* paragraph support for the global word list. To make it buffer

dependent, we need the buffer pointer. In fact we already have a
  pointer to the text inset. So as soon as inset know their buffer we
  can easily switch to a buffer local word list.
* there might be missing some more places to update the words from a
  buffer, like for example after backspace or more important when the
  cursor leaves a paragraph.


git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@23244 a592a061-630c-0410-9148-cb99ea01b6c8

src/Buffer.cpp

@@ -204,9 +204,6 @@ public:
 	/// A cache for the bibfiles (including bibfiles of loaded child
 	/// documents), needed for appropriate update of natbib labels.
 	mutable EmbeddedFileList bibfilesCache_;
-	
-	///
-	WordList words_;
 };
 
 /// Creates the per buffer temporary directory
@@ -2668,16 +2665,4 @@ void Buffer::bufferErrors(TeXErrors const & terr, ErrorList & errorList) const
 	}
 }
 
-
-void Buffer::registerWord(docstring const & word)
-{
-	d->words_.insert(word);
-}
-
-
-WordList const & Buffer::registeredWords() const
-{
-	return d->words_;
-}
-
 } // namespace lyx

src/Buffer.h

@@ -50,7 +50,6 @@ class TeXErrors;
 class TexRow;
 class TocBackend;
 class Undo;
-class WordList;
 
 namespace frontend {
 class GuiBufferDelegate;
@@ -452,11 +451,6 @@ public:
 	bool isExportable(std::string const & format) const;
 	///
 	std::vector<Format const *> exportableFormats(bool only_viewable) const;
-
-	/// Register word for completion word list.
-	void registerWord(docstring const & word);
-	///
-	WordList const & registeredWords() const;
 	
 private:
 	/// search for macro in local (buffer) table or in children

src/Paragraph.cpp

@@ -39,7 +39,9 @@
 #include "sgml.h"
 #include "TextClass.h"
 #include "TexRow.h"
+#include "Text.h"
 #include "VSpace.h"
+#include "WordList.h"
 
 #include "frontends/alert.h"
 
@@ -196,6 +198,10 @@ public:
 	typedef docstring TextContainer;
 	///
 	TextContainer text_;
+	
+	typedef std::set<docstring> Words;
+	///
+	Words words_;
 };
 
 
@@ -233,7 +239,8 @@ Paragraph::Private::Private(Paragraph * owner)
 Paragraph::Private::Private(Private const & p, Paragraph * owner)
 	: owner_(owner), inset_owner_(p.inset_owner_), fontlist_(p.fontlist_), 
 	  params_(p.params_), changes_(p.changes_), insetlist_(p.insetlist_),
-	  layout_(p.layout_), begin_of_body_(p.begin_of_body_), text_(p.text_)
+	  layout_(p.layout_), begin_of_body_(p.begin_of_body_), text_(p.text_),
+	  words_(p.words_)
 {
 	id_ = paragraph_id++;
 }
@@ -1039,6 +1046,7 @@ Paragraph::Paragraph(Paragraph const & par)
 	: itemdepth(par.itemdepth),
 	d(new Paragraph::Private(*par.d, this))
 {
+	registerWords();
 }
 
 
@@ -1048,8 +1056,10 @@ Paragraph & Paragraph::operator=(Paragraph const & par)
 	if (&par != this) {
 		itemdepth = par.itemdepth;
 
+		deregisterWords();
 		delete d;
 		d = new Private(*par.d, this);
+		registerWords();
 	}
 	return *this;
 }
@@ -1057,6 +1067,7 @@ Paragraph & Paragraph::operator=(Paragraph const & par)
 
 Paragraph::~Paragraph()
 {
+	deregisterWords();
 	delete d;
 }
 
@@ -2678,4 +2689,66 @@ bool Paragraph::isSeparator(pos_type pos) const
 }
 
 
+void Paragraph::deregisterWords()
+{
+	Private::Words::const_iterator it;
+	WordList & wl = theWordList();
+	for (it = d->words_.begin(); it != d->words_.end(); ++it)
+		wl.remove(*it);
+	d->words_.clear();
+}
+
+
+void Paragraph::collectWords(Buffer const & buf, CursorSlice const & sl)
+{
+	// find new words
+	bool inword = false;
+
+	//lyxerr << "Words: ";
+	pos_type n = size();
+	for (pos_type pos = 0; pos != n; ++pos) {
+		if (isDeleted(pos))
+			continue;
+
+		if (!isLetter(pos)) {
+			inword = false;
+			continue;
+		}
+
+		if (inword)
+			continue;
+
+		inword = true;
+		CursorSlice from = sl;
+		CursorSlice to = sl;
+		from.pos() = pos;
+		to.pos() = pos;
+		from.text()->getWord(from, to, WHOLE_WORD);
+		if (to.pos() - from.pos() < 6)
+			continue;
+		docstring word
+		= asString(buf, from.pos(), to.pos(), false);
+		d->words_.insert(word);
+		//lyxerr << word << " ";
+	}
+	//lyxerr << std::endl;
+}
+
+
+void Paragraph::registerWords()
+{
+	Private::Words::const_iterator it;
+	WordList & wl = theWordList();
+	for (it = d->words_.begin(); it != d->words_.end(); ++it)
+		wl.insert(*it);
+}
+
+
+void Paragraph::updateWords(Buffer const & buf, CursorSlice const & sl)
+{
+	deregisterWords();
+	collectWords(buf, sl);
+	registerWords();
+}
+
 } // namespace lyx

src/Paragraph.h

@@ -31,6 +31,8 @@ class Buffer;
 class BufferParams;
 class Change;
 class Counters;
+class Cursor;
+class CursorSlice;
 class Inset;
 class InsetBibitem;
 class LaTeXFeatures;
@@ -367,8 +369,18 @@ public:
 		bool mw, ///<
 		pos_type pos, ///< start from here.
 		bool del = true) const;
+	
+	///
+	void updateWords(Buffer const & buf, CursorSlice const & sl);
 
 private:
+	///
+	void deregisterWords();
+	///
+	void collectWords(Buffer const & buf, CursorSlice const & sl);
+	///
+	void registerWords();
+
 	/// Pimpl away stuff
 	class Private;
 	///

src/Text.cpp

@@ -581,19 +581,7 @@ void Text::charInserted(Cursor & cur)
 	    && !par.isLetter(cur.pos() - 1)) {
 		// get the word in front of cursor
 		BOOST_ASSERT(this == cur.text());
-		CursorSlice focus = cur.top();
-		focus.backwardPos();
-		CursorSlice from = focus;
-		CursorSlice to = focus;
-		getWord(from, to, PREVIOUS_WORD);
-		if (focus == from || to == from)
-			return;
-		docstring word
-		= par.asString(cur.buffer(), from.pos(), to.pos(), false);
-
-		// register words longer than 5 characters
-		if (word.length() > 5)
-			cur.buffer().registerWord(word);
+		cur.paragraph().updateWords(cur.buffer(), cur.top());
 	}
 }
 
@@ -1218,6 +1206,10 @@ bool Text::read(Buffer const & buf, Lexer & lex,
 			// not BufferParams
 			lyx::readParagraph(buf, pars_.back(), lex, errorList);
 
+			// register the words in the global word list
+			CursorSlice sl = CursorSlice(*insetPtr);
+			sl.pit() = pars_.size() - 1;
+			pars_.back().updateWords(buf, sl);
 		} else if (token == "\\begin_deeper") {
 			++depth;
 		} else if (token == "\\end_deeper") {

src/WordList.cpp

@@ -21,12 +21,20 @@
 
 namespace lyx {
 
+///
+WordList theGlobalWordList;
+
+WordList & theWordList()
+{
+	return theGlobalWordList;
+}
+
 ///
 struct WordList::Impl {
 	///
 	size_t c_;
 	///
-	typedef stx::weighted_btree<docstring, size_t> Words;
+	typedef stx::weighted_btree<docstring, size_t, int> Words;
 	///
 	Words words_;
 };
@@ -61,14 +69,34 @@ docstring const & WordList::word(size_t idx) const
 
 size_t WordList::size() const
 {
-	return d->words_.size();
+	return d->words_.summed_weight();
 }
 
 
 void WordList::insert(docstring const & w)
 {
-	d->words_.insert(w, size_t(1), stx::Void());
+	Impl::Words::iterator it = d->words_.find(w);
+	if (it == d->words_.end())
+		d->words_.insert(w, size_t(1), 1);
+	else {
+		it.data()++;
+		d->words_.change_weight(it, 1);
+	}
+}
+
+
+void WordList::remove(docstring const & w)
+{
+	Impl::Words::iterator it = d->words_.find(w);
+	if (it != d->words_.end()) {
+		it.data()--;
+		d->words_.change_weight(it, 0);
+		// We will not erase here, but instead we just leave it
+		// in the btree with weight 0. This avoid too much
+		// reorganisation of the tree all the time.
+		//if (it.data() == 0)
+		//	d->words_.erase(w);
+	}
 }
 
-	
 } // namespace lyx

src/WordList.h

@@ -29,12 +29,16 @@ public:
 	size_t size() const;
 	///
 	void insert(docstring const & w);
-
+	///
+	void remove(docstring const & w);
+	
 private:
 	struct Impl;
 	Impl * d;
 };
 
+WordList & theWordList();
+
 } // namespace lyx
 
 #endif // WORDLIST_H

src/frontends/qt4/GuiCompleter.h

@@ -112,7 +112,7 @@ private:
 	int updateLock_;
 	/// the BufferView::inlineCursorPos might be reset by destructive
 	/// operations like backspace. Hence, we have to keep this flag
-	/// in addition to know whether the popup is to be kept visible.
+	/// in addition to know whether the completion is to be kept visible.
 	bool inlineVisible_;
 	///
 	RtlItemDelegate * rtlItemDelegate_;

src/insets/InsetText.cpp

@@ -80,12 +80,12 @@ public:
 	///
 	virtual size_t size() const
 	{
-		return buf_.registeredWords().size();
+		return theWordList().size();
 	}
 	///
 	virtual docstring const & data(size_t idx) const
 	{
-		return buf_.registeredWords().word(idx);
+		return theWordList().word(idx);
 	}
 
 private: