#include "insets/InsetBibitem.h"
#include "insets/InsetLabel.h"
+#include "insets/InsetSpecialChar.h"
#include "support/debug.h"
#include "support/docstring_list.h"
namespace {
/// Inset identifier (above 0x10ffff, for ucs-4)
char_type const META_INSET = 0x200001;
-};
+}
/////////////////////////////////////////////////////////////////////
void setRange(FontSpan const fp, SpellChecker::Result state)
{
- eraseCoveredRanges(fp);
+ Ranges result;
+ RangesIterator et = ranges_.end();
+ RangesIterator it = ranges_.begin();
+ for (; it != et; ++it) {
+ if (!it->covered(fp))
+ result.push_back(SpellResultRange(it->range(), it->result()));
+ else if (state == SpellChecker::WORD_OK) {
+ // trim or split the current misspelled range
+ // store misspelled ranges only
+ FontSpan range = it->range();
+ if (fp.first > range.first) {
+ // misspelled area in front of WORD_OK
+ range.last = fp.first - 1;
+ result.push_back(SpellResultRange(range, it->result()));
+ range = it->range();
+ }
+ if (fp.last < range.last) {
+ // misspelled area after WORD_OK range
+ range.first = fp.last + 1;
+ result.push_back(SpellResultRange(range, it->result()));
+ }
+ }
+ }
+ ranges_ = result;
if (state != SpellChecker::WORD_OK)
ranges_.push_back(SpellResultRange(fp, state));
}
return result;
}
+ FontSpan const & getRange(pos_type pos) const
+ {
+ /// empty span to indicate mismatch
+ static FontSpan empty_;
+ RangesIterator et = ranges_.end();
+ RangesIterator it = ranges_.begin();
+ for (; it != et; ++it) {
+ if(it->inside(pos)) {
+ return it->range();
+ }
+ }
+ return empty_;
+ }
+
bool needsRefresh() const {
return needs_refresh_;
}
if (pos > refresh_.last)
refresh_.last = pos;
} else if (pos != -1) {
- refresh_.first = pos;
- refresh_.last = pos;
+ // init request check for neighbour positions too
+ refresh_.first = pos > 0 ? pos - 1 : 0;
+ // no need for special end of paragraph check
+ refresh_.last = pos + 1;
}
needs_refresh_ = pos != -1;
}
SpellChecker::ChangeNumber current_change_number_;
- void eraseCoveredRanges(FontSpan const fp)
- {
- Ranges result;
- RangesIterator et = ranges_.end();
- RangesIterator it = ranges_.begin();
- for (; it != et; ++it) {
- if (!it->covered(fp))
- result.push_back(SpellResultRange(it->range(), it->result()));
- }
- ranges_ = result;
- }
-
void correctRangesAfterPos(pos_type pos, int offset)
{
RangesIterator et = ranges_.end();
/// 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 &, TexRow & texrow,
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 &, TexRow &,
OutputParams const &) const;
/// This could go to ParagraphParameters if we want to.
- int endTeXParParams(BufferParams const &, odocstream &, TexRow &,
+ int endTeXParParams(BufferParams const &, otexstream &, TexRow &,
OutputParams const &) const;
///
void latexInset(BufferParams const &,
- odocstream &,
+ otexstream &,
TexRow & texrow, OutputParams &,
Font & running_font,
Font & basefont,
///
void latexSpecialChar(
- odocstream & os,
+ otexstream & os,
OutputParams const & runparams,
Font const & running_font,
Change const & running_change,
///
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,
unsigned int & column,
OutputParams const & runparams);
return speller_change_number > speller_state_.currentChangeNumber();
}
+ bool ignoreWord(docstring const & word) const ;
+
void setMisspelled(pos_type from, pos_type to, SpellChecker::Result state)
{
pos_type textsize = owner_->size();
// check for sane arguments
- if (to < from || from >= textsize)
+ if (to <= from || from >= textsize)
return;
- FontSpan fp = FontSpan(from, to);
- // don't mark end of paragraph
- if (fp.last >= textsize)
- fp.last = textsize - 1;
+ FontSpan fp = FontSpan(from, to - 1);
speller_state_.setRange(fp, state);
}
}
-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, TexRow & texrow,
pos_type i,
unsigned int & column,
Font const & font,
}
-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,
+ otexstream & os,
TexRow & texrow,
OutputParams & runparams,
Font & running_font,
LASSERT(inset, /**/);
if (style.pass_thru) {
- inset->plaintext(os, runparams);
+ inset->plaintext(os.os(), runparams);
return;
}
}
bool close = false;
- odocstream::pos_type const len = os.tellp();
+ odocstream::pos_type const len = os.os().tellp();
if (inset->forceLTR()
&& running_font.isRightToLeft()
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,
+ unsigned int & column)
{
char_type const c = text_[i];
}
-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,
+bool Paragraph::Private::latexSpecialPhrase(otexstream & os, pos_type & i,
unsigned int & column, OutputParams const & runparams)
{
// FIXME: if we have "LaTeX" with a font
void Paragraph::Private::validate(LaTeXFeatures & features) const
{
if (layout_->inpreamble && inset_owner_) {
+ bool const is_command = layout_->latextype == LATEX_COMMAND;
Buffer const & buf = inset_owner_->buffer();
BufferParams const & bp = buf.params();
Font f;
TexRow tr;
+ // 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;
- // we have to provide all the optional arguments here, even though
- // the last one is the only one we care about.
- owner_->latex(bp, f, ods, tr, features.runparams(), 0, -1, true);
- docstring const d = ods.str();
- if (!d.empty()) {
- // this will have "{" at the beginning, but not at the end
- string const content = to_utf8(d);
- string const cmd = layout_->latexname();
- features.addPreambleSnippet("\\" + cmd + content + "}");
+ otexstream os(ods);
+ if (is_command) {
+ ods << '\\' << 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_, os, features.runparams(),
+ layout_->reqargs, layout_->optargs);
+ else
+ 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, os, tr, features.runparams(), 0, -1, true);
+ if (ods.str().length() > length) {
+ if (is_command)
+ ods << '}';
+ string const snippet = to_utf8(ods.str());
+ features.addPreambleSnippet(snippet);
}
}
}
+bool Paragraph::isPassThru() const
+{
+ return inInset().getLayout().isPassThru() || d->layout_->pass_thru;
+}
+
namespace {
// paragraphs inside floats need different alignment tags to avoid
int Paragraph::Private::startTeXParParams(BufferParams const & bparams,
- odocstream & os, TexRow & texrow,
+ otexstream & os, TexRow & texrow,
OutputParams const & runparams) const
{
int column = 0;
- if (params_.noindent()) {
+ if (params_.noindent() && !layout_->pass_thru) {
os << "\\noindent ";
column += 10;
}
int Paragraph::Private::endTeXParParams(BufferParams const & bparams,
- odocstream & os, TexRow & texrow,
+ otexstream & os, TexRow & texrow,
OutputParams const & runparams) const
{
int column = 0;
// This one spits out the text of the paragraph
void Paragraph::latex(BufferParams const & bparams,
Font const & outerfont,
- odocstream & os, TexRow & texrow,
+ otexstream & os, TexRow & texrow,
OutputParams const & runparams,
int start_pos, int end_pos, bool force) const
{
open_font = false;
}
+ string const running_lang = runparams.use_polyglossia ?
+ running_font.language()->polyglossia() : running_font.language()->babel();
// close babel's font environment before opening CJK.
- if (!running_font.language()->babel().empty() &&
+ string const lang_end_command = runparams.use_polyglossia ?
+ "\\end{$$lang}" : lyxrc.language_command_end;
+ if (!running_lang.empty() &&
font.language()->encoding()->package() == Encoding::CJK) {
- string end_tag = subst(lyxrc.language_command_end,
+ string end_tag = subst(lang_end_command,
"$$lang",
- running_font.language()->babel());
+ running_lang);
os << from_ascii(end_tag);
column += end_tag.length();
}
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();
if (allowcust && d->endTeXParParams(bparams, os, texrow, runparams)
&& runparams.encoding != prev_encoding) {
runparams.encoding = prev_encoding;
- if (!bparams.useXetex)
+ if (!runparams.isFullUnicode())
os << setEncoding(prev_encoding->iconvName());
}
// We don't want to escape the entities. Note that
// it is safe to do this, since str can otherwise
// only be "-". E.g., it can't be "<".
- xs << XHTMLStream::NextRaw() << str;
+ xs << XHTMLStream::ESCAPE_NONE << str;
} else
xs << c;
}
{
if (Inset const * inset = getInset(pos))
return !inset->isLetter();
+ if (pos == size())
+ return true;
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) && !isDigit(c) && !contains(quote, c))
- || pos == size();
+ return (!isLetterChar(c) && !isDigitASCII(c) && !contains(quote, c));
+}
+
+
+bool Paragraph::isSameSpellRange(pos_type pos1, pos_type pos2) const
+{
+ return pos1 == pos2
+ || d->speller_state_.getRange(pos1) == d->speller_state_.getRange(pos2);
}
if (Inset const * inset = getInset(pos))
return inset->isChar();
char_type const c = d->text_[pos];
- return !isLetterChar(c) && !isDigit(c) && !lyx::isSpace(c);
+ return !isLetterChar(c) && !isDigitASCII(c) && !lyx::isSpace(c);
}
|| (c == '\n' && (options & AS_STR_NEWLINES)))
os.put(c);
else if (c == META_INSET && (options & AS_STR_INSETS)) {
- getInset(i)->tocString(os);
+ getInset(i)->toString(os);
if (getInset(i)->asInsetMath())
os << " ";
}
}
+void Paragraph::forToc(docstring & os, size_t maxlen) const
+{
+ if (!d->params_.labelString().empty())
+ os += d->params_.labelString() + ' ';
+ for (pos_type i = 0; i < size() && os.length() < maxlen; ++i) {
+ if (isDeleted(i))
+ continue;
+ char_type const c = d->text_[i];
+ if (isPrintable(c))
+ os += c;
+ else if (c == '\t' || c == '\n')
+ os += ' ';
+ else if (c == META_INSET)
+ getInset(i)->forToc(os, maxlen);
+ }
+}
+
+
docstring Paragraph::stringify(pos_type beg, pos_type end, int options, OutputParams & runparams) const
{
odocstringstream os;
{
// FIXME From JS:
// This is getting more and more a mess. ...We really should clean
- // up this bibitem issue for 1.6. See also bug 2743.
+ // up this bibitem issue for 1.6.
// Add bibitem insets if necessary
if (d->layout_->labeltype != LABEL_BIBLIO)
}
-bool Paragraph::find(docstring const & str, bool cs, bool mw,
- pos_type pos, bool del) const
+int Paragraph::find(docstring const & str, bool cs, bool mw,
+ pos_type start_pos, bool del) const
{
+ pos_type pos = start_pos;
int const strsize = str.length();
int i = 0;
pos_type const parsize = d->text_.size();
- for (i = 0; pos + i < parsize; ++i) {
- if (i >= strsize)
- break;
- if (cs && str[i] != d->text_[pos + i])
+ for (i = 0; i < strsize && pos < parsize; ++i, ++pos) {
+ // Ignore ligature break 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))
+ break;
+ pos++;
+ }
+ if (cs && str[i] != d->text_[pos])
break;
- if (!cs && uppercase(str[i]) != uppercase(d->text_[pos + i]))
+ if (!cs && uppercase(str[i]) != uppercase(d->text_[pos]))
break;
- if (!del && isDeleted(pos + i))
+ if (!del && isDeleted(pos))
break;
}
if (i != strsize)
- return false;
+ return 0;
// if necessary, check whether string matches word
if (mw) {
- if (pos > 0 && !isWordSeparator(pos - 1))
- return false;
- if (pos + strsize < parsize
- && !isWordSeparator(pos + strsize))
- return false;
+ if (start_pos > 0 && !isWordSeparator(start_pos - 1))
+ return 0;
+ if (pos < parsize
+ && !isWordSeparator(pos))
+ return 0;
}
- return true;
+ return pos - start_pos;
}
}
+bool Paragraph::Private::ignoreWord(docstring const & word) const
+{
+ // Ignore words with digits
+ // FIXME: make this customizable
+ // (note that some checkers ignore words with digits by default)
+ docstring::const_iterator cit = word.begin();
+ docstring::const_iterator const end = word.end();
+ for (; cit != end; ++cit) {
+ if (isNumber((*cit)))
+ return true;
+ }
+ return false;
+}
+
+
SpellChecker::Result Paragraph::spellCheck(pos_type & from, pos_type & to,
WordLangTuple & wl, docstring_list & suggestions,
bool do_suggestion, bool check_learned) const
if (from == to || from >= size())
return result;
- docstring word = asString(from, to, AS_STR_INSETS + AS_STR_SKIPDELETE);
+ docstring word = asString(from, to, AS_STR_INSETS | AS_STR_SKIPDELETE);
Language * lang = d->getSpellLanguage(from);
wl = WordLangTuple(word, lang);
return result;
if (needsSpellCheck() || check_learned) {
- // Ignore words with digits
- // FIXME: make this customizable
- // (note that some checkers ignore words with digits by default)
- if (!hasDigit(word)) {
+ if (!d->ignoreWord(word)) {
bool const trailing_dot = to < size() && d->text_[to] == '.';
result = speller->check(wl);
if (SpellChecker::misspelled(result) && trailing_dot) {
LYXERR(Debug::GUI, "misspelled word is correct with dot: \"" <<
word << "\" [" <<
from << ".." << to << "]");
+ } else {
+ // spell check with dot appended failed
+ // restore original word/lang value
+ word = asString(from, to, AS_STR_INSETS | AS_STR_SKIPDELETE);
+ wl = WordLangTuple(word, lang);
}
}
}
}
-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 && isWordSeparator(pos))
+ result = SpellChecker::misspelled(d->speller_state_.getState(pos - 1));
+ return result;
}