#include "support/filetools.h"
#include "support/lstrings.h"
-#include <boost/regex.hpp>
+#include "support/regex.h"
#include <algorithm>
#include <iostream>
using namespace std;
using namespace lyx::support;
-using boost::regex;
-using boost::smatch;
+
namespace lyx {
// needed to handle encodings with babel
bool one_language = true;
-// to avoid that the babel options overwrite the documentclass options
-bool documentclass_language;
-
namespace {
-const char * const known_languages[] = { "afrikaans", "american", "arabic",
-"austrian", "bahasa", "basque", "belarusian", "brazil", "breton", "british",
-"bulgarian", "canadian", "canadien", "catalan", "croatian", "czech", "danish",
-"dutch", "english", "esperanto", "estonian", "finnish", "francais", "french",
+//add this to known_languages when updating to lyxformat 266:
+// "armenian"
+//add these to known_languages when updating to lyxformat 268:
+//"chinese-simplified", "chinese-traditional", "japanese", "korean"
+// Both changes require first that support for non-babel languages (CJK,
+// armtex) is added.
+// add polutonikogreek for lyxformat 338
+// add mongolian for lyxformat 342
+// add turkmen for lyxformat 383
+const char * const known_languages[] = { "afrikaans", "albanian", "american",
+"arabic", "arabtex", "austrian", "bahasa", "bahasai", "bahasam", "basque",
+"belarusian", "brazil", "brazilian", "breton", "british", "bulgarian",
+"canadian", "canadien", "catalan", "croatian", "czech", "danish", "dutch",
+"english", "esperanto", "estonian", "finnish", "francais", "french",
"frenchb", "frenchle", "frenchpro", "galician", "german", "germanb", "greek",
-"hebrew", "icelandic", "irish", "italian", "lsorbian", "magyar", "naustrian",
-"ngerman", "ngermanb", "norsk", "nynorsk", "polish", "portuges", "romanian",
-"russian", "russianb", "scottish", "serbian", "slovak", "slovene", "spanish",
-"swedish", "thai", "turkish", "ukraineb", "ukrainian", "usorbian", "welsh", 0};
-
-//note this when updating to lyxformat 305:
-//bahasai, indonesian, and indon = equal to bahasa
-//malay, and meyalu = equal to bahasam
-
+"hebrew", "icelandic", "indon", "indonesian", "interlingua", "irish",
+"italian", "kazakh", "latin", "latvian", "lithuanian", "lsorbian", "magyar",
+"malay", "meyalu", "naustrian", "ngerman", "ngermanb", "norsk", "nynorsk",
+"polish", "portuges", "portuguese", "romanian", "russian", "russianb",
+"samin", "scottish", "serbian", "serbian-latin", "slovak", "slovene",
+"spanish", "swedish", "thai", "turkish", "ukraineb", "ukrainian",
+"usorbian", "vietnam", "welsh", 0};
+
+const char * const known_bahasa_languages[] = {"bahasa", "bahasai",
+ "indon", "indonesian", 0};
+const char * const known_bahasam_languages[] = {"bahasam", "malay",
+ "meyalu", 0};
+const char * const known_brazilian_languages[] = {"brazil", "brazilian", 0};
const char * const known_french_languages[] = {"french", "frenchb", "francais",
"frenchle", "frenchpro", 0};
const char * const known_german_languages[] = {"german", "germanb", 0};
const char * const known_ngerman_languages[] = {"ngerman", "ngermanb", 0};
+const char * const known_portuguese_languages[] = {"portuges", "portuguese", 0};
const char * const known_russian_languages[] = {"russian", "russianb", 0};
const char * const known_ukrainian_languages[] = {"ukrainian", "ukraineb", 0};
+//add these to known_english_quotes_languages when updating to lyxformat 268:
+//"chinese-simplified", "korean"
+// This requires first that support for non-babel languages (CJK) is added.
+const char * const known_english_quotes_languages[] = {"american", "canadian",
+"english", "esperanto", "hebrew", "irish", "scottish", "thai", 0};
+
+//add this to known_french_quotes_languages when updating to
+// - lyxformat 327: "spanish-mexico"
+// - lyxformat 338: "polutonikogreek"
+// - lyxformat 383: "turkmen"
+const char * const known_french_quotes_languages[] = {"albanian", "arabic",
+"basque", "canadien", "catalan", "galician", "greek", "italian", "norsk",
+"nynorsk", "spanish", "turkish", "vietnam", 0};
+
+const char * const known_german_quotes_languages[] = {"austrian", "bulgarian",
+"czech", "icelandic", "lithuanian", "lsorbian", "naustrian", "serbian",
+"serbian-latin", "slovak", "slovene", "usorbian", 0};
+
+const char * const known_polish_quotes_languages[] = {"afrikaans", "croatian",
+"dutch", "estonian", "magyar", "polish", "romanian", 0};
+
+const char * const known_swedish_quotes_languages[] = {"finnish",
+"swedish", 0};
+
char const * const known_fontsizes[] = { "10pt", "11pt", "12pt", 0 };
const char * const known_roman_fonts[] = { "ae", "bookman", "charter",
"rightmargin", "bottommargin", "headheight", "headsep", "footskip",
"columnsep", 0};
-const char * const known_lyx_commands[] = { "binom", "cedilla", "cyrtext",
-"dacute", "dgrave", "docedilla", "doogonek", "dosubhat", "dosubring",
-"dosubtilde", "greektext", "guillemotleft", "guillemotright", "guilsinglleft",
-"guilsinglright", "LyX", "lyxadded", "lyxarrow", "lyxdeleted", "lyxdot",
-"lyxgreyedout", "lyxline", "lyxmathsym", "LyXParagraphLeftIndent",
-"lyxrightaddress", "makenomenclature", "mathcircumflex", "noun", "ogonek",
-"printnomenclature", "quotedblbase", "quotesinglbase", "rcap", "subhat",
-"subring", "subtilde", "tabularnewline", "textcyr", "textgreek", 0};
-
-const char * const known_lyx_comments[] = {
-"%% Binom macro for standard LaTeX users\n",
-"%% For printing a cirumflex inside a formula\n",
-"%% Because html converters don't know tabularnewline\n",
-"%% The greyedout annotation environment\n",
-"%% A simple dot to overcome graphicx limitations\n",
-"%% Change tracking with ulem\n",
-"% the following is useful when we have the old nomencl.sty package\n",
-"%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% LyX specific LaTeX commands.\n",
-"%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% User specified LaTeX commands.\n",
-"%% LyX 2.0.0svn created this file. For more info, see http://www.lyx.org/.\n",
-0};
-
// default settings
ostringstream h_preamble;
string h_textclass = "article";
+string h_use_default_options = "false";
string h_options = string();
string h_language = "english";
string h_inputencoding = "auto";
string h_font_sf_scale = "100";
string h_font_tt_scale = "100";
string h_graphics = "default";
+string h_float_placement;
string h_paperfontsize = "default";
string h_spacing = "single";
+string h_use_hyperref = "0";
+string h_pdf_title;
+string h_pdf_author;
+string h_pdf_subject;
+string h_pdf_keywords;
+string h_pdf_bookmarks = "1";
+string h_pdf_bookmarksnumbered = "0";
+string h_pdf_bookmarksopen = "0";
+string h_pdf_bookmarksopenlevel = "1";
+string h_pdf_breaklinks = "0";
+string h_pdf_pdfborder = "0";
+string h_pdf_colorlinks = "0";
+string h_pdf_backref = "section";
+string h_pdf_pdfusetitle = "1";
+string h_pdf_pagemode;
+string h_pdf_quoted_options;
string h_papersize = "default";
string h_use_geometry = "false";
string h_use_amsmath = "1";
string h_papercolumns = "1";
string h_papersides = string();
string h_paperpagestyle = "default";
+string h_listings_params;
string h_tracking_changes = "false";
string h_output_changes = "false";
string h_margins = "";
-void handle_opt(vector<string> & opts, char const * const * what, string & target)
+/// translates a babel language name to a LyX language name
+string babel2lyx(string language)
+{
+ if (language == "arabtex")
+ return "arabic_arabtex";
+ if (language == "arabic")
+ return "arabic_arabi";
+ if (language == "lsorbian")
+ return "lowersorbian";
+ if (language == "usorbian")
+ return "uppersorbian";
+ return language;
+}
+
+
+// returns true if at least one of the options in what has been found
+bool handle_opt(vector<string> & opts, char const * const * what, string & target)
{
if (opts.empty())
- return;
+ return false;
+ bool found = false;
// the last language option is the document language (for babel and LyX)
// the last size option is the document font size
vector<string>::iterator it;
for (; *what; ++what) {
it = find(opts.begin(), opts.end(), *what);
if (it != opts.end()) {
- documentclass_language = true;
if (it >= position) {
+ found = true;
target = *what;
position = it;
}
}
}
+ return found;
}
return;
// remove found options from the list
- // do this after handle_opt to avoid potential memory leaks and to be able
- // to find in every case the last language option
+ // do this after handle_opt to avoid potential memory leaks
vector<string>::iterator it;
for (; *what; ++what) {
it = find(opts.begin(), opts.end(), *what);
}
-void handle_package(Parser &p, string const & name, string const & opts)
+void handle_package(Parser &p, string const & name, string const & opts,
+ bool in_lyx_preamble)
{
vector<string> options = split_options(opts);
add_package(name, options);
one_language = false;
h_inputencoding = "auto";
}
- // only set the document language when there was not already one set
- // via the documentclass options
- // babel takes the the last language given in the documentclass options
- // as document language. If there is no such language option, the last
- // option of its \usepackage call is used.
- if (documentclass_language == false) {
- handle_opt(options, known_languages, h_language);
- delete_opt(options, known_languages);
- if (is_known(h_language, known_french_languages))
- h_language = "french";
- else if (is_known(h_language, known_german_languages))
- h_language = "german";
- else if (is_known(h_language, known_ngerman_languages))
- h_language = "ngerman";
- else if (is_known(h_language, known_russian_languages))
- h_language = "russian";
- else if (is_known(h_language, known_ukrainian_languages))
- h_language = "ukrainian";
- h_quotes_language = h_language;
- }
+ // babel takes the last language of the option of its \usepackage
+ // call as document language. If there is no such language option, the
+ // last language in the documentclass options is used.
+ handle_opt(options, known_languages, h_language);
+ delete_opt(options, known_languages);
}
else if (name == "fontenc")
;// ignore this
else if (name == "inputenc") {
- // only set when there is not more than one inputenc
- // option therefore check for the "," character also
- // only set when there is not more then one babel
- // language option
- if (opts.find(",") == string::npos && one_language == true) {
- if (opts == "ascii")
- //change ascii to auto to be in the unicode range, see
- //http://bugzilla.lyx.org/show_bug.cgi?id=4719
- h_inputencoding = "auto";
- else if (!opts.empty())
- h_inputencoding = opts;
- }
+ // h_inputencoding is only set when there is not more than one
+ // inputenc option because otherwise h_inputencoding must be
+ // set to "auto" (the default encoding of the document language)
+ // Therefore check for the "," character.
+ // It is also only set when there is not more then one babel
+ // language option but this is handled in the routine for babel.
+ if (opts.find(",") == string::npos && one_language == true)
+ h_inputencoding = opts;
if (!options.empty())
p.setEncoding(options.back());
options.clear();
else if (name == "varioref")
; // ignore this
- else if (name == "verbatim")
+ else if (name == "verbatim")
; // ignore this
else if (name == "nomencl")
; // ignore this
+ else if (name == "textcomp")
+ ; // ignore this
+
else if (name == "url")
; // ignore this
+ else if (LYX_FORMAT >= 408 && name == "subscript")
+ ; // ignore this
+
else if (name == "color") {
// with the following command this package is only loaded when needed for
// undefined colors, since we only support the predefined colors
; // Ignore this, the geometry settings are made by the \geometry
// command. This command is handled below.
- else if (is_known(name, known_languages)) {
- if (is_known(name, known_french_languages))
- h_language = "french";
- else if (is_known(name, known_german_languages))
- h_language = "german";
- else if (is_known(name, known_ngerman_languages))
- h_language = "ngerman";
- else if (is_known(name, known_russian_languages))
- h_language = "russian";
- else if (is_known(name, known_ukrainian_languages))
- h_language = "ukrainian";
- else
- h_language = name;
- h_quotes_language = h_language;
- }
+ else if (is_known(name, known_languages))
+ h_language = name;
else if (name == "natbib") {
h_cite_engine = "natbib_authoryear";
else if (name == "jurabib")
h_cite_engine = "jurabib";
- else if (name == "babel")
- ; // ignore this
-
- else {
+ else if (!in_lyx_preamble) {
if (options.empty())
h_preamble << "\\usepackage{" << name << "}";
else {
void end_preamble(ostream & os, TextClass const & /*textclass*/)
{
+ // merge synonym languages
+ if (is_known(h_language, known_bahasa_languages))
+ h_language = "bahasa";
+ else if (is_known(h_language, known_bahasam_languages))
+ h_language = "bahasam";
+ else if (is_known(h_language, known_brazilian_languages))
+ h_language = "brazilian";
+ else if (is_known(h_language, known_french_languages))
+ h_language = "french";
+ else if (is_known(h_language, known_german_languages))
+ h_language = "german";
+ else if (is_known(h_language, known_ngerman_languages))
+ h_language = "ngerman";
+ else if (is_known(h_language, known_portuguese_languages))
+ h_language = "portuguese";
+ else if (is_known(h_language, known_russian_languages))
+ h_language = "russian";
+ else if (is_known(h_language, known_ukrainian_languages))
+ h_language = "ukrainian";
+
+ // set the quote language
+ // LyX only knows the following quotes languages:
+ // english, swedish, german, polish, french and danish
+ // (quotes for "japanese" and "chinese-traditional" are missing because
+ // they wouldn't be useful: http://www.lyx.org/trac/ticket/6383)
+ // conversion list taken from
+ // http://en.wikipedia.org/wiki/Quotation_mark,_non-English_usage
+ // (quotes for kazakh and interlingua are unknown)
+ // danish
+ if (h_language == "danish")
+ h_quotes_language = "danish";
+ // french
+ else if (is_known(h_language, known_french_quotes_languages)
+ || is_known(h_language, known_french_languages)
+ || is_known(h_language, known_russian_languages)
+ || is_known(h_language, known_ukrainian_languages))
+ h_quotes_language = "french";
+ // german
+ else if (is_known(h_language, known_german_quotes_languages)
+ || is_known(h_language, known_german_languages)
+ || is_known(h_language, known_ngerman_languages))
+ h_quotes_language = "german";
+ // polish
+ else if (is_known(h_language, known_polish_quotes_languages))
+ h_quotes_language = "polish";
+ // swedish
+ else if (is_known(h_language, known_swedish_quotes_languages))
+ h_quotes_language = "swedish";
+ //english
+ else if (is_known(h_language, known_english_quotes_languages)
+ || is_known(h_language, known_bahasa_languages)
+ || is_known(h_language, known_bahasam_languages)
+ || is_known(h_language, known_brazilian_languages)
+ || is_known(h_language, known_portuguese_languages))
+ h_quotes_language = "english";
+
+ h_language = babel2lyx(h_language);
+
+ // output the LyX file settings
os << "#LyX file created by tex2lyx " << PACKAGE_VERSION << "\n"
- << "\\lyxformat 258\n"
+ << "\\lyxformat " << LYX_FORMAT << '\n'
<< "\\begin_document\n"
<< "\\begin_header\n"
<< "\\textclass " << h_textclass << "\n";
os << "\\begin_preamble\n" << h_preamble.str() << "\n\\end_preamble\n";
if (!h_options.empty())
os << "\\options " << h_options << "\n";
- os << "\\language " << h_language << "\n"
+ os << "\\use_default_options " << h_use_default_options << "\n"
+ << "\\language " << h_language << "\n"
<< "\\inputencoding " << h_inputencoding << "\n"
<< "\\font_roman " << h_font_roman << "\n"
<< "\\font_sans " << h_font_sans << "\n"
<< "\\font_osf " << h_font_osf << "\n"
<< "\\font_sf_scale " << h_font_sf_scale << "\n"
<< "\\font_tt_scale " << h_font_tt_scale << "\n"
- << "\\graphics " << h_graphics << "\n"
- << "\\paperfontsize " << h_paperfontsize << "\n"
+ << "\\graphics " << h_graphics << "\n";
+ if (!h_float_placement.empty())
+ os << "\\float_placement " << h_float_placement << "\n";
+ os << "\\paperfontsize " << h_paperfontsize << "\n"
<< "\\spacing " << h_spacing << "\n"
- << "\\papersize " << h_papersize << "\n"
+ << "\\use_hyperref " << h_use_hyperref << '\n';
+ if (h_use_hyperref == "1") {
+ if (!h_pdf_title.empty())
+ os << "\\pdf_title \"" << h_pdf_title << "\"\n";
+ if (!h_pdf_author.empty())
+ os << "\\pdf_author \"" << h_pdf_author << "\"\n";
+ if (!h_pdf_subject.empty())
+ os << "\\pdf_subject \"" << h_pdf_subject << "\"\n";
+ if (!h_pdf_keywords.empty())
+ os << "\\pdf_keywords \"" << h_pdf_keywords << "\"\n";
+ os << "\\pdf_bookmarks " << h_pdf_bookmarks << "\n"
+ "\\pdf_bookmarksnumbered " << h_pdf_bookmarksnumbered << "\n"
+ "\\pdf_bookmarksopen " << h_pdf_bookmarksopen << "\n"
+ "\\pdf_bookmarksopenlevel " << h_pdf_bookmarksopenlevel << "\n"
+ "\\pdf_breaklinks " << h_pdf_breaklinks << "\n"
+ "\\pdf_pdfborder " << h_pdf_pdfborder << "\n"
+ "\\pdf_colorlinks " << h_pdf_colorlinks << "\n"
+ "\\pdf_backref " << h_pdf_backref << "\n"
+ "\\pdf_pdfusetitle " << h_pdf_pdfusetitle << '\n';
+ if (!h_pdf_pagemode.empty())
+ os << "\\pdf_pagemode " << h_pdf_pagemode << '\n';
+ if (!h_pdf_quoted_options.empty())
+ os << "\\pdf_quoted_options \"" << h_pdf_quoted_options << "\"\n";
+ }
+ os << "\\papersize " << h_papersize << "\n"
<< "\\use_geometry " << h_use_geometry << "\n"
<< "\\use_amsmath " << h_use_amsmath << "\n"
<< "\\use_esint " << h_use_esint << "\n"
<< "\\quotes_language " << h_quotes_language << "\n"
<< "\\papercolumns " << h_papercolumns << "\n"
<< "\\papersides " << h_papersides << "\n"
- << "\\paperpagestyle " << h_paperpagestyle << "\n"
- << "\\tracking_changes " << h_tracking_changes << "\n"
+ << "\\paperpagestyle " << h_paperpagestyle << "\n";
+ if (!h_listings_params.empty())
+ os << "\\listings_params " << h_listings_params << "\n";
+ os << "\\tracking_changes " << h_tracking_changes << "\n"
<< "\\output_changes " << h_output_changes << "\n"
<< "\\end_header\n\n"
<< "\\begin_body\n";
special_columns['D'] = 3;
bool is_full_document = false;
bool is_lyx_file = false;
- bool lyx_specific_preamble = false;
+ bool in_lyx_preamble = false;
// determine whether this is a full document or a fragment for inclusion
while (p.good()) {
//
// cat codes
//
- if ((t.cat() == catLetter ||
+ if (!in_lyx_preamble &&
+ (t.cat() == catLetter ||
t.cat() == catSuper ||
t.cat() == catSub ||
t.cat() == catOther ||
t.cat() == catEnd ||
t.cat() == catAlign ||
t.cat() == catParameter))
- h_preamble << t.character();
+ h_preamble << t.cs();
- else if (t.cat() == catSpace || t.cat() == catNewline)
+ else if (!in_lyx_preamble &&
+ (t.cat() == catSpace || t.cat() == catNewline))
h_preamble << t.asInput();
else if (t.cat() == catComment) {
- // regex to parse comments
+ // regex to parse comments (currently not used)
static regex const islyxfile("%% LyX .* created this file");
static regex const usercommands("User specified LaTeX commands");
-
+
string const comment = t.asInput();
-
+
// magically switch encoding default if it looks like XeLaTeX
static string const magicXeLaTeX =
"% This document must be compiled with XeLaTeX ";
cerr << "XeLaTeX comment found, switching to UTF8\n";
h_inputencoding = "utf8";
}
-
smatch sub;
- if (regex_search(comment, sub, islyxfile))
+ if (regex_search(comment, sub, islyxfile)) {
is_lyx_file = true;
- // don't output LyX specific comments
- if (!is_known(comment, known_lyx_comments))
+ in_lyx_preamble = true;
+ } else if (is_lyx_file
+ && regex_search(comment, sub, usercommands))
+ in_lyx_preamble = false;
+ else if (!in_lyx_preamble)
h_preamble << t.asInput();
}
// remove leading "\"
h_font_default_family = family.erase(0,1);
}
- // LyX specific commands that will automatically be set by LyX
- string lyx_command = name;
- // remove the leading "\"
- lyx_command.erase(0,1);
- if (is_known(lyx_command, known_lyx_commands))
- lyx_specific_preamble = true;
// only non-lyxspecific stuff
- if (!lyx_specific_preamble) {
+ if (!in_lyx_preamble) {
ostringstream ss;
ss << '\\' << t.cs();
if (star)
string::size_type i = h_paperfontsize.find("pt");
if (i != string::npos)
h_paperfontsize.erase(i);
- // to avoid that the babel options overwrite the documentclass options
- documentclass_language = false;
+ // The documentclass options are always parsed before the options
+ // of the babel call so that a language cannot overwrite the babel
+ // options.
handle_opt(opts, known_languages, h_language);
delete_opt(opts, known_languages);
- if (is_known(h_language, known_french_languages))
- h_language = "french";
- else if (is_known(h_language, known_german_languages))
- h_language = "german";
- else if (is_known(h_language, known_ngerman_languages))
- h_language = "ngerman";
- else if (is_known(h_language, known_russian_languages))
- h_language = "russian";
- else if (is_known(h_language, known_ukrainian_languages))
- h_language = "ukrainian";
- h_quotes_language = h_language;
+
// paper orientation
if ((it = find(opts.begin(), opts.end(), "landscape")) != opts.end()) {
h_paperorientation = "landscape";
vector<string>::const_iterator it = vecnames.begin();
vector<string>::const_iterator end = vecnames.end();
for (; it != end; ++it)
- handle_package(p, trim(*it), options);
+ handle_package(p, trim(*it), options,
+ in_lyx_preamble);
}
else if (t.cs() == "inputencoding") {
else if (t.cs() == "newenvironment") {
string const name = p.getArg('{', '}');
ostringstream ss;
- // only non LyX specific stuff is output
ss << "\\newenvironment{" << name << "}";
ss << p.getOpt();
ss << p.getOpt();
ss << '{' << p.verbatim_item() << '}';
ss << '{' << p.verbatim_item() << '}';
- if (!is_known(name, known_lyx_commands))
+ if (!in_lyx_preamble)
h_preamble << ss.str();
}
else if (t.cs() == "def") {
string name = p.get_token().cs();
while (p.next_token().cat() != catBegin)
- name += p.get_token().asString();
- if (!is_known(name, known_lyx_commands))
+ name += p.get_token().cs();
+ if (!in_lyx_preamble)
h_preamble << "\\def\\" << name << '{'
<< p.verbatim_item() << "}";
}
}
}
- else if (!t.cs().empty())
+ else if (!t.cs().empty() && !in_lyx_preamble)
h_preamble << '\\' << t.cs();
-
- // remove the whitespace
- p.skip_spaces();
}
// remove the whitespace