dbd-sqlite 0.1

data/LICENSE

@@ -0,0 +1,25 @@
+(C) 2008 Erik Hollensbe <erik@hollensbe.org>. All rights reserved.
+
+Please see "README" for earlier copyrights.
+
+Redistribution and use in source and binary forms, with or without 
+modification, are permitted provided that the following conditions 
+are met:
+1. Redistributions of source code must retain the above copyright 
+   notice, this list of conditions and the following disclaimer.
+2. Redistributions in binary form must reproduce the above copyright 
+   notice, this list of conditions and the following disclaimer in the 
+   documentation and/or other materials provided with the distribution.
+3. The name of the author may not be used to endorse or promote products
+   derived from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES,
+INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY
+AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL
+THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
+OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
+WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
+OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
+ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

data/README

@@ -0,0 +1,271 @@
+= Description
+    The DBI package is a vendor independent interface for accessing databases.
+    It is similar, but not identical to, Perl's DBI module.
+
+= Synopsis
+
+    require 'dbi'
+    
+    # Connect to a database, old style
+    dbh = DBI.connect('DBI:Mysql:test', 'testuser', 'testpwd')
+    
+    # Insert some rows, use placeholders
+    1.upto(13) do |i|
+        sql = "insert into simple01 (SongName, SongLength_s) VALUES (?, ?)"
+        dbh.do(sql, "Song #{i}", "#{i*10}")
+    end 
+    
+    # Select all rows from simple01
+    sth = dbh.prepare('select * from simple01')
+    sth.execute
+    
+    # Print out each row
+    while row=sth.fetch do
+        p row
+    end
+    
+    # Close the statement handle when done
+    sth.finish
+    
+    # Don't prepare, just do it!
+    dbh.do('delete from simple01 where internal_id > 10')
+      
+    # And finally, disconnect
+    dbh.disconnect
+        
+    # Same example, but a little more Ruby-ish
+    DBI.connect('DBI:Mysql:test', 'testuser', 'testpwd') do | dbh |
+      
+        sql = "insert into simple01 (SongName, SongLength_s) VALUES (?, ?)"
+        
+        dbh.prepare(sql) do | sth | 
+            1.upto(13) { |i| sth.execute("Song #{i}", "#{i*10}") }
+        end 
+        
+        dbh.select_all('select * from simple01') do | row |
+            p row
+        end
+        
+        dbh.do('delete from simple01 where internal_id > 10')
+      
+    end
+
+= Prerequisites
+    Ruby 1.8.6 or later is the test target, however you may have success with
+    earlier 1.8.x versions of Ruby.
+
+= RubyForge Project
+    General information: http://ruby-dbi.rubyforge.org
+    Project information: http://rubyforge.org/projects/ruby-dbi/
+    Downloads: http://rubyforge.org/frs/?group_id=234
+
+= Installation
+    There are many database drivers (DBDs) available.  You only need to install
+    the DBDs for the database software that you will be using.
+
+== Gem setup:
+
+    gem install dbi
+    # One or more of:
+    gem install dbd-mysql
+    gem install dbd-pg
+    gem install dbd-sqlite3
+    gem install dbd-sqlite
+
+== Without rubygems:
+
+    ruby setup.rb config
+    ruby setup.rb setup
+    ruby setup.rb install
+
+== The bleeding edge:
+
+    git clone git://hollensbe.org/git/ruby-dbi.git
+    git checkout -b development origin/development
+    
+    Also available at
+    
+    git clone git://github.com/erikh/ruby-dbi.git
+      
+= Available Database Drivers (DBDs)
+
+== DBD::MySQL
+    MySQL
+    Depends on the mysql-ruby package from http://www.tmtm.org/mysql or
+    available from the RAA.
+
+== DBD::ODBC
+    ODBC
+    Depends on the ruby-odbc package (0.5 or later, 0.9.3 or later recommended) at
+    http://www.ch-werner.de/rubyodbc or available from the RAA.  Works together
+    with unix-odbc.
+
+== DBD::OCI8
+    OCI8 (Oracle)
+    Depends on the the ruby-oci8 package, available on the RAA and RubyForge.
+
+== DBD::Pg
+    PostgreSQL
+    Depends on the pg package, available on RubyForge.
+
+== DBD::SQLite
+    SQLite (versions 2.x and earlier)
+    Depends on the sqlite-ruby package, available on rubyforge.
+
+== DBD::SQLite3
+    SQLite 3.x
+    Depends on the sqlite3-ruby package, available on rubyforge.
+
+= Additional Documentation
+    See the directories doc/* for DBI and DBD specific information.
+    The DBI specification is at doc/DBI_SPEC.rdoc.
+    The DBD specification is at doc/DBD_SPEC.rdoc.
+
+= Articles
+== Tutorial: Using the Ruby DBI Module
+    http://www.kitebird.com/articles/ruby-dbi.html
+   
+= Applications
+== dbi
+    The SQL command line interpreter dbi is available in directory
+    bin/. It gets installed by default.
+
+= License
+
+   Copyright (c) 2008 Erik Hollensbe
+
+   Copyright (c) 2005-2006 Kirk Haines, Francis Hwang, Patrick May and Daniel
+   Berger.
+
+   Copyright (c) 2001, 2002, 2003, 2004 Michael Neumann <mneumann@ntecs.de>
+   and others (see the beginning of each file for copyright holder information).
+
+   All rights reserved.
+
+   Redistribution and use in source and binary forms, with or without
+   modification, are permitted provided that the following conditions are met:
+
+   1. Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+   2. Redistributions in binary form must reproduce the above copyright notice,
+      this list of conditions and the following disclaimer in the documentation
+      and/or other materials provided with the distribution.
+   3. The name of the author may not be used to endorse or promote products
+      derived from this software without specific prior written permission.
+
+   THIS SOFTWARE IS PROVIDED 'AS IS' AND ANY EXPRESS OR IMPLIED WARRANTIES,
+   INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY
+   AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL
+   THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+   EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+   PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
+   OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
+   WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
+   OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
+   ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+   This is the BSD license which is less restrictive than GNU's GPL
+   (General Public License).
+
+= Contributors
+   
+   Pistos
+      Too much to specify. Infinite patience and help.
+
+   Christopher Maujean
+      Lots of initial help when reviving the project.
+
+   Jun Mukai <mukai@jmuk.org>
+      Contributed initial SQLite3 DBD.
+
+   John J. Fox IV
+      Lots of help testing on multiple platforms.
+
+   Kirk Haines
+      One of the authors of the rewrite effort (January 2006).
+
+   Francis Hwang
+      One of the authors of the rewrite effort (January 2006).
+
+   Patrick May
+      One of the authors of the rewrite effort (January 2006).
+
+   Daniel Berger
+      One of the authors of the rewrite effort (January 2006).
+
+   Michael Neumann
+      Original author of Ruby/DBI; wrote the DBI and most of the DBDs.
+
+   Rainer Perl
+      Author of Ruby/DBI 0.0.4 from which many good ideas were taken.
+
+   Jim Weirich
+      Original author of DBD::Pg.  Wrote additional code (e.g. sql.rb,
+      testcases).  Gave many helpful hints and comments. 
+
+   Eli Green
+      Implemented DatabaseHandle#columns for Mysql and Pg. 
+
+   Masatoshi SEKI
+      For his version of module BasicQuote in sql.rb. 
+
+   John Gorman
+      For his case insensitive load_driver patch and parameter parser. 
+
+   David Muse
+      For testing the DBD::SQLRelay and for his initial DBD. 
+
+   Jim Menard
+      Extended DBD::Oracle for method columns. 
+
+   Joseph McDonald
+      Fixed bug in DBD::Pg (default values in method columns). 
+
+   Norbert Gawor
+      Fixed bug in DBD::ODBC (method columns) and proxyserver. 
+
+   James F. Hranicky
+      Patch for DBD::Pg (cache PGResult#result in Tuples) which increased
+      performance by a factor around 100. 
+
+   Stephen Davies
+      Added method Statement#fetch_scroll for DBD::Pg. 
+
+   Dave Thomas
+      Several enhancements. 
+
+   Brad Hilton
+      Column coercing patch for DBD::Mysql. 
+
+   Sean Chittenden
+      Originally a co-owner of the project.  Submitted several patches
+      and helped with lots of comments.
+
+   MoonWolf
+      Provided the quote/escape_byte patch for DBD::Pg, DBD::SQLite patch and
+      Database#columns implementation. Further patches. 
+
+   Paul DuBois
+      Fixed typos and formatting. Maintains DBD::Mysql. 
+
+   Tim Bates
+      Bug fixes for Mysql and DBI.
+
+   Brian Candler
+      Zero-padding date/time/timestamps fix. 
+
+   Florian G. Pflug
+      Discussion and helpful comments/benchmarks about DBD::Pg async_exec vs.
+      exec. 
+   
+   Oliver M. Bolzer
+      Patches to support Postgres arrays for DBD::Pg. 
+
+   Stephen R. Veit
+      ruby-db2 and DBD::DB2 enhancements. 
+
+   Dennis Vshivkov
+      DBD::Pg patches 
+
+   Cail Borrell from frontbase.com
+      For the Frontbase DBD and C interface. 

data/lib/dbd/SQLite.rb

@@ -0,0 +1,97 @@
+#--
+###############################################################################
+#
+# DBD::SQLite - a DBD for SQLite for versions < 3
+#
+# Uses Jamis Buck's 'sqlite-ruby' driver to interface with SQLite directly
+#
+# (c) 2008 Erik Hollensbe & Christopher Maujean.
+#
+# TODO
+#
+# fetch_scroll implementation?
+# columns and column_info differ too much and have too much copied code, refactor
+# there are probably some edge cases with transactions
+#
+################################################################################
+#++
+
+begin
+    require 'rubygems'
+    gem 'sqlite-ruby'
+    gem 'dbi'
+rescue Exception => e
+end
+
+require 'dbi'
+require 'sqlite'
+
+module DBI
+    module DBD
+        #
+        # DBD::SQLite - Database Driver for SQLite versions 2.x and lower.
+        #
+        # Requires DBI and the 'sqlite-ruby' gem to work.
+        #
+        # Only things that extend DBI's results are documented.
+        #
+        class SQLite
+            VERSION = "0.1"
+            DESCRIPTION = "SQLite 2.x DBI DBD"
+
+            #
+            # returns 'SQLite'
+            #
+            # See DBI::TypeUtil#convert for more information.
+            #
+            def self.driver_name
+                "SQLite"
+            end
+
+            #
+            # Validates that the SQL has no literal NUL characters. (ASCII 0)
+            #
+            # SQLite apparently really hates it when you do that.
+            #
+            # It will raise DBI::DatabaseError should it find any.
+            #
+            def self.check_sql(sql)
+                # XXX I'm starting to think this is less of a problem with SQLite
+                # and more with the old C DBD
+                raise DBI::DatabaseError, "Bad SQL: SQL cannot contain nulls" if sql =~ /\0/
+            end
+
+            #
+            # Split a type definition into parts via String#match and return the whole result.
+            #
+            def self.parse_type(type_name)
+                type_name.match(/^([^\(]+)(\((\d+)(,(\d+))?\))?$/)
+            end
+
+            #
+            # See DBI::BaseDriver.
+            #
+            class Driver < DBI::BaseDriver
+                def initialize
+                    super "0.4.0"
+                end
+
+                def connect(dbname, user, auth, attr_hash)
+                    return Database.new(dbname, user, auth, attr_hash)
+                end
+            end
+        end
+    end
+end
+
+require 'dbd/sqlite/database'
+require 'dbd/sqlite/statement'
+
+DBI::TypeUtil.register_conversion(DBI::DBD::SQLite.driver_name) do |obj|
+    case obj
+    when ::NilClass
+        ["NULL", false]
+    else 
+        [obj, true]
+    end
+end

data/lib/dbd/sqlite/database.rb

@@ -0,0 +1,142 @@
+#
+# See DBI::BaseDatabase.
+#
+class DBI::DBD::SQLite::Database < DBI::BaseDatabase
+    attr_reader :db
+    attr_reader :attr_hash
+    attr_accessor :open_handles
+
+    #
+    # Constructor. Valid attributes:
+    #
+    # * AutoCommit: Commit after every statement execution.
+    #
+    def initialize(dbname, user, auth, attr_hash)
+        # FIXME why isn't this crap being done in DBI?
+        unless dbname.kind_of? String
+            raise DBI::InterfaceError, "Database Name must be a string"
+        end
+
+        unless dbname.length > 0
+            raise DBI::InterfaceError, "Database Name needs to be length > 0"
+        end
+
+        unless attr_hash.kind_of? Hash
+            raise DBI::InterfaceError, "Attributes should be a hash"
+        end
+
+        # FIXME handle busy_timeout in SQLite driver
+        # FIXME handle SQLite pragmas in SQLite driver
+        @attr_hash = attr_hash
+        @open_handles = 0
+
+        self["AutoCommit"] = true if self["AutoCommit"].nil?
+
+        # open the database
+        begin
+            @db = ::SQLite::Database.new(dbname)
+        rescue Exception => e
+            raise DBI::OperationalError, "Couldn't open database #{dbname}: #{e.message}"
+        end
+    end
+
+    def disconnect
+        rollback rescue nil
+        @db.close if @db and !@db.closed?
+        @db = nil
+    end
+
+    def prepare(stmt)
+        return DBI::DBD::SQLite::Statement.new(stmt, self)
+    end
+
+    def ping
+        return !@db.closed?
+    end
+
+    def tables
+        sth = prepare("select name from sqlite_master where type in ('table', 'view')")
+        sth.execute
+        tables = sth.fetch_all.flatten
+        sth.finish
+        return tables
+        # FIXME does sqlite use views too? not sure, but they need to be included according to spec
+    end
+
+    def commit
+        @db.commit if @db.transaction_active?
+    end
+
+    #
+    # Rollback the transaction. SQLite has some issues with open statement
+    # handles when this happens. If there are still open handles, a
+    # DBI::Warning exception will be raised.
+    #
+    def rollback
+        if @open_handles > 0
+            raise DBI::Warning, "Leaving unfinished select statement handles while rolling back a transaction can corrupt your database or crash your program"
+        end
+
+        @db.rollback if @db.transaction_active?
+    end
+
+    def [](key)
+        return @attr_hash[key]
+    end
+
+    #
+    # See DBI::BaseDatabase#[]=.
+    #
+    # If AutoCommit is set to +true+ using this method, was previously +false+,
+    # and we are currently in a transaction, The act of setting this will cause
+    # an immediate commit.
+    #
+    def []=(key, value)
+
+        old_value = @attr_hash[key]
+
+        @attr_hash[key] = value
+
+        # special handling of settings
+        case key
+        when "AutoCommit"
+            # if the value being set is true and the previous value is false,
+            # commit the current transaction (if any)
+            # FIXME I still think this is a horrible way of handling this.
+            if value and !old_value
+                begin 
+                    @dbh.commit
+                rescue Exception => e
+                end
+            end
+        end
+
+        return @attr_hash[key]
+    end
+
+    def columns(tablename)
+        return nil unless tablename and tablename.kind_of? String
+
+        sth = prepare("PRAGMA table_info(?)")
+        sth.bind_param(1, tablename)
+        sth.execute
+        columns = [ ]
+        while row = sth.fetch
+            column = { }
+            column["name"] = row[1]
+
+            m = DBI::DBD::SQLite.parse_type(row[2])
+            column["type_name"] = m[1]
+            column["precision"] = m[3].to_i if m[3]
+            column["scale"]     = m[5].to_i if m[5]
+
+            column["nullable"]  = row[3].to_i == 0
+            column["default"]   = row[4]
+            columns.push column
+        end
+
+        sth.finish
+        return columns
+        # XXX it'd be nice if the spec was changed to do this k/v with the name as the key.
+    end
+end