From bc3b9e7587e40579bfceeb448c8260a554d87854 Mon Sep 17 00:00:00 2001
From: Ben Kurtovic <ben.kurtovic@gmail.com>
Date: Thu, 17 Apr 2014 17:33:14 -0400
Subject: [PATCH] Some more database design work.

---
 bitshift/database.py       | 13 ++++++++--
 bitshift/query/__init__.py |  2 ++
 schema.sql                 | 65 +++++++++++++++++++++++++++++++++++++++-------
 3 files changed, 68 insertions(+), 12 deletions(-)

diff --git a/bitshift/database.py b/bitshift/database.py
index 647fe55..07c71c2 100644
--- a/bitshift/database.py
+++ b/bitshift/database.py
@@ -16,16 +16,25 @@ class Database(object):
         """Establish a connection to the database."""
         self._conn = oursql.connect()
 
-    def search(self, query):
+    def search(self, query, page=1):
         """
-        Search the database.
+        Search the database for a query and return the *n*\ th page of results.
 
         :param query: The query to search for.
         :type query: :py:class:`~.query.tree.Tree`
+        :param page: The result page to display.
+        :type page: int
 
         :return: A list of search results.
         :rtype: list of :py:class:`.Codelet`\ s
         """
+        # query tree hash + page -> cached?
+        #   cache HIT:
+        #       if qcache_created is too old: invalidate cache, goto cache MISS
+        #       update qcache_last_used
+        #       parse qcache_results, fetch codelets
+        #   cache MISS:
+        #       build complex search query
         pass
 
     def insert(self, codelet):
diff --git a/bitshift/query/__init__.py b/bitshift/query/__init__.py
index 7d6e0d5..6971c04 100644
--- a/bitshift/query/__init__.py
+++ b/bitshift/query/__init__.py
@@ -6,4 +6,6 @@ __all__ = ["parse_query"]
 
 def parse_query(query):
     # gets a string, returns a Tree
+    # TODO: note: resultant Trees should be normalized so that "foo OR bar"
+    # and "bar OR foo" result in equivalent trees
     pass
diff --git a/schema.sql b/schema.sql
index 3cb915c..d49fc6e 100644
--- a/schema.sql
+++ b/schema.sql
@@ -1,23 +1,68 @@
-CREATE DATABASE bitshift DEFAULT CHARACTER SET utf8 COLLATE utf8_unicode_ci;
+CREATE DATABASE `bitshift` DEFAULT CHARACTER SET utf8 COLLATE utf8_unicode_ci;
 USE `bitshift`;
 
-CREATE TABLE codelets (
+CREATE TABLE `languages` (
+    `language_id` SMALLINT UNSIGNED NOT NULL AUTO_INCREMENT UNIQUE,
+    `language_name` VARCHAR(64) NOT NULL,
+    PRIMARY KEY (`language_id`)
+) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_unicode_ci;
+
+CREATE TABLE `origins` (
+    `origin_id` TINYINT UNSIGNED NOT NULL AUTO_INCREMENT UNIQUE,
+    `origin_name` VARCHAR(64) NOT NULL,
+    `origin_url` VARCHAR(512) NOT NULL,
+    `origin_url_base` VARCHAR(512) NOT NULL,
+    `origin_image` TINYBLOB DEFAULT NULL, -- TODO: verify size (<64kB)
+    PRIMARY KEY (`origin_id`)
+) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_unicode_ci;
+
+CREATE TABLE `codelets` (
     `codelet_id` BIGINT UNSIGNED NOT NULL AUTO_INCREMENT UNIQUE,
     `codelet_name` VARCHAR(512) NOT NULL,
     `codelet_code_id` BIGINT UNSIGNED NOT NULL,
-    `codelet_lang` SMALLINT UNSIGNED DEFAULT NULL,
-    `codelet_origin` TINYINT UNSIGNED DEFAULT NULL,
+    `codelet_lang` SMALLINT UNSIGNED DEFAULT NULL, -- TODO: needs index
+    `codelet_origin` TINYINT UNSIGNED NOT NULL,
     `codelet_url` VARCHAR(512) NOT NULL,
-    `codelet_date_created` DATETIME DEFAULT NULL,
-    `codelet_date_modified` DATETIME DEFAULT NULL,
+    `codelet_date_created` DATETIME DEFAULT NULL, -- TODO: needs index
+    `codelet_date_modified` DATETIME DEFAULT NULL, -- TODO: needs index
     PRIMARY KEY (`codelet_id`)
 ) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_unicode_ci;
 
-CREATE TABLE code (
+CREATE TABLE `code` (
     `code_id` BIGINT UNSIGNED NOT NULL AUTO_INCREMENT UNIQUE,
-    `code_hash` BIGINT NOT NULL,
-    `code_code` MEDIUMTEXT NOT NULL,
+    `code_hash` BIGINT NOT NULL, -- TODO: needs index
+    `code_code` MEDIUMTEXT NOT NULL, -- TODO: verify size (16mB?)
     PRIMARY KEY (`code_id`)
 ) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_unicode_ci;
 
--- separate tables: authors, symbols, caches, search indices
+CREATE TABLE `authors` (
+    `author_id` BIGINT UNSIGNED NOT NULL AUTO_INCREMENT UNIQUE,
+    `author_codelet` BIGINT UNSIGNED NOT NULL, -- TODO: foreign index?
+    `author_name` VARCHAR(128) NOT NULL, -- TODO: needs index
+    `author_url` VARCHAR(512) DEFAULT NULL,
+    PRIMARY KEY (`author_id`)
+) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_unicode_ci;
+
+CREATE TABLE `symbols` (
+    `symbol_id` BIGINT UNSIGNED NOT NULL AUTO_INCREMENT UNIQUE,
+    `symbol_codelet` BIGINT UNSIGNED NOT NULL, -- TODO: foreign index?
+    `symbol_type` TINYINT UNSIGNED NOT NULL, -- TODO: multi-column index?
+    `symbol_name` VARCHAR(512) NOT NULL, -- TODO: needs index
+    `symbol_row` INT UNSIGNED NOT NULL,
+    `symbol_col` INT UNSIGNED NOT NULL,
+    PRIMARY KEY (`symbol_id`)
+) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_unicode_ci;
+
+CREATE TABLE `query_cache` (
+    `qcache_id` INT NOT NULL UNIQUE,
+    `qcache_query` VARCHAR(512) NOT NULL,
+    `qcache_results` BLOB NOT NULL, -- TODO: verify; perhaps use some kind of array
+    `qcache_page` TINYINT UNSIGNED NOT NULL,
+    `qcache_count_mnt` TINYINT UNSIGNED NOT NULL,
+    `qcache_count_exp` TINYINT UNSIGNED NOT NULL,
+    `qcache_created` TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP, -- TODO: verify
+    `qcache_last_used` TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP, -- TODO: verify
+    PRIMARY KEY (`cache_id`)
+) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_unicode_ci;
+
+-- TODO: full-text search index table