lyx_mirror/src/lyxlex.C
Lars Gullik Bjønnes c46b7d8955 Merge the unicode branch into trunk.
- src/support/unicode.[Ch]: new files with functions for converting
  to and fro ucs4, ucs2 and utf8.
- src/support/docstring.h: specialization of basic_string that
  holds a uint32_t internally.
- Several functions changed to use char_type instead of char or unsigned char.
- Qt3 and Qt4 sends ucs2 on to core
- Gtk sends ucs4 on to core
- Read and write utf-8 .lyx files.
- font_metrics and painter updated to handle ucs4 chars as input.
- Quite a bit of ugly compability code, conversion string->docstring, etc.
- Have fun...


git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@14661 a592a061-630c-0410-9148-cb99ea01b6c8
2006-08-13 22:54:59 +00:00

313 lines
5.1 KiB
C

/**
* \file lyxlex.C
* This file is part of LyX, the document processor.
* Licence details can be found in the file COPYING.
*
* \author Alejandro Aguilar Sierra
* \author Lars Gullik Bjønnes
* \author Jean-Marc Lasgouttes
* \author John Levon
*
* Full author contact details are available in file CREDITS.
*/
#include <config.h>
#include "lyxlex.h"
#include "debug.h"
#include "lyxlex_pimpl.h"
#include "support/convert.h"
#include "support/lstrings.h"
#include <sstream>
using lyx::support::compare_ascii_no_case;
using lyx::support::isStrDbl;
using lyx::support::isStrInt;
using lyx::support::ltrim;
using lyx::support::prefixIs;
using lyx::support::subst;
using lyx::support::trim;
using std::endl;
using std::string;
using std::istream;
using std::ostream;
LyXLex::LyXLex(keyword_item * tab, int num)
: pimpl_(new Pimpl(tab, num))
{}
LyXLex::~LyXLex()
{
delete pimpl_;
}
bool LyXLex::isOK() const
{
return pimpl_->is.good();
}
void LyXLex::setLineNo(int l)
{
pimpl_->lineno = l;
}
int LyXLex::getLineNo() const
{
return pimpl_->lineno;
}
istream & LyXLex::getStream()
{
return pimpl_->is;
}
void LyXLex::pushTable(keyword_item * tab, int num)
{
pimpl_->pushTable(tab, num);
}
void LyXLex::popTable()
{
pimpl_->popTable();
}
void LyXLex::printTable(ostream & os)
{
pimpl_->printTable(os);
}
void LyXLex::printError(string const & message) const
{
pimpl_->printError(message);
}
bool LyXLex::setFile(string const & filename)
{
return pimpl_->setFile(filename);
}
void LyXLex::setStream(istream & i)
{
pimpl_->setStream(i);
}
void LyXLex::setCommentChar(char c)
{
pimpl_->setCommentChar(c);
}
int LyXLex::lex()
{
return pimpl_->lex();
}
int LyXLex::getInteger() const
{
if (isStrInt(pimpl_->getString()))
return convert<int>(pimpl_->getString());
pimpl_->printError("Bad integer `$$Token'");
return -1;
}
double LyXLex::getFloat() const
{
// replace comma with dot in case the file was written with
// the wrong locale (should be rare, but is easy enough to
// avoid).
string const str = subst(pimpl_->getString(), ",", ".");
if (isStrDbl(str))
return convert<double>(str);
pimpl_->printError("Bad float `$$Token'");
return -1;
}
string const LyXLex::getString() const
{
return pimpl_->getString();
}
lyx::docstring const LyXLex::getDocString() const
{
return pimpl_->getDocString();
}
// I would prefer to give a tag number instead of an explicit token
// here, but it is not possible because Buffer::readDocument uses
// explicit tokens (JMarc)
string const LyXLex::getLongString(string const & endtoken)
{
string str, prefix;
bool firstline = true;
while (isOK()) {
if (!eatLine())
// blank line in the file being read
continue;
string const token = trim(getString(), " \t");
lyxerr[Debug::PARSER] << "LongString: `"
<< getString() << '\'' << endl;
// We do a case independent comparison, like search_kw does.
if (compare_ascii_no_case(token, endtoken) == 0)
break;
string tmpstr = getString();
if (firstline) {
string::size_type i(tmpstr.find_first_not_of(' '));
if (i != string::npos)
prefix = tmpstr.substr(0, i);
firstline = false;
lyxerr[Debug::PARSER]
<< "Prefix = `" << prefix << "\'" << endl;
}
// further lines in long strings may have the same
// whitespace prefix as the first line. Remove it.
if (prefix.length() && prefixIs(tmpstr, prefix)) {
tmpstr.erase(0, prefix.length() - 1);
}
str += ltrim(tmpstr, "\t") + '\n';
}
if (!isOK()) {
printError("Long string not ended by `" + endtoken + '\'');
}
return str;
}
bool LyXLex::getBool() const
{
if (pimpl_->getString() == "true") {
return true;
} else if (pimpl_->getString() != "false") {
pimpl_->printError("Bad boolean `$$Token'. "
"Use \"false\" or \"true\"");
}
return false;
}
bool LyXLex::eatLine()
{
return pimpl_->eatLine();
}
bool LyXLex::next(bool esc)
{
return pimpl_->next(esc);
}
bool LyXLex::nextToken()
{
return pimpl_->nextToken();
}
void LyXLex::pushToken(string const & pt)
{
pimpl_->pushToken(pt);
}
LyXLex::operator void const *() const
{
// This behaviour is NOT the same as the std::streams which would
// use fail() here. However, our implementation of getString() et al.
// can cause the eof() and fail() bits to be set, even though we
// haven't tried to read 'em.
return pimpl_->is.bad() ? 0 : this;
}
bool LyXLex::operator!() const
{
return pimpl_->is.bad();
}
LyXLex & LyXLex::operator>>(std::string & s)
{
if (isOK()) {
next();
s = getString();
}
return *this;
}
LyXLex & LyXLex::operator>>(double & s)
{
if (isOK()) {
next();
s = getFloat();
}
return *this;
}
LyXLex & LyXLex::operator>>(int & s)
{
if (isOK()) {
next();
s = getInteger();
}
return *this;
}
LyXLex & LyXLex::operator>>(unsigned int & s)
{
if (isOK()) {
next();
s = getInteger();
}
return *this;
}
LyXLex & LyXLex::operator>>(bool & s)
{
if (isOK()) {
next();
s = getBool();
}
return *this;
}
/// quotes a string, e.g. for use in preferences files or as an argument of the "log" dialog
string const LyXLex::quoteString(string const & arg)
{
std::ostringstream os;
os << '"' << subst(subst(arg, "\\", "\\\\"), "\"", "\\\"") << '"';
return os.str();
}