anbt-sql-formatter 0.1.0 → 0.1.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 06d3d110d976d89ae531994e7a2e504028deb41c7e0576ad977445dd12a40e97
-  data.tar.gz: 07754a3a2b3169853345b4ead8f2290f0b4279bd913359b637c2e2d87994cb80
+  metadata.gz: 84e2831d39ea0a114b7dd67a1ad69be88df0a575f017c6e9ec16ed87cdc8597b
+  data.tar.gz: b7f86dd15c4c5b421035f39b4f7bff403305e1f2f0ebd84eed5644b5be57e7a2
 SHA512:
-  metadata.gz: 760fa7076565da7636d4dd0659a7ac7f6651e61e50860b7b61cb1fdeac45a0d7836a0c3a3b832bc32b4eb8bcd939eff2e37135b0508822f028f72c4a5753ecd6
-  data.tar.gz: 3deda4f2457c216382e8c64d9343af513e60d762af57ab39f047a3b282756833ea7542b45d309375fa34f17ef4a5421ac57aee08c1c7439c94d6dcefed7c03cc
+  metadata.gz: 9b928f9e74ae4139c3f1408c6727832e352b2f78ad5479f38eb5959817c405f936faab338620536ac7c2a94dc931e839e3e58e2b460922e64fbb8da805d74aff
+  data.tar.gz: b978f80c52d14d067c150e48992dc81eaafc948d1103ff12fade72bb3d1f6835af86532393e658b5856dd75ee3d3dc03cf9498a01b66379011a09ffb9adc87cc

data/.gitignore

@@ -5,3 +5,6 @@ pkg/*
 
 # Emacs
 *~
+
+# Bundler
+/vendor/bundle/

data/CHANGELOG.md

@@ -1,3 +1,39 @@
+# 0.1.2 (2025-01-xx)
+
+## Breaking changes
+
+- Changed to exclusive branching to reduce complexity (`Formatter#format_list_main_loop`).
+  - If you have not customized the rules, there will be no change in behavior.
+    - This likely applies to most users.
+  - If you have customized the rules such that keywords overlap between conditions of the branches,
+    this update may change the behavior.
+
+```ruby
+# Example of customization that may change behavior:
+
+custom_rule = AnbtSql::Rule.new
+custom_rule.kw_nl_x << "SOME_CUSTOM_KEYWORD"
+custom_rule.kw_nl_x_plus1_indent << "SOME_CUSTOM_KEYWORD"
+```
+
+## Improvements
+
+- Some cleanups, formatting improvements
+  - cleanup: Remove encoding magic comments
+
+
+# 0.1.1 (2025-01-12)
+
+No breaking changes.
+
+## Improvements
+
+- Avoid mutative string manipulation (coarse-tokenizer.rb: tokenize)
+  - Prevents frozen literal warnings in Ruby 3.4 (Issue #17)
+- Update comments (PR #14)
+- Some small fixes, cleanups, formatting improvements
+
+
 # 0.1.0 (2018-12-16)
 
 ## Breaking changes

data/anbt-sql-formatter.gemspec

@@ -1,4 +1,3 @@
-# -*- encoding: utf-8 -*-
 $:.push File.expand_path("../lib", __FILE__)
 require "anbt-sql-formatter/version"
 
@@ -21,4 +20,6 @@ Gem::Specification.new do |s|
   # specify any dependencies here; for example:
   # s.add_development_dependency "rspec"
   # s.add_runtime_dependency "rest-client"
+
+  s.add_development_dependency "test-unit", "~> 3.6.2"
 end

data/bin/anbt-sql-formatter

@@ -1,5 +1,4 @@
 #! /usr/bin/env ruby
-# -*- coding: utf-8 -*-
 
 begin
   require "anbt-sql-formatter/formatter"

data/lib/anbt-sql-formatter/coarse-tokenizer.rb

@@ -1,5 +1,3 @@
-# -*- coding: utf-8 -*-
-
 =begin
 エスケープ文字
 =end
@@ -38,8 +36,7 @@ These are exclusive:
 =end
 
   def tokenize(str)
-    @str = str
-    @str.gsub!("\r\n", "\n")
+    @str = str.gsub("\r\n", "\n")
     out_of_quote_single   = true
     out_of_quote_double   = true
     out_of_comment_single = true
@@ -51,36 +48,36 @@ These are exclusive:
 
     while @str.size > 0
 
-      if /\A(")/ =~ str && out_of_quote_double &&
+      if /\A(")/ =~ @str && out_of_quote_double &&
           out_of_quote_single && out_of_comment_single && out_of_comment_multi
         ## begin double quote
 
         length = $1.size
         shift_token(length, :plain, :quote_double, :start)
         out_of_quote_double = false
-        
-      elsif /\A(")/ =~ str && !(out_of_quote_double) &&
+
+      elsif /\A(")/ =~ @str && !(out_of_quote_double) &&
           out_of_quote_single && out_of_comment_single && out_of_comment_multi
         ## end double quote
-         
+
         length = $1.size
-        if /\A(".")/ =~ str ## schema.table
+        if /\A(".")/ =~ @str ## schema.table
           shift_to_buf(3)
-        elsif /\A("")/ =~ str ## escaped double quote
+        elsif /\A("")/ =~ @str ## escaped double quote
           shift_to_buf(2)
         else
           shift_token(length, :quote_double, :plain, :end)
           out_of_quote_double = true
         end
 
-      elsif /\A(')/ =~ str && out_of_quote_single &&
+      elsif /\A(')/ =~ @str && out_of_quote_single &&
           out_of_quote_double && out_of_comment_single && out_of_comment_multi
         ## begin single quote
-        
+
         length = $1.size
         shift_token(length, :plain, :quote_single, :start)
         out_of_quote_single = false
-      elsif /\A(')/ =~ str && !(out_of_quote_single) &&
+      elsif /\A(')/ =~ @str && !(out_of_quote_single) &&
           out_of_quote_double && out_of_comment_single && out_of_comment_multi
         ## end single quote
 
@@ -88,49 +85,49 @@ These are exclusive:
         if /\A('')/ =~ @str ## escaped single quote
           shift_to_buf(2)
         else
-        shift_token(length, :quote_single, :plain, :end)
+          shift_token(length, :quote_single, :plain, :end)
           out_of_quote_single = true
         end
-        
-      elsif /\A(#{@comment_single_start})/ =~ str && out_of_comment_single &&
+
+      elsif /\A(#{@comment_single_start})/ =~ @str && out_of_comment_single &&
          out_of_quote_single && out_of_quote_double && out_of_comment_multi
         ## begin single line comment
-        
+
         length = $1.size
         shift_token(length, :plain, :comment_single, :start)
         out_of_comment_single = false
 
-      elsif /\A(\n)/ =~ str && !(out_of_comment_single) &&
+      elsif /\A(\n)/ =~ @str && !(out_of_comment_single) &&
           out_of_quote_single && out_of_quote_double && out_of_comment_multi
         ## end single line comment
-       
+
         length = $1.size
         shift_token(length, :comment_single, :plain, :end)
         out_of_comment_single = true
 
-      elsif /\A(#{@comment_multi_start})/ =~ str &&
+      elsif /\A(#{@comment_multi_start})/ =~ @str &&
           out_of_quote_single && out_of_quote_double && out_of_comment_single && out_of_comment_multi
         ## begin multi line comment
-        
+
         length = $1.size
         shift_token(length, :plain, :comment_multi, :start)
         out_of_comment_multi = false
 
-      elsif /\A(#{@comment_multi_end})/ =~ str &&
+      elsif /\A(#{@comment_multi_end})/ =~ @str &&
           out_of_quote_single && out_of_quote_double && out_of_comment_single && !(out_of_comment_multi)
         ## end multi line comment
-        
+
         length = $1.size
         shift_token(length, :comment_multi, :plain, :end)
         out_of_comment_multi = true
 
-      elsif /\A\\/ =~ str
+      elsif /\A\\/ =~ @str
         ## escape char
         shift_to_buf(2)
 
       else
         shift_to_buf(1)
-        
+
       end
     end
     @result << CoarseToken.new(@mode, @buf+@str) if (@buf+@str).size > 0
@@ -140,11 +137,11 @@ These are exclusive:
 
 
   def shift_to_buf(n)
-    @buf << @str[0...n]
-    @str[0...n] = ""
+    @buf += @str[0...n]
+    @str = @str[n..]
   end
-  
-  
+
+
   def shift_token(length, type, mode, flag)
     case flag
     when :start
@@ -157,7 +154,7 @@ These are exclusive:
       raise "must not happen"
     end
 
-    @str[0..(length-1)] = ""
+    @str = @str[length..]
     @mode = mode
   end
 end

data/lib/anbt-sql-formatter/constants.rb

@@ -1,12 +1,10 @@
-# -*- coding: utf-8 -*-
-
 =begin rdoc
 AnbtSqlFormatter: SQL整形ツール. SQL文を決められたルールに従い整形します。
 
 フォーマットを実施するためには、入力されるSQLがSQL文として妥当であることが前提条件となります。
 
 このクラスが準拠するSQL整形のルールについては、下記URLを参照ください。
-http://homepage2.nifty.com/igat/igapyon/diary/2005/ig050613.html
+https://www.igapyon.jp/igapyon/diary/2005/ig050613.html
 
 このクラスには ANSI SQLの予約語一覧が蓄えられます。
 
@@ -76,6 +74,7 @@ class AnbtSql
        "ORDINALITY", "OUT", "PARAMETER", "PARAMETERS", "PATH", "POSTFIX",
        "PREFIX", "PREORDER", "RAW", "READS", "RECURSIVE", "REDO",
        # ANSI SQLではないのだが とても良く使われる構文
-       "TRUNCATE" ]
+       "TRUNCATE"
+      ]
   end
 end

data/lib/anbt-sql-formatter/exception.rb

@@ -1,5 +1,3 @@
-# -*- coding: utf-8 -*-
-
 =begin rdoc
 BlancoSqlFormatterException : SQL整形ツールの例外を表します。
 

data/lib/anbt-sql-formatter/formatter.rb

@@ -1,5 +1,3 @@
-# -*- coding: utf-8 -*-
-
 require "anbt-sql-formatter/rule"
 require "anbt-sql-formatter/parser"
 require "anbt-sql-formatter/exception"
@@ -222,56 +220,47 @@ class AnbtSql
               equals_ignore_case(token.string, "UPDATE")   )
             indent += 2
             index += insert_return_and_indent(tokens, index + 1, indent, "+2")
-          end
 
           # indentを１つ増やし、キーワードの後ろで改行
-          if @rule.kw_plus1_indent_x_nl.any?{ |kw| equals_ignore_case(token.string, kw) }
+          elsif @rule.kw_plus1_indent_x_nl.any?{ |kw| equals_ignore_case(token.string, kw) }
             indent += 1
             index += insert_return_and_indent(tokens, index + 1, indent)
-          end
 
           # キーワードの前でindentを１つ減らして改行、キーワードの後ろでindentを戻して改行。
-          if @rule.kw_minus1_indent_nl_x_plus1_indent.any?{ |kw| equals_ignore_case(token.string, kw) }
+          elsif @rule.kw_minus1_indent_nl_x_plus1_indent.any?{ |kw| equals_ignore_case(token.string, kw) }
             index += insert_return_and_indent(tokens, index    , indent - 1)
             index += insert_return_and_indent(tokens, index + 1, indent    )
-          end
 
           # キーワードの前でindentを１つ減らして改行、キーワードの後ろでindentを戻して改行。
-          if (equals_ignore_case(token.string, "VALUES"))
+          elsif (equals_ignore_case(token.string, "VALUES"))
             indent -= 1
             index += insert_return_and_indent(tokens, index, indent)
-          end
 
           # キーワードの前でindentを１つ減らして改行
-          if (equals_ignore_case(token.string, "END"))
+          elsif (equals_ignore_case(token.string, "END"))
             indent -= 1
             index += insert_return_and_indent(tokens, index, indent)
-          end
 
           # キーワードの前で改行
-          if @rule.kw_nl_x.any?{ |kw| equals_ignore_case(token.string, kw) }
+          elsif @rule.kw_nl_x.any?{ |kw| equals_ignore_case(token.string, kw) }
             index += insert_return_and_indent(tokens, index, indent)
-          end
 
           # キーワードの前で改行, インデント+1
-          if @rule.kw_nl_x_plus1_indent.any?{ |kw| equals_ignore_case(token.string, kw) }
+          elsif @rule.kw_nl_x_plus1_indent.any?{ |kw| equals_ignore_case(token.string, kw) }
             index += insert_return_and_indent(tokens, index, indent + 1)
-          end
 
           # キーワードの前で改行。indentを強制的に０にする。
-          if (equals_ignore_case(token.string, "UNION"    ) ||
+          elsif (equals_ignore_case(token.string, "UNION"    ) ||
               equals_ignore_case(token.string, "INTERSECT") ||
               equals_ignore_case(token.string, "EXCEPT"   )   )
             indent -= 2
             index += insert_return_and_indent(tokens, index    , indent)
             index += insert_return_and_indent(tokens, index + 1, indent)
-          end
 
-          if equals_ignore_case(token.string, "BETWEEN")
+          elsif equals_ignore_case(token.string, "BETWEEN")
             encounter_between = true
-          end
 
-          if equals_ignore_case(token.string, "AND")
+          elsif equals_ignore_case(token.string, "AND")
             # BETWEEN のあとのANDは改行しない。
             if not encounter_between
               index += insert_return_and_indent(tokens, index, indent)
@@ -395,7 +384,7 @@ class AnbtSql
         if $DEBUG
           $stderr.puts e.message, e.backtrace
           $stderr.puts "tokens: "
-          tokens.each_with_index{|t,i|
+          tokens.each_with_index{|t, i|
             $stderr.puts "index=%d: %s" % [i, t.inspect]
           }
           $stderr.puts "index/size: %d/%d / indent: %d / opt: %s" % [index, tokens.size, indent, opt]

data/lib/anbt-sql-formatter/parser.rb

@@ -1,5 +1,3 @@
-# -*- coding: utf-8 -*-
-
 require "pp"
 
 require "anbt-sql-formatter/token"
@@ -67,7 +65,7 @@ class AnbtSql
     # これ以降の文字の扱いは保留
     def symbol?(c)
       %w(" ? % & ' \( \) | * + , - . / : ; < = > !).include? c
-      #"
+      # "
     end
 
 
@@ -98,7 +96,7 @@ class AnbtSql
           is_next_char_space = false
           if @pos + 1 < @before.size &&
             space?(char_at(@before, @pos+1))
-              is_next_char_space = true
+            is_next_char_space = true
           end
 
           if not is_next_char_space
@@ -160,7 +158,7 @@ class AnbtSql
 
         # ２文字の記号かどうか調べる
         ch2 = char_at(@before, @pos)
-        #for (int i = 0; i < two_character_symbol.length; i++) {
+        # for (int i = 0; i < two_character_symbol.length; i++) {
         for i in 0...@two_character_symbol.length
           if (char_at(@two_character_symbol[i], 0) == @char &&
               char_at(@two_character_symbol[i], 1) == ch2)

data/lib/anbt-sql-formatter/rule.rb

@@ -1,5 +1,3 @@
-# -*- coding: utf-8 -*-
-
 require "anbt-sql-formatter/helper"
 
 =begin
@@ -8,7 +6,7 @@ AnbtSqlFormatter: SQL整形ツール. SQL文を決められたルールに従い
 フォーマットを実施するためには、入力されるSQLがSQL文として妥当であることが前提条件となります。
 
 このクラスが準拠するSQL整形のルールについては、下記URLを参照ください。
-http://homepage2.nifty.com/igat/igapyon/diary/2005/ig050613.html
+https://www.igapyon.jp/igapyon/diary/2005/ig050613.html
 
 このクラスは SQLの変換規則を表します。
 

data/lib/anbt-sql-formatter/token.rb

@@ -1,5 +1,3 @@
-# -*- coding: utf-8 -*-
-
 class AnbtSql
   class TokenConstants
 
@@ -48,9 +46,9 @@ class AnbtSql
 
     #
     # このバリューオブジェクトの文字列表現を取得する。
-    # 
+    #
     # オブジェクトのシャロー範囲でしか to_s されない点に注意。
-    # 
+    #
     # @return:: バリューオブジェクトの文字列表現。
     #
     def to_s

data/lib/anbt-sql-formatter/version.rb

@@ -1,7 +1,7 @@
 module Anbt
   module Sql
     module Formatter
-      VERSION = "0.1.0"
+      VERSION = "0.1.2"
     end
   end
 end

data/misc/anbt-sql-formatter-customize-example

@@ -1,5 +1,4 @@
 #! /usr/bin/ruby
-# -*- coding: utf-8 -*-
 
 require "pp"
 
@@ -8,7 +7,7 @@ require "anbt-sql-formatter/formatter"
 =begin
 rule のプロパティを修正するだけではなく
 さらに踏み込んでカスタマイズしたい場合は
-formatter.rb 内の AnbtSql::Formatter#format_list_main_loop 
+formatter.rb 内の AnbtSql::Formatter#format_list_main_loop
 などをオーバーライドすると良いでしょう。
 
 Ruby に詳しくない場合は「Ruby オープンクラス」
@@ -39,7 +38,7 @@ if $0 == __FILE__
 
   # キーワードを大文字・小文字に変換しない
   rule.keyword = AnbtSql::Rule::KEYWORD_NONE
-  
+
   # 複数単語のキーワードを登録
   rule.kw_multi_words << "INNER JOIN"
   rule.kw_nl_x << "INNER JOIN"
@@ -50,15 +49,15 @@ if $0 == __FILE__
   rule.kw_nl_x << "WHEN"
 
   # User defined additional functions:
-  # ユーザ定義関数の追加: 
+  # ユーザ定義関数の追加:
   %w(count sum substr date).each{|func_name|
     rule.function_names << func_name.upcase
   }
 
-  #rule.indent_string = "    "
-  #rule.indent_string = "('-')"
+  # rule.indent_string = "    "
+  # rule.indent_string = "('-')"
   rule.indent_string = "|   "
-  
+
   formatter = AnbtSql::Formatter.new(rule)
   result = formatter.format(src)
   print result

data/test/helper.rb

@@ -12,8 +12,8 @@ class Helper
 end
 
 
-def assert_equals(a,b,c)
-  assert_equal(b,c,a)
+def assert_equals(a, b, c)
+  assert_equal(b, c, a)
 end
 
 def strip_indent(text)

data/test/test_coarse-tokenizer.rb

@@ -1,5 +1,3 @@
-# -*- coding: utf-8 -*-
-
 require File.join(File.expand_path(File.dirname(__FILE__)), "helper")
 
 require "anbt-sql-formatter/coarse-tokenizer"
@@ -21,7 +19,7 @@ class TestCoarseTokenizer < Test::Unit::TestCase
     }.join("\n")
   end
 
-  
+
   def test_shift_to_buf
     @tok.buf = ""
     @tok.str = "abcdefg"
@@ -52,14 +50,14 @@ class TestCoarseTokenizer < Test::Unit::TestCase
     @tok.result = []
     @tok.buf = "'ABC"
     @tok.str = "'def"
-    
+
     @tok.shift_token(1, :comment, :plain, :end)
     assert_equals( msg,  :comment, @tok.result.last._type)
     assert_equals( msg,  "'ABC'", @tok.result.last.string)
     assert_equals( msg,  "", @tok.buf)
     assert_equals( msg,  "def", @tok.str)
   end
-  
+
 
   def test_tokenize_1
     assert_equals(
@@ -327,7 +325,7 @@ class TestCoarseTokenizer < Test::Unit::TestCase
       )))
     )
   end
-  
+
 
   def test_comment_in_string_1
     assert_equals(
@@ -392,7 +390,7 @@ class TestCoarseTokenizer < Test::Unit::TestCase
       )))
     )
   end
-  
+
 
   def test_string_escape_1
     assert_equals(