Fix documentation of of chr's equivalent directive ZEB-672

And add a few tests for the behavior.

diff --git a/doc/field-structure.xml b/doc/field-structure.xml
index 1b24aae..a19838e 100644 (file)
--- a/doc/field-structure.xml
+++ b/doc/field-structure.xml
@@ -436,21 +436,22 @@
       <term>equivalent <replaceable>value-set</replaceable></term>
       <listitem>
        <para>
-       This directive introduces equivalence classes of characters
-       and/or strings for sorting purposes only. It resembles the map
-       directive, but does not affect search and retrieval indexing,
-       but only sorting order under present requests. 
+       This directive introduces equivalence classes of strings for
+       searching purposes only. It's a one-to-many
+       conversion that takes place only during search before the map
+       directive kicks in.
        </para>
        <para>
-        For example, <literal>scan.chr</literal> contains the following
-        equivalent sorting instructions, which can be uncommented:
+        For example given:
         <screen><![CDATA[
-         # equivalent æä(ae)
-         # equivalent øö(oe)
-         # equivalent å(aa)
-         # equivalent uü
+         equivalent æä(ae)
         ]]></screen>
        </para>
+       <para>
+        a search for the <literal>äsel</literal> will be be match any of
+        <literal>æsel</literal>, <literal>äsel</literal> and
+        <literal>aesel</literal>.
+       </para>
       </listitem></varlistentry>
     </variablelist>
    </para>

diff --git a/test/api/Makefile.am b/test/api/Makefile.am
index 9fc3ec5..0d5aad4 100644 (file)
--- a/test/api/Makefile.am
+++ b/test/api/Makefile.am
@@ -18,7 +18,7 @@ EXTRA_DIST=zebra.cfg test_trunc.cfg test_private_attset.cfg \
        test_icu_indexing.cfg test_icu_indexing.idx \
        test_sort1.cfg test_sort1.idx test_sort1.chr sort1.abs \
        test_sort2.cfg test_sort2.idx test_sort2.chr sort2.abs \
-       test_safari.cfg test_sort3.cfg 
+       test_safari.cfg test_sort3.cfg string.chr
 
 noinst_LIBRARIES = libtestlib.a
 

diff --git a/test/api/string.chr b/test/api/string.chr
new file mode 100644 (file)
index 0000000..637ff98
--- /dev/null
+++ b/test/api/string.chr
@@ -0,0 +1,41 @@
+# Generic character map.
+#
+
+# Define the basic value-set. *Beware* of changing this without re-indexing
+# your databases.
+
+lowercase {0-9}{a-y}üzæäøöå
+uppercase {0-9}{A-Y}ÜZÆÄØÖÅ
+
+# Breaking characters
+
+space {\001-\040}!"#$%&'\()*+,-./:;<=>?@\[\\]^_`\{|}~
+
+# Characters to be considered equivalent for searching purposes.
+
+# equivalent æä(ae)
+# equivalent øö(oe)
+equivalent å(aa)
+# equivalent uü
+
+# Supplemental mappings
+
+#map (&auml;)       ä
+#map (&aelig;)      æ
+#map (&oslash;)     ø
+#map (&aring;)      å
+#map (&ouml;)       ö
+#map (&Auml;)       Ä
+#map (&Aelig;)      Æ
+#map (&Oslash;)     Ø
+#map (&Aring;)      Å
+#map (&Ouml;)       Ö
+
+#map éÉ                e
+#map á         a
+#map ó         o
+#map í         i
+
+#map (Aa)      (AA)
+
+#map (aa)        a

diff --git a/test/api/test_search.c b/test/api/test_search.c
index a560e15..17d540e 100644 (file)
--- a/test/api/test_search.c
+++ b/test/api/test_search.c
@@ -76,6 +76,12 @@ const char *myrec[] = {
         "</title>"
         "</gils>"
         ,
+        "<gils>\n"
+        "<title>"
+        "\xe5" " og vandl" "\xf8" "b"
+        "</title>"
+        "</gils>"
+        ,
        0} ;
 
 static void tst(int argc, char **argv)
@@ -144,7 +150,7 @@ static void tst(int argc, char **argv)
     YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 5=103 titl", 3));
     YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 5=103 titlx", 3));
     YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 5=103 titlxx", 0));
-    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 5=103 x", 2));
+    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 5=103 x", 3));
     YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 5=103 le", 0));
     YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 5=103 .*le", 3));
 
@@ -219,18 +225,18 @@ static void tst(int argc, char **argv)
 
 
     /* string relations, >= */
-    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 2=4 x", 2));
-    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 2=4 tu", 2));
-    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 2=4 title", 3));
+    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 2=4 x", 3));
+    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 2=4 tu", 3));
+    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 2=4 title", 4));
 
     /* string relations, > */
-    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 2=5 x", 0));
-    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 2=5 tu", 2));
-    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 2=5 title", 2));
+    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 2=5 x", 1));
+    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 2=5 tu", 3));
+    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 2=5 title", 3));
 
     /* always-matches relation */
-    YAZ_CHECK(tl_query(zh, "@attr 1=_ALLRECORDS @attr 2=103 {ym}", 6));
-    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 2=103 {x my}", 5));
+    YAZ_CHECK(tl_query(zh, "@attr 1=_ALLRECORDS @attr 2=103 {ym}", 7));
+    YAZ_CHECK(tl_query(zh, "@attr 1=4 @attr 2=103 {x my}", 6));
     YAZ_CHECK(tl_query_x(zh, "@attr 1=1 @attr 2=103 {x my}", 0, 114));
 
     /* and searches */
@@ -434,6 +440,14 @@ static void tst(int argc, char **argv)
     YAZ_CHECK(tl_query(zh, "@attr 1=30 @attr 4=5 @attr 2=5 {2107-09-19 00:00:00}", 0));
 
 
+    YAZ_CHECK(tl_query(zh, "@attr 1=4 {\xe5" " og vandl" "\xf8" "b}", 1));
+
+    YAZ_CHECK(tl_query(zh, "@attr 1=4 {\xc5" " OG VANDL" "\xd8" "B}", 1));
+
+    /* try equivalent directive */
+    YAZ_CHECK(tl_query(zh, "@attr 1=4 {aa" " OG VANDL" "\xd8" "B}", 1));
+    YAZ_CHECK(tl_query(zh, "@attr 1=4 {AA" " OG VANDL" "\xd8" "B}", 0));
+
     YAZ_CHECK(tl_close_down(zh, zs));
 }