From 6f09db94da09282ac867d988db1a70914cd08eb9 Mon Sep 17 00:00:00 2001 From: wonder95 Date: Thu, 23 May 2013 12:54:14 -0700 Subject: [PATCH] Backport to 6.x-3.x of 1927032 - Avoid spaces between words and punctuation marks in search result snippets. --- apachesolr.module | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/apachesolr.module b/apachesolr.module index 9c1445e..b5a4dd0 100644 --- a/apachesolr.module +++ b/apachesolr.module @@ -2152,6 +2152,18 @@ function apachesolr_clean_text($text) { $text = htmlspecialchars(html_entity_decode($text, ENT_QUOTES, 'UTF-8'), ENT_QUOTES, 'UTF-8'); // Remove extra spaces. $text = preg_replace('/\s+/s', ' ', $text); + // Remove white spaces around punctuation marks probably added + // by the safety operations above. This is not a world wide perfect solution, + // but a rough attempt for at least US and Western Europe. + // Pc: Connector punctuation + // Pd: Dash punctuation + // Pe: Close punctuation + // Pf: Final punctuation + // Pi: Initial punctuation + // Po: Other punctuation, including ¿?¡!,.:; + // Ps: Open punctuation + $text = preg_replace('/\s(\p{Pc}|\p{Pd}|\p{Pe}|\p{Pf}|!|\?|,|\.|:|;)/s', '$1', $text); + $text = preg_replace('/(\p{Ps}|¿|¡)\s/s', '$1', $text); return $text; } -- 1.7.10.2 (Apple Git-33)