<tt>x</tt> for hexadcimal. An optional field width specifies the
size of elements in the array (defaults to 1). An optional precision
specifies the length of the array. In the alternate form, the
resulting output for octal and hexidecimal integer values are
prefixed with string literals <tt>0</tt> and <tt>0x</tt> respectively.
@subsection rwtest-fmtspec-ext-I Fixed-Width Integers \%{I<wn>}
The argument is a value of an integer type. The directive requires
two specifiers: the <tt>w</tt> specifier is one of 8, 16, 32, or 64
specifying the width of the integer value. The <tt>n</tt> specifier
is one of \c d, \c o, \c x, or \c X specifying the notation of the
resulting output.
@subsection rwtest-fmtspec-ext-Is Stream State \%{Is} \%{\#Is}
The argument is a value of type <tt>std::ios_base::iostate</tt>. In
the plain form, the resulting ouput is formatted as a sequence of one
or more unqualified names corresponding to the \c iostate enumerators,
separated by the OR operator (<tt>|</tt>). In the extended form, the
resulting output is prefixed with the string literal
<tt>std::ios_base::</tt> indicating the qualfied namespace.
@subsection rwtest-fmtspec-ext-Io Open Mode \%{Io} \%{\#Io}
Similar to the <tt>%{Is}</tt> directive except the argument is a
value of the <tt>std::ios_base::openmode</tt> enumeration type and
names in the resulting output correspond to enumerators of this type.
@subsection rwtest-fmtspec-ext-Iw Seek Direction \%{Iw} \%{\#Iw}
The argument is a value of type <tt>std::ios_base::seekdir</tt>. In
the plain form, the resulting ouput is formatted as the unqualified
name corresponding to one of the \c seekdir enumerators. In the
extended form, the enumerator is qualified with the enclosing
namespace.
@subsection rwtest-fmtspec-ext-If Format Flag \%{If} \%{\#Iw}
Similar to the <tt>%{Is}</tt> directive except the argument is a
value of the <tt>std::ios_base::fmtflags</tt> enumeration type and
names in the resulting output correspond to enumerators of this type.
@subsection rwtest-fmtspec-ext-Lc Locate Category \%{Lc} \%{\#Lc}
The argument is an integer type representing a constant defined by
the C <tt><locale.h></tt> header (identified by an <tt>LC_</tt>
prefix) or a <tt>std::locale::category</tt> constant defined by
the C++ <tt><locale></tt> header. The resulting output is
formatted as a character string representing the name of the C or C++
constant enclosed in square brackets (<tt>[</tt>) and <tt>]</tt>).
In the alternate form, the formatted output for C++ constants is
prefixed with the string literal <tt>std::locale::</tt> indicating
the qualified namespace.
@subsection rwtest-fmtspec-ext-LC Character Classifiers \%{LC} \%{\#LC}
The argument is an integer type representing enumerators of the
<tt>std::ctype_base::mask</tt> enumeration. The resulting output is
formatted as a character string representing the name of the
enumerator enclosed in square brackets (<tt>[</tt>) and <tt>]</tt>).
In the alternate form, the enumerator is prefixed with the string
literal <tt>std::ctype_base::</tt> indicating the qualified
namespace.
@subsection rwtest-fmtspec-ext-pid Process Identifier \%{P}
The argument is a value of the portable \c rw_pid_t type. The
resulting output is formatted as a numeric sequence of digits
representing the corresponding process identifier. The \c P
specifier may be preceded by optional width, justification, and
precision qualifiers.
@subsection rwtest-fmtspec-ext-str String \%{S}
The argument is a pointer to an object of the \c std::string class.
Otherwise, the pointer \c P is converted by calling the \c P->c_str()
function and the result is formatted identical to the <tt>%s</tt>
character string directive.
@subsection rwtest-fmtspec-ext-wstr Wide String \%{lS}
The argument is a pointer to an object of the \c std::wstring class.
Conversion and formatting is performed as stated for the \c \%{S}
directive.
@subsection rwtest-fmtspec-ext-tm Time Structure \%{t} \%{\#t}
The argument is a pointer to a \c tm structure. If the pointer is
null, the resulting output is the string literal \c (null). In the
plain form, the resulting output for a valid pointer is formatted
similar (but not necessarily identical) to the return value of the
\c asctime() function. If the value of any structure member is
outside its valid range, the resulting output is formatting using the
alternate form. In the alternate form, the resulting output is
formatted as a comma-separated sequence of members enclosed in braces
where each member is of the format "name=value [range]". The
"[range]" is only shown if the value is not in the valid range for the respective member.
The following directives are also recognized by RWTest output functions but are heretofore undocumented.
1.9.8