peptide-repo/peptide-news-plugin
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: add fallback to strip nbsp-separated publisher names from excerpts

Browse Source 
Google News RSS excerpts use double non-breaking spaces followed by the
publisher name (e.g. "...habitat  News-Medical"). When the source field
is "news.google.com" rather than the actual publisher, the exact-match
logic can't strip it. This adds a regex fallback that catches trailing
nbsp-separated capitalized names.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Terence Ong
2026-04-08 10:44:01 +07:00
parent 31df32da2d
commit d6b111ffc8
2 changed files with 20 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
includes/class-peptide-news-rest-api.php
View File

@@ -236,7 +236,7 @@ class Peptide_News_Rest_API {
} }
} }
// Fallback: strip trailing " - Short Publisher Name" patterns (1-5 words). // Fallback 1: strip trailing " - Short Publisher Name" patterns (1-5 words).
// Only strips if the part after the separator looks like a publisher name // Only strips if the part after the separator looks like a publisher name
// (starts with uppercase, no sentence-ending punctuation). // (starts with uppercase, no sentence-ending punctuation).
$pattern = '/\s+[-\x{2013}\x{2014}|]\s+[A-Z][A-Za-z0-9\s.\'\-]{0,50}$/u'; $pattern = '/\s+[-\x{2013}\x{2014}|]\s+[A-Z][A-Za-z0-9\s.\'\-]{0,50}$/u';
@@ -245,6 +245,15 @@ class Peptide_News_Rest_API {
return rtrim( $candidate ); return rtrim( $candidate );
} }
// Fallback 2: strip trailing nbsp-separated publisher names.
// Google News RSS excerpts use double non-breaking space + publisher name
// when the source field doesn't match (e.g., source = "news.google.com").
$nbsp_pattern = '/(?:\x{00A0}{1,2}|\s{2,})[A-Z][A-Za-z0-9\s.\'\-]{0,50}$/u';
$candidate = preg_replace( $nbsp_pattern, '', $text );
if ( $candidate !== $text && strlen( $candidate ) > strlen( $text ) * 0.4 ) {
return rtrim( $candidate );
}
return $text; return $text;
} }

11
peptide-news-plugin/includes/class-peptide-news-rest-api.php
View File

@@ -230,7 +230,7 @@ class Peptide_News_Rest_API {
} }
} }
// Fallback: strip trailing " - Short Publisher Name" patterns (1-5 words). // Fallback 1: strip trailing " - Short Publisher Name" patterns (1-5 words).
// Only strips if the part after the separator looks like a publisher name // Only strips if the part after the separator looks like a publisher name
// (starts with uppercase, no sentence-ending punctuation). // (starts with uppercase, no sentence-ending punctuation).
$pattern = '/\s+[-–—|]\s+[A-Z][A-Za-z0-9\s.\'-]{0,50}$/u'; $pattern = '/\s+[-–—|]\s+[A-Z][A-Za-z0-9\s.\'-]{0,50}$/u';
@@ -239,6 +239,15 @@ class Peptide_News_Rest_API {
return rtrim( $candidate ); return rtrim( $candidate );
} }
// Fallback 2: strip trailing nbsp-separated publisher names.
// Google News RSS excerpts use double non-breaking space + publisher name
// when the source field doesn't match (e.g., source = "news.google.com").
$nbsp_pattern = '/(?:\x{00A0}{1,2}|\s{2,})[A-Z][A-Za-z0-9\s.\'\-]{0,50}$/u';
$candidate = preg_replace( $nbsp_pattern, '', $text );
if ( $candidate !== $text && strlen( $candidate ) > strlen( $text ) * 0.4 ) {
return rtrim( $candidate );
}
return $text; return $text;
} }