From 17501079a0f797ef7b4735209d6e90698bac0854 Mon Sep 17 00:00:00 2001
From: konn <konn@1a406e8e-add9-4483-a2c8-d8cac5b7c224>
Date: Thu, 20 Nov 2008 09:48:42 +0000
Subject: [PATCH]   * Added illegal charactor code of SJIS   * Added illegal
 charactor code of EUC

git-svn-id: svn+ssh://svn.sourceforge.jp/svnroot/modchxj/mod_chxj/branches/RELEASE_0_12_0@4493 1a406e8e-add9-4483-a2c8-d8cac5b7c224
---
 src/chxj_encoding.c | 66 ++++++++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 63 insertions(+), 3 deletions(-)

diff --git a/src/chxj_encoding.c b/src/chxj_encoding.c
index 570c0da6..848c27be 100644
--- a/src/chxj_encoding.c
+++ b/src/chxj_encoding.c
@@ -153,10 +153,70 @@ chxj_convert_illegal_charactor_sequence(request_rec *r, chxjconvrule_entry  *ent
       *ibuf += 1;
       DBG(r, "passed 1byte.");
     }
-    if (ibuf && *ibuf) {
-      *ilen = strlen(*ibuf);
-      DBG(r, "new len = [%d].", *ilen);
+  }
+  else if (STRCASEEQ('e','E', "EUCJP",               entryp->encoding)
+      ||   STRCASEEQ('c','C', "CSEUCPKDFMTJAPANESE", entryp->encoding)
+      ||   STRCASEEQ('e','E', "EUC-JISX0213",        entryp->encoding)
+      ||   STRCASEEQ('e','E', "EUC-JP-MS",           entryp->encoding)
+      ||   STRCASEEQ('e','E', "EUC-JP",              entryp->encoding)
+      ||   STRCASEEQ('e','E', "EUCJP-MS",            entryp->encoding)
+      ||   STRCASEEQ('e','E', "EUCJP-OPEN",          entryp->encoding)
+      ||   STRCASEEQ('e','E', "EUCJP-WIN",           entryp->encoding)
+      ||   STRCASEEQ('e','E', "EUCJP",               entryp->encoding)) {
+    if ((unsigned char)**ibuf == 0x8F) {
+      /* 3byte charactor */
+      **obuf = '?';
+      *obuf += 1;
+      *olen -= 1;
+      *ibuf +=3;
+      DBG(r, "passed 3byte.");
     }
+    else {
+      /* 2byte charactor */
+      **obuf = '?';
+      *obuf += 1;
+      *olen -= 1;
+      *ibuf += 2;
+      DBG(r, "passed 2byte.");
+    }
+  }
+  else if (STRCASEEQ('c', 'C', "CP932",     entryp->encoding)
+      ||   STRCASEEQ('c', 'C', "CSIBM932",  entryp->encoding)
+      ||   STRCASEEQ('i', 'I', "IBM-932",   entryp->encoding)
+      ||   STRCASEEQ('i', 'I', "IBM932",    entryp->encoding)
+      ||   STRCASEEQ('m', 'M', "MS932",     entryp->encoding)
+      ||   STRCASEEQ('m', 'M', "MS_KANJI",  entryp->encoding)
+      ||   STRCASEEQ('s', 'S', "SJIS-OPEN", entryp->encoding)
+      ||   STRCASEEQ('s', 'S', "SJIS-WIN",  entryp->encoding)
+      ||   STRCASEEQ('s', 'S', "SJIS",      entryp->encoding)) {
+    if ( ( ((0x81 <= (unsigned char)**ibuf) && (0x9f >= (unsigned char)**ibuf))
+        || ((0xe0 <= (unsigned char)**ibuf) && (0xfc >= (unsigned char)**ibuf)))
+       &&
+       (  ((0x40 <= (unsigned char)*((*ibuf)+1)) && (0x7e >= (unsigned char)*((*ibuf)+1)))
+        ||((0x80 <= (unsigned char)*((*ibuf)+1)) && (0xfc >= (unsigned char)*((*ibuf)+1))))) {
+      /* 2byte charactor */
+      **obuf = '?';
+      *obuf += 1;
+      *olen -= 1;
+      *ibuf += 2;
+      DBG(r, "passed 2byte.");
+    }
+    else {
+      /* 1byte charactor */
+      **obuf = '?';
+      *obuf += 1;
+      *olen -= 1;
+      *ibuf += 1;
+      DBG(r, "passed 1byte.");
+    }
+  }
+  else {
+    *ilen = 0;
+    return;
+  }
+  if (ibuf && *ibuf) {
+    *ilen = strlen(*ibuf);
+    DBG(r, "new len = [%d].", *ilen);
   }
 }
 
-- 
2.11.0