tests: test the xmp metadata extraction

tests/config/fields

@@ -1,4 +1,10 @@
 [prefixes]
 testfield = TSTFLD
+pdf:producer = XYYPDFPROD
+url = XYYPDFID
+
 [stored]
 testfield=
+pdf:producer =
+url =
+

tests/config/recoll.conf

@@ -1,5 +1,6 @@
 loglevel = 6
 logfilename = /tmp/logrcltst
+idxlogfilename = /tmp/idxlogrcltst
 
 daemloglevel = 6
 daemlogfilename = /tmp/rclmontrace
@@ -35,6 +36,8 @@ daemSkippedPaths =  \
     /home/dockes/projets/fulltext/testrecoll/skipped/real* \
     /home/dockes/projets/fulltext/testrecoll/config
 
+pdfextrameta = pdf:Producer dc:identifier
+
 unac_except_trans = åå Åå ää Ää öö Öö üü Üü ßss œoe Œoe æae ÆAE fifi flfl
 
 [/home/dockes/projets/fulltext/testrecoll/utf8]

tests/pdf/pdf.sh

@@ -5,8 +5,14 @@ topdir=`dirname $0`/..
 
 initvariables $0
 
-recollq '"sequences of moveto and lineto"' OR 'ANSIX3.4' 2> $mystderr | 
-	egrep -v '^Recoll query: ' > $mystdout
+(
+    recollq '"sequences of moveto and lineto"' OR 'ANSIX3.4'
+    # pdf xmp. Note that dc:identifier is aliased to url in the
+    # defaults field file, and this can't be overruled afaics, so
+    # url is prefixed for the dc:identifier search to work
+    recollq dc:identifier:10.12345/sampledoi
+    recollq 'pdf:Producer:"GPL Ghostscript 9.18"'
+)  2> $mystderr | egrep -v '^Recoll query: ' > $mystdout
 
 diff -w ${myname}.txt $mystdout > $mydiffs 2>&1