1 files changed, 194 insertions, 0 deletions
diff --git a/ext/http11/http11.c b/ext/http11/http11.c
index 7a84db8..2c6e119 100644
--- a/ext/http11/http11.c
+++ b/ext/http11/http11.c
@@ -8,9 +8,11 @@
  #include <string.h>
  #include "http11_parser.h"
  #include <ctype.h>
+#include "tst.h"
  
  static VALUE mMongrel;
  static VALUE cHttpParser;
+static VALUE cURIClassifier;
  static VALUE eHttpParserError;
  
  #define id_handler_map rb_intern("@handler_map")
@@ -361,6 +363,191 @@ VALUE HttpParser_nread(VALUE self)
    return INT2FIX(http->nread);
  }
  
+
+void URIClassifier_free(void *data) 
+{
+  TRACE();
+
+  if(data) {
+    tst_cleanup((struct tst *)data);
+  }
+}
+
+
+
+VALUE URIClassifier_alloc(VALUE klass)
+{
+  VALUE obj;
+  struct tst *tst = tst_init(TRIE_INCREASE);
+  TRACE();
+  assert(tst && "failed to initialize trie structure");
+
+  obj = Data_Wrap_Struct(klass, NULL, URIClassifier_free, tst);
+
+  return obj;
+}
+
+/**
+ * call-seq:
+ *    URIClassifier.new -> URIClassifier
+ *
+ * Initializes a new URIClassifier object that you can use to associate URI sequences
+ * with objects.  You can actually use it with any string sequence and any objects,
+ * but it's mostly used with URIs.
+ *
+ * It uses TST from http://www.octavian.org/cs/software.html to build an ternary search
+ * trie to hold all of the URIs.  It uses this to do an initial search for the a URI
+ * prefix, and then to break the URI into SCRIPT_NAME and PATH_INFO portions.  It actually
+ * will do two searches most of the time in order to find the right handler for the
+ * registered prefix portion.
+ *
+ */
+VALUE URIClassifier_init(VALUE self)
+{
+  VALUE hash;
+
+  /* we create an internal hash to protect stuff from the GC */
+  hash = rb_hash_new();
+  rb_ivar_set(self, id_handler_map, hash);
+
+  return self;
+}
+
+
+/**
+ * call-seq:
+ *    uc.register("/someuri", SampleHandler.new) -> nil
+ *
+ * Registers the SampleHandler (one for all requests) with the "/someuri".
+ * When URIClassifier::resolve is called with "/someuri" it'll return
+ * SampleHandler immediately.  When called with "/someuri/iwant" it'll also
+ * return SomeHandler immediatly, with no additional searches, but it will
+ * return path info with "/iwant".
+ *
+ * You actually can reuse this class to register nearly anything and 
+ * quickly resolve it.  This could be used for caching, fast mapping, etc.
+ * The downside is it uses much more memory than a Hash, but it can be
+ * a lot faster.  It's main advantage is that it works on prefixes, which
+ * is damn hard to get right with a Hash.
+ */
+VALUE URIClassifier_register(VALUE self, VALUE uri, VALUE handler)
+{
+  int rc = 0;
+  void *ptr = NULL;
+  struct tst *tst = NULL;
+  DATA_GET(self, struct tst, tst);
+
+  rc = tst_insert((unsigned char *)StringValueCStr(uri), (void *)handler , tst, 0, &ptr);
+
+  if(rc == TST_DUPLICATE_KEY) {
+    rb_raise(rb_eStandardError, "Handler already registered with that name");
+  } else if(rc == TST_ERROR) {
+    rb_raise(rb_eStandardError, "Memory error registering handler");
+  } else if(rc == TST_NULL_KEY) {
+    rb_raise(rb_eStandardError, "URI was empty");
+  }
+
+  rb_hash_aset(rb_ivar_get(self, id_handler_map), uri, handler);
+
+  return Qnil;
+}
+
+
+/**
+ * call-seq:
+ *    uc.unregister("/someuri")
+ *
+ * Yep, just removes this uri and it's handler from the trie.
+ */
+VALUE URIClassifier_unregister(VALUE self, VALUE uri)
+{
+  void *handler = NULL;
+  struct tst *tst = NULL;
+  DATA_GET(self, struct tst, tst);
+
+  handler = tst_delete((unsigned char *)StringValueCStr(uri), tst);
+
+  if(handler) {
+    rb_hash_delete(rb_ivar_get(self, id_handler_map), uri);
+
+    return (VALUE)handler;
+  } else {
+    return Qnil;
+  }
+}
+
+
+/**
+ * call-seq:
+ *    uc.resolve("/someuri") -> "/someuri", "", handler
+ *    uc.resolve("/someuri/pathinfo") -> "/someuri", "/pathinfo", handler
+ *    uc.resolve("/notfound/orhere") -> nil, nil, nil
+ *    uc.resolve("/") -> "/", "/", handler  # if uc.register("/", handler)
+ *    uc.resolve("/path/from/root") -> "/", "/path/from/root", handler  # if uc.register("/", handler) 
+ * 
+ * Attempts to resolve either the whole URI or at the longest prefix, returning
+ * the prefix (as script_info), path (as path_info), and registered handler
+ * (usually an HttpHandler).  If it doesn't find a handler registered at the longest
+ * match then it returns nil,nil,nil.
+ *
+ * Because the resolver uses a trie you are able to register a handler at *any* character
+ * in the URI and it will be handled as long as it's the longest prefix.  So, if you 
+ * registered handler #1 at "/something/lik", and #2 at "/something/like/that", then a
+ * a search for "/something/like" would give you #1.  A search for "/something/like/that/too"
+ * would give you #2.
+ * 
+ * This is very powerful since it means you can also attach handlers to parts of the ; 
+ * (semi-colon) separated path params, any part of the path, use off chars, anything really.
+ * It also means that it's very efficient to do this only taking as long as the URI has
+ * characters.
+ *
+ * A slight modification to the CGI 1.2 standard is given for handlers registered to "/".
+ * CGI expects all CGI scripts to be at some script path, so it doesn't really say anything
+ * about a script that handles the root.  To make this work, the resolver will detect that
+ * the requested handler is at "/", and return that for script_name, and then simply return
+ * the full URI back as path_info.
+ *
+ * It expects strings with no embedded '\0' characters.  Don't try other string-like stuff yet.
+ */
+VALUE URIClassifier_resolve(VALUE self, VALUE uri)
+{
+  void *handler = NULL;
+  int pref_len = 0;
+  struct tst *tst = NULL;
+  VALUE result;
+  unsigned char *uri_str = NULL;
+
+  DATA_GET(self, struct tst, tst);
+  uri_str = (unsigned char *)StringValueCStr(uri);
+
+  handler = tst_search(uri_str, tst, TST_LONGEST_MATCH, &pref_len);
+
+  /* setup for multiple return values */
+  result = rb_ary_new();
+
+  if(handler) {
+    rb_ary_push(result, rb_str_substr (uri, 0, pref_len));
+    /* compensate for a script_name="/" where we need to add the "/" to path_info to keep it consistent */
+    if(pref_len == 1 && uri_str[0] == '/') {
+      /* matches the root URI so we have to use the whole URI as the path_info */
+      rb_ary_push(result, uri);
+    } else {
+      /* matches a script so process like normal */
+      rb_ary_push(result, rb_str_substr(uri, pref_len, RSTRING(uri)->len));
+    }
+
+    rb_ary_push(result, (VALUE)handler);
+  } else {
+    /* not found so push back nothing */
+    rb_ary_push(result, Qnil);
+    rb_ary_push(result, Qnil);
+    rb_ary_push(result, Qnil);
+  }
+
+  return result;
+}
+
+
  void Init_http11()
  {
  
@@ -399,4 +586,11 @@ void Init_http11()
    rb_define_method(cHttpParser, "error?", HttpParser_has_error,0);
    rb_define_method(cHttpParser, "finished?", HttpParser_is_finished,0);
    rb_define_method(cHttpParser, "nread", HttpParser_nread,0);
+
+  cURIClassifier = rb_define_class_under(mMongrel, "URIClassifier", rb_cObject);
+  rb_define_alloc_func(cURIClassifier, URIClassifier_alloc);
+  rb_define_method(cURIClassifier, "initialize", URIClassifier_init, 0);
+  rb_define_method(cURIClassifier, "register", URIClassifier_register, 2);
+  rb_define_method(cURIClassifier, "unregister", URIClassifier_unregister, 1);
+  rb_define_method(cURIClassifier, "resolve", URIClassifier_resolve, 1);
  }