Add a cut variation of Reline::Unicode.take_range method take_mbchar_range

Author: tompng

lib/reline/unicode.rb

@@ -173,16 +173,24 @@ def self.split_by_width(str, max_width, encoding = str.encoding)
 
   # Take a chunk of a String cut by width with escape sequences.
   def self.take_range(str, start_col, max_width)
+    take_mbchar_range(str, start_col, max_width).first
+  end
+
+  def self.take_mbchar_range(str, start_col, width, cover_begin: false, cover_end: false, padding: false)
     chunk = String.new(encoding: str.encoding)
     total_width = 0
     rest = str.encode(Encoding::UTF_8)
     in_zero_width = false
+    chunk_start_col = nil
+    chunk_end_col = nil
     rest.scan(WIDTH_SCANNER) do |non_printing_start, non_printing_end, csi, osc, gc|
       case
       when non_printing_start
         in_zero_width = true
+        chunk << NON_PRINTING_START
       when non_printing_end
         in_zero_width = false
+        chunk << NON_PRINTING_END
       when csi
         chunk << csi
       when osc
@@ -192,13 +200,28 @@ def self.take_range(str, start_col, max_width)
           chunk << gc
         else
           mbchar_width = get_mbchar_width(gc)
+          prev_width = total_width
           total_width += mbchar_width
-          break if (start_col + max_width) < total_width
-          chunk << gc if start_col < total_width
+          break if !cover_end && total_width > start_col + width
+          if cover_begin ? start_col < total_width : start_col <= prev_width
+            chunk << gc
+            chunk_start_col ||= prev_width
+            chunk_end_col = total_width
+          elsif padding && prev_width < start_col && start_col < total_width
+            chunk << ' '
+            chunk_start_col = start_col
+          end
+          break if total_width >= start_col + width
         end
       end
     end
-    chunk
+    chunk_start_col ||= start_col
+    chunk_end_col ||= start_col
+    if padding && chunk_end_col < start_col + width
+      chunk << ' ' * (start_col + width - chunk_end_col)
+      chunk_end_col = start_col + width
+    end
+    [chunk, chunk_start_col, chunk_end_col - chunk_start_col]
   end
 
   def self.get_next_mbchar_size(line, byte_pointer)

test/reline/test_unicode.rb

@@ -41,8 +41,8 @@ def test_split_by_width
   def test_take_range
     assert_equal 'cdef', Reline::Unicode.take_range('abcdefghi', 2, 4)
     assert_equal 'あde', Reline::Unicode.take_range('abあdef', 2, 4)
-    assert_equal 'zerocdef', Reline::Unicode.take_range("ab\1zero\2cdef", 2, 4)
-    assert_equal 'bzerocde', Reline::Unicode.take_range("ab\1zero\2cdef", 1, 4)
+    assert_equal "\1zero\2cdef", Reline::Unicode.take_range("ab\1zero\2cdef", 2, 4)
+    assert_equal "b\1zero\2cde", Reline::Unicode.take_range("ab\1zero\2cdef", 1, 4)
     assert_equal "\e[31mcd\e[42mef", Reline::Unicode.take_range("\e[31mabcd\e[42mefg", 2, 4)
     assert_equal "\e]0;1\acd", Reline::Unicode.take_range("ab\e]0;1\acd", 2, 3)
     assert_equal 'いう', Reline::Unicode.take_range('あいうえお', 2, 4)
@@ -62,4 +62,24 @@ def test_calculate_width
     assert_equal 10, Reline::Unicode.calculate_width('あいうえお')
     assert_equal 10, Reline::Unicode.calculate_width('あいうえお', true)
   end
+
+  def test_take_mbchar_range
+    assert_equal ['cdef', 2, 4], Reline::Unicode.take_mbchar_range('abcdefghi', 2, 4)
+    assert_equal ['cdef', 2, 4], Reline::Unicode.take_mbchar_range('abcdefghi', 2, 4, padding: true)
+    assert_equal ['cdef', 2, 4], Reline::Unicode.take_mbchar_range('abcdefghi', 2, 4, cover_begin: true)
+    assert_equal ['cdef', 2, 4], Reline::Unicode.take_mbchar_range('abcdefghi', 2, 4, cover_end: true)
+    assert_equal ['いう', 2, 4], Reline::Unicode.take_mbchar_range('あいうえお', 2, 4)
+    assert_equal ['いう', 2, 4], Reline::Unicode.take_mbchar_range('あいうえお', 2, 4, padding: true)
+    assert_equal ['いう', 2, 4], Reline::Unicode.take_mbchar_range('あいうえお', 2, 4, cover_begin: true)
+    assert_equal ['いう', 2, 4], Reline::Unicode.take_mbchar_range('あいうえお', 2, 4, cover_end: true)
+    assert_equal ['う', 4, 2], Reline::Unicode.take_mbchar_range('あいうえお', 3, 4)
+    assert_equal [' う ', 3, 4], Reline::Unicode.take_mbchar_range('あいうえお', 3, 4, padding: true)
+    assert_equal ['いう', 2, 4], Reline::Unicode.take_mbchar_range('あいうえお', 3, 4, cover_begin: true)
+    assert_equal ['うえ', 4, 4], Reline::Unicode.take_mbchar_range('あいうえお', 3, 4, cover_end: true)
+    assert_equal ['いう ', 2, 5], Reline::Unicode.take_mbchar_range('あいうえお', 3, 4, cover_begin: true, padding: true)
+    assert_equal [' うえ', 3, 5], Reline::Unicode.take_mbchar_range('あいうえお', 3, 4, cover_end: true, padding: true)
+    assert_equal [' うえお   ', 3, 10], Reline::Unicode.take_mbchar_range('あいうえお', 3, 10, padding: true)
+    assert_equal ["\e[31mc\1ABC\2d\e[0mef", 2, 4], Reline::Unicode.take_mbchar_range("\e[31mabc\1ABC\2d\e[0mefghi", 2, 4)
+    assert_equal ["\e[41m \e[42mい\e[43m ", 1, 4], Reline::Unicode.take_mbchar_range("\e[41mあ\e[42mい\e[43mう\e[0m", 1, 4, padding: true)
+  end
 end