diff --git a/src/selectors.cc b/src/selectors.cc
index dee4d1ef..d0135f1d 100644
--- a/src/selectors.cc
+++ b/src/selectors.cc
@@ -1,7 +1,6 @@
#include "selectors.hh"
#include "string.hh"
-#include "utf8_iterator.hh"
#include <algorithm>
@@ -10,153 +9,6 @@
namespace Kakoune
{
-using Utf8Iterator = utf8::utf8_iterator<BufferIterator, utf8::InvalidBytePolicy::Pass>;
-
-namespace
-{
-
-template<WordType word_type = Word>
-bool is_word(Codepoint c)
-{
- return Kakoune::is_word(c);
-}
-
-template<>
-bool is_word<WORD>(Codepoint c)
-{
- return !is_blank(c) and !is_eol(c);
-}
-
-static bool is_punctuation(Codepoint c)
-{
- return not (is_word(c) or is_blank(c) or is_eol(c));
-}
-
-enum class CharCategories
-{
- Blank,
- EndOfLine,
- Word,
- Punctuation,
-};
-
-template<WordType word_type = Word>
-CharCategories categorize(Codepoint c)
-{
- if (is_word(c))
- return CharCategories::Word;
- if (is_eol(c))
- return CharCategories::EndOfLine;
- if (is_blank(c))
- return CharCategories::Blank;
- return word_type == WORD ? CharCategories::Word
- : CharCategories::Punctuation;
-}
-
-template<typename Iterator, typename EndIterator, typename T>
-void skip_while(Iterator& it, const EndIterator& end, T condition)
-{
- while (it != end and condition(*it))
- ++it;
-}
-
-template<typename Iterator, typename BeginIterator, typename T>
-void skip_while_reverse(Iterator& it, const BeginIterator& begin, T condition)
-{
- while (it != begin and condition(*it))
- --it;
-}
-
-Range utf8_range(const Utf8Iterator& first, const Utf8Iterator& last)
-{
- return {first.base().coord(), last.base().coord()};
-}
-
-}
-
-typedef boost::regex_iterator<BufferIterator> RegexIterator;
-
-template<WordType word_type>
-Selection select_to_next_word(const Buffer& buffer, const Selection& selection)
-{
- Utf8Iterator begin = buffer.iterator_at(selection.last());
- if (begin+1 == buffer.end())
- return selection;
- if (categorize<word_type>(*begin) != categorize<word_type>(*(begin+1)))
- ++begin;
-
- skip_while(begin, buffer.end(), is_eol);
- if (begin == buffer.end())
- return selection;
- Utf8Iterator end = begin+1;
-
- if (word_type == Word and is_punctuation(*begin))
- skip_while(end, buffer.end(), is_punctuation);
- else if (is_word<word_type>(*begin))
- skip_while(end, buffer.end(), is_word<word_type>);
-
- skip_while(end, buffer.end(), is_blank);
-
- return utf8_range(begin, end-1);
-}
-template Selection select_to_next_word<Word>(const Buffer&, const Selection&);
-template Selection select_to_next_word<WORD>(const Buffer&, const Selection&);
-
-template<WordType word_type>
-Selection select_to_next_word_end(const Buffer& buffer, const Selection& selection)
-{
- Utf8Iterator begin = buffer.iterator_at(selection.last());
- if (begin+1 == buffer.end())
- return selection;
- if (categorize<word_type>(*begin) != categorize<word_type>(*(begin+1)))
- ++begin;
-
- skip_while(begin, buffer.end(), is_eol);
- if (begin == buffer.end())
- return selection;
- Utf8Iterator end = begin;
- skip_while(end, buffer.end(), is_blank);
-
- if (word_type == Word and is_punctuation(*end))
- skip_while(end, buffer.end(), is_punctuation);
- else if (is_word<word_type>(*end))
- skip_while(end, buffer.end(), is_word<word_type>);
-
- return utf8_range(begin, end-1);
-}
-template Selection select_to_next_word_end<Word>(const Buffer&, const Selection&);
-template Selection select_to_next_word_end<WORD>(const Buffer&, const Selection&);
-
-template<WordType word_type>
-Selection select_to_previous_word(const Buffer& buffer, const Selection& selection)
-{
- Utf8Iterator begin = buffer.iterator_at(selection.last());
- if (begin == buffer.begin())
- return selection;
- if (categorize<word_type>(*begin) != categorize<word_type>(*(begin-1)))
- --begin;
-
- skip_while_reverse(begin, buffer.begin(), is_eol);
- Utf8Iterator end = begin;
- skip_while_reverse(end, buffer.begin(), is_blank);
-
- bool with_end = false;
- if (word_type == Word and is_punctuation(*end))
- {
- skip_while_reverse(end, buffer.begin(), is_punctuation);
- with_end = is_punctuation(*end);
- }
- else if (is_word<word_type>(*end))
- {
- skip_while_reverse(end, buffer.begin(), is_word<word_type>);
- with_end = is_word<word_type>(*end);
- }
-
- return utf8_range(begin, with_end ? end : end+1);
-}
-template Selection select_to_previous_word<Word>(const Buffer&, const Selection&);
-template Selection select_to_previous_word<WORD>(const Buffer&, const Selection&);
-
Selection select_line(const Buffer& buffer, const Selection& selection)
{
Utf8Iterator first = buffer.iterator_at(selection.last());
@@ -366,50 +218,6 @@ Selection select_to_eol_reverse(const Buffer& buffer, const Selection& selection
return utf8_range(begin, end == buffer.begin() ? end : end+1);
}
-template<WordType word_type>
-Selection select_whole_word(const Buffer& buffer, const Selection& selection, ObjectFlags flags)
-{
- Utf8Iterator first = buffer.iterator_at(selection.last());
- Utf8Iterator last = first;
- if (is_word<word_type>(*first))
- {
- if (flags & ObjectFlags::ToBegin)
- {
- skip_while_reverse(first, buffer.begin(), is_word<word_type>);
- if (not is_word<word_type>(*first))
- ++first;
- }
- if (flags & ObjectFlags::ToEnd)
- {
- skip_while(last, buffer.end(), is_word<word_type>);
- if (not (flags & ObjectFlags::Inner))
- skip_while(last, buffer.end(), is_blank);
- --last;
- }
- }
- else if (not (flags & ObjectFlags::Inner))
- {
- if (flags & ObjectFlags::ToBegin)
- {
- skip_while_reverse(first, buffer.begin(), is_blank);
- if (not is_word<word_type>(*first))
- return selection;
- skip_while_reverse(first, buffer.begin(), is_word<word_type>);
- if (not is_word<word_type>(*first))
- ++first;
- }
- if (flags & ObjectFlags::ToEnd)
- {
- skip_while(last, buffer.end(), is_blank);
- --last;
- }
- }
- return (flags & ObjectFlags::ToEnd) ? utf8_range(first, last)
- : utf8_range(last, first);
-}
-template Selection select_whole_word<Word>(const Buffer&, const Selection&, ObjectFlags);
-template Selection select_whole_word<WORD>(const Buffer&, const Selection&, ObjectFlags);
-
Selection select_whole_sentence(const Buffer& buffer, const Selection& selection, ObjectFlags flags)
{
BufferIterator first = buffer.iterator_at(selection.last());
diff --git a/src/selectors.hh b/src/selectors.hh
index 64887e31..a44cf830 100644
--- a/src/selectors.hh
+++ b/src/selectors.hh
@@ -4,6 +4,7 @@
#include "selection.hh"
#include "unicode.hh"
#include "editor.hh"
+#include "utf8_iterator.hh"
namespace Kakoune
{
@@ -49,17 +50,89 @@ inline void remove_selection(const Buffer&, SelectionList& selections, int index
selections.check_invariant();
}
-enum WordType { Word, WORD };
+using Utf8Iterator = utf8::utf8_iterator<BufferIterator, utf8::InvalidBytePolicy::Pass>;
+
+inline Range utf8_range(const Utf8Iterator& first, const Utf8Iterator& last)
+{
+ return {first.base().coord(), last.base().coord()};
+}
+
+typedef boost::regex_iterator<BufferIterator> RegexIterator;
template<WordType word_type>
-Selection select_to_next_word(const Buffer& buffer,
- const Selection& selection);
+Selection select_to_next_word(const Buffer& buffer, const Selection& selection)
+{
+ Utf8Iterator begin = buffer.iterator_at(selection.last());
+ if (begin+1 == buffer.end())
+ return selection;
+ if (categorize<word_type>(*begin) != categorize<word_type>(*(begin+1)))
+ ++begin;
+
+ skip_while(begin, buffer.end(), is_eol);
+ if (begin == buffer.end())
+ return selection;
+ Utf8Iterator end = begin+1;
+
+ if (word_type == Word and is_punctuation(*begin))
+ skip_while(end, buffer.end(), is_punctuation);
+ else if (is_word<word_type>(*begin))
+ skip_while(end, buffer.end(), is_word<word_type>);
+
+ skip_while(end, buffer.end(), is_blank);
+
+ return utf8_range(begin, end-1);
+}
+
template<WordType word_type>
-Selection select_to_next_word_end(const Buffer& buffer,
- const Selection& selection);
+Selection select_to_next_word_end(const Buffer& buffer, const Selection& selection)
+{
+ Utf8Iterator begin = buffer.iterator_at(selection.last());
+ if (begin+1 == buffer.end())
+ return selection;
+ if (categorize<word_type>(*begin) != categorize<word_type>(*(begin+1)))
+ ++begin;
+
+ skip_while(begin, buffer.end(), is_eol);
+ if (begin == buffer.end())
+ return selection;
+ Utf8Iterator end = begin;
+ skip_while(end, buffer.end(), is_blank);
+
+ if (word_type == Word and is_punctuation(*end))
+ skip_while(end, buffer.end(), is_punctuation);
+ else if (is_word<word_type>(*end))
+ skip_while(end, buffer.end(), is_word<word_type>);
+
+ return utf8_range(begin, end-1);
+}
+
template<WordType word_type>
-Selection select_to_previous_word(const Buffer& buffer,
-const Selection& selection);
+Selection select_to_previous_word(const Buffer& buffer, const Selection& selection)
+{
+ Utf8Iterator begin = buffer.iterator_at(selection.last());
+ if (begin == buffer.begin())
+ return selection;
+ if (categorize<word_type>(*begin) != categorize<word_type>(*(begin-1)))
+ --begin;
+
+ skip_while_reverse(begin, buffer.begin(), is_eol);
+ Utf8Iterator end = begin;
+ skip_while_reverse(end, buffer.begin(), is_blank);
+
+ bool with_end = false;
+ if (word_type == Word and is_punctuation(*end))
+ {
+ skip_while_reverse(end, buffer.begin(), is_punctuation);
+ with_end = is_punctuation(*end);
+ }
+ else if (is_word<word_type>(*end))
+ {
+ skip_while_reverse(end, buffer.begin(), is_word<word_type>);
+ with_end = is_word<word_type>(*end);
+ }
+
+ return utf8_range(begin, with_end ? end : end+1);
+}
Selection select_line(const Buffer& buffer,
const Selection& selection);
@@ -86,8 +159,47 @@ constexpr ObjectFlags operator|(ObjectFlags lhs, ObjectFlags rhs)
{ return (ObjectFlags)((int)lhs | (int) rhs); }
template<WordType word_type>
-Selection select_whole_word(const Buffer& buffer, const Selection& selection,
- ObjectFlags flags);
+Selection select_whole_word(const Buffer& buffer, const Selection& selection, ObjectFlags flags)
+{
+ Utf8Iterator first = buffer.iterator_at(selection.last());
+ Utf8Iterator last = first;
+ if (is_word<word_type>(*first))
+ {
+ if (flags & ObjectFlags::ToBegin)
+ {
+ skip_while_reverse(first, buffer.begin(), is_word<word_type>);
+ if (not is_word<word_type>(*first))
+ ++first;
+ }
+ if (flags & ObjectFlags::ToEnd)
+ {
+ skip_while(last, buffer.end(), is_word<word_type>);
+ if (not (flags & ObjectFlags::Inner))
+ skip_while(last, buffer.end(), is_blank);
+ --last;
+ }
+ }
+ else if (not (flags & ObjectFlags::Inner))
+ {
+ if (flags & ObjectFlags::ToBegin)
+ {
+ skip_while_reverse(first, buffer.begin(), is_blank);
+ if (not is_word<word_type>(*first))
+ return selection;
+ skip_while_reverse(first, buffer.begin(), is_word<word_type>);
+ if (not is_word<word_type>(*first))
+ ++first;
+ }
+ if (flags & ObjectFlags::ToEnd)
+ {
+ skip_while(last, buffer.end(), is_blank);
+ --last;
+ }
+ }
+ return (flags & ObjectFlags::ToEnd) ? utf8_range(first, last)
+ : utf8_range(last, first);
+}
+
Selection select_whole_sentence(const Buffer& buffer, const Selection& selection,
ObjectFlags flags);
Selection select_whole_paragraph(const Buffer& buffer, const Selection& selection,
diff --git a/src/unicode.hh b/src/unicode.hh
index ce7e9be2..fb235fad 100644
--- a/src/unicode.hh
+++ b/src/unicode.hh
@@ -9,11 +9,6 @@ namespace Kakoune
using Codepoint = uint32_t;
-inline bool is_word(Codepoint c)
-{
- return c == '_' or isalnum(c);
-}
-
inline bool is_eol(Codepoint c)
{
return c == '\n';
@@ -29,6 +24,46 @@ inline bool is_horizontal_blank(Codepoint c)
return c == ' ' or c == '\t';
}
+enum WordType { Word, WORD };
+
+template<WordType word_type = Word>
+inline bool is_word(Codepoint c)
+{
+ return c == '_' or isalnum(c);
+}
+
+template<>
+inline bool is_word<WORD>(Codepoint c)
+{
+ return !is_blank(c) and !is_eol(c);
+}
+
+inline bool is_punctuation(Codepoint c)
+{
+ return not (is_word(c) or is_blank(c) or is_eol(c));
+}
+
+enum class CharCategories
+{
+ Blank,
+ EndOfLine,
+ Word,
+ Punctuation,
+};
+
+template<WordType word_type = Word>
+inline CharCategories categorize(Codepoint c)
+{
+ if (is_word(c))
+ return CharCategories::Word;
+ if (is_eol(c))
+ return CharCategories::EndOfLine;
+ if (is_blank(c))
+ return CharCategories::Blank;
+ return word_type == WORD ? CharCategories::Word
+ : CharCategories::Punctuation;
+}
+
}
#endif // unicode_hh_INCLUDED
diff --git a/src/utils.hh b/src/utils.hh
index ef0cf3d5..a08e7707 100644
--- a/src/utils.hh
+++ b/src/utils.hh
@@ -189,6 +189,20 @@ bool contains(const std::unordered_set<T1>& container, const T2& value)
return container.find(value) != container.end();
}
+template<typename Iterator, typename EndIterator, typename T>
+void skip_while(Iterator& it, const EndIterator& end, T condition)
+{
+ while (it != end and condition(*it))
+ ++it;
+}
+
+template<typename Iterator, typename BeginIterator, typename T>
+void skip_while_reverse(Iterator& it, const BeginIterator& begin, T condition)
+{
+ while (it != begin and condition(*it))
+ --it;
+}
+
// *** On scope end ***
//
// on_scope_end provides a way to register some code to be