Added skeleton for query charset conversion. Bug #977.
[yaz-moved-to-github.git] / src / srwutil.c
index c6e0c8b..c7ae1c2 100644 (file)
@@ -1,8 +1,8 @@
 /*
- * Copyright (C) 1995-2005, Index Data ApS
+ * Copyright (C) 1995-2007, Index Data ApS
  * See the file LICENSE for details.
  *
- * $Id: srwutil.c,v 1.43 2006-06-05 18:13:01 adam Exp $
+ * $Id: srwutil.c,v 1.55 2007-01-03 08:42:15 adam Exp $
  */
 /**
  * \file srwutil.c
@@ -41,9 +41,10 @@ void encode_uri_char(char *dst, char ch)
     }
 }
 
-void yaz_array_to_uri(char **path, ODR o, char **name, char **value)
+static void yaz_array_to_uri_ex(char **path, ODR o, char **name, char **value,
+                                const char *extra_args)
 {
-    size_t i, szp = 0, sz = 0;
+    size_t i, szp = 0, sz = extra_args ? 1+strlen(extra_args) : 1;
     for(i = 0; name[i]; i++)
         sz += strlen(name[i]) + 3 + strlen(value[i]) * 3;
     *path = odr_malloc(o, sz);
@@ -67,9 +68,21 @@ void yaz_array_to_uri(char **path, ODR o, char **name, char **value)
             szp += vlen;
         }
     }
+    if (extra_args)
+    {
+        if (i)
+            (*path)[szp++] = '&';
+        memcpy(*path + szp, extra_args, strlen(extra_args));
+        szp += strlen(extra_args);
+    }
     (*path)[szp] = '\0';
 }
 
+void yaz_array_to_uri(char **path, ODR o, char **name, char **value)
+{
+    yaz_array_to_uri_ex(path, o, name, value, 0);
+}
+
 int yaz_uri_array(const char *path, ODR o, char ***name, char ***val)
 {
     int no = 2;
@@ -226,8 +239,8 @@ static int yaz_base64decode(const char *in, char *out)
  * Look for authentication tokens in HTTP Basic parameters or in x-username/x-password
  * parameters. Added by SH.
  */
-static void yaz_srw_decodeauth(Z_SRW_PDU *sr, Z_HTTP_Request *hreq, char *username,
-        char *password, ODR decode)
+static void yaz_srw_decodeauth(Z_SRW_PDU *sr, Z_HTTP_Request *hreq,
+                               char *username, char *password, ODR decode)
 {
     const char *basic = z_HTTP_header_lookup(hreq->headers, "Authorization");
 
@@ -315,6 +328,36 @@ void yaz_add_srw_diagnostic(ODR o, Z_SRW_diagnostic **d,
     yaz_add_srw_diagnostic_uri(o, d, num, uri, 0, addinfo);
 }
 
+
+void yaz_add_sru_update_diagnostic(ODR o, Z_SRW_diagnostic **d,
+                                   int *num, int code, const char *addinfo)
+{
+    char uri[40];
+    
+    sprintf(uri, "info:srw/diagnostic/12/%d", code);
+    yaz_add_srw_diagnostic_uri(o, d, num, uri, 0, addinfo);
+}
+
+
+static void grab_charset(ODR o, const char *content_type, char **charset)
+{
+    if (charset)
+    { 
+        const char *charset_p = 0;
+        if (content_type && (charset_p = strstr(content_type, "; charset=")))
+        {
+            int i = 0;
+            charset_p += 10;
+            while (i < 20 && charset_p[i] &&
+                   !strchr("; \n\r", charset_p[i]))
+                i++;
+            *charset = (char*) odr_malloc(o, i+1);
+            memcpy(*charset, charset_p, i);
+            (*charset)[i] = '\0';
+        }
+    }
+}
+
 int yaz_srw_decode(Z_HTTP_Request *hreq, Z_SRW_PDU **srw_pdu,
                    Z_SOAP **soap_package, ODR decode, char **charset)
 {
@@ -330,16 +373,12 @@ int yaz_srw_decode(Z_HTTP_Request *hreq, Z_SRW_PDU **srw_pdu,
             char *db = "Default";
             const char *p0 = hreq->path, *p1;
             int ret = -1;
-            const char *charset_p = 0;
             
             static Z_SOAP_Handler soap_handlers[4] = {
-#if HAVE_XML2
-                {"http://www.loc.gov/zing/srw/", 0,
-                 (Z_SOAP_fun) yaz_srw_codec},
-                {"http://www.loc.gov/zing/srw/v1.0/", 0,
-                 (Z_SOAP_fun) yaz_srw_codec},
-                {"http://www.loc.gov/zing/srw/update/", 0,
-                 (Z_SOAP_fun) yaz_ucp_codec},
+#if YAZ_HAVE_XML2
+                { YAZ_XMLNS_SRU_v1_1, 0, (Z_SOAP_fun) yaz_srw_codec },
+                { YAZ_XMLNS_SRU_v1_0, 0, (Z_SOAP_fun) yaz_srw_codec },
+                { YAZ_XMLNS_UPDATE_v0_9, 0, (Z_SOAP_fun) yaz_ucp_codec },
 #endif
                 {0, 0, 0}
             };
@@ -356,17 +395,8 @@ int yaz_srw_decode(Z_HTTP_Request *hreq, Z_SRW_PDU **srw_pdu,
                 db[p1 - p0] = '\0';
             }
 
-            if (charset && (charset_p = strstr(content_type, "; charset=")))
-            {
-                int i = 0;
-                charset_p += 10;
-                while (i < 20 && charset_p[i] &&
-                       !strchr("; \n\r", charset_p[i]))
-                    i++;
-                *charset = (char*) odr_malloc(decode, i+1);
-                memcpy(*charset, charset_p, i);
-                (*charset)[i] = '\0';
-            }
+            grab_charset(decode, content_type, charset);
+
             ret = z_soap_codec(decode, soap_package, 
                                &hreq->content_buf, &hreq->content_len,
                                soap_handlers);
@@ -398,10 +428,10 @@ int yaz_srw_decode(Z_HTTP_Request *hreq, Z_SRW_PDU **srw_pdu,
     return 2;
 }
 
-static int yaz_sru_integer_decode(ODR odr, const char *pname, 
+static int yaz_sru_decode_integer(ODR odr, const char *pname, 
                                   const char *valstr, int **valp,
-                                  Z_SRW_diagnostic **diag, int *num_diag)
-
+                                  Z_SRW_diagnostic **diag, int *num_diag,
+                                  int min_value)
 {
     int ival;
     if (!valstr)
@@ -412,9 +442,16 @@ static int yaz_sru_integer_decode(ODR odr, const char *pname,
                                YAZ_SRW_UNSUPP_PARAMETER_VALUE, pname);
         return 0;
     }
+    if (min_value >= 0 && ival < min_value)
+    {
+        yaz_add_srw_diagnostic(odr, diag, num_diag,
+                               YAZ_SRW_UNSUPP_PARAMETER_VALUE, pname);
+        return 0;
+    }
     *valp = odr_intdup(odr, ival);
     return 1;
 }
+
 /**
   http://www.loc.gov/z3950/agency/zing/srw/service.html
 */ 
@@ -422,32 +459,29 @@ int yaz_sru_decode(Z_HTTP_Request *hreq, Z_SRW_PDU **srw_pdu,
                    Z_SOAP **soap_package, ODR decode, char **charset,
                    Z_SRW_diagnostic **diag, int *num_diag)
 {
-#if HAVE_XML2
+#if YAZ_HAVE_XML2
     static Z_SOAP_Handler soap_handlers[2] = {
-        {"http://www.loc.gov/zing/srw/", 0,
-         (Z_SOAP_fun) yaz_srw_codec},
+        {YAZ_XMLNS_SRU_v1_1, 0, (Z_SOAP_fun) yaz_srw_codec},
         {0, 0, 0}
     };
 #endif
     const char *content_type = z_HTTP_header_lookup(hreq->headers,
-                                                    "Content-Type");
+                                            "Content-Type");
+
     /*
-      SRU GET: allow any content type.
+      SRU GET: ignore content type.
       SRU POST: we support "application/x-www-form-urlencoded";
       not  "multipart/form-data" .
     */
-    if (!strcmp(hreq->method, "GET") 
-        ||
-        (!strcmp(hreq->method, "POST") 
-         && content_type &&
-         !yaz_strcmp_del("application/x-www-form-urlencoded",
-                         content_type, "; ")
-            )
-        )
+    if (!strcmp(hreq->method, "GET")
+        || 
+             (!strcmp(hreq->method, "POST") && content_type &&
+              !yaz_strcmp_del("application/x-www-form-urlencoded",
+                              content_type, "; ")))
     {
         char *db = "Default";
         const char *p0 = hreq->path, *p1;
-#if HAVE_XML2
+#if YAZ_HAVE_XML2
         const char *operation = 0;
         char *version = 0;
         char *query = 0;
@@ -470,8 +504,10 @@ int yaz_sru_decode(Z_HTTP_Request *hreq, Z_SRW_PDU **srw_pdu,
         char **uri_name;
         char **uri_val;
 
-        if (charset)
-            *charset = 0;
+        grab_charset(decode, content_type, charset);
+        if (charset && *charset == 0 && !strcmp(hreq->method, "GET"))
+            *charset = "UTF-8";
+
         if (*p0 == '/')
             p0++;
         p1 = strchr(p0, '?');
@@ -486,7 +522,7 @@ int yaz_sru_decode(Z_HTTP_Request *hreq, Z_SRW_PDU **srw_pdu,
         if (!strcmp(hreq->method, "POST"))
             p1 = hreq->content_buf;
         yaz_uri_array(p1, decode, &uri_name, &uri_val);
-#if HAVE_XML2
+#if YAZ_HAVE_XML2
         if (uri_name)
         {
             int i;
@@ -586,13 +622,13 @@ int yaz_sru_decode(Z_HTTP_Request *hreq, Z_SRW_PDU **srw_pdu,
             sr->u.request->recordPacking = recordPacking;
             sr->u.request->stylesheet = stylesheet;
 
-            yaz_sru_integer_decode(decode, "maximumRecords", maximumRecords, 
+            yaz_sru_decode_integer(decode, "maximumRecords", maximumRecords, 
                                    &sr->u.request->maximumRecords, 
-                                   diag, num_diag);
-
-            yaz_sru_integer_decode(decode, "startRecord", startRecord, 
+                                   diag, num_diag, 0);
+            
+            yaz_sru_decode_integer(decode, "startRecord", startRecord, 
                                    &sr->u.request->startRecord,
-                                   diag, num_diag);
+                                   diag, num_diag, 1);
 
             sr->u.request->database = db;
 
@@ -664,15 +700,15 @@ int yaz_sru_decode(Z_HTTP_Request *hreq, Z_SRW_PDU **srw_pdu,
                     YAZ_SRW_MANDATORY_PARAMETER_NOT_SUPPLIED, "scanClause");
             sr->u.scan_request->database = db;
             
-            yaz_sru_integer_decode(decode, "maximumTerms",
+            yaz_sru_decode_integer(decode, "maximumTerms",
                                    maximumTerms, 
                                    &sr->u.scan_request->maximumTerms,
-                                   diag, num_diag);
-
-            yaz_sru_integer_decode(decode, "responsePosition",
+                                   diag, num_diag, 0);
+            
+            yaz_sru_decode_integer(decode, "responsePosition",
                                    responsePosition, 
                                    &sr->u.scan_request->responsePosition,
-                                   diag, num_diag);
+                                   diag, num_diag, 0);
 
             sr->u.scan_request->stylesheet = stylesheet;
 
@@ -730,23 +766,48 @@ Z_SRW_extra_record *yaz_srw_get_extra_record(ODR o)
 {
     Z_SRW_extra_record *res = (Z_SRW_extra_record *)
         odr_malloc(o, sizeof(*res));
-    res->type = 1;
-    res->recordReviewCode = 0;
-    res->recordReviewNote = 0;
-    res->recordId = 0;
-    res->nonDupRecordId = 0;
-    res->recordLockStatus = 0;
-    res->recordOldVersion = 0;
+
+    res->extraRecordData_buf = 0;
+    res->extraRecordData_len = 0;
+    res->recordIdentifier = 0;
     return res;
 }
 
-Z_SRW_PDU *yaz_srw_get(ODR o, int which)
+
+Z_SRW_record *yaz_srw_get_records(ODR o, int n)
+{
+    Z_SRW_record *res = (Z_SRW_record *) odr_malloc(o, n * sizeof(*res));
+    int i;
+
+    for (i = 0; i<n; i++)
+    {
+        res[i].recordSchema = 0;
+        res[i].recordPacking = Z_SRW_recordPacking_string;
+        res[i].recordData_buf = 0;
+        res[i].recordData_len = 0;
+        res[i].recordPosition = 0;
+    }
+    return res;
+}
+
+Z_SRW_record *yaz_srw_get_record(ODR o)
 {
-    Z_SRW_PDU *sr = (Z_SRW_PDU *) odr_malloc(o, sizeof(*o));
+    return yaz_srw_get_records(o, 1);
+}
 
-    sr->username = 0;
-    sr->password = 0;
-    sr->srw_version = odr_strdup(o, "1.1");
+Z_SRW_PDU *yaz_srw_get_core_v_1_1(ODR o)
+{
+    Z_SRW_PDU *p = (Z_SRW_PDU *) odr_malloc(o, sizeof(*p));
+    p->srw_version = odr_strdup(o, "1.1");
+    p->username = 0;
+    p->password = 0;
+    p->extra_args = 0;
+    return p;
+}
+
+Z_SRW_PDU *yaz_srw_get(ODR o, int which)
+{
+    Z_SRW_PDU *sr = yaz_srw_get_core_v_1_1(o);
     sr->which = which;
     switch(which)
     {
@@ -822,15 +883,13 @@ Z_SRW_PDU *yaz_srw_get(ODR o, int which)
             odr_malloc(o, sizeof(*sr->u.update_request));
        sr->u.update_request->database = 0;
        sr->u.update_request->stylesheet = 0;
-        sr->u.update_request->record.recordSchema = 0;
-        sr->u.update_request->record.recordPacking = Z_SRW_recordPacking_XML;
+        sr->u.update_request->record = 0;
        sr->u.update_request->recordId = 0;
-       sr->u.update_request->recordVersion = 0;
-       sr->u.update_request->recordOldVersion = 0;
-        sr->u.update_request->record.recordData_buf = 0;
-        sr->u.update_request->record.recordData_len = 0;
+       sr->u.update_request->recordVersions = 0;
+       sr->u.update_request->num_recordVersions = 0;
         sr->u.update_request->extra_record = 0;
-        sr->u.update_request->extraRequestData = 0;
+        sr->u.update_request->extraRequestData_buf = 0;
+        sr->u.update_request->extraRequestData_len = 0;
        sr->u.request->database = 0;
         break;
     case Z_SRW_update_response:
@@ -838,15 +897,12 @@ Z_SRW_PDU *yaz_srw_get(ODR o, int which)
             odr_malloc(o, sizeof(*sr->u.update_response));
        sr->u.update_response->operationStatus = 0;
        sr->u.update_response->recordId = 0;
-       sr->u.update_response->recordVersion = 0;
-       sr->u.update_response->recordChecksum = 0;
-       sr->u.update_response->record.recordData_buf = 0;
-       sr->u.update_response->record.recordData_len = 0;
-       sr->u.update_response->record.recordSchema = 0;
-       sr->u.update_response->record.recordPacking =
-           Z_SRW_recordPacking_XML;
+       sr->u.update_response->recordVersions = 0;
+       sr->u.update_response->num_recordVersions = 0;
+       sr->u.update_response->record = 0;
         sr->u.update_response->extra_record = 0;
-        sr->u.update_response->extraResponseData = 0;
+        sr->u.update_response->extraResponseData_buf = 0;
+        sr->u.update_response->extraResponseData_len = 0;
        sr->u.update_response->diagnostics = 0;
        sr->u.update_response->num_diagnostics = 0;
     }
@@ -1168,11 +1224,13 @@ int yaz_sru_get_encode(Z_HTTP_Request *hreq, Z_SRW_PDU *srw_pdu,
 
     if (yaz_get_sru_parms(srw_pdu, encode, name, value))
         return -1;
-    yaz_array_to_uri(&uri_args, encode, name, value);
+    yaz_array_to_uri_ex(&uri_args, encode, name, value, srw_pdu->extra_args);
 
     hreq->method = "GET";
     
-    path = odr_malloc(encode, strlen(hreq->path) + strlen(uri_args) + 3);
+    path = odr_malloc(encode, strlen(hreq->path) + strlen(uri_args) + 4
+                      +(srw_pdu->extra_args ? strlen(srw_pdu->extra_args) : 0)
+        );
     sprintf(path, "%s?%s", hreq->path, uri_args);
     hreq->path = path;
 
@@ -1190,7 +1248,7 @@ int yaz_sru_post_encode(Z_HTTP_Request *hreq, Z_SRW_PDU *srw_pdu,
     if (yaz_get_sru_parms(srw_pdu, encode, name, value))
         return -1;
 
-    yaz_array_to_uri(&uri_args, encode, name, value);
+    yaz_array_to_uri_ex(&uri_args, encode, name, value, srw_pdu->extra_args);
 
     hreq->method = "POST";
     
@@ -1206,11 +1264,15 @@ int yaz_sru_post_encode(Z_HTTP_Request *hreq, Z_SRW_PDU *srw_pdu,
 int yaz_sru_soap_encode(Z_HTTP_Request *hreq, Z_SRW_PDU *srw_pdu,
                         ODR odr, const char *charset)
 {
-    Z_SOAP_Handler handlers[2] = {
-        {"http://www.loc.gov/zing/srw/", 0, (Z_SOAP_fun) yaz_srw_codec},
+    Z_SOAP_Handler handlers[3] = {
+#if YAZ_HAVE_XML2
+        {YAZ_XMLNS_SRU_v1_1, 0, (Z_SOAP_fun) yaz_srw_codec},
+        {YAZ_XMLNS_UPDATE_v0_9, 0, (Z_SOAP_fun) yaz_ucp_codec},
+#endif
         {0, 0, 0}
     };
     Z_SOAP *p = (Z_SOAP*) odr_malloc(odr, sizeof(*p));
+
     z_HTTP_header_add_content_type(odr,
                                    &hreq->headers,
                                    "text/xml", charset);
@@ -1223,13 +1285,55 @@ int yaz_sru_soap_encode(Z_HTTP_Request *hreq, Z_SRW_PDU *srw_pdu,
     p->u.generic->ns = 0;
     p->u.generic->p = srw_pdu;
     p->ns = "http://schemas.xmlsoap.org/soap/envelope/";
-    
+
+#if YAZ_HAVE_XML2
+    if (srw_pdu->which == Z_SRW_update_request ||
+        srw_pdu->which == Z_SRW_update_response)
+        p->u.generic->no = 1; /* second handler */
+#endif
     return z_soap_codec_enc(odr, &p,
                             &hreq->content_buf,
                             &hreq->content_len, handlers,
                             charset);
 }
 
+Z_SRW_recordVersion *yaz_srw_get_record_versions(ODR odr, int num )
+{
+    Z_SRW_recordVersion *ver 
+        = (Z_SRW_recordVersion *) odr_malloc( odr, num * sizeof(*ver) );
+    int i;
+    for ( i=0; i < num; ++i ){
+        ver[i].versionType = 0;
+        ver[i].versionValue = 0;
+    }
+    return ver;
+}
+
+const char *yaz_srw_pack_to_str(int pack)
+{
+    switch(pack)
+    {
+    case Z_SRW_recordPacking_string:
+        return "string";
+    case Z_SRW_recordPacking_XML:
+        return "xml";
+    case Z_SRW_recordPacking_URL:
+        return "url";
+    }
+    return 0;
+}
+
+int yaz_srw_str_to_pack(const char *str)
+{
+    if (!strcmp(str, "string"))
+        return Z_SRW_recordPacking_string;
+    if (!strcmp(str, "xml"))
+        return Z_SRW_recordPacking_XML;
+    if (!strcmp(str, "url"))
+        return Z_SRW_recordPacking_URL;
+    return -1;
+}
+
 /*
  * Local variables:
  * c-basic-offset: 4