haskal
/
kicad
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Rewrite class UTF8 to contain rather than extend std::string storage. 

This forces the compiler class specific features rather than borrowing
from the base class's std::string.  In some cases prior to this,
wxString( std::string ) was being called rather than UTF8::operator
wxString() leading to garbled wxStrings.

Added function UTF8::wx_str() which is of great convenience also.

Implicit conversions still work as before, and hopefully more reliably.
This commit is contained in:
Dick Hollenbeck 2017-07-25 14:14:31 -05:00 committed by Wayne Stambaugh
parent 0aedaffed9
commit 19e6bde09a
18 changed files with 170 additions and 97 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
common/fp_lib_table.cpp
View File

@ -286,7 +286,7 @@ MODULE* FP_LIB_TABLE::FootprintLoad( const wxString& aNickname, const wxString&
LIB_ID& fpid = (LIB_ID&) ret->GetFPID();
// Catch any misbehaving plugin, which should be setting internal footprint name properly:
wxASSERT( aFootprintName == (wxString) fpid.GetLibItemName() );
wxASSERT( aFootprintName == fpid.GetLibItemName().wx_str() );
// and clearing nickname
wxASSERT( !fpid.GetLibNickname().size() );

6
common/lib_id.cpp
View File

@ -83,7 +83,7 @@ int RevCmp( const char* s1, const char* s2 )
//----<Policy and field test functions>-------------------------------------
static inline int okLogical( const std::string& aField )
static inline int okLogical( const UTF8& aField )
{
// std::string::npos is largest positive number, casting to int makes it -1.
// Returning that means success.
@ -91,7 +91,7 @@ static inline int okLogical( const std::string& aField )
}
static int okRevision( const std::string& aField )
static int okRevision( const UTF8& aField )
{
char rev[32]; // C string for speed
@ -175,7 +175,7 @@ int LIB_ID::Parse( const UTF8& aId )
}
LIB_ID::LIB_ID( const std::string& aId )
LIB_ID::LIB_ID( const UTF8& aId )
{
int offset = Parse( aId );

16
common/utf8.cpp
View File

@ -40,11 +40,17 @@
UTF8::UTF8( const wxString& o ) :
std::string( (const char*) o.utf8_str() )
m_s( (const char*) o.utf8_str() )
{
}
wxString UTF8::wx_str() const
{
return wxString( c_str(), wxConvUTF8 );
}
UTF8::operator wxString () const
{
return wxString( c_str(), wxConvUTF8 );
@ -53,7 +59,7 @@ UTF8::operator wxString () const
UTF8& UTF8::operator=( const wxString& o )
{
std::string::operator=( (const char*) o.utf8_str() );
m_s = (const char*) o.utf8_str();
return *this;
}
@ -193,7 +199,7 @@ bool IsUTF8( const char* aString )
UTF8::UTF8( const wchar_t* txt ) :
// size initial string safely large enough, then shrink to known size later.
std::string( wcslen( txt ) * 4, 0 )
m_s( wcslen( txt ) * 4, 0 )
{
/*
@ -206,9 +212,9 @@ UTF8::UTF8( const wchar_t* txt ) :
*/
int sz = wxConvUTF8.WC2MB( (char*) data(), txt, size() );
int sz = wxConvUTF8.WC2MB( (char*) m_s.data(), txt, m_s.size() );
resize( sz );
m_s.resize( sz );
}

2
cvpcb/cvpcb_mainframe.cpp
View File

@ -811,7 +811,7 @@ int CVPCB_MAINFRAME::ReadSchematicNetlist( const std::string& aNetlist )
for( unsigned ii = 0; ii < m_netlist.GetCount(); ii++ )
{
if( m_netlist.GetComponent( ii )->GetFPID().GetLibItemName() == std::string( "$noname" ) )
m_netlist.GetComponent( ii )->SetFPID( LIB_ID( wxEmptyString ) );
m_netlist.GetComponent( ii )->SetFPID( LIB_ID() );
}
// Sort components by reference:

2
eeschema/class_library.cpp
View File

@ -672,7 +672,7 @@ void PART_LIBS::LoadAllLibraries( PROJECT* aProject, bool aShowProgress )
{
// Use a different exception type so catch()er can route to proper use
// of the HTML_MESSAGE_BOX.
THROW_PARSE_ERROR( wxEmptyString, UTF8( __func__ ), UTF8( libs_not_found ), 0, 0 );
THROW_PARSE_ERROR( wxEmptyString, __func__, TO_UTF8(libs_not_found), 0, 0 );
}
#if defined(DEBUG) && 1

6
eeschema/dialogs/dialog_symbol_remap.cpp
View File

@ -205,14 +205,14 @@ void DIALOG_SYMBOL_REMAP::remapSymbolsToLibTable( REPORTER& aReporter )
if( !remapSymbolToLibTable( symbol ) )
{
msg.Printf( _( "No symbol '%s' founded in symbol library table." ),
FROM_UTF8( symbol->GetLibId().GetLibItemName() ) );
symbol->GetLibId().GetLibItemName().wx_str() );
aReporter.Report( msg, REPORTER::RPT_WARNING );
}
else
{
msg.Printf( _( "Symbol '%s' mapped to symbol library '%s'." ),
FROM_UTF8( symbol->GetLibId().GetLibItemName() ),
FROM_UTF8( symbol->GetLibId().GetLibNickname() ) );
symbol->GetLibId().GetLibItemName().wx_str(),
symbol->GetLibId().GetLibNickname().wx_str() );
aReporter.Report( msg, REPORTER::RPT_ACTION );
}
}

2
eeschema/libarch.cpp
View File

@ -106,7 +106,7 @@ bool SCH_EDIT_FRAME::CreateArchiveLibrary( const wxString& aFileName )
catch( ... /* IO_ERROR ioe */ )
{
msg.Printf( _( "Failed to add symbol %s to library file '%s'" ),
wxString( component->GetLibId().GetLibItemName() ), aFileName );
component->GetLibId().GetLibItemName().wx_str(), aFileName );
DisplayError( this, msg );
return false;
}

4
eeschema/project_rescue.cpp
View File

@ -264,7 +264,7 @@ public:
for( SCH_COMPONENT* each_component : *( aRescuer.GetComponents() ) )
{
wxString part_name( each_component->GetLibId().GetLibItemName() );
wxString part_name = each_component->GetLibId().GetLibItemName();
if( last_part_name != part_name )
{
@ -373,7 +373,7 @@ public:
for( SCH_COMPONENT* each_component : *( aRescuer.GetComponents() ) )
{
wxString part_name( each_component->GetLibId().GetLibItemName() );
wxString part_name = each_component->GetLibId().GetLibItemName();
if( old_part_name != part_name )
{

2
eeschema/symbol_lib_table.cpp
View File

@ -271,7 +271,7 @@ LIB_ALIAS* SYMBOL_LIB_TABLE::LoadSymbol( const wxString& aNickname, const wxStri
LIB_ID& id = (LIB_ID&) ret->GetPart()->GetLibId();
// Catch any misbehaving plugin, which should be setting internal alias name properly:
wxASSERT( aAliasName == (wxString) id.GetLibItemName() );
wxASSERT( aAliasName == id.GetLibItemName().wx_str() );
// and clearing nickname
wxASSERT( !id.GetLibNickname().size() );

2
include/lib_id.h
View File

@ -70,7 +70,7 @@ public:
*
* @param aId is a string to be parsed into the LIB_ID object.
*/
LIB_ID( const std::string& aId );
LIB_ID( const UTF8& aId );
LIB_ID( const wxString& aId );

133
include/utf8.h
View File

@ -50,13 +50,13 @@ bool IsUTF8( const char* aString );
/**
* Class UTF8
* is an 8 bit std::string that is assuredly encoded in UTF8, and supplies special
* conversion support to and from wxString, and has iteration over unicode characters.
* is an 8 bit string that is assuredly encoded in UTF8, and supplies special
* conversion support to and from wxString, to and from std::string, and has
* non-mutating iteration over unicode characters.
*
* <p>I've been careful to supply only conversion facilities and not try
* and duplicate wxString() with many member functions. In the end it is
* to be a std::string. There are multiple ways to create text into a std::string
* without the need of too many member functions:
* and duplicate wxString() with many member functions. There are multiple ways
* to create text into a std::string without the need of too many member functions:
*
* <ul>
* <li>richio.h's StrPrintf()</li>
@ -70,7 +70,7 @@ bool IsUTF8( const char* aString );
*
* @author Dick Hollenbeck
*/
class UTF8 : public std::string
class UTF8
{
public:
@ -79,7 +79,7 @@ public:
/// This is a constructor for which you could end up with
/// non-UTF8 encoding, but that would be your fault.
UTF8( const char* txt ) :
std::string( txt )
m_s( txt )
{
MAYBE_VERIFY_UTF8( c_str() );
}
@ -89,13 +89,12 @@ public:
UTF8( const wchar_t* txt );
UTF8( const std::string& o ) :
std::string( o )
m_s( o )
{
MAYBE_VERIFY_UTF8( c_str() );
}
UTF8() :
std::string()
UTF8()
{
}
@ -103,61 +102,97 @@ public:
{
}
// expose some std::string functions publicly, since base class must be private.
const char* c_str() const { return m_s.c_str(); }
bool empty() const { return m_s.empty(); }
std::string::size_type find( char c ) const { return m_s.find( c ); }
std::string::size_type find( char c, size_t& s ) const { return m_s.find( c, s ); }
void clear() { m_s.clear(); }
std::string::size_type length() const { return m_s.length(); }
std::string::size_type size() const { return m_s.size(); }
int compare( const std::string& s ) const { return m_s.compare( s ); }
bool operator==( const UTF8& rhs ) const { return m_s == rhs.m_s; }
bool operator==( const std::string& rhs ) const { return m_s == rhs; }
bool operator==( const char* s ) const { return m_s == s; }
std::string::size_type find_first_of( const std::string& str, std::string::size_type pos = 0 ) const
{
return m_s.find_first_of( str, pos );
}
UTF8& operator+=( const UTF8& str )
{
m_s += str.m_s;
MAYBE_VERIFY_UTF8( c_str() );
return (UTF8&) *this;
}
UTF8& operator+=( char ch )
{
m_s.operator+=( ch );
MAYBE_VERIFY_UTF8( c_str() );
return (UTF8&) *this;
}
UTF8& operator+=( const char* s )
{
m_s.operator+=( s );
MAYBE_VERIFY_UTF8( c_str() );
return (UTF8&) *this;
}
static const std::string::size_type npos = std::string::npos;
UTF8& operator=( const wxString& o );
UTF8& operator=( const std::string& o )
{
std::string::operator=( o );
m_s = o;
MAYBE_VERIFY_UTF8( c_str() );
return *this;
}
UTF8& operator=( const char* s )
{
std::string::operator=( s );
m_s = s;
MAYBE_VERIFY_UTF8( c_str() );
return *this;
}
UTF8& operator=( char c )
{
std::string::operator=( c );
m_s = c;
MAYBE_VERIFY_UTF8( c_str() );
return *this;
}
UTF8 substr( size_t pos = 0, size_t len = npos ) const
// a substring of a UTF8 is not necessarily a UTF8 if a multibyte character
// was split, so return std::string not UTF8
std::string substr( size_t pos = 0, size_t len = npos ) const
{
return std::string::substr( pos, len );
return m_s.substr( pos, len );
}
operator const std::string& () const { return m_s; }
//operator std::string& () { return m_s; }
//operator std::string () const { return m_s; }
wxString wx_str() const;
operator wxString () const;
/// This one is not in std::string, and one wonders why... might be a solid
/// enough reason to remove it still.
operator char* () const
{
return (char*) c_str();
}
/**
* Function uni_forward
* advances over a single UTF8 encoded multibyte character, capturing the
* unicode character as it goes, and returning the number of bytes consumed.
*
* @param aSequence is the UTF8 byte sequence, must be aligned on start of character.
* @param aResult is where to put the unicode character, and may be NULL if no interest.
* @return int - the count of bytes consumed.
*/
static int uni_forward( const unsigned char* aSequence, unsigned* aResult = NULL );
// "Read only" iterating over bytes is done with these, use the uni_iter to iterate
// over UTF8 (multi-byte) characters
std::string::const_iterator begin() const { return m_s.begin(); }
std::string::const_iterator end() const { return m_s.end(); }
#ifndef SWIG
/**
* class uni_iter
* is a non-muting iterator that walks through unicode code points in the UTF8 encoded
* is a non-mutating iterator that walks through unicode code points in the UTF8 encoded
* string. The normal ++(), ++(int), ->(), and *() operators are all supported
* for read only access and some return an unsigned holding the unicode character
* appropriate for the respective operator.
@ -168,11 +203,10 @@ public:
const unsigned char* it;
// private constructor.
// private constructor
uni_iter( const char* start ) :
it( (const unsigned char*) start )
{
// for the human: assert( sizeof(unsigned) >= 4 );
}
@ -204,7 +238,6 @@ public:
return ret;
}
/*
/// return unicode at current position
unsigned operator->() const
{
@ -214,7 +247,6 @@ public:
uni_forward( it, &result );
return result;
}
*/
/// return unicode at current position
unsigned operator*() const
@ -226,6 +258,8 @@ public:
return result;
}
uni_iter operator-( int aVal ) const { return uni_iter( (char*) it - aVal ); }
bool operator==( const uni_iter& other ) const { return it == other.it; }
bool operator!=( const uni_iter& other ) const { return it != other.it; }
@ -243,7 +277,7 @@ public:
*/
uni_iter ubegin() const
{
return uni_iter( data() );
return uni_iter( m_s.data() );
}
/**
@ -252,9 +286,24 @@ public:
*/
uni_iter uend() const
{
return uni_iter( data() + size() );
return uni_iter( m_s.data() + m_s.size() );
}
/**
* Function uni_forward
* advances over a single UTF8 encoded multibyte character, capturing the
* unicode character as it goes, and returning the number of bytes consumed.
*
* @param aSequence is the UTF8 byte sequence, must be aligned on start of character.
* @param aResult is where to put the unicode character, and may be NULL if no interest.
* @return int - the count of bytes consumed.
*/
static int uni_forward( const unsigned char* aSequence, unsigned* aResult = NULL );
#endif // SWIG
protected:
std::string m_s;
};
#endif // UTF8_H_

2
pcbnew/eagle_plugin.cpp
View File

@ -1165,7 +1165,7 @@ MODULE* EAGLE_PLUGIN::makeModule( wxXmlNode* aPackage, const string& aPkgName )
{
std::unique_ptr<MODULE> m( new MODULE( m_board ) );
m->SetFPID( LIB_ID( aPkgName ) );
m->SetFPID( LIB_ID( UTF8(aPkgName) ) );
// Get the first package item and iterate
wxXmlNode* packageItem = aPackage->GetChildren();

4
pcbnew/exporters/gen_modules_placefile.cpp
View File

@ -116,7 +116,7 @@ private:
static int m_fileOpt;
static int m_fileFormat;
void initDialog();
void initDialog();
void OnOutputDirectoryBrowseClicked( wxCommandEvent& event ) override;
void OnOKButton( wxCommandEvent& event ) override;
@ -524,7 +524,7 @@ int PCB_EDIT_FRAME::DoGenFootprintsPositionFile( const wxString& aFullFileName,
line << csv_sep;
line << list[ii].m_Value;
line << csv_sep;
line << wxString( list[ii].m_Module->GetFPID().GetLibItemName() );
line << list[ii].m_Module->GetFPID().GetLibItemName().wx_str();
line << csv_sep;
line << wxString::Format( "%f%c%f%c%f",

19
pcbnew/github/github_plugin.cpp
View File

@ -98,9 +98,9 @@ using namespace std;
static const char* PRETTY_DIR = "allow_pretty_writing_to_this_dir";
typedef boost::ptr_map<string, wxZipEntry> MODULE_MAP;
typedef MODULE_MAP::iterator MODULE_ITER;
typedef MODULE_MAP::const_iterator MODULE_CITER;
typedef boost::ptr_map< wxString, wxZipEntry > MODULE_MAP;
typedef MODULE_MAP::iterator MODULE_ITER;
typedef MODULE_MAP::const_iterator MODULE_CITER;
/**
@ -160,7 +160,7 @@ void GITHUB_PLUGIN::FootprintEnumerate( wxArrayString& aFootprintNames,
for( MODULE_ITER it = m_gh_cache->begin(); it!=m_gh_cache->end(); ++it )
{
unique.insert( FROM_UTF8( it->first.c_str() ) );
unique.insert( it->first );
}
for( MYSET::const_iterator it = unique.begin(); it != unique.end(); ++it )
@ -205,9 +205,7 @@ MODULE* GITHUB_PLUGIN::FootprintLoad( const wxString& aLibraryPath,
}
}
UTF8 fp_name = aFootprintName;
MODULE_CITER it = m_gh_cache->find( fp_name );
MODULE_CITER it = m_gh_cache->find( aFootprintName );
if( it != m_gh_cache->end() ) // fp_name is present
{
@ -234,7 +232,7 @@ MODULE* GITHUB_PLUGIN::FootprintLoad( const wxString& aLibraryPath,
// any name found in the pretty file; any name in the pretty file
// must be ignored here. Also, the library nickname is unknown in
// this context so clear it just in case.
ret->SetFPID( fp_name );
ret->SetFPID( aFootprintName );
return ret;
}
@ -447,6 +445,7 @@ void GITHUB_PLUGIN::cacheLib( const wxString& aLibraryPath, const PROPERTIES* aP
wxZipInputStream zis( mis, wxConvUTF8 );
wxZipEntry* entry;
wxString fp_name;
while( ( entry = zis.GetNextEntry() ) != NULL )
{
@ -454,9 +453,7 @@ void GITHUB_PLUGIN::cacheLib( const wxString& aLibraryPath, const PROPERTIES* aP
if( fn.GetExt() == kicad_mod )
{
UTF8 fp_name = fn.GetName(); // omit extension & path
wxASSERT( IsUTF8( fp_name.c_str() ) );
fp_name = fn.GetName(); // omit extension & path
m_gh_cache->insert( fp_name, entry );
}

29
pcbnew/kicad_plugin.cpp
View File

@ -133,9 +133,9 @@ bool FP_CACHE_ITEM::IsModified() const
}
typedef boost::ptr_map< std::string, FP_CACHE_ITEM > MODULE_MAP;
typedef MODULE_MAP::iterator MODULE_ITER;
typedef MODULE_MAP::const_iterator MODULE_CITER;
typedef boost::ptr_map< wxString, FP_CACHE_ITEM > MODULE_MAP;
typedef MODULE_MAP::iterator MODULE_ITER;
typedef MODULE_MAP::const_iterator MODULE_CITER;
class FP_CACHE
@ -304,12 +304,13 @@ void FP_CACHE::Load()
m_owner->m_parser->SetLineReader( &reader );
UTF8 name = fullPath.GetName();
MODULE* footprint = (MODULE*) m_owner->m_parser->Parse();
MODULE* footprint = (MODULE*) m_owner->m_parser->Parse();
// The footprint name is the file name without the extension.
footprint->SetFPID( LIB_ID( fullPath.GetName() ) );
m_modules.insert( name, new FP_CACHE_ITEM( footprint, fullPath ) );
wxString fpName = fullPath.GetName();
footprint->SetFPID( LIB_ID( fpName ) );
m_modules.insert( fpName, new FP_CACHE_ITEM( footprint, fullPath ) );
}
catch( const IO_ERROR& ioe )
{
@ -333,9 +334,7 @@ void FP_CACHE::Load()
void FP_CACHE::Remove( const wxString& aFootprintName )
{
std::string footprintName = TO_UTF8( aFootprintName );
MODULE_CITER it = m_modules.find( footprintName );
MODULE_CITER it = m_modules.find( aFootprintName );
if( it == m_modules.end() )
{
@ -349,7 +348,7 @@ void FP_CACHE::Remove( const wxString& aFootprintName )
// Remove the module from the cache and delete the module file from the library.
wxString fullPath = it->second->GetFileName().GetFullPath();
m_modules.erase( footprintName );
m_modules.erase( aFootprintName );
wxRemoveFile( fullPath );
}
@ -400,7 +399,7 @@ bool FP_CACHE::IsModified( const wxString& aLibPath, const wxString& aFootprintN
}
else
{
MODULE_CITER it = m_modules.find( TO_UTF8( aFootprintName ) );
MODULE_CITER it = m_modules.find( aFootprintName );
if( it == m_modules.end() || it->second->IsModified() )
return true;
@ -1848,7 +1847,7 @@ void PCB_IO::FootprintEnumerate( wxArrayString& aFootprintNames,
for( MODULE_CITER it = mods.begin(); it != mods.end(); ++it )
{
aFootprintNames.Add( FROM_UTF8( it->first.c_str() ) );
aFootprintNames.Add( it->first );
}
if( !errorMsg.IsEmpty() )
@ -1867,7 +1866,7 @@ MODULE* PCB_IO::FootprintLoad( const wxString& aLibraryPath, const wxString& aFo
const MODULE_MAP& mods = m_cache->GetModules();
MODULE_CITER it = mods.find( TO_UTF8( aFootprintName ) );
MODULE_CITER it = mods.find( aFootprintName );
if( it == mods.end() )
{
@ -1902,7 +1901,7 @@ void PCB_IO::FootprintSave( const wxString& aLibraryPath, const MODULE* aFootpri
THROW_IO_ERROR( msg );
}
std::string footprintName = aFootprint->GetFPID().GetLibItemName();
wxString footprintName = aFootprint->GetFPID().GetLibItemName();
MODULE_MAP& mods = m_cache->GetModules();

6
pcbnew/legacy_plugin.cpp
View File

@ -442,7 +442,7 @@ void LEGACY_PLUGIN::loadAllSections( bool doAppend )
ReplaceIllegalFileNameChars( &fpName );
if( !fpName.empty() )
fpid = LIB_ID( fpName );
fpid = LIB_ID( UTF8( fpName ) );
module->SetFPID( fpid );
@ -3315,7 +3315,7 @@ void LP_CACHE::LoadModules( LINE_READER* aReader )
ReplaceIllegalFileNameChars( &footprintName );
// set the footprint name first thing, so exceptions can use name.
module->SetFPID( LIB_ID( footprintName ) );
module->SetFPID( LIB_ID( UTF8( footprintName ) ) );
m_owner->loadMODULE( module.get() );
@ -3370,7 +3370,7 @@ void LP_CACHE::LoadModules( LINE_READER* aReader )
{
nameOK = true;
m->SetFPID( LIB_ID( newName ) );
m->SetFPID( LIB_ID( UTF8( newName ) ) );
std::pair<MODULE_ITER, bool> r = m_modules.insert( newName, m );
wxASSERT_MSG( r.second, wxT( "error doing cache insert using guaranteed unique name" ) );

2
pcbnew/microwave/microwave_inductor.cpp
View File

@ -345,7 +345,7 @@ MODULE* MWAVE::CreateMicrowaveInductor( INDUCTOR_PATTERN& inductorPattern,
MODULE* module = aPcbFrame->CreateNewModule( msg );
// here the module is already in the BOARD, CreateNewModule() does that.
module->SetFPID( LIB_ID( std::string( "mw_inductor" ) ) );
module->SetFPID( LIB_ID( wxString( "mw_inductor" ) ) );
module->SetAttributes( MOD_VIRTUAL | MOD_CMS );
module->ClearFlags();
module->SetPosition( inductorPattern.m_End );

26
tools/utf8_tests.cpp
View File

@ -1,15 +1,37 @@
#include <string>
#include <utf8.h>
#include <wx/string.h>
void callee( const wxString& aString )
{
UTF8 arg = aString;
printf( "%s: '%s'\n", __func__, arg.c_str() );
}
int main( int argc, char** argv )
{
UTF8 bozo = "ü";
UTF8 bozo = "bozo";
callee( bozo );
wxString s = bozo;
wxString b = bozo;
if( s.IsEmpty() )
{
printf( "string is empty\n" );
}
if( s != bozo.wx_str() )
{
printf( "string miscompare\n" );
}
return 0;
}