22014: argument-base, insert-unicode-char

ChangeLog

@@ -1,3 +1,12 @@
+2005-11-24  Peter Stephenson  <pws@csr.com>
+
+	* 22014: Doc/Zsh/Contrib.yo, Doc/Zsh/zle.yo,
+	Functions/Zle/.distfiles, Functions/Zle/insert-unicode-char,
+	Src/Zle/iwidgets.list, Src/Zle/zle.h, Src/Zle/zle_main.c,
+	Src/Zle/zle_misc.c: internal widget argument-base sets
+	numeric base for next prefix argument, widget insert-unicode-char
+	uses this to insert Unicode character by \U........
+
 2005-11-23  Peter Stephenson  <pws@csr.com>
 
 	* 22013: INSTALL, NEWS, Completion/compinstall,

Doc/Zsh/contrib.yo

@@ -693,6 +693,9 @@ The function may be run outside zle in which case it prints the character
 (together with a newline) to standard output.  Input is still read from
 keystrokes.
 
+See tt(insert-unicode-char) for an alternative way of inserting Unicode
+characters using their hexadecimal character number.
+
 The set of accented characters is reasonably complete up to Unicode
 character U+0180, the set of special characters less so.  However, it it
 is very sporadic from that point.  Adding new characters is easy,
@@ -917,6 +920,17 @@ narrow-to-region -p $'Editing restricted region\n' \
 zle recursive-edit
 narrow-to-region -R state)
 )
+tindex(insert-unicode-char)
+item(tt(insert-unicode-char))(
+When first executed, the user inputs a set of hexadecimal digits.
+This is terminated with another call to tt(insert-unicode-char).
+The digits are then turned into the corresponding Unicode character.
+For example, if the widget is bound to tt(^XU), the character sequence
+`tt(^XU 4 c ^XU)' inserts tt(L) (Unicode U+004c).
+
+See tt(insert-composed-char) for a way of inserting characters
+using a two-character mnemonic.
+)
 tindex(predict-on)
 tindex(predict-off)
 item(tt(predict-on))(

Doc/Zsh/zle.yo

@@ -1519,6 +1519,23 @@ Inside a widget function, if passed an argument, i.e. `tt(zle
 universal-argument) var(num)', the numerical argument will be set to
 var(num); this is equivalent to `tt(NUMERIC=)var(num)'.
 )
+tindex(argument-base)
+item(tt(argument-base))(
+Use the existing numeric argument as a numeric base, which must be in the
+range 2 to 36 inclusive.  Subsequent use of tt(digit-argument) and
+tt(universal-argument) will input a new prefix in the given base.
+The usual hexadecimal convention is used: the letter tt(a) or tt(A)
+corresponds to 10, and so on.  Arguments in bases requiring digits from 10
+upwards are more conveniently input with tt(universal-argument), since
+tt(ESC-a) etc. are not usually bound to tt(digit-argument).
+
+The function can be used with a command argument inside a user-defined
+widget.  The following code sets the base to 16 and lets the user input a
+hexadecimal argument until a key out of the digit range is typed:
+
+example(zle argument-base 16
+zle universal-argument)
+)
 enditem()
 texinode(Completion)(Miscellaneous)(Arguments)(Zle Widgets)
 subsect(Completion)

Functions/Zle/.distfiles

@@ -9,6 +9,7 @@ edit-command-line              forward-word-match
 history-pattern-search         history-search-end
 incarg                         incremental-complete-word
 insert-composed-char           insert-files
+insert-unicode-char
 keeper                         kill-word-match
 match-words-by-style           narrow-to-region
 narrow-to-region-invisible     predict-on

Functions/Zle/insert-unicode-char

@@ -0,0 +1,17 @@
+# Make hex integers appear as 0x...
+setopt localoptions cbases
+
+if [[ $LASTWIDGET = insert-unicode-char ]]; then
+  # Second call; we should have a usable prefix.
+  # If we don't, give up.
+  (( ${+NUMERIC} )) || return 1
+  # Convert it back to hex, padded with zeroes to 8 digits plus the 0x...
+  local -i 16 -Z 10 arg=$NUMERIC
+  # ...and use print to turn this into a Unicode character.
+  LBUFFER+="$(print -n "\U${arg##0x}")"
+else
+  # Set the base to 16...
+  zle argument-base 16
+  # ...wait for user to type hex keys then call this widget again.
+  zle universal-argument
+fi

Src/Zle/iwidgets.list

@@ -13,6 +13,7 @@
 "accept-and-menu-complete", acceptandmenucomplete, ZLE_MENUCMP | ZLE_KEEPSUFFIX
 "accept-line", acceptline, 0
 "accept-line-and-down-history", acceptlineanddownhistory, 0
+"argument-base", argumentbase, ZLE_MENUCMP | ZLE_KEEPSUFFIX | ZLE_LASTCOL | ZLE_NOTCOMMAND
 "backward-char", backwardchar, 0
 "backward-delete-char", backwarddeletechar, ZLE_KEEPSUFFIX
 "backward-delete-word", backwarddeleteword, ZLE_KEEPSUFFIX

Src/Zle/zle.h

@@ -205,6 +205,7 @@ struct modifier {
     int mult;		/* repeat count */
     int tmult;		/* repeat count actually being edited */
     int vibuf;		/* vi cut buffer */
+    int base;		/* numeric base for digit arguments (usually 10) */
 };
 
 #define MOD_MULT  (1<<0)   /* a repeat count has been selected */

Src/Zle/zle_main.c

@@ -1153,6 +1153,7 @@ initmodifier(struct modifier *mp)
     mp->mult = 1;
     mp->tmult = 1;
     mp->vibuf = 0;
+    mp->base = 10;
 }
 
 /* Reset command modifiers, unless the command just executed was a prefix. *

Src/Zle/zle_misc.c

@@ -518,12 +518,9 @@ quotedinsert(char **args)
 	return selfinsert(args);
 }
 
-/**/
+static int
-int
+parsedigit(int inkey)
-digitargument(UNUSED(char **args))
 {
-    int sign = (zmult < 0) ? -1 : 1;
-
 #ifdef MULTIBYTE_SUPPORT
     /*
      * It's too dangerous to allow metafied input.  See
@@ -531,23 +528,48 @@ digitargument(UNUSED(char **args))
      * of digits.  We are assuming ASCII is a subset of the multibyte
      * encoding.
      */
-    if (!idigit(lastchar))
-	return 1;
 #else
     /* allow metafied as well as ordinary digits */
-    if (!idigit(lastchar & 0x7f))
+    inkey &= 0x7f;
-	return 1;
 #endif
 
+    /* remember lastchar is not a wide character */
+    if (zmod.base > 10)
+    {
+	if (lastchar >= 'a' && lastchar < 'a' + zmod.base - 10)
+	    return lastchar - 'a' + 10;
+	else if (lastchar >= 'A' && lastchar < 'A' + zmod.base - 10)
+	    return lastchar - 'A' + 10;
+	else if (idigit(lastchar))
+	    return lastchar - '0';
+	else
+	    return -1;
+    }
+    else if (lastchar >= '0' && lastchar < '0' + zmod.base)
+	return lastchar - '0';
+    else
+	return -1;
+}
+
+/**/
+int
+digitargument(UNUSED(char **args))
+{
+    int sign = (zmult < 0) ? -1 : 1;
+    int newdigit = parsedigit(lastchar);
+
+    if (newdigit < 0)
+	return 1;
+
     if (!(zmod.flags & MOD_TMULT))
 	zmod.tmult = 0;
     if (zmod.flags & MOD_NEG) {
 	/* If we just had a negative argument, this is the digit, *
 	 * rather than the -1 assumed by negargument()            */
-	zmod.tmult = sign * (lastchar & 0xf);
+	zmod.tmult = sign * newdigit;
 	zmod.flags &= ~MOD_NEG;
     } else
-	zmod.tmult = zmod.tmult * 10 + sign * (lastchar & 0xf);
+	zmod.tmult = zmod.tmult * zmod.base + sign * newdigit;
     zmod.flags |= MOD_TMULT;
     prefixflag = 1;
     return 0;
@@ -594,12 +616,16 @@ universalargument(char **args)
 	if (gotk == '-' && !digcnt) {
 	    minus = -1;
 	    digcnt++;
-	} else if (idigit(gotk)) {
-	    pref = pref * 10 + (gotk & 0xf);
-	    digcnt++;
 	} else {
-	    ungetbyte(gotk);
+	    int newdigit = parsedigit(gotk);
-	    break;
+
+	    if (newdigit >= 0) {
+		pref = pref * zmod.base + newdigit;
+		digcnt++;
+	    } else {
+		ungetbyte(gotk);
+		break;
+	    }
 	}
     }
     if (digcnt)
@@ -611,6 +637,36 @@ universalargument(char **args)
     return 0;
 }
 
+/* Set the base for a digit argument. */
+
+/**/
+int
+argumentbase(char **args)
+{
+    int multbase;
+
+    if (*args)
+	multbase = (int)zstrtol(*args, NULL, 0);
+    else
+	multbase = zmod.mult;
+
+    if (multbase < 2 || multbase > ('9' - '0' + 1) + ('z' - 'a' + 1))
+	return 1;
+
+    zmod.base = multbase;
+
+    /* reset modifier, apart from base... */
+    zmod.flags = 0;
+    zmod.mult = 1;
+    zmod.tmult = 1;
+    zmod.vibuf = 0;
+
+    /* ...but indicate we are still operating on a prefix argument. */
+    prefixflag = 1;
+
+    return 0;
+}
+
 /**/
 int
 copyprevword(UNUSED(char **args))