X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=src%2FParagraph.cpp;h=991ab14a7527d65083207e171b9e34b8fb720fce;hb=ea6aed5b8bf38366aaa1eb15ce1b9f13de76987f;hp=6aeab1aa7dce60f5cfd80b75e639be50800256d6;hpb=b922cfd3b351683fa452eeec4830258d72d47ab8;p=lyx.git

diff --git a/src/Paragraph.cpp b/src/Paragraph.cpp
index 6aeab1aa7d..991ab14a75 100644
--- a/src/Paragraph.cpp
+++ b/src/Paragraph.cpp
@@ -49,6 +49,7 @@
 #include "insets/InsetBibitem.h"
 #include "insets/InsetLabel.h"
 
+#include "support/lassert.h"
 #include "support/convert.h"
 #include "support/debug.h"
 #include "support/gettext.h"
@@ -82,6 +83,8 @@ public:
 	Private(Paragraph * owner, Layout const & layout);
 	/// "Copy constructor"
 	Private(Private const &, Paragraph * owner);
+	/// Copy constructor from \p beg  to \p end
+	Private(Private const &, Paragraph * owner, pos_type beg, pos_type end);
 
 	///
 	void insertChar(pos_type pos, char_type c, Change const & change);
@@ -246,10 +249,39 @@ Paragraph::Private::Private(Private const & p, Paragraph * owner)
 }
 
 
+Paragraph::Private::Private(Private const & p, Paragraph * owner,
+	pos_type beg, pos_type end)
+	: owner_(owner), inset_owner_(p.inset_owner_),
+	  insetlist_(p.insetlist_, beg, end),
+	  params_(p.params_), changes_(p.changes_),
+	  begin_of_body_(p.begin_of_body_), words_(p.words_),
+	  layout_(p.layout_)
+{
+	id_ = paragraph_id++;
+	if (beg >= pos_type(p.text_.size()))
+		return;
+	text_ = p.text_.substr(beg, end - beg);
+
+	FontList::const_iterator fcit = fontlist_.begin();
+	FontList::const_iterator fend = fontlist_.end();
+	for (; fcit != fend; ++fcit) {
+		if (fcit->pos() < beg)
+			continue;
+		if (fcit->pos() >= end) {
+			// Add last entry in the fontlist_.
+			fontlist_.set(text_.size() - 1, fcit->font());
+			break;
+		}
+		// Add a new entry in the fontlist_.
+		fontlist_.set(fcit->pos() - beg, fcit->font());
+	}
+}
+
+
 bool Paragraph::isChanged(pos_type start, pos_type end) const
 {
-	BOOST_ASSERT(start >= 0 && start <= size());
-	BOOST_ASSERT(end > start && end <= size() + 1);
+	LASSERT(start >= 0 && start <= size(), /**/);
+	LASSERT(end > start && end <= size() + 1, /**/);
 
 	return d->changes_.isChanged(start, end);
 }
@@ -294,7 +326,7 @@ void Paragraph::setChange(Change const & change)
 
 void Paragraph::setChange(pos_type pos, Change const & change)
 {
-	BOOST_ASSERT(pos >= 0 && pos <= size());
+	LASSERT(pos >= 0 && pos <= size(), /**/);
 	d->changes_.set(change, pos);
 
 	// see comment in setChange(Change const &) above
@@ -306,7 +338,7 @@ void Paragraph::setChange(pos_type pos, Change const & change)
 
 Change const & Paragraph::lookupChange(pos_type pos) const
 {
-	BOOST_ASSERT(pos >= 0 && pos <= size());
+	LASSERT(pos >= 0 && pos <= size(), /**/);
 	return d->changes_.lookup(pos);
 }
 
@@ -314,8 +346,8 @@ Change const & Paragraph::lookupChange(pos_type pos) const
 void Paragraph::acceptChanges(BufferParams const & bparams, pos_type start,
 		pos_type end)
 {
-	BOOST_ASSERT(start >= 0 && start <= size());
-	BOOST_ASSERT(end > start && end <= size() + 1);
+	LASSERT(start >= 0 && start <= size(), /**/);
+	LASSERT(end > start && end <= size() + 1, /**/);
 
 	for (pos_type pos = start; pos < end; ++pos) {
 		switch (lookupChange(pos).type) {
@@ -350,8 +382,8 @@ void Paragraph::acceptChanges(BufferParams const & bparams, pos_type start,
 void Paragraph::rejectChanges(BufferParams const & bparams,
 		pos_type start, pos_type end)
 {
-	BOOST_ASSERT(start >= 0 && start <= size());
-	BOOST_ASSERT(end > start && end <= size() + 1);
+	LASSERT(start >= 0 && start <= size(), /**/);
+	LASSERT(end > start && end <= size() + 1, /**/);
 
 	for (pos_type pos = start; pos < end; ++pos) {
 		switch (lookupChange(pos).type) {
@@ -386,7 +418,7 @@ void Paragraph::rejectChanges(BufferParams const & bparams,
 void Paragraph::Private::insertChar(pos_type pos, char_type c,
 		Change const & change)
 {
-	BOOST_ASSERT(pos >= 0 && pos <= int(text_.size()));
+	LASSERT(pos >= 0 && pos <= int(text_.size()), /**/);
 
 	// track change
 	changes_.insert(change, pos);
@@ -412,11 +444,11 @@ void Paragraph::Private::insertChar(pos_type pos, char_type c,
 void Paragraph::insertInset(pos_type pos, Inset * inset,
 				   Change const & change)
 {
-	BOOST_ASSERT(inset);
-	BOOST_ASSERT(pos >= 0 && pos <= size());
+	LASSERT(inset, /**/);
+	LASSERT(pos >= 0 && pos <= size(), /**/);
 
 	d->insertChar(pos, META_INSET, change);
-	BOOST_ASSERT(d->text_[pos] == META_INSET);
+	LASSERT(d->text_[pos] == META_INSET, /**/);
 
 	// Add a new entry in the insetlist_.
 	d->insetlist_.insert(inset, pos);
@@ -425,7 +457,7 @@ void Paragraph::insertInset(pos_type pos, Inset * inset,
 
 bool Paragraph::eraseChar(pos_type pos, bool trackChanges)
 {
-	BOOST_ASSERT(pos >= 0 && pos <= size());
+	LASSERT(pos >= 0 && pos <= size(), /**/);
 
 	// keep the logic here in sync with the logic of isMergedOnEndOfParDeletion()
 
@@ -474,8 +506,8 @@ bool Paragraph::eraseChar(pos_type pos, bool trackChanges)
 
 int Paragraph::eraseChars(pos_type start, pos_type end, bool trackChanges)
 {
-	BOOST_ASSERT(start >= 0 && start <= size());
-	BOOST_ASSERT(end >= start && end <= size() + 1);
+	LASSERT(start >= 0 && start <= size(), /**/);
+	LASSERT(end >= start && end <= size() + 1, /**/);
 
 	pos_type i = start;
 	for (pos_type count = end - start; count; --count) {
@@ -498,7 +530,12 @@ int Paragraph::Private::latexSurrogatePair(odocstream & os, char_type c,
 	// Is this correct WRT change tracking?
 	docstring const latex1 = encoding.latexChar(next);
 	docstring const latex2 = encoding.latexChar(c);
-	os << latex1 << '{' << latex2 << '}';
+	if (docstring(1, next) == latex1) {
+		// the encoding supports the combination
+		os << latex2 << latex1;
+		return latex1.length() + latex2.length();
+	} else
+		os << latex1 << '{' << latex2 << '}';
 	return latex1.length() + latex2.length() + 2;
 }
 
@@ -560,9 +597,10 @@ int Paragraph::Private::writeScriptChars(odocstream & os,
 
 	// We only arrive here when a proper language for character text_[i] has
 	// not been specified (i.e., it could not be translated in the current
-	// latex encoding) and it belongs to a known script.
-	// Parameter ltx contains the latex translation of text_[i] as specified in
-	// the unicodesymbols file and is something like "\textXXX{<spec>}".
+	// latex encoding) or its latex translation has been forced, and it
+	// belongs to a known script.
+	// Parameter ltx contains the latex translation of text_[i] as specified
+	// in the unicodesymbols file and is something like "\textXXX{<spec>}".
 	// The latex macro name "textXXX" specifies the script to which text_[i]
 	// belongs and we use it in order to check whether characters from the
 	// same script immediately follow, such that we can collect them in a
@@ -571,8 +609,16 @@ int Paragraph::Private::writeScriptChars(odocstream & os,
 	docstring::size_type const brace1 = ltx.find_first_of(from_ascii("{"));
 	docstring::size_type const brace2 = ltx.find_last_of(from_ascii("}"));
 	string script = to_ascii(ltx.substr(1, brace1 - 1));
-	int length = ltx.substr(0, brace2).length();
-	os << ltx.substr(0, brace2);
+	int pos = 0;
+	int length = brace2;
+	bool closing_brace = true;
+	if (script == "textgreek" && encoding.latexName() == "iso-8859-7") {
+		// Correct encoding is being used, so we can avoid \textgreek.
+		pos = brace1 + 1;
+		length -= pos;
+		closing_brace = false;
+	}
+	os << ltx.substr(pos, length);
 	int size = text_.size();
 	while (i + 1 < size) {
 		char_type const next = text_[i + 1];
@@ -606,8 +652,10 @@ int Paragraph::Private::writeScriptChars(odocstream & os,
 		length += len;
 		++i;
 	}
-	os << '}';
-	++length;
+	if (closing_brace) {
+		os << '}';
+		++length;
+	}
 	return length;
 }
 
@@ -647,7 +695,7 @@ void Paragraph::Private::latexInset(
 					     unsigned int & column)
 {
 	Inset * inset = owner_->getInset(i);
-	BOOST_ASSERT(inset);
+	LASSERT(inset, /**/);
 
 	if (style.pass_thru) {
 		inset->plaintext(os, runparams);
@@ -743,7 +791,16 @@ void Paragraph::Private::latexInset(
 		}
 	}
 
-	int tmp = inset->latex(os, runparams);
+	int tmp;
+
+	try {
+		tmp = inset->latex(os, runparams);
+	} catch (EncodingException & e) {
+		// add location information and throw again.
+		e.par_id = id_;
+		e.pos = i;
+		throw(e);
+	}
 
 	if (close) {
     	if (running_font.language()->lang() == "farsi")
@@ -787,6 +844,8 @@ void Paragraph::Private::latexSpecialChar(
 	}
 
 	if (runparams.verbatim) {
+		// FIXME UNICODE: This can fail if c cannot
+		// be encoded in the current encoding.
 		os.put(c);
 		return;
 	}
@@ -859,7 +918,6 @@ void Paragraph::Private::latexSpecialChar(
 		break;
 
 	default:
-
 		// LyX, LaTeX etc.
 		if (latexSpecialPhrase(os, i, column, runparams))
 			return;
@@ -908,11 +966,8 @@ bool Paragraph::Private::latexSpecialT1(char_type const c, odocstream & os,
 		// but we should avoid ligatures
 		if (i + 1 >= int(text_.size()) || text_[i + 1] != c)
 			return true;
-		os << "\\,{}";
-		column += 3;
-		// Alternative code:
-		//os << "\\textcompwordmark{}";
-		//column += 19;
+		os << "\\textcompwordmark{}";
+		column += 19;
 		return true;
 	case '|':
 		os.put(c);
@@ -928,37 +983,19 @@ bool Paragraph::Private::latexSpecialTypewriter(char_type const c, odocstream &
 {
 	switch (c) {
 	case '-':
+		// within \ttfamily, "--" is merged to "-" (no endash)
+		// so we avoid this rather irritating ligature
 		if (i + 1 < int(text_.size()) && text_[i + 1] == '-') {
-			// "--" in Typewriter mode -> "-{}-"
 			os << "-{}";
 			column += 2;
 		} else
 			os << '-';
 		return true;
 
-	// I assume this is hack treating typewriter as verbatim
-	// FIXME UNICODE: This can fail if c cannot be encoded
-	// in the current encoding.
-
-	case '\0':
-		return true;
-
-	// Those characters are not directly supported.
-	case '\\':
-	case '\"':
-	case '$': case '&':
-	case '%': case '#': case '{':
-	case '}': case '_':
-	case '~':
-	case '^':
-	case '*': case '[':
-	case ' ':
-		return false;
-
+	// everything else has to be checked separately
+	// (depending on the encoding)
 	default:
-		// With Typewriter font, these characters exist.
-		os.put(c);
-		return true;
+		return false;
 	}
 }
 
@@ -1054,6 +1091,14 @@ Paragraph::Paragraph(Paragraph const & par)
 }
 
 
+Paragraph::Paragraph(Paragraph const & par, pos_type beg, pos_type end)
+	: itemdepth(par.itemdepth),
+	d(new Paragraph::Private(*par.d, this, beg, end))
+{
+	registerWords();
+}
+
+
 Paragraph & Paragraph::operator=(Paragraph const & par)
 {
 	// needed as we will destroy the private part before copying it
@@ -1162,8 +1207,7 @@ void Paragraph::write(ostream & os, BufferParams const & bparams,
 			if (c != '\0')
 				os << to_utf8(docstring(1, c));
 			else
-				lyxerr << "ERROR (Paragraph::writeFile):"
-					" NULL char in structure." << endl;
+				LYXERR0("NUL char in structure.");
 			++column;
 			break;
 		}
@@ -1212,7 +1256,7 @@ void Paragraph::appendString(docstring const & s, Font const & font,
 	d->text_.append(s);
 
 	// FIXME: Optimize this!
-	for (pos_type i = 0; i != end; ++i) {
+	for (pos_type i = oldsize; i != newsize; ++i) {
 		// track change
 		d->changes_.insert(change, i);
 	}
@@ -1273,8 +1317,8 @@ Font const Paragraph::getFontSettings(BufferParams const & bparams,
 					 pos_type pos) const
 {
 	if (pos > size()) {
-		lyxerr << " pos: " << pos << " size: " << size() << endl;
-		BOOST_ASSERT(pos <= size());
+		LYXERR0("pos: " << pos << " size: " << size());
+		LASSERT(pos <= size(), /**/);
 	}
 
 	FontList::const_iterator cit = d->fontlist_.fontIterator(pos);
@@ -1290,7 +1334,7 @@ Font const Paragraph::getFontSettings(BufferParams const & bparams,
 
 FontSpan Paragraph::fontSpan(pos_type pos) const
 {
-	BOOST_ASSERT(pos <= size());
+	LASSERT(pos <= size(), /**/);
 	pos_type start = 0;
 
 	FontList::const_iterator cit = d->fontlist_.begin();
@@ -1309,8 +1353,7 @@ FontSpan Paragraph::fontSpan(pos_type pos) const
 	}
 
 	// This should not happen, but if so, we take no chances.
-	//lyxerr << "Paragraph::getEndPosOfFontSpan: This should not happen!"
-	//      << endl;
+	// LYXERR0("Paragraph::getEndPosOfFontSpan: This should not happen!");
 	return FontSpan(pos, pos);
 }
 
@@ -1333,7 +1376,7 @@ Font const Paragraph::getFirstFontSettings(BufferParams const & bparams) const
 Font const Paragraph::getFont(BufferParams const & bparams, pos_type pos,
 				 Font const & outerfont) const
 {
-	BOOST_ASSERT(pos >= 0);
+	LASSERT(pos >= 0, /**/);
 
 	Font font = getFontSettings(bparams, pos);
 
@@ -1378,8 +1421,7 @@ FontSize Paragraph::highestFontInRange
 }
 
 
-char_type
-Paragraph::getUChar(BufferParams const & bparams, pos_type pos) const
+char_type Paragraph::getUChar(BufferParams const & bparams, pos_type pos) const
 {
 	char_type c = d->text_[pos];
 	if (!lyxrc.rtl_support)
@@ -1414,14 +1456,13 @@ Paragraph::getUChar(BufferParams const & bparams, pos_type pos) const
 	}
 	if (uc != c && getFontSettings(bparams, pos).isRightToLeft())
 		return uc;
-	else
-		return c;
+	return c;
 }
 
 
 void Paragraph::setFont(pos_type pos, Font const & font)
 {
-	BOOST_ASSERT(pos <= size());
+	LASSERT(pos <= size(), /**/);
 
 	// First, reduce font against layout/label font
 	// Update: The setCharFont() routine in text2.cpp already
@@ -1614,12 +1655,12 @@ void Paragraph::setBeginOfBody()
 }
 
 
-bool Paragraph::forceEmptyLayout() const
+bool Paragraph::forcePlainLayout() const
 {
 	Inset const * const inset = inInset();
 	if (!inset)
 		return true;
-	return inset->forceEmptyLayout();
+	return inset->forcePlainLayout();
 }
 
 
@@ -1632,12 +1673,12 @@ bool Paragraph::allowParagraphCustomization() const
 }
 
 
-bool Paragraph::useEmptyLayout() const
+bool Paragraph::usePlainLayout() const
 {
 	Inset const * const inset = inInset();
 	if (!inset)
 		return false;
-	return inset->useEmptyLayout();
+	return inset->usePlainLayout();
 }
 
 
@@ -1831,10 +1872,10 @@ bool Paragraph::latex(BufferParams const & bparams,
 
 	bool return_value = false;
 
-	bool asdefault = forceEmptyLayout();
+	bool asdefault = forcePlainLayout();
 
 	Layout const & style = asdefault ?
-		bparams.documentClass().emptyLayout() :
+		bparams.documentClass().plainLayout() :
 		*d->layout_;
 
 	// Current base font for all inherited font changes, without any
@@ -1963,8 +2004,8 @@ bool Paragraph::latex(BufferParams const & bparams,
 
 		// Switch file encoding if necessary (and allowed)
 		if (!runparams.verbatim && 
-		    runparams.encoding->package() == Encoding::none &&
-		    font.language()->encoding()->package() == Encoding::none) {
+		    runparams.encoding->package() != Encoding::none &&
+		    font.language()->encoding()->package() != Encoding::none) {
 			pair<bool, int> const enc_switch = switchEncoding(os, bparams,
 					runparams, *(font.language()->encoding()));
 			if (enc_switch.first) {
@@ -2048,7 +2089,7 @@ bool Paragraph::latex(BufferParams const & bparams,
 			}
 		}
 
-		// Set the encoding to that returned from simpleTeXSpecialChars (see
+		// Set the encoding to that returned from latexSpecialChar (see
 		// comment for encoding member in OutputParams.h)
 		runparams.encoding = rp.encoding;
 	}
@@ -2257,6 +2298,24 @@ bool Paragraph::isLetter(pos_type pos) const
 }
 
 
+bool Paragraph::isChar(pos_type pos) const
+{
+	if (Inset const * inset = getInset(pos))
+		return inset->isChar();
+	char_type const c = d->text_[pos];
+	return !isLetterChar(c) && !isDigit(c) && !lyx::isSpace(c);
+}
+
+
+bool Paragraph::isSpace(pos_type pos) const
+{
+	if (Inset const * inset = getInset(pos))
+		return inset->isSpace();
+	char_type const c = d->text_[pos];
+	return lyx::isSpace(c);
+}
+
+
 Language const *
 Paragraph::getParLanguage(BufferParams const & bparams) const
 {
@@ -2305,25 +2364,26 @@ bool Paragraph::isMultiLingual(BufferParams const & bparams) const
 }
 
 
-docstring Paragraph::asString(bool label) const
+docstring Paragraph::asString(int options) const
 {
-	return asString(0, size(), label);
+	return asString(0, size(), options);
 }
 
 
-docstring Paragraph::asString(pos_type beg, pos_type end, bool label) const
+docstring Paragraph::asString(pos_type beg, pos_type end, int options) const
 {
-
 	odocstringstream os;
 
-	if (beg == 0 && label && !d->params_.labelString().empty())
+	if (beg == 0 
+		&& options & AS_STR_LABEL
+		&& !d->params_.labelString().empty())
 		os << d->params_.labelString() << ' ';
 
 	for (pos_type i = beg; i < end; ++i) {
 		char_type const c = d->text_[i];
 		if (isPrintable(c))
 			os.put(c);
-		else if (c == META_INSET)
+		else if (c == META_INSET && options & AS_STR_INSETS)
 			getInset(i)->textString(os);
 	}
 
@@ -2355,10 +2415,10 @@ void Paragraph::setLayout(Layout const & layout)
 }
 
 
-void Paragraph::setEmptyOrDefaultLayout(DocumentClass const & tclass)
+void Paragraph::setPlainOrDefaultLayout(DocumentClass const & tclass)
 {
-	if (useEmptyLayout())
-		setLayout(tclass.emptyLayout());
+	if (usePlainLayout())
+		setLayout(tclass.plainLayout());
 	else
 		setLayout(tclass.defaultLayout());
 }
@@ -2406,7 +2466,7 @@ bool Paragraph::allowEmpty() const
 
 char_type Paragraph::transformChar(char_type c, pos_type pos) const
 {
-	if (!Encodings::is_arabic(c))
+	if (!Encodings::isArabicChar(c))
 		return c;
 
 	char_type prev_char = ' ';
@@ -2414,7 +2474,7 @@ char_type Paragraph::transformChar(char_type c, pos_type pos) const
 
 	for (pos_type i = pos - 1; i >= 0; --i) {
 		char_type const par_char = d->text_[i];
-		if (!Encodings::isComposeChar_arabic(par_char)) {
+		if (!Encodings::isArabicComposeChar(par_char)) {
 			prev_char = par_char;
 			break;
 		}
@@ -2422,21 +2482,21 @@ char_type Paragraph::transformChar(char_type c, pos_type pos) const
 
 	for (pos_type i = pos + 1, end = size(); i < end; ++i) {
 		char_type const par_char = d->text_[i];
-		if (!Encodings::isComposeChar_arabic(par_char)) {
+		if (!Encodings::isArabicComposeChar(par_char)) {
 			next_char = par_char;
 			break;
 		}
 	}
 
-	if (Encodings::is_arabic(next_char)) {
-		if (Encodings::is_arabic(prev_char) &&
-			!Encodings::is_arabic_special(prev_char))
+	if (Encodings::isArabicChar(next_char)) {
+		if (Encodings::isArabicChar(prev_char) &&
+			!Encodings::isArabicSpecialChar(prev_char))
 			return Encodings::transformChar(c, Encodings::FORM_MEDIAL);
 		else
 			return Encodings::transformChar(c, Encodings::FORM_INITIAL);
 	} else {
-		if (Encodings::is_arabic(prev_char) &&
-			!Encodings::is_arabic_special(prev_char))
+		if (Encodings::isArabicChar(prev_char) &&
+			!Encodings::isArabicSpecialChar(prev_char))
 			return Encodings::transformChar(c, Encodings::FORM_FINAL);
 		else
 			return Encodings::transformChar(c, Encodings::FORM_ISOLATED);
@@ -2501,8 +2561,8 @@ int Paragraph::checkBiblio(Buffer const & buffer)
 
 	// There was no inset at the beginning, so we need to create one with
 	// the key and label of the one we erased.
-	InsetBibitem * inset = new InsetBibitem(InsetCommandParams(BIBITEM_CODE));
-	inset->setBuffer(const_cast<Buffer &>(buffer));
+	InsetBibitem * inset = 
+		new InsetBibitem(buffer, InsetCommandParams(BIBITEM_CODE));
 	// restore values of previously deleted item in this par.
 	if (!oldkey.empty())
 		inset->setParam("key", oldkey);
@@ -2544,6 +2604,12 @@ InsetList const & Paragraph::insetList() const
 }
 
 
+void Paragraph::setBuffer(Buffer & b)
+{
+	d->insetlist_.setBuffer(b);
+}
+
+
 Inset * Paragraph::releaseInset(pos_type pos)
 {
 	Inset * inset = d->insetlist_.release(pos);
@@ -2752,7 +2818,7 @@ void Paragraph::registerWords()
 
 void Paragraph::updateWords(CursorSlice const & sl)
 {
-	BOOST_ASSERT(&sl.paragraph() == this);
+	LASSERT(&sl.paragraph() == this, /**/);
 	deregisterWords();
 	collectWords(sl);
 	registerWords();