rroonga 4.0.6-x64-mingw32 → 4.0.7-x64-mingw32

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 700ef8e4151922ae4756fb908ceb89447b242a77
-  data.tar.gz: 32b88a7353fbb778f7031d72e745ff2b53e77914
+  metadata.gz: 00882b4e6fd873fc8ae74cf36d0795d36ec72aff
+  data.tar.gz: b10a63452da0149c8af43c8ae3ad132499efe2a9
 SHA512:
-  metadata.gz: 4fea5bb20d7124392327e107b8cd765bd2b91e197d378f866fc2f99465acc651a5ce82a8f2c8a8a45893a124e3a875e88985376c7a8aa3afa2408018452919ff
-  data.tar.gz: 2e74dd3f0db3c31d75015c407b6962a47a2f186f6b32df3a4a27632f57e045b1304e546dcbf02f367f439a1b027c541a8b412cbea77c3ccf71d7b416bd67fbe7
+  metadata.gz: b3e4e15326531f0b7666d70da10dfbc36a4376e2427522966d5363c8f27aee3d67d9f713a60e47896e0e2803e561c7146fa01355b889360d78c886adf70560f6
+  data.tar.gz: c8d0bcdb5126cac3a7834781cbe09bbdf470f4b62ec0fbdfc9bb55f23c6810abdc78a43323e1aa20b0461c3226587ac4c34f77f4d41bd2ecda90412f2e6132f9

data/README.md

@@ -40,7 +40,7 @@ contributed patches.)
 ## Dependencies
 
 * Ruby >= 1.9.3
-* Groonga >= 4.0.4
+* Groonga >= 4.0.7
 
 ## Install
 

data/doc/text/news.textile

@@ -1,5 +1,17 @@
 h1. NEWS
 
+h2(#4-0-7). 4.0.7: 2014-12-12
+
+h3. Improvements
+
+* Supported Groonga 4.0.8. Groonga 4.0.7 or older aren't supported.
+* Added @:reuse_posting_object@ option to
+  {Groonga::IndexCursor#each}. The option improves performance by
+  reducing memory allocation. The option is disabled by default because
+  reusing the same object isn't Rubyish.
+* Added {Groonga::IndexCursor#estimate_size}. [GitHub#38]
+* Accepted string for integer, unsigned integer and float type key.
+
 h2(#4-0-6). 4.0.6: 2014-11-06
 
 h3. Improvements
@@ -11,8 +23,8 @@ h3. Improvements
 ** Supported @:lz4@ in {Groonga::VariableSizeColumn#compressed?} options.
 ** Added {Groonga::LZ4Error}.
 ** [grndump] Supported @COMPRESS_ZLIB@ and @COMPRESS_LZ4@ flags.
-** Added {Groonga::TableKeySupport#token_filters}.
-** Added {Groonga::TableKeySupport#token_filters=}.
+** Added {Groonga::Table::KeySupport#token_filters}.
+** Added {Groonga::Table::KeySupport#token_filters=}.
 ** Supported @:token_filters@ in {Groonga::Hash.create} options.
 ** Supported @:token_filters@ in {Groonga::PatriciaTrie.create} options.
 ** Supported @:token_filters@ in {Groonga::DoubleArrayTrie.create} options.

data/ext/groonga/extconf.rb

@@ -283,9 +283,6 @@ have_func("rb_sym2str", "ruby.h")
 have_func("rb_to_symbol", "ruby.h")
 have_type("enum ruby_value_type", "ruby.h")
 
-# TODO: Remove me after Groogna 4.0.8 is released.
-have_func("grn_expr_dump_plan", "groonga.h")
-
 checking_for(checking_message("--enable-debug-log option")) do
   enable_debug_log = enable_config("debug-log", false)
   if enable_debug_log

data/ext/groonga/rb-grn-expression.c

@@ -498,13 +498,11 @@ rb_grn_expression_compile (VALUE self)
  * Dump execution plan of the `expression` in string.
  *
  * @overload dump_plan
- * @since 4.0.8
+ * @since 4.0.7
  */
 static VALUE
 rb_grn_expression_dump_plan (VALUE self)
 {
-/* TODO: Remove me after Groogna 4.0.8 is released. */
-#ifdef HAVE_GRN_EXPR_DUMP_PLAN
     grn_ctx *context = NULL;
     grn_obj *expression;
     grn_obj dumped_plan;
@@ -520,9 +518,6 @@ rb_grn_expression_dump_plan (VALUE self)
     grn_obj_unlink(context, &dumped_plan);
 
     return rb_dumped_plan;
-#else
-    return rb_str_new2("");
-#endif
 }
 
 /*

data/ext/groonga/rb-grn-index-column.c

@@ -955,6 +955,99 @@ rb_grn_index_column_open_cursor (int argc, VALUE *argv, VALUE self)
         return rb_cursor;
 }
 
+/*
+ * Estimates the number of documents found by the given token ID.
+ *
+ * @example Token ID style
+ *    # Define schema
+ *    Groonga::Schema.define do |schema|
+ *      schema.create_table("Articles") do |table|
+ *        table.text("content")
+ *      end
+ *
+ *      schema.create_table("Terms",
+ *                          :type => :hash,
+ *                          :key_type => "ShortText",
+ *                          :default_tokenizer => "TokenBigram",
+ *                          :normalizer => "NormalizerAuto") do |table|
+ *        table.index("Articles.content",
+ *                    :name => "articles_content",
+ *                    :with_position => true,
+ *                    :with_section => true)
+ *      end
+ *    end
+ *    articles = Groonga["Articles"]
+ *    terms = Groonga["Terms"]
+ *    index = Groonga["Terms.articles_content"]
+ *
+ *    # Add data
+ *    articles.add(:content => "Groonga is fast")
+ *    articles.add(:content => "Rroonga is fast")
+ *    articles.add(:content => "Mroonga is fast")
+ *
+ *    # Estimate the number of documents found by token ID
+ *    p @index.estimate_size(@terms["fast"].id)    # => 7
+ *    p @index.estimate_size(@terms["Groonga"].id) # => 1
+ *
+ * @example Token record style
+ *    # Define schema
+ *    Groonga::Schema.define do |schema|
+ *      schema.create_table("Articles") do |table|
+ *        table.text("content")
+ *      end
+ *
+ *      schema.create_table("Terms",
+ *                          :type => :hash,
+ *                          :key_type => "ShortText",
+ *                          :default_tokenizer => "TokenBigram",
+ *                          :normalizer => "NormalizerAuto") do |table|
+ *        table.index("Articles.content",
+ *                    :name => "articles_content",
+ *                    :with_position => true,
+ *                    :with_section => true)
+ *      end
+ *    end
+ *    articles = Groonga["Articles"]
+ *    terms = Groonga["Terms"]
+ *    index = Groonga["Terms.articles_content"]
+ *
+ *    # Add data
+ *    articles.add(:content => "Groonga is fast")
+ *    articles.add(:content => "Rroonga is fast")
+ *    articles.add(:content => "Mroonga is fast")
+ *
+ *    # Estimate the number of documents found by token record
+ *    p @index.estimate_size(@terms["fast"])    # => 7
+ *    p @index.estimate_size(@terms["Groonga"]) # => 1
+ *
+ * @overload estimate_size(token_id)
+ *   @param [Integer, Record] token_id The token ID to be estimated.
+ *   @return [Integer] The estimated number of documents found by the
+ *     given token ID.
+ *
+ * @since 4.0.7
+ */
+static VALUE
+rb_grn_index_column_estimate_size (VALUE self, VALUE rb_token_id)
+{
+    grn_ctx *context;
+    grn_obj *column;
+    grn_obj *domain_object;
+    grn_id token_id;
+    unsigned int size;
+
+    rb_grn_index_column_deconstruct(SELF(self), &column, &context,
+                                    NULL, &domain_object,
+                                    NULL, NULL, NULL,
+                                    NULL, NULL,
+                                    NULL, NULL);
+
+    token_id = RVAL2GRNID(rb_token_id, context, domain_object, self);
+    size = grn_ii_estimate_size(context, (grn_ii *)column, token_id);
+
+    return UINT2NUM(size);
+}
+
 void
 rb_grn_init_index_column (VALUE mGrn)
 {
@@ -990,4 +1083,7 @@ rb_grn_init_index_column (VALUE mGrn)
 
     rb_define_method(rb_cGrnIndexColumn, "open_cursor",
                      rb_grn_index_column_open_cursor, -1);
+
+    rb_define_method(rb_cGrnIndexColumn, "estimate_size",
+                     rb_grn_index_column_estimate_size, 1);
 }

data/ext/groonga/rb-grn-index-cursor.c

@@ -57,7 +57,8 @@ rb_grn_index_cursor_deconstruct (RbGrnIndexCursor *rb_grn_index_cursor,
 }
 
 static VALUE
-next_value (grn_ctx *context, grn_obj *cursor, VALUE rb_table, VALUE rb_lexicon)
+next_value (VALUE rb_posting,
+            grn_ctx *context, grn_obj *cursor, VALUE rb_table, VALUE rb_lexicon)
 {
     grn_posting *posting;
     grn_id term_id;
@@ -67,7 +68,12 @@ next_value (grn_ctx *context, grn_obj *cursor, VALUE rb_table, VALUE rb_lexicon)
         return Qnil;
     }
 
-    return rb_grn_posting_new(posting, term_id, rb_table, rb_lexicon);
+    if (NIL_P(rb_posting)) {
+        return rb_grn_posting_new(posting, term_id, rb_table, rb_lexicon);
+    } else {
+        rb_grn_posting_update(rb_posting, posting, term_id);
+        return rb_posting;
+    }
 }
 
 static VALUE
@@ -84,7 +90,7 @@ rb_grn_index_cursor_next (VALUE self)
         VALUE rb_lexicon;
         rb_table = rb_iv_get(self, "@table");
         rb_lexicon = rb_iv_get(self, "@lexicon");
-        rb_posting = next_value(context, cursor, rb_table, rb_lexicon);
+        rb_posting = next_value(Qnil, context, cursor, rb_table, rb_lexicon);
     }
 
     return rb_posting;
@@ -92,14 +98,24 @@ rb_grn_index_cursor_next (VALUE self)
 }
 
 static VALUE
-rb_grn_index_cursor_each (VALUE self)
+rb_grn_index_cursor_each (int argc, VALUE *argv, VALUE self)
 {
     grn_obj *cursor;
     grn_ctx *context;
+    grn_bool reuse_posting_object;
+    VALUE rb_options;
+    VALUE rb_reuse_posting_object;
     VALUE rb_table;
     VALUE rb_lexicon;
+    VALUE rb_posting = Qnil;
+
+    RETURN_ENUMERATOR(self, argc, argv);
+
+    rb_scan_args(argc, argv, "01", &rb_options);
 
-    RETURN_ENUMERATOR(self, 0, NULL);
+    rb_grn_scan_options(rb_options,
+                        "reuse_posting_object", &rb_reuse_posting_object,
+                        NULL);
 
     rb_grn_index_cursor_deconstruct(SELF(self), &cursor, &context,
                                     NULL, NULL, NULL, NULL);
@@ -114,9 +130,17 @@ rb_grn_index_cursor_each (VALUE self)
 
     rb_table = rb_iv_get(self, "@table");
     rb_lexicon = rb_iv_get(self, "@lexicon");
+    reuse_posting_object = RVAL2CBOOL(rb_reuse_posting_object);
+
+    if (reuse_posting_object) {
+        rb_posting = rb_grn_posting_new(NULL, GRN_ID_NIL, rb_table, rb_lexicon);
+    }
     while (GRN_TRUE) {
-        VALUE rb_posting;
-        rb_posting = next_value(context, cursor, rb_table, rb_lexicon);
+        if (!reuse_posting_object) {
+            rb_posting = Qnil;
+        }
+        rb_posting = next_value(rb_posting,
+                                context, cursor, rb_table, rb_lexicon);
         if (NIL_P(rb_posting)) {
             break;
         }
@@ -135,5 +159,5 @@ rb_grn_init_index_cursor (VALUE mGrn)
     rb_include_module(rb_cGrnIndexCursor, rb_mEnumerable);
 
     rb_define_method(rb_cGrnIndexCursor, "next", rb_grn_index_cursor_next, 0);
-    rb_define_method(rb_cGrnIndexCursor, "each", rb_grn_index_cursor_each, 0);
+    rb_define_method(rb_cGrnIndexCursor, "each", rb_grn_index_cursor_each, -1);
 }

data/ext/groonga/rb-grn-posting.c

@@ -32,13 +32,15 @@ rb_grn_posting_new (grn_posting *posting, grn_id term_id,
 #define SET_PARAMETER(key, value) \
     rb_hash_aset(parameters, RB_GRN_INTERN(key), INT2NUM((value)))
 
-    SET_PARAMETER("record_id", posting->rid);
-    SET_PARAMETER("section_id", posting->sid);
+    if (posting) {
+        SET_PARAMETER("record_id", posting->rid);
+        SET_PARAMETER("section_id", posting->sid);
+        SET_PARAMETER("position", posting->pos);
+        SET_PARAMETER("term_frequency", posting->tf);
+        SET_PARAMETER("weight", posting->weight);
+        SET_PARAMETER("n_rest_postings", posting->rest);
+    }
     SET_PARAMETER("term_id", term_id);
-    SET_PARAMETER("position", posting->pos);
-    SET_PARAMETER("term_frequency", posting->tf);
-    SET_PARAMETER("weight", posting->weight);
-    SET_PARAMETER("n_rest_postings", posting->rest);
 
 #undef SET_PARAMETER
 
@@ -49,6 +51,24 @@ rb_grn_posting_new (grn_posting *posting, grn_id term_id,
                       parameters);
 }
 
+void
+rb_grn_posting_update (VALUE self, grn_posting *posting, grn_id term_id)
+{
+#define SET_PARAMETER(key, value) \
+    rb_funcall(self, rb_intern(key "="), 1, INT2NUM(value));
+
+    SET_PARAMETER("record_id",       posting->rid);
+    SET_PARAMETER("section_id",      posting->sid);
+    SET_PARAMETER("position",        posting->pos);
+    SET_PARAMETER("term_frequency",  posting->tf);
+    SET_PARAMETER("weight",          posting->weight);
+    SET_PARAMETER("n_rest_postings", posting->rest);
+
+    SET_PARAMETER("term_id",         term_id);
+
+#undef SET_PARAMETER
+}
+
 void
 rb_grn_init_posting (VALUE mGrn)
 {

data/ext/groonga/rb-grn-utils.c

@@ -481,46 +481,73 @@ rb_grn_bulk_from_ruby_object_with_type (VALUE object, grn_ctx *context,
 
     switch (type_id) {
     case GRN_DB_INT8:
+        if (string_p) {
+            object = rb_Integer(object);
+        }
         value.int8_value = NUM2SHORT(object);
         string = (const char *)&(value.int8_value);
         size = sizeof(value.int8_value);
         break;
     case GRN_DB_UINT8:
+        if (string_p) {
+            object = rb_Integer(object);
+        }
         value.uint8_value = NUM2USHORT(object);
         string = (const char *)&(value.uint8_value);
         size = sizeof(value.uint8_value);
         break;
     case GRN_DB_INT16:
+        if (string_p) {
+            object = rb_Integer(object);
+        }
         value.int16_value = NUM2SHORT(object);
         string = (const char *)&(value.int16_value);
         size = sizeof(value.int16_value);
         break;
     case GRN_DB_UINT16:
+        if (string_p) {
+            object = rb_Integer(object);
+        }
         value.uint16_value = NUM2USHORT(object);
         string = (const char *)&(value.uint16_value);
         size = sizeof(value.uint16_value);
         break;
     case GRN_DB_INT32:
+        if (string_p) {
+            object = rb_Integer(object);
+        }
         value.int32_value = NUM2INT(object);
         string = (const char *)&(value.int32_value);
         size = sizeof(value.int32_value);
         break;
     case GRN_DB_UINT32:
+        if (string_p) {
+            object = rb_Integer(object);
+        }
         value.uint32_value = NUM2UINT(object);
         string = (const char *)&(value.uint32_value);
         size = sizeof(value.uint32_value);
         break;
     case GRN_DB_INT64:
+        if (string_p) {
+            object = rb_Integer(object);
+        }
         value.int64_value = NUM2LL(object);
         string = (const char *)&(value.int64_value);
         size = sizeof(value.int64_value);
         break;
     case GRN_DB_UINT64:
+        if (string_p) {
+            object = rb_Integer(object);
+        }
         value.uint64_value = NUM2ULL(object);
         string = (const char *)&(value.uint64_value);
         size = sizeof(value.uint64_value);
         break;
     case GRN_DB_FLOAT:
+        if (string_p) {
+            object = rb_Float(object);
+        }
         value.double_value = NUM2DBL(object);
         string = (const char *)&(value.double_value);
         size = sizeof(value.double_value);

data/ext/groonga/rb-grn.h

@@ -92,7 +92,7 @@ RB_GRN_BEGIN_DECLS
 
 #define RB_GRN_MAJOR_VERSION 4
 #define RB_GRN_MINOR_VERSION 0
-#define RB_GRN_MICRO_VERSION 6
+#define RB_GRN_MICRO_VERSION 7
 
 #define RB_GRN_QUERY_DEFAULT_MAX_EXPRESSIONS 32
 
@@ -568,6 +568,9 @@ VALUE          rb_grn_posting_new                   (grn_posting *posting,
                                                      grn_id term_id,
                                                      VALUE rb_table,
                                                      VALUE rb_lexicon);
+void           rb_grn_posting_update                (VALUE rb_posting,
+                                                     grn_posting *posting,
+                                                     grn_id term_id);
 
 VALUE          rb_grn_tokyo_geo_point_new           (int   latitude,
                                                      int   longitude);

data/rroonga-build.rb

@@ -19,15 +19,15 @@ module RroongaBuild
   module RequiredGroongaVersion
     MAJOR = 4
     MINOR = 0
-    MICRO = 7
+    MICRO = 8
     VERSION = [MAJOR, MINOR, MICRO]
-    RELEASED_DATE = Time.utc(2014, 10, 29)
+    RELEASED_DATE = Time.utc(2014, 11, 29)
   end
 
   module LatestGroongaVersion
     MAJOR = 4
     MINOR = 0
-    MICRO = 7
+    MICRO = 8
     VERSION = [MAJOR, MINOR, MICRO]
   end