X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=src%2FLexer.cpp;h=06a3408f41e9591d77afabfbb3074d2f3b18d9de;hb=75ffcc8e866de04b076c81d78003a86efca98e4e;hp=5136c672073d6dc451360a22f6b088915c1a66f3;hpb=e7645f24391648fb08cf6b6d808fcabec76be744;p=lyx.git diff --git a/src/Lexer.cpp b/src/Lexer.cpp index 5136c67207..06a3408f41 100644 --- a/src/Lexer.cpp +++ b/src/Lexer.cpp @@ -4,7 +4,7 @@ * Licence details can be found in the file COPYING. * * \author Alejandro Aguilar Sierra - * \author Lars Gullik Bjønnes + * \author Lars Gullik Bjønnes * \author Jean-Marc Lasgouttes * \author John Levon * @@ -14,12 +14,14 @@ #include #include "Lexer.h" +#include "Format.h" #include "support/convert.h" #include "support/debug.h" #include "support/FileName.h" #include "support/filetools.h" #include "support/gzstream.h" +#include "support/lassert.h" #include "support/lstrings.h" #include "support/lyxalgo.h" #include "support/types.h" @@ -173,7 +175,9 @@ void Lexer::Pimpl::printError(string const & message) const { string const tmpmsg = subst(message, "$$Token", getString()); lyxerr << "LyX: " << tmpmsg << " [around line " << lineno - << " of file " << to_utf8(makeDisplayPath(name)) << ']' << endl; + << " of file " << to_utf8(makeDisplayPath(name)) + << " current token: '" << getString() << "'" + << " context: '" << context << "']" << endl; } @@ -234,10 +238,14 @@ void Lexer::Pimpl::popTable() bool Lexer::Pimpl::setFile(FileName const & filename) { +#ifdef TEX2LYX + // tex2lyx does not read lyxrc and therefore can't really check for + // zipped formats. + if (false) { +#else // Check the format of the file. - string const format = filename.guessFormatFromContents(); - - if (format == "gzip" || format == "zip" || format == "compress") { + if (formats.isZippedFile(filename)) { +#endif LYXERR(Debug::LYXLEX, "lyxlex: compressed"); // The check only outputs a debug message, because it triggers // a bug in compaq cxx 6.2, where is_open() returns 'true' for @@ -247,9 +255,10 @@ bool Lexer::Pimpl::setFile(FileName const & filename) "file or stream already set."); gz_.open(filename.toFilesystemEncoding().c_str(), ios::in); is.rdbuf(&gz_); - name = filename.absFilename(); + name = filename.absFileName(); lineno = 0; - return gz_.is_open() && is.good(); + if (!gz_.is_open() || !is.good()) + return false; } else { LYXERR(Debug::LYXLEX, "lyxlex: UNcompressed"); @@ -260,12 +269,25 @@ bool Lexer::Pimpl::setFile(FileName const & filename) LYXERR(Debug::LYXLEX, "Error in Lexer::setFile: " "file or stream already set."); } - fb_.open(filename.toFilesystemEncoding().c_str(), ios::in); + fb_.open(filename.toSafeFilesystemEncoding().c_str(), ios::in); is.rdbuf(&fb_); - name = filename.absFilename(); + name = filename.absFileName(); lineno = 0; - return fb_.is_open() && is.good(); + if (!fb_.is_open() || !is.good()) + return false; + } + + // Skip byte order mark. + if (is.peek() == 0xef) { + is.get(); + if (is.peek() == 0xbb) { + is.get(); + LASSERT(is.get() == 0xbf, /**/); + } else + is.unget(); } + + return true; } @@ -311,11 +333,6 @@ bool Lexer::Pimpl::next(bool esc /* = false */) is.get(cc); c = cc; - // skip ','s - if (esc && c == ',') - continue; - - if (c == commentChar) { // Read rest of line (fast :-) #if 1 @@ -380,13 +397,13 @@ bool Lexer::Pimpl::next(bool esc /* = false */) break; } - if (!esc && c == ',') + if (c == ',') continue; /* Skip ','s */ - // using relational operators with chars other - // than == and != is not safe. And if it is done - // the type _have_ to be unsigned. It usually a - // lot better to use the functions from cctype + // using relational operators with chars other + // than == and != is not safe. And if it is done + // the type _have_ to be unsigned. It usually a + // lot better to use the functions from cctype if (c > ' ' && is) { buff.clear(); @@ -404,7 +421,7 @@ bool Lexer::Pimpl::next(bool esc /* = false */) status = LEX_TOKEN; } - if (!esc && c == '\r' && is) { + if (c == '\r' && is) { // The Windows support has lead to the // possibility of "\r\n" at the end of // a line. This will stop LyX choking @@ -461,7 +478,7 @@ bool Lexer::Pimpl::eatLine() is.get(cc); c = cc; //LYXERR(Debug::LYXLEX, "Lexer::EatLine read char: `" << c << '\''); - if (c != '\r') + if (c != '\r' && is) buff.push_back(c); } @@ -502,7 +519,7 @@ bool Lexer::Pimpl::nextToken() char cc = 0; is.get(cc); c = cc; - if (c >= ' ' && is) { + if ((c >= ' ' || c == '\t') && is) { buff.clear(); if (c == '\\') { // first char == '\\' @@ -516,7 +533,7 @@ bool Lexer::Pimpl::nextToken() buff.push_back(c); is.get(cc); c = cc; - } while (c >= ' ' && c != '\\' && is); + } while ((c >= ' ' || c == '\t') && c != '\\' && is); } if (c == '\\') @@ -639,6 +656,7 @@ void Lexer::setCommentChar(char c) pimpl_->setCommentChar(c); } + int Lexer::lex() { return pimpl_->lex(); @@ -765,6 +783,7 @@ bool Lexer::getBool() const pimpl_->printError("Bad boolean `$$Token'. " "Use \"false\" or \"true\""); lastReadOk_ = false; + return false; } @@ -902,6 +921,18 @@ string Lexer::quoteString(string const & arg) } +// same for docstring +docstring Lexer::quoteString(docstring const & arg) +{ + docstring res; + res += '"'; + res += subst(subst(arg, from_ascii("\\"), from_ascii("\\\\")), + from_ascii("\""), from_ascii("\\\"")); + res += '"'; + return res; +} + + Lexer & Lexer::operator>>(char const * required) { string token;