#include "TextClass.h"
#include "TexRow.h"
#include "Text.h"
-#include "VSpace.h"
#include "WordLangTuple.h"
#include "WordList.h"
if (range_.first > pos) {
range_.first += offset;
range_.last += offset;
- } else if (range_.last > pos) {
+ } else if (range_.last >= pos) {
range_.last += offset;
}
}
/// Output the surrogate pair formed by \p c and \p next to \p os.
/// \return the number of characters written.
- int latexSurrogatePair(odocstream & os, char_type c, char_type next,
+ int latexSurrogatePair(otexstream & os, char_type c, char_type next,
OutputParams const &);
/// Output a space in appropriate formatting (or a surrogate pair
/// if the next character is a combining character).
/// \return whether a surrogate pair was output.
bool simpleTeXBlanks(OutputParams const &,
- odocstream &, TexRow & texrow,
+ otexstream &,
pos_type i,
unsigned int & column,
Font const & font,
/// Output consecutive unicode chars, belonging to the same script as
/// specified by the latex macro \p ltx, to \p os starting from \p i.
/// \return the number of characters written.
- int writeScriptChars(odocstream & os, docstring const & ltx,
+ int writeScriptChars(otexstream & os, docstring const & ltx,
Change const &, Encoding const &, pos_type & i);
/// This could go to ParagraphParameters if we want to.
- int startTeXParParams(BufferParams const &, odocstream &, TexRow &,
+ int startTeXParParams(BufferParams const &, otexstream &,
OutputParams const &) const;
/// This could go to ParagraphParameters if we want to.
- int endTeXParParams(BufferParams const &, odocstream &, TexRow &,
- OutputParams const &) const;
+ bool endTeXParParams(BufferParams const &, otexstream &,
+ OutputParams const &) const;
///
void latexInset(BufferParams const &,
- odocstream &,
- TexRow & texrow, OutputParams &,
+ otexstream &,
+ OutputParams &,
Font & running_font,
Font & basefont,
Font const & outerfont,
///
void latexSpecialChar(
- odocstream & os,
+ otexstream & os,
OutputParams const & runparams,
Font const & running_font,
Change const & running_change,
Layout const & style,
pos_type & i,
+ pos_type end_pos,
unsigned int & column);
///
bool latexSpecialT1(
char_type const c,
- odocstream & os,
+ otexstream & os,
pos_type i,
unsigned int & column);
///
bool latexSpecialTypewriter(
char_type const c,
- odocstream & os,
+ otexstream & os,
pos_type i,
unsigned int & column);
///
bool latexSpecialPhrase(
- odocstream & os,
+ otexstream & os,
pos_type & i,
+ pos_type end_pos,
unsigned int & column,
OutputParams const & runparams);
}
void requestSpellCheck(pos_type pos) {
- speller_state_.needsRefresh(pos);
+ if (pos == -1)
+ speller_state_.needsCompleteRefresh(speller_state_.currentChangeNumber());
+ else
+ speller_state_.needsRefresh(pos);
}
void readySpellCheck() {
}
-int Paragraph::Private::latexSurrogatePair(odocstream & os, char_type c,
+int Paragraph::Private::latexSurrogatePair(otexstream & os, char_type c,
char_type next, OutputParams const & runparams)
{
// Writing next here may circumvent a possible font change between
bool Paragraph::Private::simpleTeXBlanks(OutputParams const & runparams,
- odocstream & os, TexRow & texrow,
+ otexstream & os,
pos_type i,
unsigned int & column,
Font const & font,
|| text_[i - 1] == ':'
|| text_[i - 1] == '!'))) {
os << '\n';
- texrow.newline();
- texrow.start(owner_->id(), i + 1);
+ os.texrow().start(owner_->id(), i + 1);
column = 0;
} else if (style.free_spacing) {
os << '~';
}
-int Paragraph::Private::writeScriptChars(odocstream & os,
+int Paragraph::Private::writeScriptChars(otexstream & os,
docstring const & ltx,
Change const & runningChange,
Encoding const & encoding,
void Paragraph::Private::latexInset(BufferParams const & bparams,
- odocstream & os,
- TexRow & texrow,
+ otexstream & os,
OutputParams & runparams,
Font & running_font,
Font & basefont,
LASSERT(inset, /**/);
if (style.pass_thru) {
- inset->plaintext(os, runparams);
+ inset->plaintext(os.os(), runparams);
return;
}
os << "\\protect ";
}
- texrow.newline();
- texrow.start(owner_->id(), i + 1);
+ os.texrow().start(owner_->id(), i + 1);
column = 0;
}
}
bool close = false;
- odocstream::pos_type const len = os.tellp();
+ odocstream::pos_type const len = os.os().tellp();
if (inset->forceLTR()
&& running_font.isRightToLeft()
// ArabTeX, though, cannot handle this special behavior, it seems.
bool arabtex = basefont.language()->lang() == "arabic_arabtex"
|| running_font.language()->lang() == "arabic_arabtex";
- if (open_font && inset->noFontChange()) {
+ if (open_font && !inset->inheritFont()) {
bool closeLanguage = arabtex
|| basefont.isRightToLeft() == running_font.isRightToLeft();
unsigned int count = running_font.latexWriteEndChanges(os,
}
}
- int tmp;
+ int prev_rows = os.texrow().rows();
try {
- tmp = inset->latex(os, runparams);
+ runparams.lastid = id_;
+ runparams.lastpos = i;
+ inset->latex(os, runparams);
} catch (EncodingException & e) {
// add location information and throw again.
e.par_id = id_;
os << '}';
}
- if (tmp) {
- texrow.newlines(tmp);
- texrow.start(owner_->id(), i + 1);
+ if (os.texrow().rows() > prev_rows) {
+ os.texrow().start(owner_->id(), i + 1);
column = 0;
} else {
- column += (unsigned int)(os.tellp() - len);
+ column += (unsigned int)(os.os().tellp() - len);
}
if (owner_->isDeleted(i))
}
-void Paragraph::Private::latexSpecialChar(
- odocstream & os,
- OutputParams const & runparams,
- Font const & running_font,
- Change const & running_change,
- Layout const & style,
- pos_type & i,
- unsigned int & column)
+void Paragraph::Private::latexSpecialChar(otexstream & os,
+ OutputParams const & runparams,
+ Font const & running_font,
+ Change const & running_change,
+ Layout const & style,
+ pos_type & i,
+ pos_type end_pos,
+ unsigned int & column)
{
char_type const c = text_[i];
if (style.pass_thru || runparams.pass_thru) {
- if (c != '\0')
- // FIXME UNICODE: This can fail if c cannot
- // be encoded in the current encoding.
+ if (c != '\0') {
+ Encoding const * const enc = runparams.encoding;
+ if (enc && enc->latexChar(c, true).empty())
+ throw EncodingException(c);
os.put(c);
+ }
return;
}
default:
// LyX, LaTeX etc.
- if (latexSpecialPhrase(os, i, column, runparams))
+ if (latexSpecialPhrase(os, i, end_pos, column, runparams))
return;
if (c == '\0')
}
-bool Paragraph::Private::latexSpecialT1(char_type const c, odocstream & os,
+bool Paragraph::Private::latexSpecialT1(char_type const c, otexstream & os,
pos_type i, unsigned int & column)
{
switch (c) {
}
-bool Paragraph::Private::latexSpecialTypewriter(char_type const c, odocstream & os,
+bool Paragraph::Private::latexSpecialTypewriter(char_type const c, otexstream & os,
pos_type i, unsigned int & column)
{
switch (c) {
}
-bool Paragraph::Private::latexSpecialPhrase(odocstream & os, pos_type & i,
+/// \param end_pos
+/// If [start_pos, end_pos) does not include entirely the special phrase, then
+/// do not apply the macro transformation.
+bool Paragraph::Private::latexSpecialPhrase(otexstream & os, pos_type & i, pos_type end_pos,
unsigned int & column, OutputParams const & runparams)
{
// FIXME: if we have "LaTeX" with a font
// "words" for some definition of word
for (size_t pnr = 0; pnr < phrases_nr; ++pnr) {
- if (!isTextAt(special_phrases[pnr].phrase, i))
+ if (!isTextAt(special_phrases[pnr].phrase, i)
+ || (end_pos != -1 && i + int(special_phrases[pnr].phrase.size()) > end_pos))
continue;
if (runparams.moving_arg)
os << "\\protect";
Buffer const & buf = inset_owner_->buffer();
BufferParams const & bp = buf.params();
Font f;
- TexRow tr;
+ TexRow texrow;
// Using a string stream here circumvents the encoding
// switching machinery of odocstream. Therefore the
// output is wrong if this paragraph contains content
// that needs to switch encoding.
odocstringstream ods;
+ otexstream os(ods, texrow);
if (is_command) {
- ods << '\\' << from_ascii(layout_->latexname());
+ os << '\\' << from_ascii(layout_->latexname());
// we have to provide all the optional arguments here, even though
// the last one is the only one we care about.
// Separate handling of optional argument inset.
if (layout_->optargs != 0 || layout_->reqargs != 0)
- latexArgInsets(*owner_, ods, features.runparams(),
- layout_->reqargs, layout_->optargs);
+ latexArgInsets(*owner_, os, features.runparams(),
+ layout_->reqargs, layout_->optargs);
else
- ods << from_ascii(layout_->latexparam());
+ os << from_ascii(layout_->latexparam());
}
docstring::size_type const length = ods.str().length();
// this will output "{" at the beginning, but not at the end
- owner_->latex(bp, f, ods, tr, features.runparams(), 0, -1, true);
+ owner_->latex(bp, f, os, features.runparams(), 0, -1, true);
if (ods.str().length() > length) {
if (is_command)
ods << '}';
// when appending characters, no need to update tables
d->text_.push_back(c);
setFont(d->text_.size() - 1, font);
+ d->requestSpellCheck(d->text_.size() - 1);
}
for (size_t i = oldsize; i != newsize; ++i) {
// track change
d->changes_.insert(change, i);
+ d->requestSpellCheck(i);
}
d->fontlist_.set(oldsize, font);
d->fontlist_.set(newsize - 1, font);
}
-void adjust_row_column(string const & str, TexRow & texrow, int & column)
+void adjust_column(string const & str, int & column)
{
if (!contains(str, "\n"))
column += str.size();
else {
string tmp;
- texrow.newline();
column = rsplit(str, tmp, '\n').size();
}
}
int Paragraph::Private::startTeXParParams(BufferParams const & bparams,
- odocstream & os, TexRow & texrow,
- OutputParams const & runparams) const
+ otexstream & os, OutputParams const & runparams) const
{
int column = 0;
else
output = corrected_env(begin_tag, "flushright", code, lastpar);
os << from_ascii(output);
- adjust_row_column(output, texrow, column);
+ adjust_column(output, column);
break;
} case LYX_ALIGN_RIGHT: {
string output;
else
output = corrected_env(begin_tag, "flushleft", code, lastpar);
os << from_ascii(output);
- adjust_row_column(output, texrow, column);
+ adjust_column(output, column);
break;
} case LYX_ALIGN_CENTER: {
string output;
output = corrected_env(begin_tag, "center", code, lastpar);
os << from_ascii(output);
- adjust_row_column(output, texrow, column);
+ adjust_column(output, column);
break;
}
}
}
-int Paragraph::Private::endTeXParParams(BufferParams const & bparams,
- odocstream & os, TexRow & texrow,
- OutputParams const & runparams) const
+bool Paragraph::Private::endTeXParParams(BufferParams const & bparams,
+ otexstream & os, OutputParams const & runparams) const
{
- int column = 0;
-
LyXAlignment const curAlign = params_.align();
if (curAlign == layout_->align)
- return column;
+ return false;
switch (curAlign) {
case LYX_ALIGN_NONE:
case LYX_ALIGN_LEFT:
case LYX_ALIGN_RIGHT:
case LYX_ALIGN_CENTER:
- if (runparams.moving_arg) {
+ if (runparams.moving_arg)
os << "\\protect";
- column = 8;
- }
break;
}
+ string output;
string const end_tag = "\n\\par\\end";
InsetCode code = ownerCode();
bool const lastpar = runparams.isLastPar;
case LYX_ALIGN_DECIMAL:
break;
case LYX_ALIGN_LEFT: {
- string output;
if (owner_->getParLanguage(bparams)->babel() != "hebrew")
output = corrected_env(end_tag, "flushleft", code, lastpar);
else
output = corrected_env(end_tag, "flushright", code, lastpar);
os << from_ascii(output);
- adjust_row_column(output, texrow, column);
break;
} case LYX_ALIGN_RIGHT: {
- string output;
if (owner_->getParLanguage(bparams)->babel() != "hebrew")
output = corrected_env(end_tag, "flushright", code, lastpar);
else
output = corrected_env(end_tag, "flushleft", code, lastpar);
os << from_ascii(output);
- adjust_row_column(output, texrow, column);
break;
} case LYX_ALIGN_CENTER: {
- string output;
output = corrected_env(end_tag, "center", code, lastpar);
os << from_ascii(output);
- adjust_row_column(output, texrow, column);
break;
}
}
- return column;
+ return !output.empty() || lastpar;
}
// This one spits out the text of the paragraph
void Paragraph::latex(BufferParams const & bparams,
Font const & outerfont,
- odocstream & os, TexRow & texrow,
+ otexstream & os,
OutputParams const & runparams,
int start_pos, int end_pos, bool force) const
{
if (body_pos > 0) {
// the optional argument is kept in curly brackets in
// case it contains a ']'
+ // This is not strictly needed, but if this is changed it
+ // would be a file format change, and tex2lyx would need
+ // to be adjusted, since it unconditionally removes the
+ // braces when it parses \item.
os << "[{";
column += 2;
basefont = getLabelFont(bparams, outerfont);
Encoding const * const prev_encoding = runparams.encoding;
- texrow.start(id(), 0);
+ os.texrow().start(id(), 0);
// if the paragraph is empty, the loop will not be entered at all
if (empty()) {
++column;
}
if (allowcust)
- column += d->startTeXParParams(bparams, os, texrow,
- runparams);
+ column += d->startTeXParParams(bparams, os, runparams);
}
for (pos_type i = 0; i < size(); ++i) {
if (allowcust)
column += d->startTeXParParams(bparams, os,
- texrow,
runparams);
}
- Change const & change = runparams.inDeletedInset ? runparams.changeOfDeletedInset
- : lookupChange(i);
+ Change const & change = runparams.inDeletedInset
+ ? runparams.changeOfDeletedInset : lookupChange(i);
if (bparams.outputChanges && runningChange != change) {
if (open_font) {
if (!runparams.pass_thru && !style.pass_thru &&
runparams.encoding->package() != Encoding::none &&
font.language()->encoding()->package() != Encoding::none) {
- pair<bool, int> const enc_switch = switchEncoding(os, bparams,
- runparams, *(font.language()->encoding()));
+ pair<bool, int> const enc_switch =
+ switchEncoding(os.os(), bparams, runparams,
+ *(font.language()->encoding()));
if (enc_switch.first) {
column += enc_switch.second;
runparams.encoding = font.language()->encoding();
// latexSpecialChar ignores spaces if
// style.pass_thru is false.
if (i != body_pos - 1) {
- if (d->simpleTeXBlanks(
- runparams, os, texrow,
+ if (d->simpleTeXBlanks(runparams, os,
i, column, font, style)) {
// A surrogate pair was output. We
// must not call latexSpecialChar
// and then split to handle the two modes separately.
if (c == META_INSET) {
if (i >= start_pos && (end_pos == -1 || i < end_pos)) {
- d->latexInset(bparams, os,
- texrow, rp, running_font,
+ d->latexInset(bparams, os, rp, running_font,
basefont, outerfont, open_font,
runningChange, style, i, column);
}
if (i >= start_pos && (end_pos == -1 || i < end_pos)) {
try {
d->latexSpecialChar(os, rp, running_font, runningChange,
- style, i, column);
+ style, i, end_pos, column);
} catch (EncodingException & e) {
if (runparams.dryrun) {
os << "<" << _("LyX Warning: ")
os << "}]~";
}
- if (allowcust && d->endTeXParParams(bparams, os, texrow, runparams)
+ if (allowcust && d->endTeXParParams(bparams, os, runparams)
&& runparams.encoding != prev_encoding) {
runparams.encoding = prev_encoding;
if (!runparams.isFullUnicode())
bool emph_flag = false;
bool bold_flag = false;
- string closing_tag;
Layout const & style = *d->layout_;
+ xs.startParagraph(allowEmpty());
+
if (!runparams.for_toc && runparams.html_make_pars) {
// generate a magic label for this paragraph
string const attr = "id='" + magicLabel() + "'";
}
xs.closeFontTags();
+ xs.endParagraph();
return retval;
}
bool Paragraph::isWordSeparator(pos_type pos) const
{
+ if (pos == size())
+ return true;
if (Inset const * inset = getInset(pos))
return !inset->isLetter();
+ // if we have a hard hyphen (no en- or emdash) or apostrophe
+ // we pass this to the spell checker
+ // FIXME: this method is subject to change, visit
+ // https://bugzilla.mozilla.org/show_bug.cgi?id=355178
+ // to get an impression how complex this is.
+ if (isHardHyphenOrApostrophe(pos))
+ return false;
char_type const c = d->text_[pos];
- // We want to pass the ' and escape chars to the spellchecker
- static docstring const quote = from_utf8(lyxrc.spellchecker_esc_chars + '\'');
- return (!isLetterChar(c) && !isDigitASCII(c) && !contains(quote, c))
- || pos == size();
+ // We want to pass the escape chars to the spellchecker
+ docstring const escape_chars = from_utf8(lyxrc.spellchecker_esc_chars);
+ return !isLetterChar(c) && !isDigitASCII(c) && !contains(escape_chars, c);
+}
+
+
+bool Paragraph::isHardHyphenOrApostrophe(pos_type pos) const
+{
+ pos_type const psize = size();
+ if (pos >= psize)
+ return false;
+ char_type const c = d->text_[pos];
+ if (c != '-' && c != '\'')
+ return false;
+ int nextpos = pos + 1;
+ int prevpos = pos > 0 ? pos - 1 : 0;
+ if ((nextpos == psize || isSpace(nextpos))
+ && (pos == 0 || isSpace(prevpos)))
+ return false;
+ return c == '\''
+ || ((nextpos == psize || d->text_[nextpos] != '-')
+ && (pos == 0 || d->text_[prevpos] != '-'));
}
if (font.language() == from) {
font.setLanguage(to);
setFont(i, font);
+ d->requestSpellCheck(i);
}
}
- d->requestSpellCheck(size());
}
int i = 0;
pos_type const parsize = d->text_.size();
for (i = 0; i < strsize && pos < parsize; ++i, ++pos) {
- // Ignore ligature break and hyphenation chars while searching
+ // Ignore "invisible" letters such as ligature breaks
+ // and hyphenation chars while searching
while (pos < parsize - 1 && isInset(pos)) {
- const InsetSpecialChar *isc = dynamic_cast<const InsetSpecialChar*>(getInset(pos));
- if (isc == 0
- || (isc->kind() != InsetSpecialChar::HYPHENATION
- && isc->kind() != InsetSpecialChar::LIGATURE_BREAK))
+ odocstringstream os;
+ getInset(pos)->toString(os);
+ if (!getInset(pos)->isLetter() || !os.str().empty())
break;
pos++;
}
void Paragraph::collectWords()
{
- // This is the value that needs to be exposed in the preferences
- // to resolve bug #6760.
- static int minlength = 6;
pos_type n = size();
for (pos_type pos = 0; pos < n; ++pos) {
if (isWordSeparator(pos))
continue;
pos_type from = pos;
locateWord(from, pos, WHOLE_WORD);
- if (pos - from >= minlength) {
+ if ((pos - from) >= (int)lyxrc.completion_minlength) {
docstring word = asString(from, pos, AS_STR_NONE);
FontList::const_iterator cit = d->fontlist_.fontIterator(pos);
if (cit == d->fontlist_.end())
void Paragraph::requestSpellCheck(pos_type pos)
{
- d->requestSpellCheck(pos == -1 ? size() : pos);
+ d->requestSpellCheck(pos);
}
return result;
if (needsSpellCheck() || check_learned) {
+ pos_type end = to;
if (!d->ignoreWord(word)) {
bool const trailing_dot = to < size() && d->text_[to] == '.';
result = speller->check(wl);
word << "\" [" <<
from << ".." << to << "]");
} else {
- // spell check with dot appended failed
+ // spell check with dot appended failed too
// restore original word/lang value
word = asString(from, to, AS_STR_INSETS | AS_STR_SKIPDELETE);
wl = WordLangTuple(word, lang);
}
}
}
- d->setMisspelled(from, to, result);
+ if (!SpellChecker::misspelled(result)) {
+ // area up to the begin of the next word is not misspelled
+ while (end < size() && isWordSeparator(end))
+ ++end;
+ }
+ d->setMisspelled(from, end, result);
} else {
result = d->speller_state_.getState(from);
}
- bool const misspelled_ = SpellChecker::misspelled(result) ;
- if (misspelled_ && do_suggestion)
- speller->suggest(wl, suggestions);
- else if (misspelled_)
+ if (do_suggestion)
+ suggestions.clear();
+
+ if (SpellChecker::misspelled(result)) {
LYXERR(Debug::GUI, "misspelled word: \"" <<
word << "\" [" <<
from << ".." << to << "]");
- else
- suggestions.clear();
-
+ if (do_suggestion)
+ speller->suggest(wl, suggestions);
+ }
return result;
}
}
-bool Paragraph::isMisspelled(pos_type pos) const
+bool Paragraph::isMisspelled(pos_type pos, bool check_boundary) const
{
- return SpellChecker::misspelled(d->speller_state_.getState(pos));
+ bool result = SpellChecker::misspelled(d->speller_state_.getState(pos));
+ if (result || pos <= 0 || pos > size())
+ return result;
+ if (check_boundary && (pos == size() || isWordSeparator(pos)))
+ result = SpellChecker::misspelled(d->speller_state_.getState(pos - 1));
+ return result;
}