You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
tqt3/src/widgets/qvalidator.cpp

673 lines
18 KiB

/****************************************************************************
**
** Implementation of validator classes
**
** Created : 970610
**
** Copyright (C) 1992-2008 Trolltech ASA. All rights reserved.
**
** This file is part of the widgets module of the TQt GUI Toolkit.
**
** This file may be used under the terms of the GNU General
** Public License versions 2.0 or 3.0 as published by the Free
** Software Foundation and appearing in the files LICENSE.GPL2
** and LICENSE.GPL3 included in the packaging of this file.
** Alternatively you may (at your option) use any later version
** of the GNU General Public License if such license has been
** publicly approved by Trolltech ASA (or its successors, if any)
** and the KDE Free TQt Foundation.
**
** Please review the following information to ensure GNU General
** Public Licensing requirements will be met:
** http://trolltech.com/products/qt/licenses/licensing/opensource/.
** If you are unsure which license is appropriate for your use, please
** review the following information:
** http://trolltech.com/products/qt/licenses/licensing/licensingoverview
** or contact the sales department at sales@trolltech.com.
**
** This file may be used under the terms of the Q Public License as
** defined by Trolltech ASA and appearing in the file LICENSE.TQPL
** included in the packaging of this file. Licensees holding valid TQt
** Commercial licenses may use this file in accordance with the TQt
** Commercial License Agreement provided with the Software.
**
** This file is provided "AS IS" with NO WARRANTY OF ANY KIND,
** INCLUDING THE WARRANTIES OF DESIGN, MERCHANTABILITY AND FITNESS FOR
** A PARTICULAR PURPOSE. Trolltech reserves all rights not granted
** herein.
**
**********************************************************************/
#include "ntqvalidator.h"
#ifndef TQT_NO_VALIDATOR
#include <limits.h>
#include <math.h>
/*!
\class TQValidator
\brief The TQValidator class provides validation of input text.
\ingroup misc
\mainclass
The class itself is abstract. Two subclasses, \l TQIntValidator and
\l TQDoubleValidator, provide basic numeric-range checking, and \l
TQRegExpValidator provides general checking using a custom regular
expression.
If the built-in validators aren't sufficient, you can subclass
TQValidator. The class has two virtual functions: validate() and
fixup().
\l validate() must be implemented by every subclass. It returns
\c Invalid, \c Intermediate or \c Acceptable depending on whether
its argument is valid (for the subclass's definition of valid).
These three states require some explanation. An \c Invalid string
is \e clearly invalid. \c Intermediate is less obvious: the
concept of validity is slippery when the string is incomplete
(still being edited). TQValidator defines \c Intermediate as the
property of a string that is neither clearly invalid nor
acceptable as a final result. \c Acceptable means that the string
is acceptable as a final result. One might say that any string
that is a plausible intermediate state during entry of an \c
Acceptable string is \c Intermediate.
Here are some examples:
\list
\i For a line edit that accepts integers from 0 to 999 inclusive,
42 and 123 are \c Acceptable, the empty string and 1114 are \c
Intermediate and asdf is \c Invalid.
\i For an editable combobox that accepts URLs, any well-formed URL
is \c Acceptable, "http://www.trolltech.com/," is \c Intermediate
(it might be a cut and paste operation that accidentally took in a
comma at the end), the empty string is \c Intermediate (the user
might select and delete all of the text in preparation for entering
a new URL), and "http:///./" is \c Invalid.
\i For a spin box that accepts lengths, "11cm" and "1in" are \c
Acceptable, "11" and the empty string are \c Intermediate and
"http://www.trolltech.com" and "hour" are \c Invalid.
\endlist
\l fixup() is provided for validators that can repair some user
errors. The default implementation does nothing. TQLineEdit, for
example, will call fixup() if the user presses Enter (or Return)
and the content is not currently valid. This allows the fixup()
function the opportunity of performing some magic to make an \c
Invalid string \c Acceptable.
TQValidator is typically used with TQLineEdit, TQSpinBox and
TQComboBox.
*/
/*!
\enum TQValidator::State
This enum type defines the states in which a validated string can
exist.
\value Invalid the string is \e clearly invalid.
\value Intermediate the string is a plausible intermediate value
during editing.
\value Acceptable the string is acceptable as a final result,
i.e. it is valid.
*/
/*!
Sets up the validator. The \a parent and \a name parameters are
passed on to the TQObject constructor.
*/
TQValidator::TQValidator( TQObject * parent, const char *name )
: TQObject( parent, name )
{
}
/*!
Destroys the validator, freeing any storage and other resources
used.
*/
TQValidator::~TQValidator()
{
}
/*!
\fn TQValidator::State TQValidator::validate( TQString& input, int& pos ) const
This pure virtual function returns \c Invalid if \a input is
invalid according to this validator's rules, \c Intermediate if it
is likely that a little more editing will make the input
acceptable (e.g. the user types '4' into a widget which accepts
integers between 10 and 99) and \c Acceptable if the input is
valid.
The function can change \a input and \a pos (the cursor position)
if it wants to.
*/
/*!
\fn void TQValidator::fixup( TQString & input ) const
This function attempts to change \a input to be valid according to
this validator's rules. It need not result in a valid string:
callers of this function must re-test afterwards; the default does
nothing.
Reimplementations of this function can change \a input even if
they do not produce a valid string. For example, an ISBN validator
might want to delete every character except digits and "-", even
if the result is still not a valid ISBN; a surname validator might
want to remove whitespace from the start and end of the string,
even if the resulting string is not in the list of accepted
surnames.
*/
void TQValidator::fixup( TQString & ) const
{
}
/*!
\class TQIntValidator
\brief The TQIntValidator class provides a validator which ensures
that a string contains a valid integer within a specified range.
\ingroup misc
Example of use:
\code
TQValidator* validator = new TQIntValidator( 100, 999, this );
TQLineEdit* edit = new TQLineEdit( this );
// the edit lineedit will only accept integers between 100 and 999
edit->setValidator( validator );
\endcode
Below we present some examples of validators. In practice they would
normally be associated with a widget as in the example above.
\code
TQString str;
int pos = 0;
TQIntValidator v( 100, 999, this );
str = "1";
v.validate( str, pos ); // returns Intermediate
str = "12";
v.validate( str, pos ); // returns Intermediate
str = "123";
v.validate( str, pos ); // returns Acceptable
str = "678";
v.validate( str, pos ); // returns Acceptable
str = "1234";
v.validate( str, pos ); // returns Invalid
str = "-123";
v.validate( str, pos ); // returns Invalid
str = "abc";
v.validate( str, pos ); // returns Invalid
str = "12cm";
v.validate( str, pos ); // returns Invalid
\endcode
The minimum and maximum values are set in one call with setRange()
or individually with setBottom() and setTop().
\sa TQDoubleValidator TQRegExpValidator
*/
/*!
Constructs a validator called \a name with parent \a parent, that
accepts all integers.
*/
TQIntValidator::TQIntValidator( TQObject * parent, const char *name )
: TQValidator( parent, name )
{
b = INT_MIN;
t = INT_MAX;
}
/*!
Constructs a validator called \a name with parent \a parent, that
accepts integers from \a minimum to \a maximum inclusive.
*/
TQIntValidator::TQIntValidator( int minimum, int maximum,
TQObject * parent, const char* name )
: TQValidator( parent, name )
{
b = minimum;
t = maximum;
}
/*!
Destroys the validator, freeing any resources allocated.
*/
TQIntValidator::~TQIntValidator()
{
// nothing
}
/*!
Returns \c Acceptable if the \a input is an integer within the
valid range, \c Intermediate if the \a input is an integer outside
the valid range and \c Invalid if the \a input is not an integer.
Note: If the valid range consists of just positive integers (e.g. 32 - 100)
and \a input is a negative integer then Invalid is returned.
\code
int pos = 0;
s = "abc";
v.validate( s, pos ); // returns Invalid
s = "5";
v.validate( s, pos ); // returns Intermediate
s = "50";
v.validate( s, pos ); // returns Valid
\endcode
*/
TQValidator::State TQIntValidator::validate( TQString & input, int & ) const
{
TQString stripped = input.stripWhiteSpace();
if ( stripped.isEmpty() || (b < 0 && stripped == "-") )
return Intermediate;
bool ok;
long entered = input.toLong( &ok );
if ( !ok || (entered < 0 && b >= 0) ) {
return Invalid;
} else if ( entered >= b && entered <= t ) {
return Acceptable;
} else {
if ( entered >= 0 )
return ( entered > t ) ? Invalid : Intermediate;
else
return ( entered < b ) ? Invalid : Intermediate;
}
}
/*!
Sets the range of the validator to only accept integers between \a
bottom and \a top inclusive.
*/
void TQIntValidator::setRange( int bottom, int top )
{
b = bottom;
t = top;
}
/*!
\property TQIntValidator::bottom
\brief the validator's lowest acceptable value
\sa setRange()
*/
void TQIntValidator::setBottom( int bottom )
{
setRange( bottom, top() );
}
/*!
\property TQIntValidator::top
\brief the validator's highest acceptable value
\sa setRange()
*/
void TQIntValidator::setTop( int top )
{
setRange( bottom(), top );
}
#ifndef TQT_NO_REGEXP
/*!
\class TQDoubleValidator
\brief The TQDoubleValidator class provides range checking of
floating-point numbers.
\ingroup misc
TQDoubleValidator provides an upper bound, a lower bound and a
limit on the number of digits after the decimal point. It does not
provide a fixup() function.
You can set the acceptable range in one call with setRange(), or
with setBottom() and setTop(). Set the number of decimal places
with setDecimals(). The validate() function returns the validation
state.
\sa TQIntValidator TQRegExpValidator
*/
/*!
Constructs a validator object with parent \a parent, called \a
name, which accepts any double.
*/
TQDoubleValidator::TQDoubleValidator( TQObject * parent, const char *name )
: TQValidator( parent, name )
{
b = -HUGE_VAL;
t = HUGE_VAL;
d = 1000;
}
/*!
Constructs a validator object with parent \a parent, called \a
name. This validator will accept doubles from \a bottom to \a top
inclusive, with up to \a decimals digits after the decimal point.
*/
TQDoubleValidator::TQDoubleValidator( double bottom, double top, int decimals,
TQObject * parent, const char* name )
: TQValidator( parent, name )
{
b = bottom;
t = top;
d = decimals;
}
/*!
Destroys the validator, freeing any resources used.
*/
TQDoubleValidator::~TQDoubleValidator()
{
}
/*!
Returns \c Acceptable if the string \a input contains a double
that is within the valid range and is in the correct format.
Returns \c Intermediate if \a input contains a double that is
outside the range or is in the wrong format, e.g. with too many
digits after the decimal point or is empty.
Returns \c Invalid if the \a input is not a double.
Note: If the valid range consists of just positive doubles (e.g. 0.0 - 100.0)
and \a input is a negative double then Invalid is returned.
*/
TQValidator::State TQDoubleValidator::validate( TQString & input, int & ) const
{
TQRegExp empty( TQString::fromLatin1(" *-?\\.? *") );
if ( b >= 0 &&
input.stripWhiteSpace().startsWith(TQString::fromLatin1("-")) )
return Invalid;
if ( empty.exactMatch(input) )
return Intermediate;
bool ok = TRUE;
double entered = input.toDouble( &ok );
int nume = input.contains( 'e', FALSE );
if ( !ok ) {
// explicit exponent regexp
TQRegExp expexpexp( TQString::fromLatin1("[Ee][+-]?\\d*$") );
int eeePos = expexpexp.search( input );
if ( eeePos > 0 && nume == 1 ) {
TQString mantissa = input.left( eeePos );
entered = mantissa.toDouble( &ok );
if ( !ok )
return Invalid;
} else if ( eeePos == 0 ) {
return Intermediate;
} else {
return Invalid;
}
}
int i = input.find( '.' );
if ( i >= 0 && nume == 0 ) {
// has decimal point (but no E), now count digits after that
i++;
int j = i;
while( input[j].isDigit() )
j++;
if ( j - i > d )
return Intermediate;
}
if ( entered < b || entered > t )
return Intermediate;
else
return Acceptable;
}
/*!
Sets the validator to accept doubles from \a minimum to \a maximum
inclusive, with at most \a decimals digits after the decimal
point.
*/
void TQDoubleValidator::setRange( double minimum, double maximum, int decimals )
{
b = minimum;
t = maximum;
d = decimals;
}
/*!
\property TQDoubleValidator::bottom
\brief the validator's minimum acceptable value
\sa setRange()
*/
void TQDoubleValidator::setBottom( double bottom )
{
setRange( bottom, top(), decimals() );
}
/*!
\property TQDoubleValidator::top
\brief the validator's maximum acceptable value
\sa setRange()
*/
void TQDoubleValidator::setTop( double top )
{
setRange( bottom(), top, decimals() );
}
/*!
\property TQDoubleValidator::decimals
\brief the validator's maximum number of digits after the decimal point
\sa setRange()
*/
void TQDoubleValidator::setDecimals( int decimals )
{
setRange( bottom(), top(), decimals );
}
/*!
\class TQRegExpValidator
\brief The TQRegExpValidator class is used to check a string
against a regular expression.
\ingroup misc
TQRegExpValidator contains a regular expression, "regexp", used to
determine whether an input string is \c Acceptable, \c
Intermediate or \c Invalid.
The regexp is treated as if it begins with the start of string
assertion, <b>^</b>, and ends with the end of string assertion
<b>$</b> so the match is against the entire input string, or from
the given position if a start position greater than zero is given.
For a brief introduction to TQt's regexp engine see \l TQRegExp.
Example of use:
\code
// regexp: optional '-' followed by between 1 and 3 digits
TQRegExp rx( "-?\\d{1,3}" );
TQValidator* validator = new TQRegExpValidator( rx, this );
TQLineEdit* edit = new TQLineEdit( this );
edit->setValidator( validator );
\endcode
Below we present some examples of validators. In practice they would
normally be associated with a widget as in the example above.
\code
// integers 1 to 9999
TQRegExp rx( "[1-9]\\d{0,3}" );
// the validator treats the regexp as "^[1-9]\\d{0,3}$"
TQRegExpValidator v( rx, 0 );
TQString s;
int pos = 0;
s = "0"; v.validate( s, pos ); // returns Invalid
s = "12345"; v.validate( s, pos ); // returns Invalid
s = "1"; v.validate( s, pos ); // returns Acceptable
rx.setPattern( "\\S+" ); // one or more non-whitespace characters
v.setRegExp( rx );
s = "myfile.txt"; v.validate( s, pos ); // Returns Acceptable
s = "my file.txt"; v.validate( s, pos ); // Returns Invalid
// A, B or C followed by exactly five digits followed by W, X, Y or Z
rx.setPattern( "[A-C]\\d{5}[W-Z]" );
v.setRegExp( rx );
s = "a12345Z"; v.validate( s, pos ); // Returns Invalid
s = "A12345Z"; v.validate( s, pos ); // Returns Acceptable
s = "B12"; v.validate( s, pos ); // Returns Intermediate
// match most 'readme' files
rx.setPattern( "read\\S?me(\.(txt|asc|1st))?" );
rx.setCaseSensitive( FALSE );
v.setRegExp( rx );
s = "readme"; v.validate( s, pos ); // Returns Acceptable
s = "README.1ST"; v.validate( s, pos ); // Returns Acceptable
s = "read me.txt"; v.validate( s, pos ); // Returns Invalid
s = "readm"; v.validate( s, pos ); // Returns Intermediate
\endcode
\sa TQRegExp TQIntValidator TQDoubleValidator
*/
/*!
Constructs a validator that accepts any string (including an empty
one) as valid. The object's parent is \a parent and its name is \a
name.
*/
TQRegExpValidator::TQRegExpValidator( TQObject *parent, const char *name )
: TQValidator( parent, name ), r( TQString::fromLatin1(".*") )
{
}
/*!
Constructs a validator which accepts all strings that match the
regular expression \a rx. The object's parent is \a parent and its
name is \a name.
The match is made against the entire string, e.g. if the regexp is
<b>[A-Fa-f0-9]+</b> it will be treated as <b>^[A-Fa-f0-9]+$</b>.
*/
TQRegExpValidator::TQRegExpValidator( const TQRegExp& rx, TQObject *parent,
const char *name )
: TQValidator( parent, name ), r( rx )
{
}
/*!
Destroys the validator, freeing any resources allocated.
*/
TQRegExpValidator::~TQRegExpValidator()
{
}
/*!
Returns \c Acceptable if \a input is matched by the regular
expression for this validator, \c Intermediate if it has matched
partially (i.e. could be a valid match if additional valid
characters are added), and \c Invalid if \a input is not matched.
The \a pos parameter is set to the length of the \a input parameter.
For example, if the regular expression is <b>\\w\\d\\d</b> (that
is, word-character, digit, digit) then "A57" is \c Acceptable,
"E5" is \c Intermediate and "+9" is \c Invalid.
\sa TQRegExp::match() TQRegExp::search()
*/
TQValidator::State TQRegExpValidator::validate( TQString& input, int& pos ) const
{
if ( r.exactMatch(input) ) {
return Acceptable;
} else {
if ( ((TQRegExp&) r).matchedLength() == (int) input.length() ) {
return Intermediate;
} else {
pos = input.length();
return Invalid;
}
}
}
/*!
Sets the regular expression used for validation to \a rx.
\sa regExp()
*/
void TQRegExpValidator::setRegExp( const TQRegExp& rx )
{
r = rx;
}
/*!
\fn const TQRegExp& TQRegExpValidator::regExp() const
Returns the regular expression used for validation.
\sa setRegExp()
*/
#endif
#endif