sgml filter doesn't interpret entities

diff --git a/debian/changelog b/debian/changelog
index 8134884..40a2a10 100644 (file)
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,9 @@
+yaz (1.9.1-3) unstable; urgency=low
+
+  * don't let grs.sgml filter interpret &-entities.
+
+ -- Adam Dickmeiss <adam@indexdata.dk>  Tue,  8 Oct 2002 22:13:30 +0100
+
 yaz (1.9.1-2) unstable; urgency=low
 
   * official release.

diff --git a/debian/rules b/debian/rules
index ca06c9f..4370644 100755 (executable)
--- a/debian/rules
+++ b/debian/rules
@@ -84,7 +84,7 @@ stamp-install: stamp-build
 binary-indep: build install
        dh_testdir
        dh_testroot
-       dh_installdocs -p yaz-doc doc/*.html doc/*.pdf
+       dh_installdocs -p yaz-doc doc/*.html doc/*.pdf doc/*.png
        dh_installchangelogs -p yaz-doc
        dh_link -p yaz-doc
        dh_strip -p yaz-doc

diff --git a/include/yaz/data1.h b/include/yaz/data1.h
index e214fa5..0db650f 100644 (file)
--- a/include/yaz/data1.h
+++ b/include/yaz/data1.h
@@ -23,7 +23,7 @@
  * LIABILITY, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE
  * OF THIS SOFTWARE.
  *
- * $Id: data1.h,v 1.17 2002-09-24 07:58:59 adam Exp $
+ * $Id: data1.h,v 1.18 2002-10-08 20:14:44 adam Exp $
  */
 
 #ifndef DATA1_H
@@ -197,6 +197,7 @@ typedef struct data1_xattr {
     char *name;
     char *value;
     struct data1_xattr *next;
+    unsigned short what;  /* DATA1I_text, .. see data1_node.u.data */
 } data1_xattr;
 
 typedef struct data1_absyn
@@ -257,16 +258,18 @@ typedef struct data1_node
        {
            char *data;      /* filename or data */
            int len;
-        /* text inclusion */
+            /* text inclusion */
 #define DATA1I_inctxt 1
-        /* binary data inclusion */
+            /* binary data inclusion */
 #define DATA1I_incbin 2
         /* text data */
 #define DATA1I_text 3 
-        /* numerical data */
+            /* numerical data */
 #define DATA1I_num 4
-        /* object identifier */
-#define DATA1I_oid 5         
+            /* object identifier */
+#define DATA1I_oid 5
+            /* XML text */
+#define DATA1I_xmltext 6
             unsigned what:7;
            unsigned formatted_text : 1;   /* newlines are significant */
        } data;

diff --git a/retrieval/d1_prtree.c b/retrieval/d1_prtree.c
index 6984659..4354b30 100644 (file)
--- a/retrieval/d1_prtree.c
+++ b/retrieval/d1_prtree.c
@@ -3,7 +3,7 @@
  * See the file LICENSE for details.
  * Sebastian Hammer, Adam Dickmeiss
  *
- * $Id: d1_prtree.c,v 1.10 2002-09-24 13:58:13 adam Exp $
+ * $Id: d1_prtree.c,v 1.11 2002-10-08 20:14:44 adam Exp $
  */
 
 #include <yaz/log.h>
@@ -71,6 +71,11 @@ static void pr_tree (data1_handle dh, data1_node *n, FILE *out, int level)
            pr_string (out, n->u.data.data, n->u.data.len);
            fprintf (out, "'\n");
            break;
+       case DATA1I_xmltext:
+           fprintf (out, "xml text '");
+           pr_string (out, n->u.data.data, n->u.data.len);
+           fprintf (out, "'\n");
+            break;
        default:
            fprintf (out, "unknown(%d)\n", n->u.data.what);
            break;

diff --git a/retrieval/d1_read.c b/retrieval/d1_read.c
index 2660283..934bd87 100644 (file)
--- a/retrieval/d1_read.c
+++ b/retrieval/d1_read.c
@@ -3,7 +3,7 @@
  * See the file LICENSE for details.
  * Sebastian Hammer, Adam Dickmeiss
  *
- * $Id: d1_read.c,v 1.52 2002-09-24 07:58:59 adam Exp $
+ * $Id: d1_read.c,v 1.53 2002-10-08 20:14:44 adam Exp $
  */
 
 #include <assert.h>
@@ -187,6 +187,8 @@ data1_node *data1_mk_preprocess_n (data1_handle dh, NMEM nmem,
         *p = (data1_xattr*) nmem_malloc (nmem, sizeof(**p));
         (*p)->name = nmem_strdup (nmem, *attr++);
         (*p)->value = nmem_strdup (nmem, *attr++);
+        (*p)->what = DATA1I_text;
+
         p = &(*p)->next;
     }
     *p = 0;
@@ -223,6 +225,7 @@ data1_node *data1_mk_tag_n (data1_handle dh, NMEM nmem,
         *p = (data1_xattr*) nmem_malloc (nmem, sizeof(**p));
         (*p)->name = nmem_strdup (nmem, *attr++);
         (*p)->value = nmem_strdup (nmem, *attr++);
+        (*p)->what = DATA1I_text;
         p = &(*p)->next;
     }
     *p = 0;
@@ -246,6 +249,7 @@ void data1_tag_add_attr (data1_handle dh, NMEM nmem,
         *p = (data1_xattr*) nmem_malloc (nmem, sizeof(**p));
         (*p)->name = nmem_strdup (nmem, *attr++);
         (*p)->value = nmem_strdup (nmem, *attr++);
+        (*p)->what = DATA1I_text;
         p = &(*p)->next;
     }
     *p = 0;
@@ -489,6 +493,11 @@ data1_node *data1_mk_tag_data_text_uni (data1_handle dh, data1_node *at,
 
 static int ampr (int (*get_byte)(void *fh), void *fh, int *amp)
 {
+#if 1
+    int c = (*get_byte)(fh);
+    *amp = 0;
+    return c;
+#else
     int c = (*get_byte)(fh);
     *amp = 0;
     if (c == '&')
@@ -524,6 +533,7 @@ static int ampr (int (*get_byte)(void *fh), void *fh, int *amp)
         }
     }
     return c;
+#endif
 }
 
 data1_xattr *data1_read_xattr (data1_handle dh, NMEM m,
@@ -545,6 +555,7 @@ data1_xattr *data1_read_xattr (data1_handle dh, NMEM m,
        p->next = 0;
        pp = &p->next;
        p->value = 0;
+        p->what = DATA1I_xmltext;
        
        wrbuf_rewind(wrbuf);
        while (c && c != '=' && c != '>' && c != '/' && !d1_isspace(c))
@@ -837,7 +848,7 @@ data1_node *data1_read_nodex (data1_handle dh, NMEM m,
                continue;
            }
            res = data1_mk_node2 (dh, m, DATA1N_data, parent);
-           res->u.data.what = DATA1I_text;
+           res->u.data.what = DATA1I_xmltext;
            res->u.data.formatted_text = 0;
            d1_stack[level] = res;
            

diff --git a/retrieval/d1_write.c b/retrieval/d1_write.c
index 3114e40..b884248 100644 (file)
--- a/retrieval/d1_write.c
+++ b/retrieval/d1_write.c
@@ -3,7 +3,7 @@
  * See the file LICENSE for details.
  * Sebastian Hammer, Adam Dickmeiss
  *
- * $Id: d1_write.c,v 1.18 2002-09-24 13:58:13 adam Exp $
+ * $Id: d1_write.c,v 1.19 2002-10-08 20:14:44 adam Exp $
  */
 
 #include <string.h>
@@ -65,6 +65,28 @@ static void wrbuf_put_cdata(WRBUF b, const char *msg)
     wrbuf_write_cdata (b, msg, strlen(msg));
 }
 
+static void wrbuf_put_xattr(WRBUF b, data1_xattr *p)
+{
+    for (; p; p = p->next)
+    {
+        wrbuf_putc (b, ' ');
+        if (p->what == DATA1I_xmltext)
+            wrbuf_puts (b, p->name);
+        else
+            wrbuf_put_cdata (b, p->name);
+        if (p->value)
+        {
+            wrbuf_putc (b, '=');
+            wrbuf_putc (b, '"');
+            if (p->what == DATA1I_text)
+                wrbuf_put_cdata (b, p->value);
+            else
+                wrbuf_puts (b, p->value);
+            wrbuf_putc (b, '"');
+        }
+    }
+}
+
 static int nodetoidsgml(data1_node *n, int select, WRBUF b, int col,
                         int pretty_format)
 {
@@ -82,15 +104,7 @@ static int nodetoidsgml(data1_node *n, int select, WRBUF b, int col,
                 indent (b, col);
            wrbuf_puts (b, "<?");
             wrbuf_put_cdata (b, c->u.preprocess.target);
-            for (p = c->u.preprocess.attributes; p; p = p->next)
-            {
-                wrbuf_putc (b, ' ');
-                wrbuf_put_cdata (b, p->name);
-                wrbuf_putc (b, '=');
-                wrbuf_putc (b, '"');
-                wrbuf_put_cdata (b, p->value);
-                wrbuf_putc (b, '"');
-            }
+            wrbuf_put_xattr (b, c->u.preprocess.attributes);
             if (c->child)
                 wrbuf_puts(b, " ");
             if (nodetoidsgml(c, select, b, (col > 40) ? 40 : col+2,
@@ -116,15 +130,7 @@ static int nodetoidsgml(data1_node *n, int select, WRBUF b, int col,
                     indent (b, col);
                wrbuf_puts (b, "<");    
                wrbuf_put_cdata (b, tag);
-               for (p = c->u.tag.attributes; p; p = p->next)
-               {
-                   wrbuf_putc (b, ' ');
-                   wrbuf_put_cdata (b, p->name);
-                   wrbuf_putc (b, '=');
-                   wrbuf_putc (b, '"');
-                   wrbuf_put_cdata (b, p->value);
-                   wrbuf_putc (b, '"');
-               }
+                wrbuf_put_xattr (b, c->u.tag.attributes);
                wrbuf_puts(b, ">");
                 if (pretty_format)
                     wrbuf_puts(b, "\n");
@@ -153,6 +159,9 @@ static int nodetoidsgml(data1_node *n, int select, WRBUF b, int col,
                 wrbuf_puts (b, "<!--");
            switch (c->u.data.what)
            {
+            case DATA1I_xmltext:
+               wrbuf_write(b, c->u.data.data, c->u.data.len);
+                break;
            case DATA1I_text:
                 if (!pretty_format || c->u.data.formatted_text)
                 {