Add commentary about where these code conversion rules came from (Bug#3428).

diff --git a/lisp/international/eucjp-ms.el b/lisp/international/eucjp-ms.el
index e658d2e41a6bcbf60d80c9ac05c80b28b2f5f59c..9dc73a31968f2a85e2749df35158f2dfcba22366 100644 (file)
--- a/lisp/international/eucjp-ms.el
+++ b/lisp/international/eucjp-ms.el
@@ -1,5 +1,16 @@
 ;;; eucjp-ms.el -- translation table for eucJP-ms. -*- no-byte-compile: t -*-
-;;; Automatically genrated from eucJP-13th.txt, eucJP-udc.txt, eucJP-ibmext.txt
+
+;; This translation table was originally generated from the code set
+;; conversion rules given in "Problems and Solutions for Unicode and
+;; User/Vendor Defined Characters" (1996), by the TOG/JVC CDE/Motif
+;; Technical Working Group.  The Working Group's website is now
+;; defunct, but the original files specifying these conversion rules
+;; are archived here:
+;;
+;; http://home.m05.itscom.net/numa/cde/ucs-conv-e/eucJP-13th.txt
+;; http://home.m05.itscom.net/numa/cde/ucs-conv-e/eucJP-udc.txt
+;; http://home.m05.itscom.net/numa/cde/ucs-conv-e/eucJP-ibmext.txt
+
 (let ((map
        '(;JISEXT<->UNICODE
         (#x2d21 . #x2460)