[lyx.git] / src / tex2lyx / text.cpp

/**
 * \file tex2lyx/text.cpp
 * This file is part of LyX, the document processor.
 * Licence details can be found in the file COPYING.
 *
 * \author André Pönitz
 * \author Jean-Marc Lasgouttes
 * \author Uwe Stöhr
 *
 * Full author contact details are available in file CREDITS.
 */

// {[(

#include <config.h>

#include "tex2lyx.h"

#include "Context.h"
#include "Encoding.h"
#include "FloatList.h"
#include "LaTeXPackages.h"
#include "Layout.h"
#include "Length.h"
#include "Preamble.h"

#include "insets/ExternalTemplate.h"

#include "support/lassert.h"
#include "support/convert.h"
#include "support/FileName.h"
#include "support/filetools.h"
#include "support/lstrings.h"
#include "support/lyxtime.h"

#include <algorithm>
#include <iostream>
#include <map>
#include <sstream>
#include <vector>

using namespace std;
using namespace lyx::support;

namespace lyx {


void parse_text_in_inset(Parser & p, ostream & os, unsigned flags, bool outer,
                Context const & context, InsetLayout const * layout)
{
        bool const forcePlainLayout =
                layout ? layout->forcePlainLayout() : false;
        Context newcontext(true, context.textclass);
        if (forcePlainLayout)
                newcontext.layout = &context.textclass.plainLayout();
        else
                newcontext.font = context.font;
        parse_text(p, os, flags, outer, newcontext);
        newcontext.check_end_layout(os);
}


namespace {

void parse_text_in_inset(Parser & p, ostream & os, unsigned flags, bool outer,
                Context const & context, string const & name)
{
        InsetLayout const * layout = 0;
        DocumentClass::InsetLayouts::const_iterator it =
                context.textclass.insetLayouts().find(from_ascii(name));
        if (it != context.textclass.insetLayouts().end())
                layout = &(it->second);
        parse_text_in_inset(p, os, flags, outer, context, layout);
}

/// parses a paragraph snippet, useful for example for \\emph{...}
void parse_text_snippet(Parser & p, ostream & os, unsigned flags, bool outer,
                Context & context)
{
        Context newcontext(context);
        // Don't inherit the paragraph-level extra stuff
        newcontext.par_extra_stuff.clear();
        parse_text(p, os, flags, outer, newcontext);
        // Make sure that we don't create invalid .lyx files
        context.need_layout = newcontext.need_layout;
        context.need_end_layout = newcontext.need_end_layout;
}


/*!
 * Thin wrapper around parse_text_snippet() using a string.
 *
 * We completely ignore \c context.need_layout and \c context.need_end_layout,
 * because our return value is not used directly (otherwise the stream version
 * of parse_text_snippet() could be used). That means that the caller needs
 * to do layout management manually.
 * This is intended to parse text that does not create any layout changes.
 */
string parse_text_snippet(Parser & p, unsigned flags, const bool outer,
                  Context & context)
{
        Context newcontext(context);
        newcontext.need_layout = false;
        newcontext.need_end_layout = false;
        newcontext.new_layout_allowed = false;
        // Avoid warning by Context::~Context()
        newcontext.par_extra_stuff.clear();
        ostringstream os;
        parse_text_snippet(p, os, flags, outer, newcontext);
        return os.str();
}


char const * const known_ref_commands[] = { "ref", "pageref", "vref",
 "vpageref", "prettyref", "eqref", 0 };

char const * const known_coded_ref_commands[] = { "ref", "pageref", "vref",
 "vpageref", "formatted", "eqref", 0 };

/*!
 * natbib commands.
 * The starred forms are also known except for "citefullauthor",
 * "citeyear" and "citeyearpar".
 */
char const * const known_natbib_commands[] = { "cite", "citet", "citep",
"citealt", "citealp", "citeauthor", "citeyear", "citeyearpar",
"citefullauthor", "Citet", "Citep", "Citealt", "Citealp", "Citeauthor", 0 };

/*!
 * jurabib commands.
 * No starred form other than "cite*" known.
 */
char const * const known_jurabib_commands[] = { "cite", "citet", "citep",
"citealt", "citealp", "citeauthor", "citeyear", "citeyearpar",
// jurabib commands not (yet) supported by LyX:
// "fullcite",
// "footcite", "footcitet", "footcitep", "footcitealt", "footcitealp",
// "footciteauthor", "footciteyear", "footciteyearpar",
"citefield", "citetitle", 0 };

/// LaTeX names for quotes
char const * const known_quotes[] = { "dq", "guillemotleft", "flqq", "og",
"guillemotright", "frqq", "fg", "glq", "glqq", "textquoteleft", "grq", "grqq",
"quotedblbase", "textquotedblleft", "quotesinglbase", "textquoteright", "flq",
"guilsinglleft", "frq", "guilsinglright", 0};

/// the same as known_quotes with .lyx names
char const * const known_coded_quotes[] = { "prd", "ard", "ard", "ard",
"ald", "ald", "ald", "gls", "gld", "els", "els", "grd",
"gld", "grd", "gls", "ers", "fls",
"fls", "frs", "frs", 0};

/// LaTeX names for font sizes
char const * const known_sizes[] = { "tiny", "scriptsize", "footnotesize",
"small", "normalsize", "large", "Large", "LARGE", "huge", "Huge", 0};

/// the same as known_sizes with .lyx names
char const * const known_coded_sizes[] = { "tiny", "scriptsize", "footnotesize",
"small", "normal", "large", "larger", "largest", "huge", "giant", 0};

/// LaTeX 2.09 names for font families
char const * const known_old_font_families[] = { "rm", "sf", "tt", 0};

/// LaTeX names for font families
char const * const known_font_families[] = { "rmfamily", "sffamily",
"ttfamily", 0};

/// LaTeX names for font family changing commands
char const * const known_text_font_families[] = { "textrm", "textsf",
"texttt", 0};

/// The same as known_old_font_families, known_font_families and
/// known_text_font_families with .lyx names
char const * const known_coded_font_families[] = { "roman", "sans",
"typewriter", 0};

/// LaTeX 2.09 names for font series
char const * const known_old_font_series[] = { "bf", 0};

/// LaTeX names for font series
char const * const known_font_series[] = { "bfseries", "mdseries", 0};

/// LaTeX names for font series changing commands
char const * const known_text_font_series[] = { "textbf", "textmd", 0};

/// The same as known_old_font_series, known_font_series and
/// known_text_font_series with .lyx names
char const * const known_coded_font_series[] = { "bold", "medium", 0};

/// LaTeX 2.09 names for font shapes
char const * const known_old_font_shapes[] = { "it", "sl", "sc", 0};

/// LaTeX names for font shapes
char const * const known_font_shapes[] = { "itshape", "slshape", "scshape",
"upshape", 0};

/// LaTeX names for font shape changing commands
char const * const known_text_font_shapes[] = { "textit", "textsl", "textsc",
"textup", 0};

/// The same as known_old_font_shapes, known_font_shapes and
/// known_text_font_shapes with .lyx names
char const * const known_coded_font_shapes[] = { "italic", "slanted",
"smallcaps", "up", 0};

/// Known special characters which need skip_spaces_braces() afterwards
char const * const known_special_chars[] = {"ldots", "lyxarrow",
"textcompwordmark", "slash", 0};

/// the same as known_special_chars with .lyx names
char const * const known_coded_special_chars[] = {"ldots{}", "menuseparator",
"textcompwordmark{}", "slash{}", 0};

/*!
 * Graphics file extensions known by the dvips driver of the graphics package.
 * These extensions are used to complete the filename of an included
 * graphics file if it does not contain an extension.
 * The order must be the same that latex uses to find a file, because we
 * will use the first extension that matches.
 * This is only an approximation for the common cases. If we would want to
 * do it right in all cases, we would need to know which graphics driver is
 * used and know the extensions of every driver of the graphics package.
 */
char const * const known_dvips_graphics_formats[] = {"eps", "ps", "eps.gz",
"ps.gz", "eps.Z", "ps.Z", 0};

/*!
 * Graphics file extensions known by the pdftex driver of the graphics package.
 * \sa known_dvips_graphics_formats
 */
char const * const known_pdftex_graphics_formats[] = {"png", "pdf", "jpg",
"mps", "tif", 0};

/*!
 * Known file extensions for TeX files as used by \\include.
 */
char const * const known_tex_extensions[] = {"tex", 0};

/// spaces known by InsetSpace
char const * const known_spaces[] = { " ", "space", ",",
"thinspace", "quad", "qquad", "enspace", "enskip",
"negthinspace", "negmedspace", "negthickspace", "textvisiblespace",
"hfill", "dotfill", "hrulefill", "leftarrowfill", "rightarrowfill",
"upbracefill", "downbracefill", 0};

/// the same as known_spaces with .lyx names
char const * const known_coded_spaces[] = { "space{}", "space{}",
"thinspace{}", "thinspace{}", "quad{}", "qquad{}", "enspace{}", "enskip{}",
"negthinspace{}", "negmedspace{}", "negthickspace{}", "textvisiblespace{}",
"hfill{}", "dotfill{}", "hrulefill{}", "leftarrowfill{}", "rightarrowfill{}",
"upbracefill{}", "downbracefill{}", 0};

/// These are translated by LyX to commands like "\\LyX{}", so we have to put
/// them in ERT. "LaTeXe" must come before "LaTeX"!
char const * const known_phrases[] = {"LyX", "TeX", "LaTeXe", "LaTeX", 0};
char const * const known_coded_phrases[] = {"LyX", "TeX", "LaTeX2e", "LaTeX", 0};
int const known_phrase_lengths[] = {3, 5, 7, 0};

// string to store the float type to be able to determine the type of subfloats
string float_type = "";


/// splits "x=z, y=b" into a map and an ordered keyword vector
void split_map(string const & s, map<string, string> & res, vector<string> & keys)
{
        vector<string> v;
        split(s, v);
        res.clear();
        keys.resize(v.size());
        for (size_t i = 0; i < v.size(); ++i) {
                size_t const pos   = v[i].find('=');
                string const index = trimSpaceAndEol(v[i].substr(0, pos));
                string const value = trimSpaceAndEol(v[i].substr(pos + 1, string::npos));
                res[index] = value;
                keys[i] = index;
        }
}


/*!
 * Split a LaTeX length into value and unit.
 * The latter can be a real unit like "pt", or a latex length variable
 * like "\textwidth". The unit may contain additional stuff like glue
 * lengths, but we don't care, because such lengths are ERT anyway.
 * \returns true if \p value and \p unit are valid.
 */
bool splitLatexLength(string const & len, string & value, string & unit)
{
        if (len.empty())
                return false;
        const string::size_type i = len.find_first_not_of(" -+0123456789.,");
        //'4,5' is a valid LaTeX length number. Change it to '4.5'
        string const length = subst(len, ',', '.');
        if (i == string::npos)
                return false;
        if (i == 0) {
                if (len[0] == '\\') {
                        // We had something like \textwidth without a factor
                        value = "1.0";
                } else {
                        return false;
                }
        } else {
                value = trimSpaceAndEol(string(length, 0, i));
        }
        if (value == "-")
                value = "-1.0";
        // 'cM' is a valid LaTeX length unit. Change it to 'cm'
        if (contains(len, '\\'))
                unit = trimSpaceAndEol(string(len, i));
        else
                unit = ascii_lowercase(trimSpaceAndEol(string(len, i)));
        return true;
}


/// A simple function to translate a latex length to something LyX can
/// understand. Not perfect, but rather best-effort.
bool translate_len(string const & length, string & valstring, string & unit)
{
        if (!splitLatexLength(length, valstring, unit))
                return false;
        // LyX uses percent values
        double value;
        istringstream iss(valstring);
        iss >> value;
        value *= 100;
        ostringstream oss;
        oss << value;
        string const percentval = oss.str();
        // a normal length
        if (unit.empty() || unit[0] != '\\')
                return true;
        string::size_type const i = unit.find(' ');
        string const endlen = (i == string::npos) ? string() : string(unit, i);
        if (unit == "\\textwidth") {
                valstring = percentval;
                unit = "text%" + endlen;
        } else if (unit == "\\columnwidth") {
                valstring = percentval;
                unit = "col%" + endlen;
        } else if (unit == "\\paperwidth") {
                valstring = percentval;
                unit = "page%" + endlen;
        } else if (unit == "\\linewidth") {
                valstring = percentval;
                unit = "line%" + endlen;
        } else if (unit == "\\paperheight") {
                valstring = percentval;
                unit = "pheight%" + endlen;
        } else if (unit == "\\textheight") {
                valstring = percentval;
                unit = "theight%" + endlen;
        }
        return true;
}

}


string translate_len(string const & length)
{
        string unit;
        string value;
        if (translate_len(length, value, unit))
                return value + unit;
        // If the input is invalid, return what we have.
        return length;
}


namespace {

/*!
 * Translates a LaTeX length into \p value, \p unit and
 * \p special parts suitable for a box inset.
 * The difference from translate_len() is that a box inset knows about
 * some special "units" that are stored in \p special.
 */
void translate_box_len(string const & length, string & value, string & unit, string & special)
{
        if (translate_len(length, value, unit)) {
                if (unit == "\\height" || unit == "\\depth" ||
                    unit == "\\totalheight" || unit == "\\width") {
                        special = unit.substr(1);
                        // The unit is not used, but LyX requires a dummy setting
                        unit = "in";
                } else
                        special = "none";
        } else {
                value.clear();
                unit = length;
                special = "none";
        }
}


/*!
 * Find a file with basename \p name in path \p path and an extension
 * in \p extensions.
 */
string find_file(string const & name, string const & path,
                 char const * const * extensions)
{
        for (char const * const * what = extensions; *what; ++what) {
                string const trial = addExtension(name, *what);
                if (makeAbsPath(trial, path).exists())
                        return trial;
        }
        return string();
}


void begin_inset(ostream & os, string const & name)
{
        os << "\n\\begin_inset " << name;
}


void begin_command_inset(ostream & os, string const & name,
                         string const & latexname)
{
        begin_inset(os, "CommandInset ");
        os << name << "\nLatexCommand " << latexname << '\n';
}


void end_inset(ostream & os)
{
        os << "\n\\end_inset\n\n";
}


bool skip_braces(Parser & p)
{
        if (p.next_token().cat() != catBegin)
                return false;
        p.get_token();
        if (p.next_token().cat() == catEnd) {
                p.get_token();
                return true;
        }
        p.putback();
        return false;
}


/// replace LaTeX commands in \p s from the unicodesymbols file with their
/// unicode points
docstring convert_unicodesymbols(docstring s)
{
        odocstringstream os;
        for (size_t i = 0; i < s.size();) {
                if (s[i] != '\\') {
                        os.put(s[i++]);
                        continue;
                }
                s = s.substr(i);
                docstring rem;
                set<string> req;
                docstring parsed = encodings.fromLaTeXCommand(s,
                                Encodings::TEXT_CMD, rem, &req);
                for (set<string>::const_iterator it = req.begin(); it != req.end(); it++)
                        preamble.registerAutomaticallyLoadedPackage(*it);
                os << parsed;
                s = rem;
                if (s.empty() || s[0] != '\\')
                        i = 0;
                else
                        i = 1;
        }
        return os.str();
}


/// try to convert \p s to a valid InsetCommand argument
string convert_command_inset_arg(string s)
{
        if (isAscii(s))
                // since we don't know the input encoding we can't use from_utf8
                s = to_utf8(convert_unicodesymbols(from_ascii(s)));
        // LyX cannot handle newlines in a latex command
        return subst(s, "\n", " ");
}


void handle_backslash(ostream & os, string const & s)
{
        for (string::const_iterator it = s.begin(), et = s.end(); it != et; ++it) {
                if (*it == '\\')
                        os << "\n\\backslash\n";
                else
                        os << *it;
        }
}


void handle_ert(ostream & os, string const & s, Context & context)
{
        // We must have a valid layout before outputting the ERT inset.
        context.check_layout(os);
        Context newcontext(true, context.textclass);
        begin_inset(os, "ERT");
        os << "\nstatus collapsed\n";
        newcontext.check_layout(os);
        for (string::const_iterator it = s.begin(), et = s.end(); it != et; ++it) {
                if (*it == '\\')
                        os << "\n\\backslash\n";
                else if (*it == '\n') {
                        newcontext.new_paragraph(os);
                        newcontext.check_layout(os);
                } else
                        os << *it;
        }
        newcontext.check_end_layout(os);
        end_inset(os);
}


void handle_comment(ostream & os, string const & s, Context & context)
{
        // TODO: Handle this better
        Context newcontext(true, context.textclass);
        begin_inset(os, "ERT");
        os << "\nstatus collapsed\n";
        newcontext.check_layout(os);
        handle_backslash(os, s);
        // make sure that our comment is the last thing on the line
        newcontext.new_paragraph(os);
        newcontext.check_layout(os);
        newcontext.check_end_layout(os);
        end_inset(os);
}


Layout const * findLayout(TextClass const & textclass, string const & name, bool command)
{
        Layout const * layout = findLayoutWithoutModule(textclass, name, command);
        if (layout)
                return layout;
        if (checkModule(name, command))
                return findLayoutWithoutModule(textclass, name, command);
        return layout;
}


InsetLayout const * findInsetLayout(TextClass const & textclass, string const & name, bool command)
{
        InsetLayout const * insetlayout = findInsetLayoutWithoutModule(textclass, name, command);
        if (insetlayout)
                return insetlayout;
        if (checkModule(name, command))
                return findInsetLayoutWithoutModule(textclass, name, command);
        return insetlayout;
}


void eat_whitespace(Parser &, ostream &, Context &, bool);


/*!
 * Skips whitespace and braces.
 * This should be called after a command has been parsed that is not put into
 * ERT, and where LyX adds "{}" if needed.
 */
void skip_spaces_braces(Parser & p, bool keepws = false)
{
        /* The following four examples produce the same typeset output and
           should be handled by this function:
           - abc \j{} xyz
           - abc \j {} xyz
           - abc \j
             {} xyz
           - abc \j %comment
             {} xyz
         */
        // Unfortunately we need to skip comments, too.
        // We can't use eat_whitespace since writing them after the {}
        // results in different output in some cases.
        bool const skipped_spaces = p.skip_spaces(true);
        bool const skipped_braces = skip_braces(p);
        if (keepws && skipped_spaces && !skipped_braces)
                // put back the space (it is better handled by check_space)
                p.unskip_spaces(true);
}


void output_command_layout(ostream & os, Parser & p, bool outer,
                           Context & parent_context,
                           Layout const * newlayout)
{
        TeXFont const oldFont = parent_context.font;
        // save the current font size
        string const size = oldFont.size;
        // reset the font size to default, because the font size switches
        // don't affect section headings and the like
        parent_context.font.size = Context::normalfont.size;
        // we only need to write the font change if we have an open layout
        if (!parent_context.atParagraphStart())
                output_font_change(os, oldFont, parent_context.font);
        parent_context.check_end_layout(os);
        Context context(true, parent_context.textclass, newlayout,
                        parent_context.layout, parent_context.font);
        if (parent_context.deeper_paragraph) {
                // We are beginning a nested environment after a
                // deeper paragraph inside the outer list environment.
                // Therefore we don't need to output a "begin deeper".
                context.need_end_deeper = true;
        }
        context.check_deeper(os);
        context.check_layout(os);
        unsigned int optargs = 0;
        while (optargs < context.layout->optargs) {
                eat_whitespace(p, os, context, false);
                if (p.next_token().cat() == catEscape ||
                    p.next_token().character() != '[')
                        break;
                p.get_token(); // eat '['
                begin_inset(os, "Argument\n");
                os << "status collapsed\n\n";
                parse_text_in_inset(p, os, FLAG_BRACK_LAST, outer, context);
                end_inset(os);
                eat_whitespace(p, os, context, false);
                ++optargs;
        }
        unsigned int reqargs = 0;
        while (reqargs < context.layout->reqargs) {
                eat_whitespace(p, os, context, false);
                if (p.next_token().cat() != catBegin)
                        break;
                p.get_token(); // eat '{'
                begin_inset(os, "Argument\n");
                os << "status collapsed\n\n";
                parse_text_in_inset(p, os, FLAG_BRACE_LAST, outer, context);
                end_inset(os);
                eat_whitespace(p, os, context, false);
                ++reqargs;
        }
        parse_text(p, os, FLAG_ITEM, outer, context);
        context.check_end_layout(os);
        if (parent_context.deeper_paragraph) {
                // We must suppress the "end deeper" because we
                // suppressed the "begin deeper" above.
                context.need_end_deeper = false;
        }
        context.check_end_deeper(os);
        // We don't need really a new paragraph, but
        // we must make sure that the next item gets a \begin_layout.
        parent_context.new_paragraph(os);
        // Set the font size to the original value. No need to output it here
        // (Context::begin_layout() will do that if needed)
        parent_context.font.size = size;
}


/*!
 * Output a space if necessary.
 * This function gets called for every whitespace token.
 *
 * We have three cases here:
 * 1. A space must be suppressed. Example: The lyxcode case below
 * 2. A space may be suppressed. Example: Spaces before "\par"
 * 3. A space must not be suppressed. Example: A space between two words
 *
 * We currently handle only 1. and 3 and from 2. only the case of
 * spaces before newlines as a side effect.
 *
 * 2. could be used to suppress as many spaces as possible. This has two effects:
 * - Reimporting LyX generated LaTeX files changes almost no whitespace
 * - Superflous whitespace from non LyX generated LaTeX files is removed.
 * The drawback is that the logic inside the function becomes
 * complicated, and that is the reason why it is not implemented.
 */
void check_space(Parser & p, ostream & os, Context & context)
{
        Token const next = p.next_token();
        Token const curr = p.curr_token();
        // A space before a single newline and vice versa must be ignored
        // LyX emits a newline before \end{lyxcode}.
        // This newline must be ignored,
        // otherwise LyX will add an additional protected space.
        if (next.cat() == catSpace ||
            next.cat() == catNewline ||
            (next.cs() == "end" && context.layout->free_spacing && curr.cat() == catNewline)) {
                return;
        }
        context.check_layout(os);
        os << ' ';
}


/*!
 * Parse all arguments of \p command
 */
void parse_arguments(string const & command,
                     vector<ArgumentType> const & template_arguments,
                     Parser & p, ostream & os, bool outer, Context & context)
{
        string ert = command;
        size_t no_arguments = template_arguments.size();
        for (size_t i = 0; i < no_arguments; ++i) {
                switch (template_arguments[i]) {
                case required:
                case req_group:
                        // This argument contains regular LaTeX
                        handle_ert(os, ert + '{', context);
                        eat_whitespace(p, os, context, false);
                        if (template_arguments[i] == required)
                                parse_text(p, os, FLAG_ITEM, outer, context);
                        else
                                parse_text_snippet(p, os, FLAG_ITEM, outer, context);
                        ert = "}";
                        break;
                case item:
                        // This argument consists only of a single item.
                        // The presence of '{' or not must be preserved.
                        p.skip_spaces();
                        if (p.next_token().cat() == catBegin)
                                ert += '{' + p.verbatim_item() + '}';
                        else
                                ert += p.verbatim_item();
                        break;
                case displaymath:
                case verbatim:
                        // This argument may contain special characters
                        ert += '{' + p.verbatim_item() + '}';
                        break;
                case optional:
                case opt_group:
                        // true because we must not eat whitespace
                        // if an optional arg follows we must not strip the
                        // brackets from this one
                        if (i < no_arguments - 1 &&
                            template_arguments[i+1] == optional)
                                ert += p.getFullOpt(true);
                        else
                                ert += p.getOpt(true);
                        break;
                }
        }
        handle_ert(os, ert, context);
}


/*!
 * Check whether \p command is a known command. If yes,
 * handle the command with all arguments.
 * \return true if the command was parsed, false otherwise.
 */
bool parse_command(string const & command, Parser & p, ostream & os,
                   bool outer, Context & context)
{
        if (known_commands.find(command) != known_commands.end()) {
                parse_arguments(command, known_commands[command], p, os,
                                outer, context);
                return true;
        }
        return false;
}


/// Parses a minipage or parbox
void parse_box(Parser & p, ostream & os, unsigned outer_flags,
               unsigned inner_flags, bool outer, Context & parent_context,
               string const & outer_type, string const & special,
               string const & inner_type)
{
        string position;
        string inner_pos;
        string hor_pos = "c";
        // We need to set the height to the LaTeX default of 1\\totalheight
        // for the case when no height argument is given
        string height_value = "1";
        string height_unit = "in";
        string height_special = "totalheight";
        string latex_height;
        string width_value;
        string width_unit;
        string latex_width;
        string width_special = "none";
        if (!inner_type.empty() && p.hasOpt()) {
                if (inner_type != "makebox")
                        position = p.getArg('[', ']');
                else {
                        latex_width = p.getArg('[', ']');
                        translate_box_len(latex_width, width_value, width_unit, width_special);
                        position = "t";
                }
                if (position != "t" && position != "c" && position != "b") {
                        cerr << "invalid position " << position << " for "
                             << inner_type << endl;
                        position = "c";
                }
                if (p.hasOpt()) {
                        if (inner_type != "makebox") {
                                latex_height = p.getArg('[', ']');
                                translate_box_len(latex_height, height_value, height_unit, height_special);
                        } else
                                hor_pos = p.getArg('[', ']');

                        if (p.hasOpt()) {
                                inner_pos = p.getArg('[', ']');
                                if (inner_pos != "c" && inner_pos != "t" &&
                                    inner_pos != "b" && inner_pos != "s") {
                                        cerr << "invalid inner_pos "
                                             << inner_pos << " for "
                                             << inner_type << endl;
                                        inner_pos = position;
                                }
                        }
                }
        }
        if (inner_type.empty()) {
                if (special.empty() && outer_type != "framebox")
                        latex_width = "1\\columnwidth";
                else {
                        Parser p2(special);
                        latex_width = p2.getArg('[', ']');
                        string const opt = p2.getArg('[', ']');
                        if (!opt.empty()) {
                                hor_pos = opt;
                                if (hor_pos != "l" && hor_pos != "c" &&
                                    hor_pos != "r") {
                                        cerr << "invalid hor_pos " << hor_pos
                                             << " for " << outer_type << endl;
                                        hor_pos = "c";
                                }
                        }
                }
        } else if (inner_type != "makebox")
                latex_width = p.verbatim_item();
        // if e.g. only \ovalbox{content} was used, set the width to 1\columnwidth
        // as this is LyX's standard for such cases (except for makebox)
        // \framebox is more special and handled below
        if (latex_width.empty() && inner_type != "makebox"
                && outer_type != "framebox")
                latex_width = "1\\columnwidth";

        translate_len(latex_width, width_value, width_unit);

        bool shadedparbox = false;
        if (inner_type == "shaded") {
                eat_whitespace(p, os, parent_context, false);
                if (outer_type == "parbox") {
                        // Eat '{'
                        if (p.next_token().cat() == catBegin)
                                p.get_token();
                        eat_whitespace(p, os, parent_context, false);
                        shadedparbox = true;
                }
                p.get_token();
                p.getArg('{', '}');
        }
        // If we already read the inner box we have to push the inner env
        if (!outer_type.empty() && !inner_type.empty() &&
            (inner_flags & FLAG_END))
                active_environments.push_back(inner_type);
        // LyX can't handle length variables
        bool use_ert = contains(width_unit, '\\') || contains(height_unit, '\\');
        if (!use_ert && !outer_type.empty() && !inner_type.empty()) {
                // Look whether there is some content after the end of the
                // inner box, but before the end of the outer box.
                // If yes, we need to output ERT.
                p.pushPosition();
                if (inner_flags & FLAG_END)
                        p.verbatimEnvironment(inner_type);
                else
                        p.verbatim_item();
                p.skip_spaces(true);
                bool const outer_env(outer_type == "framed" || outer_type == "minipage");
                if ((outer_env && p.next_token().asInput() != "\\end") ||
                    (!outer_env && p.next_token().cat() != catEnd)) {
                        // something is between the end of the inner box and
                        // the end of the outer box, so we need to use ERT.
                        use_ert = true;
                }
                p.popPosition();
        }
        // if only \makebox{content} was used we can set its width to 1\width
        // because this identic and also identic to \mbox
        // this doesn't work for \framebox{content}, thus we have to use ERT for this
        if (latex_width.empty() && inner_type == "makebox") {
                width_value = "1";
                width_unit = "in";
                width_special = "width";
        } else if (latex_width.empty() && outer_type == "framebox") {
                use_ert = true;
        }
        if (use_ert) {
                ostringstream ss;
                if (!outer_type.empty()) {
                        if (outer_flags & FLAG_END)
                                ss << "\\begin{" << outer_type << '}';
                        else {
                                ss << '\\' << outer_type << '{';
                                if (!special.empty())
                                        ss << special;
                        }
                }
                if (!inner_type.empty()) {
                        if (inner_type != "shaded") {
                                if (inner_flags & FLAG_END)
                                        ss << "\\begin{" << inner_type << '}';
                                else
                                        ss << '\\' << inner_type;
                        }
                        if (!position.empty())
                                ss << '[' << position << ']';
                        if (!latex_height.empty())
                                ss << '[' << latex_height << ']';
                        if (!inner_pos.empty())
                                ss << '[' << inner_pos << ']';
                        ss << '{' << latex_width << '}';
                        if (!(inner_flags & FLAG_END))
                                ss << '{';
                }
                if (inner_type == "shaded")
                        ss << "\\begin{shaded}";
                handle_ert(os, ss.str(), parent_context);
                if (!inner_type.empty()) {
                        parse_text(p, os, inner_flags, outer, parent_context);
                        if (inner_flags & FLAG_END)
                                handle_ert(os, "\\end{" + inner_type + '}',
                                           parent_context);
                        else
                                handle_ert(os, "}", parent_context);
                }
                if (!outer_type.empty()) {
                        // If we already read the inner box we have to pop
                        // the inner env
                        if (!inner_type.empty() && (inner_flags & FLAG_END))
                                active_environments.pop_back();

                        // Ensure that the end of the outer box is parsed correctly:
                        // The opening brace has been eaten by parse_outer_box()
                        if (!outer_type.empty() && (outer_flags & FLAG_ITEM)) {
                                outer_flags &= ~FLAG_ITEM;
                                outer_flags |= FLAG_BRACE_LAST;
                        }
                        parse_text(p, os, outer_flags, outer, parent_context);
                        if (outer_flags & FLAG_END)
                                handle_ert(os, "\\end{" + outer_type + '}',
                                           parent_context);
                        else if (inner_type.empty() && outer_type == "framebox")
                                // in this case it is already closed later
                                ;
                        else
                                handle_ert(os, "}", parent_context);
                }
        } else {
                // LyX does not like empty positions, so we have
                // to set them to the LaTeX default values here.
                if (position.empty())
                        position = "c";
                if (inner_pos.empty())
                        inner_pos = position;
                parent_context.check_layout(os);
                begin_inset(os, "Box ");
                if (outer_type == "framed")
                        os << "Framed\n";
                else if (outer_type == "framebox")
                        os << "Boxed\n";
                else if (outer_type == "shadowbox")
                        os << "Shadowbox\n";
                else if ((outer_type == "shaded" && inner_type.empty()) ||
                             (outer_type == "minipage" && inner_type == "shaded") ||
                             (outer_type == "parbox" && inner_type == "shaded")) {
                        os << "Shaded\n";
                        preamble.registerAutomaticallyLoadedPackage("color");
                } else if (outer_type == "doublebox")
                        os << "Doublebox\n";
                else if (outer_type.empty())
                        os << "Frameless\n";
                else
                        os << outer_type << '\n';
                os << "position \"" << position << "\"\n";
                os << "hor_pos \"" << hor_pos << "\"\n";
                os << "has_inner_box " << !inner_type.empty() << "\n";
                os << "inner_pos \"" << inner_pos << "\"\n";
                os << "use_parbox " << (inner_type == "parbox" || shadedparbox)
                   << '\n';
                os << "use_makebox " << (inner_type == "makebox") << '\n';
                os << "width \"" << width_value << width_unit << "\"\n";
                os << "special \"" << width_special << "\"\n";
                os << "height \"" << height_value << height_unit << "\"\n";
                os << "height_special \"" << height_special << "\"\n";
                os << "status open\n\n";

                // Unfortunately we can't use parse_text_in_inset:
                // InsetBox::forcePlainLayout() is hard coded and does not
                // use the inset layout. Apart from that do we call parse_text
                // up to two times, but need only one check_end_layout.
                bool const forcePlainLayout =
                        (!inner_type.empty() || inner_type == "makebox") &&
                        outer_type != "shaded" && outer_type != "framed";
                Context context(true, parent_context.textclass);
                if (forcePlainLayout)
                        context.layout = &context.textclass.plainLayout();
                else
                        context.font = parent_context.font;

                // If we have no inner box the contents will be read with the outer box
                if (!inner_type.empty())
                        parse_text(p, os, inner_flags, outer, context);

                // Ensure that the end of the outer box is parsed correctly:
                // The opening brace has been eaten by parse_outer_box()
                if (!outer_type.empty() && (outer_flags & FLAG_ITEM)) {
                        outer_flags &= ~FLAG_ITEM;
                        outer_flags |= FLAG_BRACE_LAST;
                }

                // Find end of outer box, output contents if inner_type is
                // empty and output possible comments
                if (!outer_type.empty()) {
                        // If we already read the inner box we have to pop
                        // the inner env
                        if (!inner_type.empty() && (inner_flags & FLAG_END))
                                active_environments.pop_back();
                        // This does not output anything but comments if
                        // inner_type is not empty (see use_ert)
                        parse_text(p, os, outer_flags, outer, context);
                }

                context.check_end_layout(os);
                end_inset(os);
#ifdef PRESERVE_LAYOUT
                // LyX puts a % after the end of the minipage
                if (p.next_token().cat() == catNewline && p.next_token().cs().size() > 1) {
                        // new paragraph
                        //handle_comment(os, "%dummy", parent_context);
                        p.get_token();
                        p.skip_spaces();
                        parent_context.new_paragraph(os);
                }
                else if (p.next_token().cat() == catSpace || p.next_token().cat() == catNewline) {
                        //handle_comment(os, "%dummy", parent_context);
                        p.get_token();
                        p.skip_spaces();
                        // We add a protected space if something real follows
                        if (p.good() && p.next_token().cat() != catComment) {
                                begin_inset(os, "space ~\n");
                                end_inset(os);
                        }
                }
#endif
        }
}


void parse_outer_box(Parser & p, ostream & os, unsigned flags, bool outer,
                     Context & parent_context, string const & outer_type,
                     string const & special)
{
        eat_whitespace(p, os, parent_context, false);
        if (flags & FLAG_ITEM) {
                // Eat '{'
                if (p.next_token().cat() == catBegin)
                        p.get_token();
                else
                        cerr << "Warning: Ignoring missing '{' after \\"
                             << outer_type << '.' << endl;
                eat_whitespace(p, os, parent_context, false);
        }
        string inner;
        unsigned int inner_flags = 0;
        p.pushPosition();
        if (outer_type == "minipage" || outer_type == "parbox") {
                p.skip_spaces(true);
                while (p.hasOpt()) {
                        p.getArg('[', ']');
                        p.skip_spaces(true);
                }
                p.getArg('{', '}');
                p.skip_spaces(true);
                if (outer_type == "parbox") {
                        // Eat '{'
                        if (p.next_token().cat() == catBegin)
                                p.get_token();
                        p.skip_spaces(true);
                }
        }
        if (outer_type == "shaded") {
                // These boxes never have an inner box
                ;
        } else if (p.next_token().asInput() == "\\parbox") {
                inner = p.get_token().cs();
                inner_flags = FLAG_ITEM;
        } else if (p.next_token().asInput() == "\\begin") {
                // Is this a minipage or shaded box?
                p.pushPosition();
                p.get_token();
                inner = p.getArg('{', '}');
                p.popPosition();
                if (inner == "minipage" || inner == "shaded")
                        inner_flags = FLAG_END;
                else
                        inner = "";
        }
        p.popPosition();
        if (inner_flags == FLAG_END) {
                if (inner != "shaded")
                {
                        p.get_token();
                        p.getArg('{', '}');
                        eat_whitespace(p, os, parent_context, false);
                }
                parse_box(p, os, flags, FLAG_END, outer, parent_context,
                          outer_type, special, inner);
        } else {
                if (inner_flags == FLAG_ITEM) {
                        p.get_token();
                        eat_whitespace(p, os, parent_context, false);
                }
                parse_box(p, os, flags, inner_flags, outer, parent_context,
                          outer_type, special, inner);
        }
}


void parse_listings(Parser & p, ostream & os, Context & parent_context)
{
        parent_context.check_layout(os);
        begin_inset(os, "listings\n");
        os << "inline false\n"
           << "status collapsed\n";
        Context context(true, parent_context.textclass);
        context.layout = &parent_context.textclass.plainLayout();
        context.check_layout(os);
        string const s = p.verbatimEnvironment("lstlisting");
        for (string::const_iterator it = s.begin(), et = s.end(); it != et; ++it) {
                if (*it == '\\')
                        os << "\n\\backslash\n";
                else if (*it == '\n') {
                        // avoid adding an empty paragraph at the end
                        if (it + 1 != et) {
                                context.new_paragraph(os);
                                context.check_layout(os);
                        }
                } else
                        os << *it;
        }
        context.check_end_layout(os);
        end_inset(os);
}


/// parse an unknown environment
void parse_unknown_environment(Parser & p, string const & name, ostream & os,
                               unsigned flags, bool outer,
                               Context & parent_context)
{
        if (name == "tabbing")
                // We need to remember that we have to handle '\=' specially
                flags |= FLAG_TABBING;

        // We need to translate font changes and paragraphs inside the
        // environment to ERT if we have a non standard font.
        // Otherwise things like
        // \large\begin{foo}\huge bar\end{foo}
        // will not work.
        bool const specialfont =
                (parent_context.font != parent_context.normalfont);
        bool const new_layout_allowed = parent_context.new_layout_allowed;
        if (specialfont)
                parent_context.new_layout_allowed = false;
        handle_ert(os, "\\begin{" + name + "}", parent_context);
        parse_text_snippet(p, os, flags, outer, parent_context);
        handle_ert(os, "\\end{" + name + "}", parent_context);
        if (specialfont)
                parent_context.new_layout_allowed = new_layout_allowed;
}


void parse_environment(Parser & p, ostream & os, bool outer,
                       string & last_env, bool & title_layout_found,
                       Context & parent_context)
{
        Layout const * newlayout;
        InsetLayout const * newinsetlayout = 0;
        string const name = p.getArg('{', '}');
        const bool is_starred = suffixIs(name, '*');
        string const unstarred_name = rtrim(name, "*");
        active_environments.push_back(name);

        if (is_math_env(name)) {
                parent_context.check_layout(os);
                begin_inset(os, "Formula ");
                os << "\\begin{" << name << "}";
                parse_math(p, os, FLAG_END, MATH_MODE);
                os << "\\end{" << name << "}";
                end_inset(os);
                if (is_display_math_env(name)) {
                        // Prevent the conversion of a line break to a space
                        // (bug 7668). This does not change the output, but
                        // looks ugly in LyX.
                        eat_whitespace(p, os, parent_context, false);
                }
        }

        else if (unstarred_name == "tabular" || name == "longtable") {
                eat_whitespace(p, os, parent_context, false);
                string width = "0pt";
                if (name == "tabular*") {
                        width = lyx::translate_len(p.getArg('{', '}'));
                        eat_whitespace(p, os, parent_context, false);
                }
                parent_context.check_layout(os);
                begin_inset(os, "Tabular ");
                handle_tabular(p, os, name, width, parent_context);
                end_inset(os);
                p.skip_spaces();
        }

        else if (parent_context.textclass.floats().typeExist(unstarred_name)) {
                eat_whitespace(p, os, parent_context, false);
                string const opt = p.hasOpt() ? p.getArg('[', ']') : string();
                eat_whitespace(p, os, parent_context, false);
                parent_context.check_layout(os);
                begin_inset(os, "Float " + unstarred_name + "\n");
                // store the float type for subfloats
                // subfloats only work with figures and tables
                if (unstarred_name == "figure")
                        float_type = unstarred_name;
                else if (unstarred_name == "table")
                        float_type = unstarred_name;
                else
                        float_type = "";
                if (!opt.empty())
                        os << "placement " << opt << '\n';
                if (contains(opt, "H"))
                        preamble.registerAutomaticallyLoadedPackage("float");
                else {
                        Floating const & fl = parent_context.textclass.floats()
                                .getType(unstarred_name);
                        if (!fl.floattype().empty() && fl.usesFloatPkg())
                                preamble.registerAutomaticallyLoadedPackage("float");
                }

                os << "wide " << convert<string>(is_starred)
                   << "\nsideways false"
                   << "\nstatus open\n\n";
                parse_text_in_inset(p, os, FLAG_END, outer, parent_context);
                end_inset(os);
                // We don't need really a new paragraph, but
                // we must make sure that the next item gets a \begin_layout.
                parent_context.new_paragraph(os);
                p.skip_spaces();
                // the float is parsed thus delete the type
                float_type = "";
        }

        else if (unstarred_name == "sidewaysfigure"
                || unstarred_name == "sidewaystable") {
                eat_whitespace(p, os, parent_context, false);
                parent_context.check_layout(os);
                if (unstarred_name == "sidewaysfigure")
                        begin_inset(os, "Float figure\n");
                else
                        begin_inset(os, "Float table\n");
                os << "wide " << convert<string>(is_starred)
                   << "\nsideways true"
                   << "\nstatus open\n\n";
                parse_text_in_inset(p, os, FLAG_END, outer, parent_context);
                end_inset(os);
                // We don't need really a new paragraph, but
                // we must make sure that the next item gets a \begin_layout.
                parent_context.new_paragraph(os);
                p.skip_spaces();
                preamble.registerAutomaticallyLoadedPackage("rotfloat");
        }

        else if (name == "wrapfigure" || name == "wraptable") {
                // syntax is \begin{wrapfigure}[lines]{placement}[overhang]{width}
                eat_whitespace(p, os, parent_context, false);
                parent_context.check_layout(os);
                // default values
                string lines = "0";
                string overhang = "0col%";
                // parse
                if (p.hasOpt())
                        lines = p.getArg('[', ']');
                string const placement = p.getArg('{', '}');
                if (p.hasOpt())
                        overhang = p.getArg('[', ']');
                string const width = p.getArg('{', '}');
                // write
                if (name == "wrapfigure")
                        begin_inset(os, "Wrap figure\n");
                else
                        begin_inset(os, "Wrap table\n");
                os << "lines " << lines
                   << "\nplacement " << placement
                   << "\noverhang " << lyx::translate_len(overhang)
                   << "\nwidth " << lyx::translate_len(width)
                   << "\nstatus open\n\n";
                parse_text_in_inset(p, os, FLAG_END, outer, parent_context);
                end_inset(os);
                // We don't need really a new paragraph, but
                // we must make sure that the next item gets a \begin_layout.
                parent_context.new_paragraph(os);
                p.skip_spaces();
                preamble.registerAutomaticallyLoadedPackage("wrapfig");
        }

        else if (name == "minipage") {
                eat_whitespace(p, os, parent_context, false);
                // Test whether this is an outer box of a shaded box
                p.pushPosition();
                // swallow arguments
                while (p.hasOpt()) {
                        p.getArg('[', ']');
                        p.skip_spaces(true);
                }
                p.getArg('{', '}');
                p.skip_spaces(true);
                Token t = p.get_token();
                bool shaded = false;
                if (t.asInput() == "\\begin") {
                        p.skip_spaces(true);
                        if (p.getArg('{', '}') == "shaded")
                                shaded = true;
                }
                p.popPosition();
                if (shaded)
                        parse_outer_box(p, os, FLAG_END, outer,
                                        parent_context, name, "shaded");
                else
                        parse_box(p, os, 0, FLAG_END, outer, parent_context,
                                  "", "", name);
                p.skip_spaces();
        }

        else if (name == "comment") {
                eat_whitespace(p, os, parent_context, false);
                parent_context.check_layout(os);
                begin_inset(os, "Note Comment\n");
                os << "status open\n";
                parse_text_in_inset(p, os, FLAG_END, outer, parent_context);
                end_inset(os);
                p.skip_spaces();
                skip_braces(p); // eat {} that might by set by LyX behind comments
                preamble.registerAutomaticallyLoadedPackage("verbatim");
        }

        else if (name == "verbatim") {
                eat_whitespace(p, os, parent_context, false);
                os << "\n\\begin_layout Verbatim\n";
                string const s = p.verbatimEnvironment("verbatim");
                string::const_iterator it2 = s.begin();
                for (string::const_iterator it = s.begin(), et = s.end(); it != et; ++it) {
                        if (*it == '\n') {
                                it2 = it + 1;
                                // avoid adding an empty paragraph at the end
                                // if there are 2 consecutive spaces at the end ignore it
                                // because LyX will re-add a \n
                                if ((it + 1 != et) && (it + 2 != et || *it2 != '\n'))
                                        os << "\n\\end_layout\n\\begin_layout Verbatim\n";
                        } else
                                os << *it;
                }
                os << "\n\\end_layout\n\n";
                p.skip_spaces();
                skip_braces(p); // eat {} that might by set by LyX behind comments
                // reset to Standard layout
                os << "\n\\begin_layout Standard\n";
        }

        else if (name == "lyxgreyedout") {
                eat_whitespace(p, os, parent_context, false);
                parent_context.check_layout(os);
                begin_inset(os, "Note Greyedout\n");
                os << "status open\n";
                parse_text_in_inset(p, os, FLAG_END, outer, parent_context);
                end_inset(os);
                p.skip_spaces();
                if (!preamble.notefontcolor().empty())
                        preamble.registerAutomaticallyLoadedPackage("color");
        }

        else if (name == "framed" || name == "shaded") {
                eat_whitespace(p, os, parent_context, false);
                parse_outer_box(p, os, FLAG_END, outer, parent_context, name, "");
                p.skip_spaces();
        }

        else if (name == "lstlisting") {
                eat_whitespace(p, os, parent_context, false);
                // FIXME handle listings with parameters
                //       If this is added, don't forgot to handle the
                //       automatic color package loading
                if (p.hasOpt())
                        parse_unknown_environment(p, name, os, FLAG_END,
                                                  outer, parent_context);
                else
                        parse_listings(p, os, parent_context);
                p.skip_spaces();
        }

        else if (!parent_context.new_layout_allowed)
                parse_unknown_environment(p, name, os, FLAG_END, outer,
                                          parent_context);

        // Alignment and spacing settings
        // FIXME (bug xxxx): These settings can span multiple paragraphs and
        //                                       therefore are totally broken!
        // Note that \centering, raggedright, and raggedleft cannot be handled, as
        // they are commands not environments. They are furthermore switches that
        // can be ended by another switches, but also by commands like \footnote or
        // \parbox. So the only safe way is to leave them untouched.
        else if (name == "center" || name == "centering" ||
                 name == "flushleft" || name == "flushright" ||
                 name == "singlespace" || name == "onehalfspace" ||
                 name == "doublespace" || name == "spacing") {
                eat_whitespace(p, os, parent_context, false);
                // We must begin a new paragraph if not already done
                if (! parent_context.atParagraphStart()) {
                        parent_context.check_end_layout(os);
                        parent_context.new_paragraph(os);
                }
                if (name == "flushleft")
                        parent_context.add_extra_stuff("\\align left\n");
                else if (name == "flushright")
                        parent_context.add_extra_stuff("\\align right\n");
                else if (name == "center" || name == "centering")
                        parent_context.add_extra_stuff("\\align center\n");
                else if (name == "singlespace")
                        parent_context.add_extra_stuff("\\paragraph_spacing single\n");
                else if (name == "onehalfspace") {
                        parent_context.add_extra_stuff("\\paragraph_spacing onehalf\n");
                        preamble.registerAutomaticallyLoadedPackage("setspace");
                } else if (name == "doublespace") {
                        parent_context.add_extra_stuff("\\paragraph_spacing double\n");
                        preamble.registerAutomaticallyLoadedPackage("setspace");
                } else if (name == "spacing") {
                        parent_context.add_extra_stuff("\\paragraph_spacing other " + p.verbatim_item() + "\n");
                        preamble.registerAutomaticallyLoadedPackage("setspace");
                }
                parse_text(p, os, FLAG_END, outer, parent_context);
                // Just in case the environment is empty
                parent_context.extra_stuff.erase();
                // We must begin a new paragraph to reset the alignment
                parent_context.new_paragraph(os);
                p.skip_spaces();
        }

        // The single '=' is meant here.
        else if ((newlayout = findLayout(parent_context.textclass, name, false))) {
                eat_whitespace(p, os, parent_context, false);
                Context context(true, parent_context.textclass, newlayout,
                                parent_context.layout, parent_context.font);
                if (parent_context.deeper_paragraph) {
                        // We are beginning a nested environment after a
                        // deeper paragraph inside the outer list environment.
                        // Therefore we don't need to output a "begin deeper".
                        context.need_end_deeper = true;
                }
                parent_context.check_end_layout(os);
                if (last_env == name) {
                        // we need to output a separator since LyX would export
                        // the two environments as one otherwise (bug 5716)
                        docstring const sep = from_ascii("--Separator--");
                        TeX2LyXDocClass const & textclass(parent_context.textclass);
                        if (textclass.hasLayout(sep)) {
                                Context newcontext(parent_context);
                                newcontext.layout = &(textclass[sep]);
                                newcontext.check_layout(os);
                                newcontext.check_end_layout(os);
                        } else {
                                parent_context.check_layout(os);
                                begin_inset(os, "Note Note\n");
                                os << "status closed\n";
                                Context newcontext(true, textclass,
                                                &(textclass.defaultLayout()));
                                newcontext.check_layout(os);
                                newcontext.check_end_layout(os);
                                end_inset(os);
                                parent_context.check_end_layout(os);
                        }
                }
                switch (context.layout->latextype) {
                case  LATEX_LIST_ENVIRONMENT:
                        context.add_par_extra_stuff("\\labelwidthstring "
                                                    + p.verbatim_item() + '\n');
                        p.skip_spaces();
                        break;
                case  LATEX_BIB_ENVIRONMENT:
                        p.verbatim_item(); // swallow next arg
                        p.skip_spaces();
                        break;
                default:
                        break;
                }
                context.check_deeper(os);
                // handle known optional and required arguments
                // layouts require all optional arguments before the required ones
                // Unfortunately LyX can't handle arguments of list arguments (bug 7468):
                // It is impossible to place anything after the environment name,
                // but before the first \\item.
                if (context.layout->latextype == LATEX_ENVIRONMENT) {
                        bool need_layout = true;
                        unsigned int optargs = 0;
                        while (optargs < context.layout->optargs) {
                                eat_whitespace(p, os, context, false);
                                if (p.next_token().cat() == catEscape ||
                                    p.next_token().character() != '[')
                                        break;
                                p.get_token(); // eat '['
                                if (need_layout) {
                                        context.check_layout(os);
                                        need_layout = false;
                                }
                                begin_inset(os, "Argument\n");
                                os << "status collapsed\n\n";
                                parse_text_in_inset(p, os, FLAG_BRACK_LAST, outer, context);
                                end_inset(os);
                                eat_whitespace(p, os, context, false);
                                ++optargs;
                        }
                        unsigned int reqargs = 0;
                        while (reqargs < context.layout->reqargs) {
                                eat_whitespace(p, os, context, false);
                                if (p.next_token().cat() != catBegin)
                                        break;
                                p.get_token(); // eat '{'
                                if (need_layout) {
                                        context.check_layout(os);
                                        need_layout = false;
                                }
                                begin_inset(os, "Argument\n");
                                os << "status collapsed\n\n";
                                parse_text_in_inset(p, os, FLAG_BRACE_LAST, outer, context);
                                end_inset(os);
                                eat_whitespace(p, os, context, false);
                                ++reqargs;
                        }
                }
                parse_text(p, os, FLAG_END, outer, context);
                context.check_end_layout(os);
                if (parent_context.deeper_paragraph) {
                        // We must suppress the "end deeper" because we
                        // suppressed the "begin deeper" above.
                        context.need_end_deeper = false;
                }
                context.check_end_deeper(os);
                parent_context.new_paragraph(os);
                p.skip_spaces();
                if (!title_layout_found)
                        title_layout_found = newlayout->intitle;
                set<string> const & req = newlayout->requires();
                for (set<string>::const_iterator it = req.begin(); it != req.end(); it++)
                        preamble.registerAutomaticallyLoadedPackage(*it);
        }

        // The single '=' is meant here.
        else if ((newinsetlayout = findInsetLayout(parent_context.textclass, name, false))) {
                eat_whitespace(p, os, parent_context, false);
                parent_context.check_layout(os);
                begin_inset(os, "Flex ");
                os << to_utf8(newinsetlayout->name()) << '\n'
                   << "status collapsed\n";
                parse_text_in_inset(p, os, FLAG_END, false, parent_context, newinsetlayout);
                end_inset(os);
        }

        else if (name == "appendix") {
                // This is no good latex style, but it works and is used in some documents...
                eat_whitespace(p, os, parent_context, false);
                parent_context.check_end_layout(os);
                Context context(true, parent_context.textclass, parent_context.layout,
                                parent_context.layout, parent_context.font);
                context.check_layout(os);
                os << "\\start_of_appendix\n";
                parse_text(p, os, FLAG_END, outer, context);
                context.check_end_layout(os);
                p.skip_spaces();
        }

        else if (known_environments.find(name) != known_environments.end()) {
                vector<ArgumentType> arguments = known_environments[name];
                // The last "argument" denotes wether we may translate the
                // environment contents to LyX
                // The default required if no argument is given makes us
                // compatible with the reLyXre environment.
                ArgumentType contents = arguments.empty() ?
                        required :
                        arguments.back();
                if (!arguments.empty())
                        arguments.pop_back();
                // See comment in parse_unknown_environment()
                bool const specialfont =
                        (parent_context.font != parent_context.normalfont);
                bool const new_layout_allowed =
                        parent_context.new_layout_allowed;
                if (specialfont)
                        parent_context.new_layout_allowed = false;
                parse_arguments("\\begin{" + name + "}", arguments, p, os,
                                outer, parent_context);
                if (contents == verbatim)
                        handle_ert(os, p.verbatimEnvironment(name),
                                   parent_context);
                else
                        parse_text_snippet(p, os, FLAG_END, outer,
                                           parent_context);
                handle_ert(os, "\\end{" + name + "}", parent_context);
                if (specialfont)
                        parent_context.new_layout_allowed = new_layout_allowed;
        }

        else
                parse_unknown_environment(p, name, os, FLAG_END, outer,
                                          parent_context);

        last_env = name;
        active_environments.pop_back();
}


/// parses a comment and outputs it to \p os.
void parse_comment(Parser & p, ostream & os, Token const & t, Context & context)
{
        LASSERT(t.cat() == catComment, return);
        if (!t.cs().empty()) {
                context.check_layout(os);
                handle_comment(os, '%' + t.cs(), context);
                if (p.next_token().cat() == catNewline) {
                        // A newline after a comment line starts a new
                        // paragraph
                        if (context.new_layout_allowed) {
                                if(!context.atParagraphStart())
                                        // Only start a new paragraph if not already
                                        // done (we might get called recursively)
                                        context.new_paragraph(os);
                        } else
                                handle_ert(os, "\n", context);
                        eat_whitespace(p, os, context, true);
                }
        } else {
                // "%\n" combination
                p.skip_spaces();
        }
}


/*!
 * Reads spaces and comments until the first non-space, non-comment token.
 * New paragraphs (double newlines or \\par) are handled like simple spaces
 * if \p eatParagraph is true.
 * Spaces are skipped, but comments are written to \p os.
 */
void eat_whitespace(Parser & p, ostream & os, Context & context,
                    bool eatParagraph)
{
        while (p.good()) {
                Token const & t = p.get_token();
                if (t.cat() == catComment)
                        parse_comment(p, os, t, context);
                else if ((! eatParagraph && p.isParagraph()) ||
                         (t.cat() != catSpace && t.cat() != catNewline)) {
                        p.putback();
                        return;
                }
        }
}


/*!
 * Set a font attribute, parse text and reset the font attribute.
 * \param attribute Attribute name (e.g. \\family, \\shape etc.)
 * \param currentvalue Current value of the attribute. Is set to the new
 * value during parsing.
 * \param newvalue New value of the attribute
 */
void parse_text_attributes(Parser & p, ostream & os, unsigned flags, bool outer,
                           Context & context, string const & attribute,
                           string & currentvalue, string const & newvalue)
{
        context.check_layout(os);
        string const oldvalue = currentvalue;
        currentvalue = newvalue;
        os << '\n' << attribute << ' ' << newvalue << "\n";
        parse_text_snippet(p, os, flags, outer, context);
        context.check_layout(os);
        os << '\n' << attribute << ' ' << oldvalue << "\n";
        currentvalue = oldvalue;
}


/// get the arguments of a natbib or jurabib citation command
void get_cite_arguments(Parser & p, bool natbibOrder,
        string & before, string & after)
{
        // We need to distinguish "" and "[]", so we can't use p.getOpt().

        // text before the citation
        before.clear();
        // text after the citation
        after = p.getFullOpt();

        if (!after.empty()) {
                before = p.getFullOpt();
                if (natbibOrder && !before.empty())
                        swap(before, after);
        }
}


/// Convert filenames with TeX macros and/or quotes to something LyX
/// can understand
string const normalize_filename(string const & name)
{
        Parser p(trim(name, "\""));
        ostringstream os;
        while (p.good()) {
                Token const & t = p.get_token();
                if (t.cat() != catEscape)
                        os << t.asInput();
                else if (t.cs() == "lyxdot") {
                        // This is used by LyX for simple dots in relative
                        // names
                        os << '.';
                        p.skip_spaces();
                } else if (t.cs() == "space") {
                        os << ' ';
                        p.skip_spaces();
                } else
                        os << t.asInput();
        }
        return os.str();
}


/// Convert \p name from TeX convention (relative to master file) to LyX
/// convention (relative to .lyx file) if it is relative
void fix_relative_filename(string & name)
{
        if (FileName::isAbsolute(name))
                return;

        name = to_utf8(makeRelPath(from_utf8(makeAbsPath(name, getMasterFilePath()).absFileName()),
                                   from_utf8(getParentFilePath())));
}


/// Parse a NoWeb Scrap section. The initial "<<" is already parsed.
void parse_noweb(Parser & p, ostream & os, Context & context)
{
        // assemble the rest of the keyword
        string name("<<");
        bool scrap = false;
        while (p.good()) {
                Token const & t = p.get_token();
                if (t.asInput() == ">" && p.next_token().asInput() == ">") {
                        name += ">>";
                        p.get_token();
                        scrap = (p.good() && p.next_token().asInput() == "=");
                        if (scrap)
                                name += p.get_token().asInput();
                        break;
                }
                name += t.asInput();
        }

        if (!scrap || !context.new_layout_allowed ||
            !context.textclass.hasLayout(from_ascii("Scrap"))) {
                cerr << "Warning: Could not interpret '" << name
                     << "'. Ignoring it." << endl;
                return;
        }

        // We use new_paragraph instead of check_end_layout because the stuff
        // following the noweb chunk needs to start with a \begin_layout.
        // This may create a new paragraph even if there was none in the
        // noweb file, but the alternative is an invalid LyX file. Since
        // noweb code chunks are implemented with a layout style in LyX they
        // always must be in an own paragraph.
        context.new_paragraph(os);
        Context newcontext(true, context.textclass,
                &context.textclass[from_ascii("Scrap")]);
        newcontext.check_layout(os);
        os << name;
        while (p.good()) {
                Token const & t = p.get_token();
                // We abuse the parser a bit, because this is no TeX syntax
                // at all.
                if (t.cat() == catEscape)
                        os << subst(t.asInput(), "\\", "\n\\backslash\n");
                else {
                        ostringstream oss;
                        Context tmp(false, context.textclass,
                                    &context.textclass[from_ascii("Scrap")]);
                        tmp.need_end_layout = true;
                        tmp.check_layout(oss);
                        os << subst(t.asInput(), "\n", oss.str());
                }
                // The scrap chunk is ended by an @ at the beginning of a line.
                // After the @ the line may contain a comment and/or
                // whitespace, but nothing else.
                if (t.asInput() == "@" && p.prev_token().cat() == catNewline &&
                    (p.next_token().cat() == catSpace ||
                     p.next_token().cat() == catNewline ||
                     p.next_token().cat() == catComment)) {
                        while (p.good() && p.next_token().cat() == catSpace)
                                os << p.get_token().asInput();
                        if (p.next_token().cat() == catComment)
                                // The comment includes a final '\n'
                                os << p.get_token().asInput();
                        else {
                                if (p.next_token().cat() == catNewline)
                                        p.get_token();
                                os << '\n';
                        }
                        break;
                }
        }
        newcontext.check_end_layout(os);
}


/// detects \\def, \\long\\def and \\global\\long\\def with ws and comments
bool is_macro(Parser & p)
{
        Token first = p.curr_token();
        if (first.cat() != catEscape || !p.good())
                return false;
        if (first.cs() == "def")
                return true;
        if (first.cs() != "global" && first.cs() != "long")
                return false;
        Token second = p.get_token();
        int pos = 1;
        while (p.good() && !p.isParagraph() && (second.cat() == catSpace ||
               second.cat() == catNewline || second.cat() == catComment)) {
                second = p.get_token();
                pos++;
        }
        bool secondvalid = second.cat() == catEscape;
        Token third;
        bool thirdvalid = false;
        if (p.good() && first.cs() == "global" && secondvalid &&
            second.cs() == "long") {
                third = p.get_token();
                pos++;
                while (p.good() && !p.isParagraph() &&
                       (third.cat() == catSpace ||
                        third.cat() == catNewline ||
                        third.cat() == catComment)) {
                        third = p.get_token();
                        pos++;
                }
                thirdvalid = third.cat() == catEscape;
        }
        for (int i = 0; i < pos; ++i)
                p.putback();
        if (!secondvalid)
                return false;
        if (!thirdvalid)
                return (first.cs() == "global" || first.cs() == "long") &&
                       second.cs() == "def";
        return first.cs() == "global" && second.cs() == "long" &&
               third.cs() == "def";
}


/// Parse a macro definition (assumes that is_macro() returned true)
void parse_macro(Parser & p, ostream & os, Context & context)
{
        context.check_layout(os);
        Token first = p.curr_token();
        Token second;
        Token third;
        string command = first.asInput();
        if (first.cs() != "def") {
                p.get_token();
                eat_whitespace(p, os, context, false);
                second = p.curr_token();
                command += second.asInput();
                if (second.cs() != "def") {
                        p.get_token();
                        eat_whitespace(p, os, context, false);
                        third = p.curr_token();
                        command += third.asInput();
                }
        }
        eat_whitespace(p, os, context, false);
        string const name = p.get_token().cs();
        eat_whitespace(p, os, context, false);

        // parameter text
        bool simple = true;
        string paramtext;
        int arity = 0;
        while (p.next_token().cat() != catBegin) {
                if (p.next_token().cat() == catParameter) {
                        // # found
                        p.get_token();
                        paramtext += "#";

                        // followed by number?
                        if (p.next_token().cat() == catOther) {
                                char c = p.getChar();
                                paramtext += c;
                                // number = current arity + 1?
                                if (c == arity + '0' + 1)
                                        ++arity;
                                else
                                        simple = false;
                        } else
                                paramtext += p.get_token().cs();
                } else {
                        paramtext += p.get_token().cs();
                        simple = false;
                }
        }

        // only output simple (i.e. compatible) macro as FormulaMacros
        string ert = '\\' + name + ' ' + paramtext + '{' + p.verbatim_item() + '}';
        if (simple) {
                context.check_layout(os);
                begin_inset(os, "FormulaMacro");
                os << "\n\\def" << ert;
                end_inset(os);
        } else
                handle_ert(os, command + ert, context);
}


void registerExternalTemplatePackages(string const & name)
{
        external::TemplateManager const & etm = external::TemplateManager::get();
        external::Template const * const et = etm.getTemplateByName(name);
        if (!et)
                return;
        external::Template::Formats::const_iterator cit = et->formats.end();
        if (pdflatex)
                cit = et->formats.find("PDFLaTeX");
        if (cit == et->formats.end())
                // If the template has not specified a PDFLaTeX output,
                // we try the LaTeX format.
                cit = et->formats.find("LaTeX");
        if (cit == et->formats.end())
                return;
        vector<string>::const_iterator qit = cit->second.requirements.begin();
        vector<string>::const_iterator qend = cit->second.requirements.end();
        for (; qit != qend; ++qit)
                preamble.registerAutomaticallyLoadedPackage(*qit);
}

} // anonymous namespace


void parse_text(Parser & p, ostream & os, unsigned flags, bool outer,
                Context & context)
{
        Layout const * newlayout = 0;
        InsetLayout const * newinsetlayout = 0;
        char const * const * where = 0;
        // Store the latest bibliographystyle and nocite{*} option
        // (needed for bibtex inset)
        string btprint;
        string bibliographystyle = "default";
        bool const use_natbib = preamble.isPackageUsed("natbib");
        bool const use_jurabib = preamble.isPackageUsed("jurabib");
        string last_env;
        bool title_layout_found = false;
        while (p.good()) {
                Token const & t = p.get_token();

#ifdef FILEDEBUG
                debugToken(cerr, t, flags);
#endif

                if (flags & FLAG_ITEM) {
                        if (t.cat() == catSpace)
                                continue;

                        flags &= ~FLAG_ITEM;
                        if (t.cat() == catBegin) {
                                // skip the brace and collect everything to the next matching
                                // closing brace
                                flags |= FLAG_BRACE_LAST;
                                continue;
                        }

                        // handle only this single token, leave the loop if done
                        flags |= FLAG_LEAVE;
                }

                if (t.cat() != catEscape && t.character() == ']' &&
                    (flags & FLAG_BRACK_LAST))
                        return;
                if (t.cat() == catEnd && (flags & FLAG_BRACE_LAST))
                        return;

                // If there is anything between \end{env} and \begin{env} we
                // don't need to output a separator.
                if (t.cat() != catSpace && t.cat() != catNewline &&
                    t.asInput() != "\\begin")
                        last_env = "";

                //
                // cat codes
                //
                if (t.cat() == catMath) {
                        // we are inside some text mode thingy, so opening new math is allowed
                        context.check_layout(os);
                        begin_inset(os, "Formula ");
                        Token const & n = p.get_token();
                        bool const display(n.cat() == catMath && outer);
                        if (display) {
                                // TeX's $$...$$ syntax for displayed math
                                os << "\\[";
                                parse_math(p, os, FLAG_SIMPLE, MATH_MODE);
                                os << "\\]";
                                p.get_token(); // skip the second '$' token
                        } else {
                                // simple $...$  stuff
                                p.putback();
                                os << '$';
                                parse_math(p, os, FLAG_SIMPLE, MATH_MODE);
                                os << '$';
                        }
                        end_inset(os);
                        if (display) {
                                // Prevent the conversion of a line break to a
                                // space (bug 7668). This does not change the
                                // output, but looks ugly in LyX.
                                eat_whitespace(p, os, context, false);
                        }
                }

                else if (t.cat() == catSuper || t.cat() == catSub)
                        cerr << "catcode " << t << " illegal in text mode\n";

                // Basic support for english quotes. This should be
                // extended to other quotes, but is not so easy (a
                // left english quote is the same as a right german
                // quote...)
                else if (t.asInput() == "`" && p.next_token().asInput() == "`") {
                        context.check_layout(os);
                        begin_inset(os, "Quotes ");
                        os << "eld";
                        end_inset(os);
                        p.get_token();
                        skip_braces(p);
                }
                else if (t.asInput() == "'" && p.next_token().asInput() == "'") {
                        context.check_layout(os);
                        begin_inset(os, "Quotes ");
                        os << "erd";
                        end_inset(os);
                        p.get_token();
                        skip_braces(p);
                }

                else if (t.asInput() == ">" && p.next_token().asInput() == ">") {
                        context.check_layout(os);
                        begin_inset(os, "Quotes ");
                        os << "ald";
                        end_inset(os);
                        p.get_token();
                        skip_braces(p);
                }

                else if (t.asInput() == "<" && p.next_token().asInput() == "<") {
                        context.check_layout(os);
                        begin_inset(os, "Quotes ");
                        os << "ard";
                        end_inset(os);
                        p.get_token();
                        skip_braces(p);
                }

                else if (t.asInput() == "<"
                         && p.next_token().asInput() == "<" && noweb_mode) {
                        p.get_token();
                        parse_noweb(p, os, context);
                }

                else if (t.cat() == catSpace || (t.cat() == catNewline && ! p.isParagraph()))
                        check_space(p, os, context);

                else if (t.character() == '[' && noweb_mode &&
                         p.next_token().character() == '[') {
                        // These can contain underscores
                        p.putback();
                        string const s = p.getFullOpt() + ']';
                        if (p.next_token().character() == ']')
                                p.get_token();
                        else
                                cerr << "Warning: Inserting missing ']' in '"
                                     << s << "'." << endl;
                        handle_ert(os, s, context);
                }

                else if (t.cat() == catLetter) {
                        context.check_layout(os);
                        // Workaround for bug 4752.
                        // FIXME: This whole code block needs to be removed
                        //        when the bug is fixed and tex2lyx produces
                        //        the updated file format.
                        // The replacement algorithm in LyX is so stupid that
                        // it even translates a phrase if it is part of a word.
                        bool handled = false;
                        for (int const * l = known_phrase_lengths; *l; ++l) {
                                string phrase = t.cs();
                                for (int i = 1; i < *l && p.next_token().isAlnumASCII(); ++i)
                                        phrase += p.get_token().cs();
                                if (is_known(phrase, known_coded_phrases)) {
                                        handle_ert(os, phrase, context);
                                        handled = true;
                                        break;
                                } else {
                                        for (size_t i = 1; i < phrase.length(); ++i)
                                                p.putback();
                                }
                        }
                        if (!handled)
                                os << t.cs();
                }

                else if (t.cat() == catOther ||
                               t.cat() == catAlign ||
                               t.cat() == catParameter) {
                        // This translates "&" to "\\&" which may be wrong...
                        context.check_layout(os);
                        os << t.cs();
                }

                else if (p.isParagraph()) {
                        if (context.new_layout_allowed)
                                context.new_paragraph(os);
                        else
                                handle_ert(os, "\\par ", context);
                        eat_whitespace(p, os, context, true);
                }

                else if (t.cat() == catActive) {
                        context.check_layout(os);
                        if (t.character() == '~') {
                                if (context.layout->free_spacing)
                                        os << ' ';
                                else {
                                        begin_inset(os, "space ~\n");
                                        end_inset(os);
                                }
                        } else
                                os << t.cs();
                }

                else if (t.cat() == catBegin) {
                        Token const next = p.next_token();
                        Token const end = p.next_next_token();
                        if (next.cat() == catEnd) {
                        // {}
                        Token const prev = p.prev_token();
                        p.get_token();
                        if (p.next_token().character() == '`' ||
                            (prev.character() == '-' &&
                             p.next_token().character() == '-'))
                                ; // ignore it in {}`` or -{}-
                        else
                                handle_ert(os, "{}", context);
                        } else if (next.cat() == catEscape &&
                                   is_known(next.cs(), known_quotes) &&
                                   end.cat() == catEnd) {
                                // Something like {\textquoteright} (e.g.
                                // from writer2latex). LyX writes
                                // \textquoteright{}, so we may skip the
                                // braces here for better readability.
                                parse_text_snippet(p, os, FLAG_BRACE_LAST,
                                                   outer, context);
                        } else {
                        context.check_layout(os);
                        // special handling of font attribute changes
                        Token const prev = p.prev_token();
                        TeXFont const oldFont = context.font;
                        if (next.character() == '[' ||
                            next.character() == ']' ||
                            next.character() == '*') {
                                p.get_token();
                                if (p.next_token().cat() == catEnd) {
                                        os << next.cs();
                                        p.get_token();
                                } else {
                                        p.putback();
                                        handle_ert(os, "{", context);
                                        parse_text_snippet(p, os,
                                                        FLAG_BRACE_LAST,
                                                        outer, context);
                                        handle_ert(os, "}", context);
                                }
                        } else if (! context.new_layout_allowed) {
                                handle_ert(os, "{", context);
                                parse_text_snippet(p, os, FLAG_BRACE_LAST,
                                                   outer, context);
                                handle_ert(os, "}", context);
                        } else if (is_known(next.cs(), known_sizes)) {
                                // next will change the size, so we must
                                // reset it here
                                parse_text_snippet(p, os, FLAG_BRACE_LAST,
                                                   outer, context);
                                if (!context.atParagraphStart())
                                        os << "\n\\size "
                                           << context.font.size << "\n";
                        } else if (is_known(next.cs(), known_font_families)) {
                                // next will change the font family, so we
                                // must reset it here
                                parse_text_snippet(p, os, FLAG_BRACE_LAST,
                                                   outer, context);
                                if (!context.atParagraphStart())
                                        os << "\n\\family "
                                           << context.font.family << "\n";
                        } else if (is_known(next.cs(), known_font_series)) {
                                // next will change the font series, so we
                                // must reset it here
                                parse_text_snippet(p, os, FLAG_BRACE_LAST,
                                                   outer, context);
                                if (!context.atParagraphStart())
                                        os << "\n\\series "
                                           << context.font.series << "\n";
                        } else if (is_known(next.cs(), known_font_shapes)) {
                                // next will change the font shape, so we
                                // must reset it here
                                parse_text_snippet(p, os, FLAG_BRACE_LAST,
                                                   outer, context);
                                if (!context.atParagraphStart())
                                        os << "\n\\shape "
                                           << context.font.shape << "\n";
                        } else if (is_known(next.cs(), known_old_font_families) ||
                                   is_known(next.cs(), known_old_font_series) ||
                                   is_known(next.cs(), known_old_font_shapes)) {
                                // next will change the font family, series
                                // and shape, so we must reset it here
                                parse_text_snippet(p, os, FLAG_BRACE_LAST,
                                                   outer, context);
                                if (!context.atParagraphStart())
                                        os <<  "\n\\family "
                                           << context.font.family
                                           << "\n\\series "
                                           << context.font.series
                                           << "\n\\shape "
                                           << context.font.shape << "\n";
                        } else {
                                handle_ert(os, "{", context);
                                parse_text_snippet(p, os, FLAG_BRACE_LAST,
                                                   outer, context);
                                handle_ert(os, "}", context);
                                }
                        }
                }

                else if (t.cat() == catEnd) {
                        if (flags & FLAG_BRACE_LAST) {
                                return;
                        }
                        cerr << "stray '}' in text\n";
                        handle_ert(os, "}", context);
                }

                else if (t.cat() == catComment)
                        parse_comment(p, os, t, context);

                //
                // control sequences
                //

                else if (t.cs() == "(") {
                        context.check_layout(os);
                        begin_inset(os, "Formula");
                        os << " \\(";
                        parse_math(p, os, FLAG_SIMPLE2, MATH_MODE);
                        os << "\\)";
                        end_inset(os);
                }

                else if (t.cs() == "[") {
                        context.check_layout(os);
                        begin_inset(os, "Formula");
                        os << " \\[";
                        parse_math(p, os, FLAG_EQUATION, MATH_MODE);
                        os << "\\]";
                        end_inset(os);
                        // Prevent the conversion of a line break to a space
                        // (bug 7668). This does not change the output, but
                        // looks ugly in LyX.
                        eat_whitespace(p, os, context, false);
                }

                else if (t.cs() == "begin")
                        parse_environment(p, os, outer, last_env,
                                          title_layout_found, context);

                else if (t.cs() == "end") {
                        if (flags & FLAG_END) {
                                // eat environment name
                                string const name = p.getArg('{', '}');
                                if (name != active_environment())
                                        cerr << "\\end{" + name + "} does not match \\begin{"
                                                + active_environment() + "}\n";
                                return;
                        }
                        p.error("found 'end' unexpectedly");
                }

                else if (t.cs() == "item") {
                        string s;
                        bool const optarg = p.hasOpt();
                        if (optarg) {
                                // FIXME: This swallows comments, but we cannot use
                                //        eat_whitespace() since we must not output
                                //        anything before the item.
                                p.skip_spaces(true);
                                s = p.verbatimOption();
                        } else
                                p.skip_spaces(false);
                        context.set_item();
                        context.check_layout(os);
                        if (context.has_item) {
                                // An item in an unknown list-like environment
                                // FIXME: Do this in check_layout()!
                                context.has_item = false;
                                if (optarg)
                                        handle_ert(os, "\\item", context);
                                else
                                        handle_ert(os, "\\item ", context);
                        }
                        if (optarg) {
                                if (context.layout->labeltype != LABEL_MANUAL) {
                                        // LyX does not support \item[\mybullet]
                                        // in itemize environments
                                        Parser p2(s + ']');
                                        os << parse_text_snippet(p2,
                                                FLAG_BRACK_LAST, outer, context);
                                } else if (!s.empty()) {
                                        // LyX adds braces around the argument,
                                        // so we need to remove them here.
                                        if (s.size() > 2 && s[0] == '{' &&
                                            s[s.size()-1] == '}')
                                                s = s.substr(1, s.size()-2);
                                        // If the argument contains a space we
                                        // must put it into ERT: Otherwise LyX
                                        // would misinterpret the space as
                                        // item delimiter (bug 7663)
                                        if (contains(s, ' ')) {
                                                handle_ert(os, s, context);
                                        } else {
                                                Parser p2(s + ']');
                                                os << parse_text_snippet(p2,
                                                        FLAG_BRACK_LAST,
                                                        outer, context);
                                        }
                                        // The space is needed to separate the
                                        // item from the rest of the sentence.
                                        os << ' ';
                                        eat_whitespace(p, os, context, false);
                                }
                        }
                }

                else if (t.cs() == "bibitem") {
                        context.set_item();
                        context.check_layout(os);
                        eat_whitespace(p, os, context, false);
                        string label = convert_command_inset_arg(p.verbatimOption());
                        string key = convert_command_inset_arg(p.verbatim_item());
                        if (contains(label, '\\') || contains(key, '\\')) {
                                // LyX can't handle LaTeX commands in labels or keys
                                handle_ert(os, t.asInput() + '[' + label +
                                               "]{" + p.verbatim_item() + '}',
                                           context);
                        } else {
                                begin_command_inset(os, "bibitem", "bibitem");
                                os << "label \"" << label << "\"\n"
                                      "key \"" << key << "\"\n";
                                end_inset(os);
                        }
                }

                else if (is_macro(p)) {
                        // catch the case of \def\inputGnumericTable
                        bool macro = true;
                        if (t.cs() == "def") {
                                Token second = p.next_token();
                                if (second.cs() == "inputGnumericTable") {
                                        p.pushPosition();
                                        p.get_token();
                                        skip_braces(p);
                                        Token third = p.get_token();
                                        p.popPosition();
                                        if (third.cs() == "input") {
                                                p.get_token();
                                                skip_braces(p);
                                                p.get_token();
                                                string name = normalize_filename(p.verbatim_item());
                                                string const path = getMasterFilePath();
                                                // We want to preserve relative / absolute filenames,
                                                // therefore path is only used for testing
                                                // The file extension is in every case ".tex".
                                                // So we need to remove this extension and check for
                                                // the original one.
                                                name = removeExtension(name);
                                                if (!makeAbsPath(name, path).exists()) {
                                                        char const * const Gnumeric_formats[] = {"gnumeric",
                                                                "ods", "xls", 0};
                                                        string const Gnumeric_name =
                                                                find_file(name, path, Gnumeric_formats);
                                                        if (!Gnumeric_name.empty())
                                                                name = Gnumeric_name;
                                                }
                                                if (makeAbsPath(name, path).exists())
                                                        fix_relative_filename(name);
                                                else
                                                        cerr << "Warning: Could not find file '"
                                                             << name << "'." << endl;
                                                context.check_layout(os);
                                                begin_inset(os, "External\n\ttemplate ");
                                                os << "GnumericSpreadsheet\n\tfilename "
                                                   << name << "\n";
                                                end_inset(os);
                                                context.check_layout(os);
                                                macro = false;
                                                // register the packages that are automatically reloaded
                                                // by the Gnumeric template
                                                registerExternalTemplatePackages("GnumericSpreadsheet");
                                        }
                                }
                        }
                        if (macro)
                                parse_macro(p, os, context);
                }

                else if (t.cs() == "noindent") {
                        p.skip_spaces();
                        context.add_par_extra_stuff("\\noindent\n");
                }

                else if (t.cs() == "appendix") {
                        context.add_par_extra_stuff("\\start_of_appendix\n");
                        // We need to start a new paragraph. Otherwise the
                        // appendix in 'bla\appendix\chapter{' would start
                        // too late.
                        context.new_paragraph(os);
                        // We need to make sure that the paragraph is
                        // generated even if it is empty. Otherwise the
                        // appendix in '\par\appendix\par\chapter{' would
                        // start too late.
                        context.check_layout(os);
                        // FIXME: This is a hack to prevent paragraph
                        // deletion if it is empty. Handle this better!
                        handle_comment(os,
                                "%dummy comment inserted by tex2lyx to "
                                "ensure that this paragraph is not empty",
                                context);
                        // Both measures above may generate an additional
                        // empty paragraph, but that does not hurt, because
                        // whitespace does not matter here.
                        eat_whitespace(p, os, context, true);
                }

                // Must catch empty dates before findLayout is called below
                else if (t.cs() == "date") {
                        string const date = p.verbatim_item();
                        if (date.empty())
                                preamble.suppressDate(true);
                        else {
                                preamble.suppressDate(false);
                                if (context.new_layout_allowed &&
                                    (newlayout = findLayout(context.textclass,
                                                            t.cs(), true))) {
                                        // write the layout
                                        output_command_layout(os, p, outer,
                                                        context, newlayout);
                                        p.skip_spaces();
                                        if (!title_layout_found)
                                                title_layout_found = newlayout->intitle;
                                        set<string> const & req = newlayout->requires();
                                        for (set<string>::const_iterator it = req.begin();
                                             it != req.end(); it++)
                                                preamble.registerAutomaticallyLoadedPackage(*it);
                                } else
                                        handle_ert(os, "\\date{" + date + '}',
                                                        context);
                        }
                }

                // Starred section headings
                // Must attempt to parse "Section*" before "Section".
                else if ((p.next_token().asInput() == "*") &&
                         context.new_layout_allowed &&
                         (newlayout = findLayout(context.textclass, t.cs() + '*', true))) {
                        // write the layout
                        p.get_token();
                        output_command_layout(os, p, outer, context, newlayout);
                        p.skip_spaces();
                        if (!title_layout_found)
                                title_layout_found = newlayout->intitle;
                        set<string> const & req = newlayout->requires();
                        for (set<string>::const_iterator it = req.begin(); it != req.end(); it++)
                                preamble.registerAutomaticallyLoadedPackage(*it);
                }

                // Section headings and the like
                else if (context.new_layout_allowed &&
                         (newlayout = findLayout(context.textclass, t.cs(), true))) {
                        // write the layout
                        output_command_layout(os, p, outer, context, newlayout);
                        p.skip_spaces();
                        if (!title_layout_found)
                                title_layout_found = newlayout->intitle;
                        set<string> const & req = newlayout->requires();
                        for (set<string>::const_iterator it = req.begin(); it != req.end(); it++)
                                preamble.registerAutomaticallyLoadedPackage(*it);
                }

                else if (t.cs() == "caption") {
                        p.skip_spaces();
                        context.check_layout(os);
                        p.skip_spaces();
                        begin_inset(os, "Caption\n");
                        Context newcontext(true, context.textclass);
                        newcontext.font = context.font;
                        newcontext.check_layout(os);
                        if (p.next_token().cat() != catEscape &&
                            p.next_token().character() == '[') {
                                p.get_token(); // eat '['
                                begin_inset(os, "Argument\n");
                                os << "status collapsed\n";
                                parse_text_in_inset(p, os, FLAG_BRACK_LAST, outer, context);
                                end_inset(os);
                                eat_whitespace(p, os, context, false);
                        }
                        parse_text(p, os, FLAG_ITEM, outer, context);
                        context.check_end_layout(os);
                        // We don't need really a new paragraph, but
                        // we must make sure that the next item gets a \begin_layout.
                        context.new_paragraph(os);
                        end_inset(os);
                        p.skip_spaces();
                        newcontext.check_end_layout(os);
                }

                else if (t.cs() == "subfloat") {
                        // the syntax is \subfloat[caption]{content}
                        // if it is a table of figure depends on the surrounding float
                        bool has_caption = false;
                        p.skip_spaces();
                        // do nothing if there is no outer float
                        if (!float_type.empty()) {
                                context.check_layout(os);
                                p.skip_spaces();
                                begin_inset(os, "Float " + float_type + "\n");
                                os << "wide false"
                                   << "\nsideways false"
                                   << "\nstatus collapsed\n\n";
                                // test for caption
                                string caption;
                                if (p.next_token().cat() != catEscape &&
                                                p.next_token().character() == '[') {
                                                        p.get_token(); // eat '['
                                                        caption = parse_text_snippet(p, FLAG_BRACK_LAST, outer, context);
                                                        has_caption = true;
                                }
                                // the content
                                parse_text_in_inset(p, os, FLAG_ITEM, outer, context);
                                // the caption comes always as the last
                                if (has_caption) {
                                        // we must make sure that the caption gets a \begin_layout
                                        os << "\n\\begin_layout Plain Layout";
                                        p.skip_spaces();
                                        begin_inset(os, "Caption\n");
                                        Context newcontext(true, context.textclass);
                                        newcontext.font = context.font;
                                        newcontext.check_layout(os);
                                        os << caption << "\n";
                                        newcontext.check_end_layout(os);
                                        // We don't need really a new paragraph, but
                                        // we must make sure that the next item gets a \begin_layout.
                                        //newcontext.new_paragraph(os);
                                        end_inset(os);
                                        p.skip_spaces();
                                }
                                // We don't need really a new paragraph, but
                                // we must make sure that the next item gets a \begin_layout.
                                if (has_caption)
                                        context.new_paragraph(os);
                                end_inset(os);
                                p.skip_spaces();
                                context.check_end_layout(os);
                                // close the layout we opened
                                if (has_caption)
                                        os << "\n\\end_layout\n";
                        } else {
                                // if the float type is not supported or there is no surrounding float
                                // output it as ERT
                                if (p.hasOpt()) {
                                        string opt_arg = convert_command_inset_arg(p.getArg('[', ']'));
                                        handle_ert(os, t.asInput() + '[' + opt_arg +
                                               "]{" + p.verbatim_item() + '}', context);
                                } else
                                        handle_ert(os, t.asInput() + "{" + p.verbatim_item() + '}', context);
                        }
                }

                else if (t.cs() == "includegraphics") {
                        bool const clip = p.next_token().asInput() == "*";
                        if (clip)
                                p.get_token();
                        string const arg = p.getArg('[', ']');
                        map<string, string> opts;
                        vector<string> keys;
                        split_map(arg, opts, keys);
                        if (clip)
                                opts["clip"] = string();
                        string name = normalize_filename(p.verbatim_item());

                        string const path = getMasterFilePath();
                        // We want to preserve relative / absolute filenames,
                        // therefore path is only used for testing
                        if (!makeAbsPath(name, path).exists()) {
                                // The file extension is probably missing.
                                // Now try to find it out.
                                string const dvips_name =
                                        find_file(name, path,
                                                  known_dvips_graphics_formats);
                                string const pdftex_name =
                                        find_file(name, path,
                                                  known_pdftex_graphics_formats);
                                if (!dvips_name.empty()) {
                                        if (!pdftex_name.empty()) {
                                                cerr << "This file contains the "
                                                        "latex snippet\n"
                                                        "\"\\includegraphics{"
                                                     << name << "}\".\n"
                                                        "However, files\n\""
                                                     << dvips_name << "\" and\n\""
                                                     << pdftex_name << "\"\n"
                                                        "both exist, so I had to make a "
                                                        "choice and took the first one.\n"
                                                        "Please move the unwanted one "
                                                        "someplace else and try again\n"
                                                        "if my choice was wrong."
                                                     << endl;
                                        }
                                        name = dvips_name;
                                } else if (!pdftex_name.empty()) {
                                        name = pdftex_name;
                                        pdflatex = true;
                                }
                        }

                        if (makeAbsPath(name, path).exists())
                                fix_relative_filename(name);
                        else
                                cerr << "Warning: Could not find graphics file '"
                                     << name << "'." << endl;

                        context.check_layout(os);
                        begin_inset(os, "Graphics ");
                        os << "\n\tfilename " << name << '\n';
                        if (opts.find("width") != opts.end())
                                os << "\twidth "
                                   << translate_len(opts["width"]) << '\n';
                        if (opts.find("height") != opts.end())
                                os << "\theight "
                                   << translate_len(opts["height"]) << '\n';
                        if (opts.find("scale") != opts.end()) {
                                istringstream iss(opts["scale"]);
                                double val;
                                iss >> val;
                                val = val*100;
                                os << "\tscale " << val << '\n';
                        }
                        if (opts.find("angle") != opts.end()) {
                                os << "\trotateAngle "
                                   << opts["angle"] << '\n';
                                vector<string>::const_iterator a =
                                        find(keys.begin(), keys.end(), "angle");
                                vector<string>::const_iterator s =
                                        find(keys.begin(), keys.end(), "width");
                                if (s == keys.end())
                                        s = find(keys.begin(), keys.end(), "height");
                                if (s == keys.end())
                                        s = find(keys.begin(), keys.end(), "scale");
                                if (s != keys.end() && distance(s, a) > 0)
                                        os << "\tscaleBeforeRotation\n";
                        }
                        if (opts.find("origin") != opts.end()) {
                                ostringstream ss;
                                string const opt = opts["origin"];
                                if (opt.find('l') != string::npos) ss << "left";
                                if (opt.find('r') != string::npos) ss << "right";
                                if (opt.find('c') != string::npos) ss << "center";
                                if (opt.find('t') != string::npos) ss << "Top";
                                if (opt.find('b') != string::npos) ss << "Bottom";
                                if (opt.find('B') != string::npos) ss << "Baseline";
                                if (!ss.str().empty())
                                        os << "\trotateOrigin " << ss.str() << '\n';
                                else
                                        cerr << "Warning: Ignoring unknown includegraphics origin argument '" << opt << "'\n";
                        }
                        if (opts.find("keepaspectratio") != opts.end())
                                os << "\tkeepAspectRatio\n";
                        if (opts.find("clip") != opts.end())
                                os << "\tclip\n";
                        if (opts.find("draft") != opts.end())
                                os << "\tdraft\n";
                        if (opts.find("bb") != opts.end())
                                os << "\tBoundingBox "
                                   << opts["bb"] << '\n';
                        int numberOfbbOptions = 0;
                        if (opts.find("bbllx") != opts.end())
                                numberOfbbOptions++;
                        if (opts.find("bblly") != opts.end())
                                numberOfbbOptions++;
                        if (opts.find("bburx") != opts.end())
                                numberOfbbOptions++;
                        if (opts.find("bbury") != opts.end())
                                numberOfbbOptions++;
                        if (numberOfbbOptions == 4)
                                os << "\tBoundingBox "
                                   << opts["bbllx"] << " " << opts["bblly"] << " "
                                   << opts["bburx"] << " " << opts["bbury"] << '\n';
                        else if (numberOfbbOptions > 0)
                                cerr << "Warning: Ignoring incomplete includegraphics boundingbox arguments.\n";
                        numberOfbbOptions = 0;
                        if (opts.find("natwidth") != opts.end())
                                numberOfbbOptions++;
                        if (opts.find("natheight") != opts.end())
                                numberOfbbOptions++;
                        if (numberOfbbOptions == 2)
                                os << "\tBoundingBox 0bp 0bp "
                                   << opts["natwidth"] << " " << opts["natheight"] << '\n';
                        else if (numberOfbbOptions > 0)
                                cerr << "Warning: Ignoring incomplete includegraphics boundingbox arguments.\n";
                        ostringstream special;
                        if (opts.find("hiresbb") != opts.end())
                                special << "hiresbb,";
                        if (opts.find("trim") != opts.end())
                                special << "trim,";
                        if (opts.find("viewport") != opts.end())
                                special << "viewport=" << opts["viewport"] << ',';
                        if (opts.find("totalheight") != opts.end())
                                special << "totalheight=" << opts["totalheight"] << ',';
                        if (opts.find("type") != opts.end())
                                special << "type=" << opts["type"] << ',';
                        if (opts.find("ext") != opts.end())
                                special << "ext=" << opts["ext"] << ',';
                        if (opts.find("read") != opts.end())
                                special << "read=" << opts["read"] << ',';
                        if (opts.find("command") != opts.end())
                                special << "command=" << opts["command"] << ',';
                        string s_special = special.str();
                        if (!s_special.empty()) {
                                // We had special arguments. Remove the trailing ','.
                                os << "\tspecial " << s_special.substr(0, s_special.size() - 1) << '\n';
                        }
                        // TODO: Handle the unknown settings better.
                        // Warn about invalid options.
                        // Check whether some option was given twice.
                        end_inset(os);
                        preamble.registerAutomaticallyLoadedPackage("graphicx");
                }

                else if (t.cs() == "footnote" ||
                         (t.cs() == "thanks" && context.layout->intitle)) {
                        p.skip_spaces();
                        context.check_layout(os);
                        begin_inset(os, "Foot\n");
                        os << "status collapsed\n\n";
                        parse_text_in_inset(p, os, FLAG_ITEM, false, context);
                        end_inset(os);
                }

                else if (t.cs() == "marginpar") {
                        p.skip_spaces();
                        context.check_layout(os);
                        begin_inset(os, "Marginal\n");
                        os << "status collapsed\n\n";
                        parse_text_in_inset(p, os, FLAG_ITEM, false, context);
                        end_inset(os);
                }

                else if (t.cs() == "ensuremath") {
                        p.skip_spaces();
                        context.check_layout(os);
                        string const s = p.verbatim_item();
                        //FIXME: this never triggers in UTF8
                        if (s == "\xb1" || s == "\xb3" || s == "\xb2" || s == "\xb5")
                                os << s;
                        else
                                handle_ert(os, "\\ensuremath{" + s + "}",
                                           context);
                }

                else if (t.cs() == "makeindex" || t.cs() == "maketitle") {
                        if (title_layout_found) {
                                // swallow this
                                skip_spaces_braces(p);
                        } else
                                handle_ert(os, t.asInput(), context);
                }

                else if (t.cs() == "tableofcontents") {
                        context.check_layout(os);
                        begin_command_inset(os, "toc", "tableofcontents");
                        end_inset(os);
                        skip_spaces_braces(p);
                }

                else if (t.cs() == "listoffigures") {
                        context.check_layout(os);
                        begin_inset(os, "FloatList figure\n");
                        end_inset(os);
                        skip_spaces_braces(p);
                }

                else if (t.cs() == "listoftables") {
                        context.check_layout(os);
                        begin_inset(os, "FloatList table\n");
                        end_inset(os);
                        skip_spaces_braces(p);
                }

                else if (t.cs() == "listof") {
                        p.skip_spaces(true);
                        string const name = p.get_token().cs();
                        if (context.textclass.floats().typeExist(name)) {
                                context.check_layout(os);
                                begin_inset(os, "FloatList ");
                                os << name << "\n";
                                end_inset(os);
                                p.get_token(); // swallow second arg
                        } else
                                handle_ert(os, "\\listof{" + name + "}", context);
                }

                else if ((where = is_known(t.cs(), known_text_font_families)))
                        parse_text_attributes(p, os, FLAG_ITEM, outer,
                                context, "\\family", context.font.family,
                                known_coded_font_families[where - known_text_font_families]);

                else if ((where = is_known(t.cs(), known_text_font_series)))
                        parse_text_attributes(p, os, FLAG_ITEM, outer,
                                context, "\\series", context.font.series,
                                known_coded_font_series[where - known_text_font_series]);

                else if ((where = is_known(t.cs(), known_text_font_shapes)))
                        parse_text_attributes(p, os, FLAG_ITEM, outer,
                                context, "\\shape", context.font.shape,
                                known_coded_font_shapes[where - known_text_font_shapes]);

                else if (t.cs() == "textnormal" || t.cs() == "normalfont") {
                        context.check_layout(os);
                        TeXFont oldFont = context.font;
                        context.font.init();
                        context.font.size = oldFont.size;
                        os << "\n\\family " << context.font.family << "\n";
                        os << "\n\\series " << context.font.series << "\n";
                        os << "\n\\shape " << context.font.shape << "\n";
                        if (t.cs() == "textnormal") {
                                parse_text_snippet(p, os, FLAG_ITEM, outer, context);
                                output_font_change(os, context.font, oldFont);
                                context.font = oldFont;
                        } else
                                eat_whitespace(p, os, context, false);
                }

                else if (t.cs() == "textcolor") {
                        // scheme is \textcolor{color name}{text}
                        string const color = p.verbatim_item();
                        // we only support the predefined colors of the color package
                        if (color == "black" || color == "blue" || color == "cyan"
                                || color == "green" || color == "magenta" || color == "red"
                                || color == "white" || color == "yellow") {
                                        context.check_layout(os);
                                        os << "\n\\color " << color << "\n";
                                        parse_text_snippet(p, os, FLAG_ITEM, outer, context);
                                        context.check_layout(os);
                                        os << "\n\\color inherit\n";
                                        preamble.registerAutomaticallyLoadedPackage("color");
                        } else
                                // for custom defined colors
                                handle_ert(os, t.asInput() + "{" + color + "}", context);
                }

                else if (t.cs() == "underbar" || t.cs() == "uline") {
                        // \underbar is not 100% correct (LyX outputs \uline
                        // of ulem.sty). The difference is that \ulem allows
                        // line breaks, and \underbar does not.
                        // Do NOT handle \underline.
                        // \underbar cuts through y, g, q, p etc.,
                        // \underline does not.
                        context.check_layout(os);
                        os << "\n\\bar under\n";
                        parse_text_snippet(p, os, FLAG_ITEM, outer, context);
                        context.check_layout(os);
                        os << "\n\\bar default\n";
                        preamble.registerAutomaticallyLoadedPackage("ulem");
                }

                else if (t.cs() == "sout") {
                        context.check_layout(os);
                        os << "\n\\strikeout on\n";
                        parse_text_snippet(p, os, FLAG_ITEM, outer, context);
                        context.check_layout(os);
                        os << "\n\\strikeout default\n";
                        preamble.registerAutomaticallyLoadedPackage("ulem");
                }

                else if (t.cs() == "uuline" || t.cs() == "uwave" ||
                         t.cs() == "emph" || t.cs() == "noun") {
                        context.check_layout(os);
                        os << "\n\\" << t.cs() << " on\n";
                        parse_text_snippet(p, os, FLAG_ITEM, outer, context);
                        context.check_layout(os);
                        os << "\n\\" << t.cs() << " default\n";
                        if (t.cs() == "uuline" || t.cs() == "uwave")
                                preamble.registerAutomaticallyLoadedPackage("ulem");
                }

                else if (t.cs() == "lyxadded" || t.cs() == "lyxdeleted") {
                        context.check_layout(os);
                        string name = p.getArg('{', '}');
                        string localtime = p.getArg('{', '}');
                        preamble.registerAuthor(name);
                        Author const & author = preamble.getAuthor(name);
                        // from_ctime() will fail if LyX decides to output the
                        // time in the text language. It might also use a wrong
                        // time zone (if the original LyX document was exported
                        // with a different time zone).
                        time_t ptime = from_ctime(localtime);
                        if (ptime == static_cast<time_t>(-1)) {
                                cerr << "Warning: Could not parse time `" << localtime
                                     << "´ for change tracking, using current time instead.\n";
                                ptime = current_time();
                        }
                        if (t.cs() == "lyxadded")
                                os << "\n\\change_inserted ";
                        else
                                os << "\n\\change_deleted ";
                        os << author.bufferId() << ' ' << ptime << '\n';
                        parse_text_snippet(p, os, FLAG_ITEM, outer, context);
                        bool dvipost    = LaTeXPackages::isAvailable("dvipost");
                        bool xcolorulem = LaTeXPackages::isAvailable("ulem") &&
                                          LaTeXPackages::isAvailable("xcolor");
                        // No need to test for luatex, since luatex comes in
                        // two flavours (dvi and pdf), like latex, and those
                        // are detected by pdflatex.
                        if (pdflatex || xetex) {
                                if (xcolorulem) {
                                        preamble.registerAutomaticallyLoadedPackage("ulem");
                                        preamble.registerAutomaticallyLoadedPackage("xcolor");
                                        preamble.registerAutomaticallyLoadedPackage("pdfcolmk");
                                }
                        } else {
                                if (dvipost) {
                                        preamble.registerAutomaticallyLoadedPackage("dvipost");
                                } else if (xcolorulem) {
                                        preamble.registerAutomaticallyLoadedPackage("ulem");
                                        preamble.registerAutomaticallyLoadedPackage("xcolor");
                                }
                        }
                }

                else if (t.cs() == "phantom" || t.cs() == "hphantom" ||
                             t.cs() == "vphantom") {
                        context.check_layout(os);
                        if (t.cs() == "phantom")
                                begin_inset(os, "Phantom Phantom\n");
                        if (t.cs() == "hphantom")
                                begin_inset(os, "Phantom HPhantom\n");
                        if (t.cs() == "vphantom")
                                begin_inset(os, "Phantom VPhantom\n");
                        os << "status open\n";
                        parse_text_in_inset(p, os, FLAG_ITEM, outer, context,
                                            "Phantom");
                        end_inset(os);
                }

                else if (t.cs() == "href") {
                        context.check_layout(os);
                        string target = p.getArg('{', '}');
                        string name = p.getArg('{', '}');
                        string type;
                        size_t i = target.find(':');
                        if (i != string::npos) {
                                type = target.substr(0, i + 1);
                                if (type == "mailto:" || type == "file:")
                                        target = target.substr(i + 1);
                                // handle the case that name is equal to target, except of "http://"
                                else if (target.substr(i + 3) == name && type == "http:")
                                        target = name;
                        }
                        begin_command_inset(os, "href", "href");
                        if (name != target)
                                os << "name \"" << name << "\"\n";
                        os << "target \"" << target << "\"\n";
                        if (type == "mailto:" || type == "file:")
                                os << "type \"" << type << "\"\n";
                        end_inset(os);
                        skip_spaces_braces(p);
                }

                else if (t.cs() == "lyxline") {
                        // swallow size argument (it is not used anyway)
                        p.getArg('{', '}');
                        if (!context.atParagraphStart()) {
                                // so our line is in the middle of a paragraph
                                // we need to add a new line, lest this line
                                // follow the other content on that line and
                                // run off the side of the page
                                // FIXME: This may create an empty paragraph,
                                //        but without that it would not be
                                //        possible to set noindent below.
                                //        Fortunately LaTeX does not care
                                //        about the empty paragraph.
                                context.new_paragraph(os);
                        }
                        if (preamble.indentParagraphs()) {
                                // we need to unindent, lest the line be too long
                                context.add_par_extra_stuff("\\noindent\n");
                        }
                        context.check_layout(os);
                        begin_command_inset(os, "line", "rule");
                        os << "offset \"0.5ex\"\n"
                              "width \"100line%\"\n"
                              "height \"1pt\"\n";
                        end_inset(os);
                }

                else if (t.cs() == "rule") {
                        string const offset = (p.hasOpt() ? p.getArg('[', ']') : string());
                        string const width = p.getArg('{', '}');
                        string const thickness = p.getArg('{', '}');
                        context.check_layout(os);
                        begin_command_inset(os, "line", "rule");
                        if (!offset.empty())
                                os << "offset \"" << translate_len(offset) << "\"\n";
                        os << "width \"" << translate_len(width) << "\"\n"
                                  "height \"" << translate_len(thickness) << "\"\n";
                        end_inset(os);
                }

                else if (is_known(t.cs(), known_phrases) ||
                         (t.cs() == "protect" &&
                          p.next_token().cat() == catEscape &&
                          is_known(p.next_token().cs(), known_phrases))) {
                        // LyX sometimes puts a \protect in front, so we have to ignore it
                        // FIXME: This needs to be changed when bug 4752 is fixed.
                        where = is_known(
                                t.cs() == "protect" ? p.get_token().cs() : t.cs(),
                                known_phrases);
                        context.check_layout(os);
                        os << known_coded_phrases[where - known_phrases];
                        skip_spaces_braces(p);
                }

                else if ((where = is_known(t.cs(), known_ref_commands))) {
                        string const opt = p.getOpt();
                        if (opt.empty()) {
                                context.check_layout(os);
                                begin_command_inset(os, "ref",
                                        known_coded_ref_commands[where - known_ref_commands]);
                                os << "reference \""
                                   << convert_command_inset_arg(p.verbatim_item())
                                   << "\"\n";
                                end_inset(os);
                                if (t.cs() == "vref" || t.cs() == "vpageref")
                                        preamble.registerAutomaticallyLoadedPackage("varioref");

                        } else {
                                // LyX does not support optional arguments of ref commands
                                handle_ert(os, t.asInput() + '[' + opt + "]{" +
                                               p.verbatim_item() + "}", context);
                        }
                }

                else if (use_natbib &&
                         is_known(t.cs(), known_natbib_commands) &&
                         ((t.cs() != "citefullauthor" &&
                           t.cs() != "citeyear" &&
                           t.cs() != "citeyearpar") ||
                          p.next_token().asInput() != "*")) {
                        context.check_layout(os);
                        string command = t.cs();
                        if (p.next_token().asInput() == "*") {
                                command += '*';
                                p.get_token();
                        }
                        if (command == "citefullauthor")
                                // alternative name for "\\citeauthor*"
                                command = "citeauthor*";

                        // text before the citation
                        string before;
                        // text after the citation
                        string after;
                        get_cite_arguments(p, true, before, after);

                        if (command == "cite") {
                                // \cite without optional argument means
                                // \citet, \cite with at least one optional
                                // argument means \citep.
                                if (before.empty() && after.empty())
                                        command = "citet";
                                else
                                        command = "citep";
                        }
                        if (before.empty() && after == "[]")
                                // avoid \citet[]{a}
                                after.erase();
                        else if (before == "[]" && after == "[]") {
                                // avoid \citet[][]{a}
                                before.erase();
                                after.erase();
                        }
                        // remove the brackets around after and before
                        if (!after.empty()) {
                                after.erase(0, 1);
                                after.erase(after.length() - 1, 1);
                                after = convert_command_inset_arg(after);
                        }
                        if (!before.empty()) {
                                before.erase(0, 1);
                                before.erase(before.length() - 1, 1);
                                before = convert_command_inset_arg(before);
                        }
                        begin_command_inset(os, "citation", command);
                        os << "after " << '"' << after << '"' << "\n";
                        os << "before " << '"' << before << '"' << "\n";
                        os << "key \""
                           << convert_command_inset_arg(p.verbatim_item())
                           << "\"\n";
                        end_inset(os);
                }

                else if (use_jurabib &&
                         is_known(t.cs(), known_jurabib_commands) &&
                         (t.cs() == "cite" || p.next_token().asInput() != "*")) {
                        context.check_layout(os);
                        string command = t.cs();
                        if (p.next_token().asInput() == "*") {
                                command += '*';
                                p.get_token();
                        }
                        char argumentOrder = '\0';
                        vector<string> const options =
                                preamble.getPackageOptions("jurabib");
                        if (find(options.begin(), options.end(),
                                      "natbiborder") != options.end())
                                argumentOrder = 'n';
                        else if (find(options.begin(), options.end(),
                                           "jurabiborder") != options.end())
                                argumentOrder = 'j';

                        // text before the citation
                        string before;
                        // text after the citation
                        string after;
                        get_cite_arguments(p, argumentOrder != 'j', before, after);

                        string const citation = p.verbatim_item();
                        if (!before.empty() && argumentOrder == '\0') {
                                cerr << "Warning: Assuming argument order "
                                        "of jurabib version 0.6 for\n'"
                                     << command << before << after << '{'
                                     << citation << "}'.\n"
                                        "Add 'jurabiborder' to the jurabib "
                                        "package options if you used an\n"
                                        "earlier jurabib version." << endl;
                        }
                        if (!after.empty()) {
                                after.erase(0, 1);
                                after.erase(after.length() - 1, 1);
                        }
                        if (!before.empty()) {
                                before.erase(0, 1);
                                before.erase(before.length() - 1, 1);
                        }
                        begin_command_inset(os, "citation", command);
                        os << "after " << '"' << after << '"' << "\n";
                        os << "before " << '"' << before << '"' << "\n";
                        os << "key " << '"' << citation << '"' << "\n";
                        end_inset(os);
                }

                else if (t.cs() == "cite"
                        || t.cs() == "nocite") {
                        context.check_layout(os);
                        string after = convert_command_inset_arg(p.getArg('[', ']'));
                        string key = convert_command_inset_arg(p.verbatim_item());
                        // store the case that it is "\nocite{*}" to use it later for
                        // the BibTeX inset
                        if (key != "*") {
                                begin_command_inset(os, "citation", t.cs());
                                os << "after " << '"' << after << '"' << "\n";
                                os << "key " << '"' << key << '"' << "\n";
                                end_inset(os);
                        } else if (t.cs() == "nocite")
                                btprint = key;
                }

                else if (t.cs() == "index") {
                        context.check_layout(os);
                        begin_inset(os, "Index idx\n");
                        os << "status collapsed\n";
                        parse_text_in_inset(p, os, FLAG_ITEM, false, context, "Index");
                        end_inset(os);
                }

                else if (t.cs() == "nomenclature") {
                        context.check_layout(os);
                        begin_command_inset(os, "nomenclature", "nomenclature");
                        string prefix = convert_command_inset_arg(p.getArg('[', ']'));
                        if (!prefix.empty())
                                os << "prefix " << '"' << prefix << '"' << "\n";
                        os << "symbol " << '"'
                           << convert_command_inset_arg(p.verbatim_item());
                        os << "\"\ndescription \""
                           << convert_command_inset_arg(p.verbatim_item())
                           << "\"\n";
                        end_inset(os);
                        preamble.registerAutomaticallyLoadedPackage("nomencl");
                }

                else if (t.cs() == "label") {
                        context.check_layout(os);
                        begin_command_inset(os, "label", "label");
                        os << "name \""
                           << convert_command_inset_arg(p.verbatim_item())
                           << "\"\n";
                        end_inset(os);
                }

                else if (t.cs() == "printindex") {
                        context.check_layout(os);
                        begin_command_inset(os, "index_print", "printindex");
                        os << "type \"idx\"\n";
                        end_inset(os);
                        skip_spaces_braces(p);
                        preamble.registerAutomaticallyLoadedPackage("makeidx");
                        if (preamble.use_indices() == "true")
                                preamble.registerAutomaticallyLoadedPackage("splitidx");
                }

                else if (t.cs() == "printnomenclature") {
                        string width = "";
                        string width_type = "";
                        context.check_layout(os);
                        begin_command_inset(os, "nomencl_print", "printnomenclature");
                        // case of a custom width
                        if (p.hasOpt()) {
                                width = p.getArg('[', ']');
                                width = translate_len(width);
                                width_type = "custom";
                        }
                        // case of no custom width
                        // the case of no custom width but the width set
                        // via \settowidth{\nomlabelwidth}{***} cannot be supported
                        // because the user could have set anything, not only the width
                        // of the longest label (which would be width_type = "auto")
                        string label = convert_command_inset_arg(p.getArg('{', '}'));
                        if (label.empty() && width_type.empty())
                                width_type = "none";
                        os << "set_width \"" << width_type << "\"\n";
                        if (width_type == "custom")
                                os << "width \"" << width << '\"';
                        end_inset(os);
                        skip_spaces_braces(p);
                        preamble.registerAutomaticallyLoadedPackage("nomencl");
                }

                else if ((t.cs() == "textsuperscript" || t.cs() == "textsubscript")) {
                        context.check_layout(os);
                        begin_inset(os, "script ");
                        os << t.cs().substr(4) << '\n';
                        parse_text_in_inset(p, os, FLAG_ITEM, false, context);
                        end_inset(os);
                        if (t.cs() == "textsubscript")
                                preamble.registerAutomaticallyLoadedPackage("subscript");
                }

                else if ((where = is_known(t.cs(), known_quotes))) {
                        context.check_layout(os);
                        begin_inset(os, "Quotes ");
                        os << known_coded_quotes[where - known_quotes];
                        end_inset(os);
                        // LyX adds {} after the quote, so we have to eat
                        // spaces here if there are any before a possible
                        // {} pair.
                        eat_whitespace(p, os, context, false);
                        skip_braces(p);
                }

                else if ((where = is_known(t.cs(), known_sizes)) &&
                         context.new_layout_allowed) {
                        context.check_layout(os);
                        TeXFont const oldFont = context.font;
                        context.font.size = known_coded_sizes[where - known_sizes];
                        output_font_change(os, oldFont, context.font);
                        eat_whitespace(p, os, context, false);
                }

                else if ((where = is_known(t.cs(), known_font_families)) &&
                         context.new_layout_allowed) {
                        context.check_layout(os);
                        TeXFont const oldFont = context.font;
                        context.font.family =
                                known_coded_font_families[where - known_font_families];
                        output_font_change(os, oldFont, context.font);
                        eat_whitespace(p, os, context, false);
                }

                else if ((where = is_known(t.cs(), known_font_series)) &&
                         context.new_layout_allowed) {
                        context.check_layout(os);
                        TeXFont const oldFont = context.font;
                        context.font.series =
                                known_coded_font_series[where - known_font_series];
                        output_font_change(os, oldFont, context.font);
                        eat_whitespace(p, os, context, false);
                }

                else if ((where = is_known(t.cs(), known_font_shapes)) &&
                         context.new_layout_allowed) {
                        context.check_layout(os);
                        TeXFont const oldFont = context.font;
                        context.font.shape =
                                known_coded_font_shapes[where - known_font_shapes];
                        output_font_change(os, oldFont, context.font);
                        eat_whitespace(p, os, context, false);
                }
                else if ((where = is_known(t.cs(), known_old_font_families)) &&
                         context.new_layout_allowed) {
                        context.check_layout(os);
                        TeXFont const oldFont = context.font;
                        context.font.init();
                        context.font.size = oldFont.size;
                        context.font.family =
                                known_coded_font_families[where - known_old_font_families];
                        output_font_change(os, oldFont, context.font);
                        eat_whitespace(p, os, context, false);
                }

                else if ((where = is_known(t.cs(), known_old_font_series)) &&
                         context.new_layout_allowed) {
                        context.check_layout(os);
                        TeXFont const oldFont = context.font;
                        context.font.init();
                        context.font.size = oldFont.size;
                        context.font.series =
                                known_coded_font_series[where - known_old_font_series];
                        output_font_change(os, oldFont, context.font);
                        eat_whitespace(p, os, context, false);
                }

                else if ((where = is_known(t.cs(), known_old_font_shapes)) &&
                         context.new_layout_allowed) {
                        context.check_layout(os);
                        TeXFont const oldFont = context.font;
                        context.font.init();
                        context.font.size = oldFont.size;
                        context.font.shape =
                                known_coded_font_shapes[where - known_old_font_shapes];
                        output_font_change(os, oldFont, context.font);
                        eat_whitespace(p, os, context, false);
                }

                else if (t.cs() == "selectlanguage") {
                        context.check_layout(os);
                        // save the language for the case that a
                        // \foreignlanguage is used

                        context.font.language = babel2lyx(p.verbatim_item());
                        os << "\n\\lang " << context.font.language << "\n";
                }

                else if (t.cs() == "foreignlanguage") {
                        string const lang = babel2lyx(p.verbatim_item());
                        parse_text_attributes(p, os, FLAG_ITEM, outer,
                                              context, "\\lang",
                                              context.font.language, lang);
                }

                else if (t.cs() == "inputencoding") {
                        // nothing to write here
                        string const enc = subst(p.verbatim_item(), "\n", " ");
                        p.setEncoding(enc);
                }

                else if ((where = is_known(t.cs(), known_special_chars))) {
                        context.check_layout(os);
                        os << "\\SpecialChar \\"
                           << known_coded_special_chars[where - known_special_chars]
                           << '\n';
                        skip_spaces_braces(p);
                }

                else if (t.cs() == "nobreakdash" && p.next_token().asInput() == "-") {
                        context.check_layout(os);
                        os << "\\SpecialChar \\nobreakdash-\n";
                        p.get_token();
                }

                else if (t.cs() == "textquotedbl") {
                        context.check_layout(os);
                        os << "\"";
                        skip_braces(p);
                }

                else if (t.cs() == "@" && p.next_token().asInput() == ".") {
                        context.check_layout(os);
                        os << "\\SpecialChar \\@.\n";
                        p.get_token();
                }

                else if (t.cs() == "-") {
                        context.check_layout(os);
                        os << "\\SpecialChar \\-\n";
                }

                else if (t.cs() == "textasciitilde") {
                        context.check_layout(os);
                        os << '~';
                        skip_spaces_braces(p);
                }

                else if (t.cs() == "textasciicircum") {
                        context.check_layout(os);
                        os << '^';
                        skip_spaces_braces(p);
                }

                else if (t.cs() == "textbackslash") {
                        context.check_layout(os);
                        os << "\n\\backslash\n";
                        skip_spaces_braces(p);
                }

                else if (t.cs() == "_" || t.cs() == "&" || t.cs() == "#"
                            || t.cs() == "$" || t.cs() == "{" || t.cs() == "}"
                            || t.cs() == "%") {
                        context.check_layout(os);
                        os << t.cs();
                }

                else if (t.cs() == "char") {
                        context.check_layout(os);
                        if (p.next_token().character() == '`') {
                                p.get_token();
                                if (p.next_token().cs() == "\"") {
                                        p.get_token();
                                        os << '"';
                                        skip_braces(p);
                                } else {
                                        handle_ert(os, "\\char`", context);
                                }
                        } else {
                                handle_ert(os, "\\char", context);
                        }
                }

                else if (t.cs() == "verb") {
                        context.check_layout(os);
                        char const delimiter = p.next_token().character();
                        string const arg = p.getArg(delimiter, delimiter);
                        ostringstream oss;
                        oss << "\\verb" << delimiter << arg << delimiter;
                        handle_ert(os, oss.str(), context);
                }

                // Problem: \= creates a tabstop inside the tabbing environment
                // and else an accent. In the latter case we really would want
                // \={o} instead of \= o.
                else if (t.cs() == "=" && (flags & FLAG_TABBING))
                        handle_ert(os, t.asInput(), context);

                // accents (see Table 6 in Comprehensive LaTeX Symbol List)
                else if (t.cs().size() == 1
                         && contains("\"'.=^`bcdHkrtuv~", t.cs())) {
                        context.check_layout(os);
                        // try to see whether the string is in unicodesymbols
                        docstring rem;
                        string command = t.asInput() + "{"
                                + trimSpaceAndEol(p.verbatim_item())
                                + "}";
                        set<string> req;
                        docstring s = encodings.fromLaTeXCommand(from_utf8(command),
                                Encodings::TEXT_CMD | Encodings::MATH_CMD, rem, &req);
                        if (!s.empty()) {
                                if (!rem.empty())
                                        cerr << "When parsing " << command
                                             << ", result is " << to_utf8(s)
                                             << "+" << to_utf8(rem) << endl;
                                os << to_utf8(s);
                                for (set<string>::const_iterator it = req.begin(); it != req.end(); it++)
                                        preamble.registerAutomaticallyLoadedPackage(*it);
                        } else
                                // we did not find a non-ert version
                                handle_ert(os, command, context);
                }

                else if (t.cs() == "\\") {
                        context.check_layout(os);
                        if (p.hasOpt())
                                handle_ert(os, "\\\\" + p.getOpt(), context);
                        else if (p.next_token().asInput() == "*") {
                                p.get_token();
                                // getOpt() eats the following space if there
                                // is no optional argument, but that is OK
                                // here since it has no effect in the output.
                                handle_ert(os, "\\\\*" + p.getOpt(), context);
                        }
                        else {
                                begin_inset(os, "Newline newline");
                                end_inset(os);
                        }
                }

                else if (t.cs() == "newline" ||
                         (t.cs() == "linebreak" && !p.hasOpt())) {
                        context.check_layout(os);
                        begin_inset(os, "Newline ");
                        os << t.cs();
                        end_inset(os);
                        skip_spaces_braces(p);
                }

                else if (t.cs() == "input" || t.cs() == "include"
                         || t.cs() == "verbatiminput") {
                        string name = t.cs();
                        if (t.cs() == "verbatiminput"
                            && p.next_token().asInput() == "*")
                                name += p.get_token().asInput();
                        context.check_layout(os);
                        string filename(normalize_filename(p.getArg('{', '}')));
                        string const path = getMasterFilePath();
                        // We want to preserve relative / absolute filenames,
                        // therefore path is only used for testing
                        if ((t.cs() == "include" || t.cs() == "input") &&
                            !makeAbsPath(filename, path).exists()) {
                                // The file extension is probably missing.
                                // Now try to find it out.
                                string const tex_name =
                                        find_file(filename, path,
                                                  known_tex_extensions);
                                if (!tex_name.empty())
                                        filename = tex_name;
                        }
                        bool external = false;
                        string outname;
                        if (makeAbsPath(filename, path).exists()) {
                                string const abstexname =
                                        makeAbsPath(filename, path).absFileName();
                                string const abslyxname =
                                        changeExtension(abstexname, ".lyx");
                                string const absfigname =
                                        changeExtension(abstexname, ".fig");
                                fix_relative_filename(filename);
                                string const lyxname =
                                        changeExtension(filename, ".lyx");
                                bool xfig = false;
                                external = FileName(absfigname).exists();
                                if (t.cs() == "input") {
                                        string const ext = getExtension(abstexname);

                                        // Combined PS/LaTeX:
                                        // x.eps, x.pstex_t (old xfig)
                                        // x.pstex, x.pstex_t (new xfig, e.g. 3.2.5)
                                        FileName const absepsname(
                                                changeExtension(abstexname, ".eps"));
                                        FileName const abspstexname(
                                                changeExtension(abstexname, ".pstex"));
                                        bool const xfigeps =
                                                (absepsname.exists() ||
                                                 abspstexname.exists()) &&
                                                ext == "pstex_t";

                                        // Combined PDF/LaTeX:
                                        // x.pdf, x.pdftex_t (old xfig)
                                        // x.pdf, x.pdf_t (new xfig, e.g. 3.2.5)
                                        FileName const abspdfname(
                                                changeExtension(abstexname, ".pdf"));
                                        bool const xfigpdf =
                                                abspdfname.exists() &&
                                                (ext == "pdftex_t" || ext == "pdf_t");
                                        if (xfigpdf)
                                                pdflatex = true;

                                        // Combined PS/PDF/LaTeX:
                                        // x_pspdftex.eps, x_pspdftex.pdf, x.pspdftex
                                        string const absbase2(
                                                removeExtension(abstexname) + "_pspdftex");
                                        FileName const abseps2name(
                                                addExtension(absbase2, ".eps"));
                                        FileName const abspdf2name(
                                                addExtension(absbase2, ".pdf"));
                                        bool const xfigboth =
                                                abspdf2name.exists() &&
                                                abseps2name.exists() && ext == "pspdftex";

                                        xfig = xfigpdf || xfigeps || xfigboth;
                                        external = external && xfig;
                                }
                                if (external) {
                                        outname = changeExtension(filename, ".fig");
                                } else if (xfig) {
                                        // Don't try to convert, the result
                                        // would be full of ERT.
                                        outname = filename;
                                } else if (t.cs() != "verbatiminput" &&
                                    tex2lyx(abstexname, FileName(abslyxname),
                                            p.getEncoding())) {
                                        outname = lyxname;
                                } else {
                                        outname = filename;
                                }
                        } else {
                                cerr << "Warning: Could not find included file '"
                                     << filename << "'." << endl;
                                outname = filename;
                        }
                        if (external) {
                                begin_inset(os, "External\n");
                                os << "\ttemplate XFig\n"
                                   << "\tfilename " << outname << '\n';
                                registerExternalTemplatePackages("XFig");
                        } else {
                                begin_command_inset(os, "include", name);
                                os << "preview false\n"
                                      "filename \"" << outname << "\"\n";
                                if (t.cs() == "verbatiminput")
                                        preamble.registerAutomaticallyLoadedPackage("verbatim");
                        }
                        end_inset(os);
                }

                else if (t.cs() == "bibliographystyle") {
                        // store new bibliographystyle
                        bibliographystyle = p.verbatim_item();
                        // If any other command than \bibliography and
                        // \nocite{*} follows, we need to output the style
                        // (because it might be used by that command).
                        // Otherwise, it will automatically be output by LyX.
                        p.pushPosition();
                        bool output = true;
                        for (Token t2 = p.get_token(); p.good(); t2 = p.get_token()) {
                                if (t2.cat() == catBegin)
                                        break;
                                if (t2.cat() != catEscape)
                                        continue;
                                if (t2.cs() == "nocite") {
                                        if (p.getArg('{', '}') == "*")
                                                continue;
                                } else if (t2.cs() == "bibliography")
                                        output = false;
                                break;
                        }
                        p.popPosition();
                        if (output) {
                                handle_ert(os,
                                        "\\bibliographystyle{" + bibliographystyle + '}',
                                        context);
                        }
                }

                else if (t.cs() == "bibliography") {
                        context.check_layout(os);
                        begin_command_inset(os, "bibtex", "bibtex");
                        if (!btprint.empty()) {
                                os << "btprint " << '"' << "btPrintAll" << '"' << "\n";
                                // clear the string because the next BibTeX inset can be without the
                                // \nocite{*} option
                                btprint.clear();
                        }
                        os << "bibfiles " << '"' << p.verbatim_item() << '"' << "\n";
                        // Do we have a bibliographystyle set?
                        if (!bibliographystyle.empty())
                                os << "options " << '"' << bibliographystyle << '"' << "\n";
                        end_inset(os);
                }

                else if (t.cs() == "parbox") {
                        // Test whether this is an outer box of a shaded box
                        p.pushPosition();
                        // swallow arguments
                        while (p.hasOpt()) {
                                p.getArg('[', ']');
                                p.skip_spaces(true);
                        }
                        p.getArg('{', '}');
                        p.skip_spaces(true);
                        // eat the '{'
                        if (p.next_token().cat() == catBegin)
                                p.get_token();
                        p.skip_spaces(true);
                        Token to = p.get_token();
                        bool shaded = false;
                        if (to.asInput() == "\\begin") {
                                p.skip_spaces(true);
                                if (p.getArg('{', '}') == "shaded")
                                        shaded = true;
                        }
                        p.popPosition();
                        if (shaded) {
                                parse_outer_box(p, os, FLAG_ITEM, outer,
                                                context, "parbox", "shaded");
                        } else
                                parse_box(p, os, 0, FLAG_ITEM, outer, context,
                                          "", "", t.cs());
                }

                else if (t.cs() == "ovalbox" || t.cs() == "Ovalbox" ||
                         t.cs() == "shadowbox" || t.cs() == "doublebox")
                        parse_outer_box(p, os, FLAG_ITEM, outer, context, t.cs(), "");

                else if (t.cs() == "framebox") {
                        if (p.next_token().character() == '(') {
                                //the syntax is: \framebox(x,y)[position]{content}
                                string arg = t.asInput();
                                arg += p.getFullParentheseArg();
                                arg += p.getFullOpt();
                                eat_whitespace(p, os, context, false);
                                handle_ert(os, arg + '{', context);
                                eat_whitespace(p, os, context, false);
                                parse_text(p, os, FLAG_ITEM, outer, context);
                                handle_ert(os, "}", context);
                        } else {
                                string special = p.getFullOpt();
                                special += p.getOpt();
                                parse_outer_box(p, os, FLAG_ITEM, outer,
                                                context, t.cs(), special);
                        }
                }

                //\makebox() is part of the picture environment and different from \makebox{}
                //\makebox{} will be parsed by parse_box
                else if (t.cs() == "makebox") {
                        if (p.next_token().character() == '(') {
                                //the syntax is: \makebox(x,y)[position]{content}
                                string arg = t.asInput();
                                arg += p.getFullParentheseArg();
                                arg += p.getFullOpt();
                                eat_whitespace(p, os, context, false);
                                handle_ert(os, arg + '{', context);
                                eat_whitespace(p, os, context, false);
                                parse_text(p, os, FLAG_ITEM, outer, context);
                                handle_ert(os, "}", context);
                        } else
                                //the syntax is: \makebox[width][position]{content}
                                parse_box(p, os, 0, FLAG_ITEM, outer, context,
                                          "", "", t.cs());
                }

                else if (t.cs() == "smallskip" ||
                         t.cs() == "medskip" ||
                         t.cs() == "bigskip" ||
                         t.cs() == "vfill") {
                        context.check_layout(os);
                        begin_inset(os, "VSpace ");
                        os << t.cs();
                        end_inset(os);
                        skip_spaces_braces(p);
                }

                else if ((where = is_known(t.cs(), known_spaces))) {
                        context.check_layout(os);
                        begin_inset(os, "space ");
                        os << '\\' << known_coded_spaces[where - known_spaces]
                           << '\n';
                        end_inset(os);
                        // LaTeX swallows whitespace after all spaces except
                        // "\\,". We have to do that here, too, because LyX
                        // adds "{}" which would make the spaces significant.
                        if (t.cs() !=  ",")
                                eat_whitespace(p, os, context, false);
                        // LyX adds "{}" after all spaces except "\\ " and
                        // "\\,", so we have to remove "{}".
                        // "\\,{}" is equivalent to "\\," in LaTeX, so we
                        // remove the braces after "\\,", too.
                        if (t.cs() != " ")
                                skip_braces(p);
                }

                else if (t.cs() == "newpage" ||
                         (t.cs() == "pagebreak" && !p.hasOpt()) ||
                         t.cs() == "clearpage" ||
                         t.cs() == "cleardoublepage") {
                        context.check_layout(os);
                        begin_inset(os, "Newpage ");
                        os << t.cs();
                        end_inset(os);
                        skip_spaces_braces(p);
                }

                else if (t.cs() == "DeclareRobustCommand" ||
                         t.cs() == "DeclareRobustCommandx" ||
                         t.cs() == "newcommand" ||
                         t.cs() == "newcommandx" ||
                         t.cs() == "providecommand" ||
                         t.cs() == "providecommandx" ||
                         t.cs() == "renewcommand" ||
                         t.cs() == "renewcommandx") {
                        // DeclareRobustCommand, DeclareRobustCommandx,
                        // providecommand and providecommandx could be handled
                        // by parse_command(), but we need to call
                        // add_known_command() here.
                        string name = t.asInput();
                        if (p.next_token().asInput() == "*") {
                                // Starred form. Eat '*'
                                p.get_token();
                                name += '*';
                        }
                        string const command = p.verbatim_item();
                        string const opt1 = p.getFullOpt();
                        string const opt2 = p.getFullOpt();
                        add_known_command(command, opt1, !opt2.empty());
                        string const ert = name + '{' + command + '}' +
                                           opt1 + opt2 +
                                           '{' + p.verbatim_item() + '}';

                        if (t.cs() == "DeclareRobustCommand" ||
                            t.cs() == "DeclareRobustCommandx" ||
                            t.cs() == "providecommand" ||
                            t.cs() == "providecommandx" ||
                            name[name.length()-1] == '*')
                                handle_ert(os, ert, context);
                        else {
                                context.check_layout(os);
                                begin_inset(os, "FormulaMacro");
                                os << "\n" << ert;
                                end_inset(os);
                        }
                }

                else if (t.cs() == "let" && p.next_token().asInput() != "*") {
                        // let could be handled by parse_command(),
                        // but we need to call add_known_command() here.
                        string ert = t.asInput();
                        string name;
                        p.skip_spaces();
                        if (p.next_token().cat() == catBegin) {
                                name = p.verbatim_item();
                                ert += '{' + name + '}';
                        } else {
                                name = p.verbatim_item();
                                ert += name;
                        }
                        string command;
                        p.skip_spaces();
                        if (p.next_token().cat() == catBegin) {
                                command = p.verbatim_item();
                                ert += '{' + command + '}';
                        } else {
                                command = p.verbatim_item();
                                ert += command;
                        }
                        // If command is known, make name known too, to parse
                        // its arguments correctly. For this reason we also
                        // have commands in syntax.default that are hardcoded.
                        CommandMap::iterator it = known_commands.find(command);
                        if (it != known_commands.end())
                                known_commands[t.asInput()] = it->second;
                        handle_ert(os, ert, context);
                }

                else if (t.cs() == "hspace" || t.cs() == "vspace") {
                        bool starred = false;
                        if (p.next_token().asInput() == "*") {
                                p.get_token();
                                starred = true;
                        }
                        string name = t.asInput();
                        string const length = p.verbatim_item();
                        string unit;
                        string valstring;
                        bool valid = splitLatexLength(length, valstring, unit);
                        bool known_hspace = false;
                        bool known_vspace = false;
                        bool known_unit = false;
                        double value;
                        if (valid) {
                                istringstream iss(valstring);
                                iss >> value;
                                if (value == 1.0) {
                                        if (t.cs()[0] == 'h') {
                                                if (unit == "\\fill") {
                                                        if (!starred) {
                                                                unit = "";
                                                                name = "\\hfill";
                                                        }
                                                        known_hspace = true;
                                                }
                                        } else {
                                                if (unit == "\\smallskipamount") {
                                                        unit = "smallskip";
                                                        known_vspace = true;
                                                } else if (unit == "\\medskipamount") {
                                                        unit = "medskip";
                                                        known_vspace = true;
                                                } else if (unit == "\\bigskipamount") {
                                                        unit = "bigskip";
                                                        known_vspace = true;
                                                } else if (unit == "\\fill") {
                                                        unit = "vfill";
                                                        known_vspace = true;
                                                }
                                        }
                                }
                                if (!known_hspace && !known_vspace) {
                                        switch (unitFromString(unit)) {
                                        case Length::SP:
                                        case Length::PT:
                                        case Length::BP:
                                        case Length::DD:
                                        case Length::MM:
                                        case Length::PC:
                                        case Length::CC:
                                        case Length::CM:
                                        case Length::IN:
                                        case Length::EX:
                                        case Length::EM:
                                        case Length::MU:
                                                known_unit = true;
                                                break;
                                        default:
                                                break;
                                        }
                                }
                        }

                        if (t.cs()[0] == 'h' && (known_unit || known_hspace)) {
                                // Literal horizontal length or known variable
                                context.check_layout(os);
                                begin_inset(os, "space ");
                                os << name;
                                if (starred)
                                        os << '*';
                                os << '{';
                                if (known_hspace)
                                        os << unit;
                                os << "}";
                                if (known_unit && !known_hspace)
                                        os << "\n\\length "
                                           << translate_len(length);
                                end_inset(os);
                        } else if (known_unit || known_vspace) {
                                // Literal vertical length or known variable
                                context.check_layout(os);
                                begin_inset(os, "VSpace ");
                                if (known_unit)
                                        os << value;
                                os << unit;
                                if (starred)
                                        os << '*';
                                end_inset(os);
                        } else {
                                // LyX can't handle other length variables in Inset VSpace/space
                                if (starred)
                                        name += '*';
                                if (valid) {
                                        if (value == 1.0)
                                                handle_ert(os, name + '{' + unit + '}', context);
                                        else if (value == -1.0)
                                                handle_ert(os, name + "{-" + unit + '}', context);
                                        else
                                                handle_ert(os, name + '{' + valstring + unit + '}', context);
                                } else
                                        handle_ert(os, name + '{' + length + '}', context);
                        }
                }

                // The single '=' is meant here.
                else if ((newinsetlayout = findInsetLayout(context.textclass, t.cs(), true))) {
                        p.skip_spaces();
                        context.check_layout(os);
                        begin_inset(os, "Flex ");
                        os << to_utf8(newinsetlayout->name()) << '\n'
                           << "status collapsed\n";
                        parse_text_in_inset(p, os, FLAG_ITEM, false, context, newinsetlayout);
                        end_inset(os);
                }

                else if (t.cs() == "includepdf") {
                        p.skip_spaces();
                        string const arg = p.getArg('[', ']');
                        map<string, string> opts;
                        vector<string> keys;
                        split_map(arg, opts, keys);
                        string name = normalize_filename(p.verbatim_item());
                        string const path = getMasterFilePath();
                        // We want to preserve relative / absolute filenames,
                        // therefore path is only used for testing
                        if (!makeAbsPath(name, path).exists()) {
                                // The file extension is probably missing.
                                // Now try to find it out.
                                char const * const pdfpages_format[] = {"pdf", 0};
                                string const pdftex_name =
                                        find_file(name, path, pdfpages_format);
                                if (!pdftex_name.empty()) {
                                        name = pdftex_name;
                                        pdflatex = true;
                                }
                        }
                        if (makeAbsPath(name, path).exists())
                                fix_relative_filename(name);
                        else
                                cerr << "Warning: Could not find file '"
                                     << name << "'." << endl;
                        // write output
                        context.check_layout(os);
                        begin_inset(os, "External\n\ttemplate ");
                        os << "PDFPages\n\tfilename "
                           << name << "\n";
                        // parse the options
                        if (opts.find("pages") != opts.end())
                                os << "\textra LaTeX \"pages="
                                   << opts["pages"] << "\"\n";
                        if (opts.find("angle") != opts.end())
                                os << "\trotateAngle "
                                   << opts["angle"] << '\n';
                        if (opts.find("origin") != opts.end()) {
                                ostringstream ss;
                                string const opt = opts["origin"];
                                if (opt == "tl") ss << "topleft";
                                if (opt == "bl") ss << "bottomleft";
                                if (opt == "Bl") ss << "baselineleft";
                                if (opt == "c") ss << "center";
                                if (opt == "tc") ss << "topcenter";
                                if (opt == "bc") ss << "bottomcenter";
                                if (opt == "Bc") ss << "baselinecenter";
                                if (opt == "tr") ss << "topright";
                                if (opt == "br") ss << "bottomright";
                                if (opt == "Br") ss << "baselineright";
                                if (!ss.str().empty())
                                        os << "\trotateOrigin " << ss.str() << '\n';
                                else
                                        cerr << "Warning: Ignoring unknown includegraphics origin argument '" << opt << "'\n";
                        }
                        if (opts.find("width") != opts.end())
                                os << "\twidth "
                                   << translate_len(opts["width"]) << '\n';
                        if (opts.find("height") != opts.end())
                                os << "\theight "
                                   << translate_len(opts["height"]) << '\n';
                        if (opts.find("keepaspectratio") != opts.end())
                                os << "\tkeepAspectRatio\n";
                        end_inset(os);
                        context.check_layout(os);
                        registerExternalTemplatePackages("PDFPages");
                }

                else if (t.cs() == "loadgame") {
                        p.skip_spaces();
                        string name = normalize_filename(p.verbatim_item());
                        string const path = getMasterFilePath();
                        // We want to preserve relative / absolute filenames,
                        // therefore path is only used for testing
                        if (!makeAbsPath(name, path).exists()) {
                                // The file extension is probably missing.
                                // Now try to find it out.
                                char const * const lyxskak_format[] = {"fen", 0};
                                string const lyxskak_name =
                                        find_file(name, path, lyxskak_format);
                                if (!lyxskak_name.empty())
                                        name = lyxskak_name;
                        }
                        if (makeAbsPath(name, path).exists())
                                fix_relative_filename(name);
                        else
                                cerr << "Warning: Could not find file '"
                                     << name << "'." << endl;
                        context.check_layout(os);
                        begin_inset(os, "External\n\ttemplate ");
                        os << "ChessDiagram\n\tfilename "
                           << name << "\n";
                        end_inset(os);
                        context.check_layout(os);
                        // after a \loadgame follows a \showboard
                        if (p.get_token().asInput() == "showboard")
                                p.get_token();
                        registerExternalTemplatePackages("ChessDiagram");
                }

                else {
                        // try to see whether the string is in unicodesymbols
                        // Only use text mode commands, since we are in text mode here,
                        // and math commands may be invalid (bug 6797)
                        docstring rem;
                        set<string> req;
                        docstring s = encodings.fromLaTeXCommand(from_utf8(t.asInput()),
                                                                 Encodings::TEXT_CMD, rem, &req);
                        if (!s.empty()) {
                                if (!rem.empty())
                                        cerr << "When parsing " << t.cs()
                                             << ", result is " << to_utf8(s)
                                             << "+" << to_utf8(rem) << endl;
                                context.check_layout(os);
                                os << to_utf8(s);
                                skip_spaces_braces(p);
                                for (set<string>::const_iterator it = req.begin(); it != req.end(); it++)
                                        preamble.registerAutomaticallyLoadedPackage(*it);
                        }
                        //cerr << "#: " << t << " mode: " << mode << endl;
                        // heuristic: read up to next non-nested space
                        /*
                        string s = t.asInput();
                        string z = p.verbatim_item();
                        while (p.good() && z != " " && z.size()) {
                                //cerr << "read: " << z << endl;
                                s += z;
                                z = p.verbatim_item();
                        }
                        cerr << "found ERT: " << s << endl;
                        handle_ert(os, s + ' ', context);
                        */
                        else {
                                string name = t.asInput();
                                if (p.next_token().asInput() == "*") {
                                        // Starred commands like \vspace*{}
                                        p.get_token();  // Eat '*'
                                        name += '*';
                                }
                                if (!parse_command(name, p, os, outer, context))
                                        handle_ert(os, name, context);
                        }
                }

                if (flags & FLAG_LEAVE) {
                        flags &= ~FLAG_LEAVE;
                        break;
                }
        }
}

// }])


} // namespace lyx