From: Pierre Choffet <peuc@wanadoo.fr>
Date: Fri, 27 Mar 2026 14:51:40 +0000 (-0400)
Subject: Fix known error in source data
X-Git-Url: https://git.choffet.net/?a=commitdiff_plain;h=d2ec7f0b1294ef365dfd20cebefb9a58a3599806;p=banq_wikidata.git

Fix known error in source data
---

diff --git a/filter_marcxml.xslt b/filter_marcxml.xslt
index 8c0c6bf..457465e 100644
--- a/filter_marcxml.xslt
+++ b/filter_marcxml.xslt
@@ -25,6 +25,8 @@
 		Remove the following authority entries:
 		  - never referenced from any bibliographic record
 		  - with no 100, 110 or 111 zones
+		
+		Also fixes some known errors in source records.
 	-->
 	
 	
@@ -61,7 +63,9 @@
 		
 		<!-- Keep only authorities not referenced in any "no" filter -->
 		<xsl:if test="$keep='yes'">
-			<xsl:copy-of select="." />
+			<xsl:copy>
+				<xsl:apply-templates />
+			</xsl:copy>
 		</xsl:if>
 	</xsl:template>
 	
@@ -124,7 +128,9 @@
 			
 			<!-- Keep only authorities with at least one reference -->
 			<xsl:if test="$has-valid-reference='yes'">
-				<xsl:copy-of select="." />
+				<xsl:copy>
+					<xsl:apply-templates />
+				</xsl:copy>
 			</xsl:if>
 		</xsl:if>
 	</xsl:template>
@@ -139,6 +145,11 @@
 		</xsl:message>
 	</xsl:template>
 	
+	<!-- NOTE: In BAnQ data, there is a known error in data where full "Pablo" serie
+	           from "Michel Lancelot" has "Pablo Picasso" as contributor. Here,
+	           we’re removing this information -->
+	<xsl:template match="/marc:collection/marc:record[marc:datafield[@tag = '100']/marc:subfield[@code = 'a'] = 'Lancelot, Michel,']/marc:datafield[@tag = '700' and marc:subfield[@code = 'a'] = 'Picasso, Pablo,']" />
+	
 	<!-- Return "no" if it contains any value to filter out record in 655$a, "yes" otherwise -->
 	<xsl:template name="keep-655a">
 		<xsl:variable name="filters">