| @@ -102,8 +102,38 @@ | ||
| 102 | 102 | return html_entity_decode($value, ENT_QUOTES | ENT_HTML5, 'UTF-8'); |
| 103 | 103 | } |
| 104 | 104 | |
| 105 | 105 | /** |
| 106 | + * Strip leading and trailing whitespace and title separators. | |
| 107 | + * | |
| 108 | + * The separator is removed as a whole string, never byte by byte. | |
| 109 | + * `trim($title, " \t..." . $separator)` reads its second argument as a | |
| 110 | + * list of single bytes, so with the Bullet separator (`•`, E2 80 A2) it | |
| 111 | + * also stripped any E2, 80 or A2 byte at either end of the title: the | |
| 112 | + * middle of €, ✅, curly quotes and many Cyrillic letters. The broken | |
| 113 | + * UTF-8 left behind made `<title>` invalid and `esc_attr()` blank out | |
| 114 | + * og:title and twitter:title (#972). | |
| 115 | + * | |
| 116 | + * @since 2.15.0 | |
| 117 | + * | |
| 118 | + * @param string $value Title or description. | |
| 119 | + * @param string $separator The active separator, possibly multibyte. | |
| 120 | + * @return string | |
| 121 | + */ | |
| 122 | + public static function trim_separator(string $value, string $separator): string { | |
| 123 | + $alternatives = '[\s\x00]'; | |
| 124 | + if ('' !== $separator) { | |
| 125 | + $alternatives .= '|' . preg_quote($separator, '/'); | |
| 126 | + } | |
| 127 | + | |
| 128 | + $trimmed = preg_replace('/^(?:' . $alternatives . ')+|(?:' . $alternatives . ')+$/u', '', $value); | |
| 129 | + | |
| 130 | + // Invalid UTF-8 makes a /u pattern fail. Fall back to whitespace only, | |
| 131 | + // which cannot cut a character in half. | |
| 132 | + return null === $trimmed ? trim($value) : $trimmed; | |
| 133 | + } | |
| 134 | + | |
| 135 | + /** | |
| 106 | 136 | * Trim a description to a character budget, multibyte-safe. |
| 107 | 137 | * |
| 108 | 138 | * Cuts on a word boundary when one is available inside the budget, so the |
| 109 | 139 | * result does not end mid-word; falls back to a hard character cut for |
| @@ -175,8 +205,37 @@ | ||
| 175 | 205 | '/\s*(\[\s*(\x{2026}|\.\.\.)\s*\]|\x{2026})\s*$/u', |
| 176 | 206 | '', |
| 177 | 207 | $text |
| 178 | 208 | ); |
| 209 | + | |
| 210 | + return trim((string) preg_replace('/\s+/u', ' ', $text)); | |
| 211 | + } | |
| 212 | + | |
| 213 | + /** | |
| 214 | + * Make a title or name fit to appear in JSON-LD. | |
| 215 | + * | |
| 216 | + * get_the_title() runs wptexturize and keeps any markup in the title, and | |
| 217 | + * WordPress stores term names and the site name esc_html()'d, so a page | |
| 218 | + * called "R&D at Tom's" reached the graph as `R&D at Tom’s` and | |
| 219 | + * a category "Cats & Dogs" as `Cats & Dogs`. Nothing downstream | |
| 220 | + * decodes JSON-LD, so those were published literally (#976). | |
| 221 | + * | |
| 222 | + * normalize_schema_text() without its excerpt-marker strip: that marker | |
| 223 | + * is an artefact of a derived description, while a title that ends in an | |
| 224 | + * ellipsis was written that way and keeps it. A decoded `<` stays safe | |
| 225 | + * because the graph is encoded with JSON_HEX_TAG (#118). | |
| 226 | + * | |
| 227 | + * @since 2.15.0 | |
| 228 | + * | |
| 229 | + * @param string $text Raw title or name. | |
| 230 | + * @return string Plain text with tags removed and entities decoded. | |
| 231 | + */ | |
| 232 | + public static function normalize_schema_name(string $text): string { | |
| 233 | + if ('' === trim($text)) { | |
| 234 | + return ''; | |
| 235 | + } | |
| 236 | + | |
| 237 | + $text = self::decode_schema_entities(wp_strip_all_tags($text)); | |
| 179 | 238 | |
| 180 | 239 | return trim((string) preg_replace('/\s+/u', ' ', $text)); |
| 181 | 240 | } |
| 182 | 241 | |
