Fix regexp branches that subsume other branches

Problems reported by Mattias Engdegård in:
https://lists.gnu.org/r/emacs-devel/2019-04/msg00803.html
* lisp/arc-mode.el (archive-rar-summarize):
* lisp/eshell/em-hist.el (eshell-hist-word-designator):
* lisp/info.el (Info-dir-remove-duplicates):
* lisp/international/ja-dic-cnv.el (skkdic-convert-postfix)
(skkdic-convert-prefix, skkdic-collect-okuri-nasi):
* lisp/progmodes/cc-awk.el (c-awk-esc-pair-re):
* lisp/xml.el (xml-att-type-re):
Omit regexp branches that subsume other branches.
* lisp/progmodes/cperl-mode.el (cperl-beautify-regexp-piece):
$ and ^ aren’t simple-codes.

1 files changed, 1 insertions, 1 deletions

diff --git a/lisp/eshell/em-hist.el b/lisp/eshell/em-hist.el
index 614faaa131e..adb028002be 100644
--- a/lisp/eshell/em-hist.el
+++ b/lisp/eshell/em-hist.el
@@ -153,7 +153,7 @@ element, regardless of any text on the command line.  In that case,
   :group 'eshell-hist)
 
 (defcustom eshell-hist-word-designator
-  "^:?\\([0-9]+\\|[$^%*]\\)?\\(\\*\\|-[0-9]*\\|[$^%*]\\)?"
+  "^:?\\([0-9]+\\|[$^%*]\\)?\\(-[0-9]*\\|[$^%*]\\)?"
   "The regexp used to identify history word designators."
   :type 'regexp
   :group 'eshell-hist)