-/** The math parser
- \author André Pönitz (2001)
+/**
+ * \file math_parser.C
+ * This file is part of LyX, the document processor.
+ * Licence details can be found in the file COPYING.
+ *
+ * \author André Pönitz
+ *
+ * Full author contact details are available in file CREDITS.
*/
/*
...
\[\begin{array}{ccc}
- 1 & 2\b & 3^2\\
- 4 & 5\e & 6\\
- 7 & 8 & 9
+1
+&
+
\end{array}\]
*/
#include <config.h>
-#ifdef __GNUG__
-#pragma implementation
-#endif
-
#include "math_parser.h"
-#include "math_inset.h"
#include "math_arrayinset.h"
#include "math_braceinset.h"
-#include "math_boxinset.h"
#include "math_charinset.h"
#include "math_commentinset.h"
#include "math_deliminset.h"
#include "math_envinset.h"
-#include "math_extern.h"
#include "math_factory.h"
#include "math_kerninset.h"
#include "math_macro.h"
#include "math_macrotemplate.h"
-#include "math_hullinset.h"
#include "math_parboxinset.h"
#include "math_parinset.h"
#include "math_rootinset.h"
#include "math_scriptinset.h"
-#include "math_sizeinset.h"
#include "math_sqrtinset.h"
-#include "math_stringinset.h"
#include "math_support.h"
-#include "math_xyarrowinset.h"
+#include "math_tabularinset.h"
//#include "insets/insetref.h"
#include "ref_inset.h"
#include "lyxlex.h"
+#include "support/std_sstream.h"
#include "debug.h"
-#include "support/LAssert.h"
-#include "support/lstrings.h"
-#include <cctype>
-#include <algorithm>
using std::istream;
using std::ostream;
namespace {
-MathInset::mode_type asMode(string const & str)
+MathInset::mode_type asMode(MathInset::mode_type oldmode, string const & str)
{
+ //lyxerr << "handling mode: '" << str << "'" << endl;
if (str == "mathmode")
return MathInset::MATH_MODE;
if (str == "textmode" || str == "forcetext")
return MathInset::TEXT_MODE;
- if (str == "verbatimmode")
- return MathInset::VERBATIM_MODE;
- return MathInset::UNDECIDED_MODE;
+ return oldmode;
}
fill(theCatcode + 'a', theCatcode + 'z' + 1, catLetter);
fill(theCatcode + 'A', theCatcode + 'Z' + 1, catLetter);
- theCatcode['\\'] = catEscape;
- theCatcode['{'] = catBegin;
- theCatcode['}'] = catEnd;
- theCatcode['$'] = catMath;
- theCatcode['&'] = catAlign;
- theCatcode['\n'] = catNewline;
- theCatcode['#'] = catParameter;
- theCatcode['^'] = catSuper;
- theCatcode['_'] = catSub;
- theCatcode['\7f'] = catIgnore;
- theCatcode[' '] = catSpace;
- theCatcode['\t'] = catSpace;
- theCatcode['\r'] = catNewline;
- theCatcode['~'] = catActive;
- theCatcode['%'] = catComment;
+ theCatcode[int('\\')] = catEscape;
+ theCatcode[int('{')] = catBegin;
+ theCatcode[int('}')] = catEnd;
+ theCatcode[int('$')] = catMath;
+ theCatcode[int('&')] = catAlign;
+ theCatcode[int('\n')] = catNewline;
+ theCatcode[int('#')] = catParameter;
+ theCatcode[int('^')] = catSuper;
+ theCatcode[int('_')] = catSub;
+ theCatcode[int(0x7f)] = catIgnore;
+ theCatcode[int(' ')] = catSpace;
+ theCatcode[int('\t')] = catSpace;
+ theCatcode[int('\r')] = catNewline;
+ theCatcode[int('~')] = catActive;
+ theCatcode[int('%')] = catComment;
}
ostream & operator<<(ostream & os, Token const & t)
{
if (t.cs().size())
- os << "\\" << t.cs();
+ os << '\\' << t.cs();
+ else if (t.cat() == catLetter)
+ os << t.character();
else
- os << "[" << t.character() << "," << t.cat() << "]";
+ os << '[' << t.character() << ',' << t.cat() << ']';
return os;
}
void lex(string const & s);
///
bool good() const;
+ ///
+ string parse_verbatim_item();
+ ///
+ string parse_verbatim_option();
///
int lineno_;
Token const & Parser::getToken()
{
static const Token dummy;
- //lyxerr << "looking at token " << tokens_[pos_] << " pos: " << pos_ << '\n';
+ //lyxerr << "looking at token " << tokens_[pos_] << " pos: " << pos_ << endl;
return good() ? tokens_[pos_++] : dummy;
}
while (catcode(c) == catSpace || catcode(c) == catNewline)
if (!is.get(c))
break;
- //lyxerr << "putting back: " << c << "\n";
+ //lyxerr << "putting back: " << c << endl;
is.putback(c);
}
break;
}
}
+ // Remove the space after \end_inset
+ if (is.get(c) && c != ' ')
+ is.unget();
// tokenize buffer
tokenize(s);
char c;
while (is.get(c)) {
- //lyxerr << "reading c: " << c << "\n";
+ //lyxerr << "reading c: " << c << endl;
switch (catcode(c)) {
case catNewline: {
}
case catIgnore: {
- lyxerr << "ignoring a char: " << int(c) << "\n";
+ lyxerr << "ignoring a char: " << int(c) << endl;
break;
}
lyxerr << " <#> ";
lyxerr << tokens_[i];
}
- lyxerr << " pos: " << pos_ << "\n";
+ lyxerr << " pos: " << pos_ << endl;
}
}
+string Parser::parse_verbatim_option()
+{
+ skipSpaces();
+ string res;
+ if (nextToken().character() == '[') {
+ Token t = getToken();
+ for (Token t = getToken(); t.character() != ']' && good(); t = getToken()) {
+ if (t.cat() == catBegin) {
+ putback();
+ res += '{' + parse_verbatim_item() + '}';
+ } else
+ res += t.asString();
+ }
+ }
+ return res;
+}
+
+
+string Parser::parse_verbatim_item()
+{
+ skipSpaces();
+ string res;
+ if (nextToken().cat() == catBegin) {
+ Token t = getToken();
+ for (Token t = getToken(); t.cat() != catEnd && good(); t = getToken()) {
+ if (t.cat() == catBegin) {
+ putback();
+ res += '{' + parse_verbatim_item() + '}';
+ }
+ else
+ res += t.asString();
+ }
+ }
+ return res;
+}
+
+
MathArray Parser::parse(unsigned flags, mode_type mode)
{
MathArray ar;
}
-void Parser::parse2(MathAtom & at, unsigned flags, mode_type mode,
- bool numbered)
+void Parser::parse2(MathAtom & at, const unsigned flags, const mode_type mode,
+ const bool numbered)
{
parse1(*(at.nucleus()->asGridInset()), flags, mode, numbered);
}
void Parser::parse1(MathGridInset & grid, unsigned flags,
- mode_type mode, bool numbered)
+ const mode_type mode, const bool numbered)
{
int limits = 0;
MathGridInset::row_type cellrow = 0;
grid.asHullInset()->numbered(cellrow, numbered);
//dump();
+ //lyxerr << " flags: " << flags << endl;
+ //lyxerr << " mode: " << mode << endl;
//lyxerr << "grid: " << grid << endl;
while (good()) {
Token const & t = getToken();
#ifdef FILEDEBUG
- lyxerr << "t: " << t << " flags: " << flags << "\n";
+ lyxerr << "t: " << t << " flags: " << flags << endl;
+ lyxerr << "mode: " << mode << endl;
cell->dump();
- lyxerr << "\n";
+ lyxerr << endl;
#endif
if (flags & FLAG_ITEM) {
- if (t.cat() == catSpace)
- continue;
- flags &= ~FLAG_ITEM;
- if (t.cat() == catBegin) {
+ if (t.cat() == catBegin) {
// skip the brace and collect everything to the next matching
// closing brace
- flags |= FLAG_BRACE_LAST;
- continue;
+ parse1(grid, FLAG_BRACE_LAST, mode, numbered);
+ return;
}
// handle only this single token, leave the loop if done
- flags |= FLAG_LEAVE;
+ flags = FLAG_LEAVE;
}
}
else {
- error("something strange in the parser\n");
+ error("something strange in the parser");
break;
}
}
cell->push_back(MathAtom(new MathCharInset(t.character())));
}
- else if (t.cat() == catNewline && mode != MathInset::MATH_MODE)
- cell->push_back(MathAtom(new MathCharInset(t.character())));
+ else if (t.cat() == catNewline && mode != MathInset::MATH_MODE) {
+ if (cell->empty() || cell->back()->getChar() != ' ')
+ cell->push_back(MathAtom(new MathCharInset(' ')));
+ }
else if (t.cat() == catParameter) {
Token const & n = getToken();
else if (t.cat() == catAlign) {
++cellcol;
- //lyxerr << " column now " << cellcol << " max: " << grid.ncols() << "\n";
+ //lyxerr << " column now " << cellcol << " max: " << grid.ncols() << endl;
if (cellcol == grid.ncols()) {
- //lyxerr << "adding column " << cellcol << "\n";
+ //lyxerr << "adding column " << cellcol << endl;
grid.addCol(cellcol - 1);
}
cell = &grid.cell(grid.index(cellrow, cellcol));
}
else if (t.cat() == catSuper || t.cat() == catSub) {
- if (mode == MathInset::VERBATIM_MODE)
- cell->push_back(MathAtom(new MathStringInset(t.asString())));
- else {
- bool up = (t.cat() == catSuper);
- // we need no new script inset if the last thing was a scriptinset,
- // which has that script already not the same script already
- if (!cell->size())
- cell->push_back(MathAtom(new MathScriptInset(up)));
- else if (cell->back()->asScriptInset() &&
- !cell->back()->asScriptInset()->has(up))
- cell->back().nucleus()->asScriptInset()->ensure(up);
- else if (cell->back()->asScriptInset())
- cell->push_back(MathAtom(new MathScriptInset(up)));
- else
- cell->back() = MathAtom(new MathScriptInset(cell->back(), up));
- MathScriptInset * p = cell->back().nucleus()->asScriptInset();
- // special handling of {}-bases
- // is this always correct?
- if (p->nuc().size() == 1 && p->nuc().back()->asNestInset() &&
- p->nuc().back()->extraBraces())
- p->nuc() = p->nuc().back()->asNestInset()->cell(0);
- parse(p->cell(up), FLAG_ITEM, mode);
- if (limits) {
- p->limits(limits);
- limits = 0;
- }
+ bool up = (t.cat() == catSuper);
+ // we need no new script inset if the last thing was a scriptinset,
+ // which has that script already not the same script already
+ if (!cell->size())
+ cell->push_back(MathAtom(new MathScriptInset(up)));
+ else if (cell->back()->asScriptInset() &&
+ !cell->back()->asScriptInset()->has(up))
+ cell->back().nucleus()->asScriptInset()->ensure(up);
+ else if (cell->back()->asScriptInset())
+ cell->push_back(MathAtom(new MathScriptInset(up)));
+ else
+ cell->back() = MathAtom(new MathScriptInset(cell->back(), up));
+ MathScriptInset * p = cell->back().nucleus()->asScriptInset();
+ // special handling of {}-bases
+ // is this always correct?
+ // It appears that this is wrong (Dekel)
+ //if (p->nuc().size() == 1 && p->nuc().back()->asNestInset() &&
+ // p->nuc().back()->extraBraces())
+ // p->nuc() = p->nuc().back()->asNestInset()->cell(0);
+ parse(p->cell(up), FLAG_ITEM, mode);
+ if (limits) {
+ p->limits(limits);
+ limits = 0;
}
}
else if (t.character() == ']' && (flags & FLAG_BRACK_LAST)) {
- //lyxerr << "finished reading option\n";
+ //lyxerr << "finished reading option" << endl;
return;
}
cell->back().nucleus()->lock(true);
}
- else if (t.cs() == "def" || t.cs() == "newcommand") {
+ else if (t.cs() == "def" ||
+ t.cs() == "newcommand" ||
+ t.cs() == "renewcommand")
+ {
+ string const type = t.cs();
string name;
int nargs = 0;
if (t.cs() == "def") {
++nargs;
}
nargs /= 2;
- //lyxerr << "read \\def parameter list '" << pars << "'\n";
+ //lyxerr << "read \\def parameter list '" << pars << "'" << endl;
- } else { // t.cs() == "newcommand"
+ } else { // t.cs() == "newcommand" || t.cs() == "renewcommand"
if (getToken().cat() != catBegin) {
- error("'{' in \\newcommand expected (1) \n");
+ error("'{' in \\newcommand expected (1) ");
return;
}
name = getToken().cs();
if (getToken().cat() != catEnd) {
- error("'}' in \\newcommand expected\n");
+ error("'}' in \\newcommand expected");
return;
}
//MathArray test;
//test.push_back(createMathInset(name));
//if (ar1.contains(test)) {
- // error("we cannot handle recursive macros at all.\n");
+ // error("we cannot handle recursive macros at all.");
// return;
//}
// is a version for display attached?
skipSpaces();
MathArray ar2;
- if (nextToken().cat() == catBegin) {
+ if (nextToken().cat() == catBegin)
parse(ar2, FLAG_ITEM, MathInset::MATH_MODE);
- }
- cell->push_back(MathAtom(new MathMacroTemplate(name, nargs, ar1, ar2)));
+ cell->push_back(MathAtom(new MathMacroTemplate(name, nargs, type,
+ ar1, ar2)));
}
else if (t.cs() == "(") {
parse(count, FLAG_ITEM, mode);
int cols = 1;
if (!extractNumber(count, cols)) {
- lyxerr << " can't extract number of cells from " << count << "\n";
+ lyxerr << " can't extract number of cells from " << count << endl;
}
// resize the table if necessary
for (int i = 0; i < cols; ++i) {
++cellcol;
if (cellcol == grid.ncols()) {
- //lyxerr << "adding column " << cellcol << "\n";
+ //lyxerr << "adding column " << cellcol << endl;
grid.addCol(cellcol - 1);
}
cell = &grid.cell(grid.index(cellrow, cellcol));
}
else if (t.cs() == "hline") {
- if (grid.asHullInset())
- grid.asHullInset()->rowinfo(cellrow + 1);
+ grid.rowinfo(cellrow).lines_ ++;
}
else if (t.cs() == "sqrt") {
}
}
- else if (t.cs() == "ref") {
- cell->push_back(MathAtom(new RefInset));
+ else if (t.cs() == "xrightarrow" || t.cs() == "xleftarrow") {
+ cell->push_back(createMathInset(t.cs()));
+ parse(cell->back().nucleus()->cell(1), FLAG_OPTION, mode);
+ parse(cell->back().nucleus()->cell(0), FLAG_ITEM, mode);
+ }
+
+ else if (t.cs() == "ref" || t.cs() == "prettyref" ||
+ t.cs() == "pageref" || t.cs() == "vpageref" || t.cs() == "vref") {
+ cell->push_back(MathAtom(new RefInset(t.cs())));
parse(cell->back().nucleus()->cell(1), FLAG_OPTION, mode);
parse(cell->back().nucleus()->cell(0), FLAG_ITEM, mode);
}
else if (t.cs() == "left") {
+ skipSpaces();
string l = getToken().asString();
MathArray ar;
parse(ar, FLAG_RIGHT, mode);
+ skipSpaces();
string r = getToken().asString();
cell->push_back(MathAtom(new MathDelimInset(l, r, ar)));
}
else if (t.cs() == "right") {
if (flags & FLAG_RIGHT)
return;
- //lyxerr << "got so far: '" << cell << "'\n";
+ //lyxerr << "got so far: '" << cell << "'" << endl;
error("Unmatched right delimiter");
return;
}
string const name = getArg('{', '}');
if (name == "array" || name == "subarray") {
- string const valign =
- asString(parse(FLAG_OPTION, MathInset::VERBATIM_MODE)) + 'c';
- string const halign =
- asString(parse(FLAG_ITEM, MathInset::VERBATIM_MODE));
+ string const valign = parse_verbatim_option() + 'c';
+ string const halign = parse_verbatim_item();
cell->push_back(MathAtom(new MathArrayInset(name, valign[0], halign)));
parse2(cell->back(), FLAG_END, mode, false);
}
+ else if (name == "tabular") {
+ string const valign = parse_verbatim_option() + 'c';
+ string const halign = parse_verbatim_item();
+ cell->push_back(MathAtom(new MathTabularInset(name, valign[0], halign)));
+ parse2(cell->back(), FLAG_END, MathInset::TEXT_MODE, false);
+ }
+
else if (name == "split" || name == "cases" ||
name == "gathered" || name == "aligned") {
cell->push_back(createMathInset(name));
parse2(cell->back(), FLAG_END, MathInset::MATH_MODE, !stared(name));
}
+ else if (name == "flalign" || name == "flalign*") {
+ cell->push_back(MathAtom(new MathHullInset("flalign")));
+ parse2(cell->back(), FLAG_END, MathInset::MATH_MODE, !stared(name));
+ }
+
else if (name == "alignat" || name == "alignat*") {
// ignore this for a while
getArg('{', '}');
}
else {
- // lyxerr << "unknow math inset begin '" << name << "'\n";
+ dump();
+ lyxerr << "found unknown math environment '" << name << "'" << endl;
// create generic environment inset
cell->push_back(MathAtom(new MathEnvInset(name)));
- parse(cell->back().nucleus()->cell(0), FLAG_END, mode);
+ parse(cell->back().nucleus()->cell(0), FLAG_ITEM, mode);
}
}
#warning A hack...
#endif
string s;
- while (1) {
+ while (true) {
Token const & t = getToken();
if (!good()) {
putback();
}
else if (t.cs() == "label") {
- MathArray ar = parse(FLAG_ITEM, MathInset::VERBATIM_MODE);
+ string label = parse_verbatim_item();
+ MathArray ar;
+ asArray(label, ar);
if (grid.asHullInset()) {
- grid.asHullInset()->label(cellrow, asString(ar));
+ grid.asHullInset()->label(cellrow, label);
} else {
cell->push_back(createMathInset(t.cs()));
cell->push_back(MathAtom(new MathBraceInset(ar)));
parse2(cell->back(), FLAG_ITEM, mode, false);
}
- else if (t.cs() == "xymatrix") {
+ else if (t.cs() == "framebox" || t.cs() == "makebox") {
cell->push_back(createMathInset(t.cs()));
- parse2(cell->back(), FLAG_ITEM, mode, false);
- }
-
- else if (t.cs() == "framebox") {
- cell->push_back(createMathInset(t.cs()));
- parse(cell->back().nucleus()->cell(0), FLAG_OPTION, mode);
- parse(cell->back().nucleus()->cell(1), FLAG_ITEM, mode);
+ parse(cell->back().nucleus()->cell(0), FLAG_OPTION, MathInset::TEXT_MODE);
+ parse(cell->back().nucleus()->cell(1), FLAG_OPTION, MathInset::TEXT_MODE);
+ parse(cell->back().nucleus()->cell(2), FLAG_ITEM, MathInset::TEXT_MODE);
}
#if 0
p->up_ = nextToken().cat() == catSuper;
getToken();
parse(p->cell(1), FLAG_ITEM, mode);
- //lyxerr << "read label: " << p->cell(1) << "\n";
+ //lyxerr << "read label: " << p->cell(1) << endl;
}
cell->push_back(MathAtom(p));
- //lyxerr << "read cell: " << cell << "\n";
+ //lyxerr << "read cell: " << cell << endl;
}
#endif
if (l) {
if (l->inset == "font") {
cell->push_back(createMathInset(t.cs()));
- parse(cell->back().nucleus()->cell(0), FLAG_ITEM, asMode(l->extra));
+ parse(cell->back().nucleus()->cell(0), FLAG_ITEM, asMode(mode, l->extra));
}
else if (l->inset == "oldfont") {
cell->push_back(createMathInset(t.cs()));
- parse(cell->back().nucleus()->cell(0), flags, asMode(l->extra));
+ parse(cell->back().nucleus()->cell(0), flags, asMode(mode, l->extra));
return;
}
else if (l->inset == "parbox") {
// read optional positioning and width
- MathArray pos, width;
- parse(pos, FLAG_OPTION, MathInset::VERBATIM_MODE);
-
- parse(width, FLAG_ITEM, MathInset::VERBATIM_MODE);
+ string pos = parse_verbatim_option();
+ string width = parse_verbatim_item();
cell->push_back(createMathInset(t.cs()));
parse(cell->back().nucleus()->cell(0), FLAG_ITEM, MathInset::TEXT_MODE);
- cell->back().nucleus()->asParboxInset()->setPosition(asString(pos));
- cell->back().nucleus()->asParboxInset()->setWidth(asString(width));
+ cell->back().nucleus()->asParboxInset()->setPosition(pos);
+ cell->back().nucleus()->asParboxInset()->setWidth(width);
}
else {
MathAtom at = createMathInset(t.cs());
for (MathInset::idx_type i = 0; i < at->nargs(); ++i)
- parse(at.nucleus()->cell(i), FLAG_ITEM, asMode(l->extra));
+ parse(at.nucleus()->cell(i), FLAG_ITEM, asMode(mode, l->extra));
cell->push_back(at);
}
}
else {
MathAtom at = createMathInset(t.cs());
MathInset::mode_type m = mode;
- if (m == MathInset::UNDECIDED_MODE)
+ //if (m == MathInset::UNDECIDED_MODE)
+ //lyxerr << "default creation: m1: " << m << endl;
+ if (at->currentMode() != MathInset::UNDECIDED_MODE)
m = at->currentMode();
+ //lyxerr << "default creation: m2: " << m << endl;
MathInset::idx_type start = 0;
// this fails on \bigg[...\bigg]
//MathArray opt;
istringstream is(str.c_str());
Parser(is).parse1(grid, 0, MathInset::MATH_MODE, false);
}
-