Fix html entity conversion

yunkya2 · yunkya2 · commit e2255ba879f9 · 2025-09-28T17:54:35.000+09:00
diff --git a/httpsget.c b/httpsget.c
@@ -152,76 +152,56 @@ void parse_html(char *html_content) {
  * HTML実体参照を文字コードに変換
  */
 void decode_html_entities(char *text) {
+    // よく使われる実体参照の変換テーブル
+    struct {
+        const char *entity;
+        int len;
+        const char *replacement;
+        int replacement_len;
+    } entities[] = {
+        {"&lt;", 4, "<", 1},
+        {"&gt;", 4, ">", 1},
+        {"&amp;", 5, "&", 1},
+        {"&quot;", 6, "\"", 1},
+        {"&apos;", 6, "'", 1},
+        {"&laquo;", 7, "<<", 2},
+        {"&raquo;", 7, ">>", 2},
+        {"&nbsp;", 6, "\xC2\xA0", 2},        // 改行なしスペース (U+00A0)
+        {"&copy;", 6, "\xC2\xA9", 2},        // copyright symbol (U+00A9)
+        {"&reg;", 5, "\xC2\xAE", 2},         // registered trademark symbol (U+00AE)
+        {"&trade;", 7, "\xE2\x84\xA2", 3},   // trademark symbol (U+2122)
+        {"&yen;", 5, "\xC2\xA5", 2},         // yen sign (U+00A5)
+        {"&euro;", 6, "\xE2\x82\xAC", 3},    // euro sign (U+20AC)
+        {"&pound;", 7, "\xC2\xA3", 2},       // pound sign (U+00A3)
+        {"&cent;", 6, "\xC2\xA2", 2},        // cent sign (U+00A2)
+        {"&deg;", 5, "\xC2\xB0", 2},         // degree sign (U+00B0)
+        {NULL, 0, NULL, 0}
+    };
+    
     char *src = text;
     char *dst = text;
     
     while (*src) {
         if (*src == '&') {
             // よく使われる実体参照を変換
-            if (strncmp(src, "&lt;", 4) == 0) {
-                *dst++ = '<';
-                src += 4;
-            } else if (strncmp(src, "&gt;", 4) == 0) {
-                *dst++ = '>';
-                src += 4;
-            } else if (strncmp(src, "&amp;", 5) == 0) {
-                *dst++ = '&';
-                src += 5;
-            } else if (strncmp(src, "&quot;", 6) == 0) {
-                *dst++ = '"';
-                src += 6;
-            } else if (strncmp(src, "&apos;", 6) == 0) {
-                *dst++ = '\'';
-                src += 6;
-            } else if (strncmp(src, "&nbsp;", 6) == 0) {
-                // 改行なしスペース (U+00A0) -> UTF-8: 0xC2 0xA0
-                *dst++ = 0xC2;
-                *dst++ = 0xA0;
-                src += 6;
-            } else if (strncmp(src, "&copy;", 6) == 0) {
-                // copyright symbol (U+00A9) -> UTF-8: 0xC2 0xA9
-                *dst++ = 0xC2;
-                *dst++ = 0xA9;
-                src += 6;
-            } else if (strncmp(src, "&reg;", 5) == 0) {
-                // registered trademark symbol (U+00AE) -> UTF-8: 0xC2 0xAE
-                *dst++ = 0xC2;
-                *dst++ = 0xAE;
-                src += 5;
-            } else if (strncmp(src, "&trade;", 7) == 0) {
-                // trademark symbol (U+2122) -> UTF-8: 0xE2 0x84 0xA2
-                *dst++ = 0xE2;
-                *dst++ = 0x84;
-                *dst++ = 0xA2;
-                src += 7;
-            } else if (strncmp(src, "&yen;", 5) == 0) {
-                // yen sign (U+00A5) -> UTF-8: 0xC2 0xA5
-                *dst++ = 0xC2;
-                *dst++ = 0xA5;
-                src += 5;
-            } else if (strncmp(src, "&euro;", 6) == 0) {
-                // euro sign (U+20AC) -> UTF-8: 0xE2 0x82 0xAC
-                *dst++ = 0xE2;
-                *dst++ = 0x82;
-                *dst++ = 0xAC;
-                src += 6;
-            } else if (strncmp(src, "&pound;", 7) == 0) {
-                // pound sign (U+00A3) -> UTF-8: 0xC2 0xA3
-                *dst++ = 0xC2;
-                *dst++ = 0xA3;
-                src += 7;
-            } else if (strncmp(src, "&cent;", 6) == 0) {
-                // cent sign (U+00A2) -> UTF-8: 0xC2 0xA2
-                *dst++ = 0xC2;
-                *dst++ = 0xA2;
-                src += 6;
-            } else if (strncmp(src, "&deg;", 5) == 0) {
-                // degree sign (U+00B0) -> UTF-8: 0xC2 0xB0
-                *dst++ = 0xC2;
-                *dst++ = 0xB0;
-                src += 5;
-                src += 7;
-            } else if (src[1] == '#') {
+            int found = 0;
+            int i;
+            for (i = 0; entities[i].entity != NULL; i++) {
+                if (strncmp(src, entities[i].entity, entities[i].len) == 0) {
+                    int j;
+                    for (j = 0; j < entities[i].replacement_len; j++) {
+                        *dst++ = entities[i].replacement[j];
+                    }
+                    src += entities[i].len;
+                    found = 1;
+                    break;
+                }
+            }
+            if (found) {
+                continue;
+            }
+
+            if (src[1] == '#') {
                 // 数値文字参照 &#nnn; または &#xhex;
                 char *end_ptr;
                 unsigned long code;
