Re: bug with camel case and delete-whole-word-match function

[Oliver Kiddle <okiddle@xxxxxxxxxxx>]

On 5 Jul, Peter wrote:
> If you have "ThisIsSomeWords"

> However, if you're on the "S", you get "Is" before and "Some" after.
> Again there's no white space, so there's nothing to indicate to the
> calling function that these are two separate words rather than bits of
> the same word.  So I think we'd need to add some extra signalling from
> match-words-by-style to indicate "I'm at a word start" whether or not
> there's white space, which needs some thinking about.

Do we need to keep the existing seven elements of matched_words
unchanged for backwards compatibility? Not that I can think of a
particularly obvious way to augment it for this case. May be just
a 1/0 indicator for start of word is the simplest. It also seems
to lack having a <whitespace before the word> field.

I've attached a patch for a select-word-match function which uses the
mechanism for a vim style text object. In the process, I've found a
couple of other issues with match-words-by-style:

One is that if the cursor is in the middle of a block of whitespace
at the end of the line, the 4th element (whitespace after cursor)
is empty while element 7 contains the whitespace.
A similar issue occurs at the start of the line - element 1 contains
whitespace while element 3 doesn't.

The other issue is that with the shell word style, it'll put whitespace
at the end of element 5 instead of in element 6.

Oliver

diff --git a/Doc/Zsh/contrib.yo b/Doc/Zsh/contrib.yo
index 1d2b7ca..c3dec34 100644
--- a/Doc/Zsh/contrib.yo
+++ b/Doc/Zsh/contrib.yo
@@ -1940,6 +1940,8 @@ tindex(transpose-words-match)
 tindex(capitalize-word-match)
 tindex(up-case-word-match)
 tindex(down-case-word-match)
+tindex(delete-whole-word-match)
+tindex(select-word-match)
 tindex(select-word-style)
 tindex(match-word-context)
 tindex(match-words-by-style)
@@ -1947,12 +1949,14 @@ xitem(tt(forward-word-match), tt(backward-word-match))
 xitem(tt(kill-word-match), tt(backward-kill-word-match))
 xitem(tt(transpose-words-match), tt(capitalize-word-match))
 xitem(tt(up-case-word-match), tt(down-case-word-match))
+xitem(tt(delete-whole-word-match), tt(select-word-match))
 item(tt(select-word-style), tt(match-word-context), tt(match-words-by-style))(
-The eight `tt(-match)' functions are drop-in replacements for the
+The first eight `tt(-match)' functions are drop-in replacements for the
 builtin widgets without the suffix.  By default they behave in a similar
 way.  However, by the use of styles and the function tt(select-word-style),
-the way words are matched can be altered.  For comparison, the widgets
-described in ifzman(zmanref(zshzle) under Text Objects)\
+the way words are matched can be altered. tt(select-word-match) is intended
+to be used as a text object in vi mode but with custom word styles. For
+comparison, the widgets described in ifzman(zmanref(zshzle) under Text Objects)\
 ifnzman(noderef(Text Objects)) use fixed definitions of words, compatible
 with the tt(vim) editor.
 
@@ -1960,7 +1964,7 @@ The simplest way of configuring the functions is to use
 tt(select-word-style), which can either be called as a normal function with
 the appropriate argument, or invoked as a user-defined widget that will
 prompt for the first character of the word style to be used.  The first
-time it is invoked, the eight tt(-match) functions will automatically
+time it is invoked, the first eight tt(-match) functions will automatically
 replace the builtin versions, so they do not need to be loaded explicitly.
 
 The word styles available are as follows.  Only the first character
diff --git a/Functions/Zle/select-word-match b/Functions/Zle/select-word-match
new file mode 100644
index 0000000..24620c9
--- /dev/null
+++ b/Functions/Zle/select-word-match
@@ -0,0 +1,121 @@
+# Select the entire word around the cursor. Intended for use as
+# a vim-style text object in vi mode but with customisable
+# word boundaries.
+#
+# For example:
+#   autoload -U select-word-match
+#   zle -N select-in-camel select-word-match
+#   bindkey -M viopp ic select-in-camel
+#   zstyle ':zle:*-camel' word-style normal-subword
+
+emulate -L zsh
+setopt extendedglob
+
+local curcontext=:zle:$WIDGET
+local -a matched_words
+# Start and end of range of characters
+integer pos1 pos2 num=${NUMERIC:-1}
+local style word
+
+# choose between inner word or a word style of widget
+for style in $1 ${${WIDGET#*-}[1]} $KEYS[1] "i"; do
+  [[ $style = [ai] ]] && break
+done
+
+autoload -Uz match-words-by-style
+
+while (( num-- )); do
+  if (( MARK > CURSOR )); then
+    # if cursor is at the start of the selection, just move back a word
+    match-words-by-style
+    if [[ $style = i && -n $matched_words[3] ]]; then
+      word=$matched_words[3]
+    else
+      word=$matched_words[2]$matched_words[3]
+    fi
+    if [[ -n $word ]]; then
+      (( CURSOR -= ${#word} ))
+    else
+      return 1
+    fi
+  elif (( MARK >= 0 && MARK < CURSOR )); then
+    # cursor at the end, move forward a word
+    (( CURSOR+1 == $#BUFFER )) && return 1
+    (( CURSOR++ ))
+    match-words-by-style
+    if [[ -n $matched_words[4] ]]; then
+      if [[ $style = i ]]; then
+	# just skip the whitespace
+	word=$matched_words[4]
+      else
+	# skip the whitespace plus word
+	word=$matched_words[4]$matched_words[5]
+      fi
+    else
+      if [[ $style = i ]]; then
+	# skip the word
+	word=$matched_words[5]
+      else
+	# skip word and following whitespace
+	word=$matched_words[5]$matched_words[6]
+      fi
+    fi
+    (( CURSOR += ${#word} - 1 ))
+  else
+    match-words-by-style
+
+    if [[ -n "${matched_words[3]}" ]]; then
+      # There's whitespace before the cursor, so the word we are selecting
+      # starts at the cursor position.
+      pos1=$CURSOR
+    else
+      # No whitespace before us, so select any wordcharacters there.
+      pos1="${#matched_words[1]}"
+    fi
+
+    if [[ -n "${matched_words[4]}" ]]; then
+      if [[ -n "${matched_words[3]}" ]] || (( CURSOR == 0 )); then
+        # whitespace either side, select it
+	(( pos1 = CURSOR - ${#matched_words[3]} ))
+	(( pos2 = CURSOR + ${#matched_words[4]} ))
+      else
+	# There's whitespace at the cursor position, so only select
+	# up to the cursor position.
+	(( pos2 = CURSOR + 1 ))
+      fi
+    else
+      # No whitespace at the cursor position, so select the
+      # current character and any following wordcharacters.
+      (( pos2 = CURSOR + ${#matched_words[5]} ))
+    fi
+
+    if [[ $style = a ]]; then
+      if [[ -n "${matched_words[4]}"  && ( -n "${matched_words[3]}" || CURSOR -eq 0 ) ]]; then
+	# in the middle of whitespace so grab a word
+	if [[ -n "${matched_words[5]}" ]]; then
+	  (( pos2 += ${#matched_words[5]} )) # preferably the one after
+	else
+	  (( pos1 -= ${#matched_words[2]} )) # otherwise the one before
+	fi
+      elif [[ -n "${matched_words[6]}" ]]; then
+	(( pos2 += ${#matched_words[6]} ))
+      elif [[ -n "${matched_words[3]}" ]]; then
+	# couldn't grab whitespace forwards so try backwards
+	(( pos1 -= ${#matched_words[3]} ))
+      elif (( pos1 > 0 )); then
+	# There might have been whitespace before the word
+	(( CURSOR = pos1 ))
+	match-words-by-style
+	if [[ -n "${matched_words[3]}" ]]; then
+	  (( pos1 -= ${#matched_words[3]} ))
+	fi
+      fi
+    fi
+
+    (( MARK = pos1, CURSOR = pos2-1 ))
+  fi
+done
+
+if [[ $KEYMAP == vicmd ]] && (( !REGION_ACTIVE )); then
+  (( CURSOR++ )) # Need to include cursor position for operators
+fi