4 files changed, 376 insertions, 0 deletions

diff --git a/src/load/load_features.c b/src/load/load_features.c
new file mode 100644
index 0000000..b18031a
--- a/dev/null
+++ b/src/load/load_features.c
@@ -0,0 +1,167 @@
+#include "load_features.h"
+#include <libxml/parser.h>
+#include <stdbool.h>
+#include <asn.h>
+#include <objgbseq.h>
+
+/*
+ * An NCBI GBSeq structure to hold the data for the current record.
+ */
+GBSeqPtr g;
+
+bool in_element;
+
+static void
+lf_startDocument (void *ctx ATTRIBUTE_UNUSED)
+{
+  printf ("SAX.startDocument()\n");
+
+  return;
+}
+
+static void
+lf_endDocument (void *ctx ATTRIBUTE_UNUSED)
+{
+  printf ("SAX.endDocument()\n");
+
+  return;
+}
+
+static xmlEntityPtr
+lf_getEntity (void *ctx ATTRIBUTE_UNUSED, const xmlChar *name)
+{
+  printf("SAX.getEntity(%s)\n", name);
+
+  return (NULL);
+}
+
+static void
+lf_startElement(void *ctx ATTRIBUTE_UNUSED,
+                const xmlChar *name, const xmlChar **atts)
+{
+  int i;
+
+  fprintf(stdout, "SAX.startElement(%s", (char *) name);
+  if (atts != NULL) {
+    for (i = 0;(atts[i] != NULL);i++) {
+      fprintf(stdout, ", %s='", atts[i++]);
+      if (atts[i] != NULL)
+        fprintf(stdout, "%s'", atts[i]);
+    }
+  }
+  fprintf(stdout, ")\n");
+
+  in_element = true;
+
+  return;
+}
+
+static void
+lf_endElement(void *ctx ATTRIBUTE_UNUSED, const xmlChar *name)
+{
+  printf("SAX.endElement(%s)\n", (char *) name);
+  in_element = false;
+
+  return;
+}
+
+static void
+lf_characters(void *ctx ATTRIBUTE_UNUSED, const xmlChar *ch, int len)
+{
+  if (!in_element)
+    return;
+
+  char output[40];
+  int i;
+
+  for (i = 0;(i<len) && (i < 30);i++)
+    output[i] = ch[i];
+  output[i] = 0;
+
+  printf("SAX.characters(%s, %d)\n", output, len);
+}
+
+/*
+ * 1. Parse and load the XML file into memory.
+ * 2. Insert the XML into HDF5.
+ * For an example of parsing XML with libxml2 and SAX see:
+ * [http://git.gnome.org/browse/libxml2/tree/testSAX.c].
+ */
+void
+load_features (hid_t file_id, const char* file_name)
+{
+  g = GBSeqNew ();
+
+  LIBXML_TEST_VERSION;
+
+  in_element = false;
+  
+  static xmlSAXHandler emptySAXHandlerStruct = {
+    NULL, /* internalSubset */
+    NULL, /* isStandalone */
+    NULL, /* hasInternalSubset */
+    NULL, /* hasExternalSubset */
+    NULL, /* resolveEntity */
+    lf_getEntity, /* getEntity */
+    NULL, /* entityDecl */
+    NULL, /* notationDecl */
+    NULL, /* attributeDecl */
+    NULL, /* elementDecl */
+    NULL, /* unparsedEntityDecl */
+    NULL, /* setDocumentLocator */
+    lf_startDocument, /* startDocument */
+    lf_endDocument, /* endDocument */
+    lf_startElement, /* startElement */
+    lf_endElement, /* endElement */
+    NULL, /* reference */
+    lf_characters, /* characters */
+    NULL, /* ignorableWhitespace */
+    NULL, /* processingInstruction */
+    NULL, /* comment */
+    NULL, /* xmlParserWarning */
+    NULL, /* xmlParserError */
+    NULL, /* xmlParserError */
+    NULL, /* getParameterEntity */
+    NULL, /* cdataBlock; */
+    NULL, /* externalSubset; */
+    1,
+    NULL,
+    NULL, /* startElementNs */
+    NULL, /* endElementNs */
+    NULL  /* xmlStructuredErrorFunc */
+  };
+ 
+  static xmlSAXHandlerPtr emptySAXHandler = &emptySAXHandlerStruct;
+
+  FILE *f = fopen (file_name, "r");
+
+  if (f != NULL)
+    {
+      int ret;
+      char chars[10];
+      xmlParserCtxtPtr ctxt;
+  
+      ret = fread (chars, 1, 4, f);
+      if (ret > 0)
+        {
+          ctxt = xmlCreatePushParserCtxt (emptySAXHandler, NULL, 
+                                          chars, ret, file_name);
+          while ((ret = fread(chars, 1, 3, f)) > 0)
+            {
+              xmlParseChunk (ctxt, chars, ret, 0);
+            }
+          xmlParseChunk (ctxt, chars, 0, 1);
+          xmlFreeParserCtxt(ctxt);
+        }
+      fclose (f);
+    }
+  else
+    {
+      xmlGenericError (xmlGenericErrorContext,
+                       "Cannot read file.");
+    }
+
+  GBSeqFree (g);
+
+  return;
+}