ﺍﺨﺘﻴﺎﺭ ﺍلCharacter Set:

ﻴﺠﺏ ﻋﻠﻲ ﻤﺩﻴﺭ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ ﻟﺤﻅﺔ ﺇﻨﺸﺎﺀ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ ﺍﺘﺨﺎﺫ ﻗﺭﺍﺭﻴﻥ ﻤﻬﻤﻴﻥ:

ﺍﻟﻘﺭﺍﺭ ﺍﻷﻭل ﻫﻭ ﺍﺨﺘﻴﺎﺭ ﺤﺠﻡ ﺍلBlock Database.

(DB_BLOCK_SIZE) ﺍﻟﺫﻱ ﻻ ﻴﻤﻜﻥ ﺘﻐﻴﻴﺭﻩ ﻭﺫﻟﻙ ﻷﻨﻪ ﻴﺴﺘﺨﺩﻡ ﻓﻲ ﺇﻨﺸﺎﺀ ﺍلSystem Tablespace ﻓﻼ ﻴﻤﻜﻥ ﺘﻐﻴﻴﺭ ﺤﺠﻡ ﺍلDatabase Block ﻤﺎ ﻟﻡ ﻴﺘﻡ ﺇﻋﺎﺩﺓ ﺇﻨﺸﺎﺀ ﺍلData Dictionary

ﺃﻭ ﺒﻤﻌﻨﻲ ﺁﺨﺭ ﺇﻋﺎﺩﺓ ﺇﻨﺸﺎﺀ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ.

ﺃﻤﺎ ﺍﻟﻘﺭﺍﺭ ﺍﻟﺜﺎﻨﻲ ﻓﻬﻭ ﺍﺨﺘﻴﺎﺭ ﺍلCharacter Set، ﻗﺩ ﻴﻜﻭﻥ ﻤﻥ ﺍﻟﻤﺤﺘﻤل ﻟﻤﺩﻴﺭ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ ﺘﻐﻴﻴﺭ ﺍلCharacter Set ﻭﻟﻜﻥ ﻟﻴﺱ ﺩﺍﺌﻤﺎﹰ ﻭﻻ ﻋﻤﻠﻴﺎﹰ.

ﺍلSet Character Database ﻟﺘﺨﺯﻴﻥ ﺍﻟﺒﻴﺎﻨﺎﺕ ﻓﻲ ﺍﻷﻋﻤﺩﺓ ﻤﻥ ﺍﻟﻨﻭﻉ VARCHAR2 and CHAR and CLOB and LONG.

ﺇﺫﺍ ﻗﺎﻡ ﻤﺩﻴﺭ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ ﺒﺘﻐﻴﻴﺭ ﺍلSet Character ﺭﺒﻤﺎ ﻴﺅﺩﻱ ﺫﻟﻙ ﻟﺘﺤﻁﻴﻡ ﺍﻟﺒﻴﺎﻨﺎﺕ ﻓﻲ ﺍﻷﻋﻤﺩﺓ ﻤﻥ ﻫﺫﺍ ﺍﻟﻨﻭﻉ ﻟﺫﺍ ﻴﺠﺏ ﺍﺨﺘﻴﺎﺭ ﺍلCharacter Set

ﺍﻟﺘﻲ ﺴﺘﻨﺠﺯ ﺍﺤﺘﻴﺎﺠﺎﺘﻙ ﺍﻟﺤﺎﻟﻴﺔ ﻭﺍﻟﻤﺴﺘﻘﺒﻠﻴﺔ ﻋﻨﺩ ﺇﻨﺸﺎﺀ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ.

ﻓﻠﻭ ﻜﺎﻨﺕ ﻟﺩﻴﻙ ﺒﻴﺎﻨﺎﺕ ﺒﺎﻟﻠﻐﺔ ﺍﻟﻔﺭﻨﺴﻴﺔ ﺃﻭ ﺍﻷﺴﺒﺎﻨﻴﺔ ﻓﺄﻨﺕ ﺘﺤﺘﺎﺝ ﻟلWestern European Character Set ، ﺃﻤﺎ ﺇﺫﺍ ﻜﺎﻨﺕ ﺍﻟﺒﻴﺎﻨﺎﺕ ﺒﺎﻟﻠﻐﺔ ﺍﻟﺭﻭﺴﻴﺔ ﺃﻭ ﺍﻟﺘﺸﻴﻜﻴﺔ ﻓﺄﻨﺕ ﺘﺤﺘﺎﺝ

ﻟلEastern European Character Set ، ﻟﻜﻥ ﻤﺎﺫﺍ ﻟﻭ ﻜﺎﻨﺕ ﻟﺩﻴﻙ ﺒﻴﺎﻨﺎﺕ ﺒﻠﻐﺎﺕ ﺃﻭﺭﺒﺎ ﺍﻟﺸﺭﻗﻴﺔ ﻭﺍﻟﻐﺭﺒﻴﺔ ﻤﻌﺎﹰ؟ ﺒﺎﻹﻀﺎﻓﺔ ﺇﻟﻲ ﺃﻨﻙ ﻗﺩ ﺘﺤﺘﺎﺝ ﻟﺘﺨﺯﻴﻥ ﺒﻴﺎﻨﺎﺕ ﺒﻠﻐﺎﺕ ﻜﻭﺭﻴﺔ ﺃﻭ ﺘﺎﻴﻭﺍﻨﻴﺔ،

ﺸﺭﻜﺔ ﺃﻭﺭﻜل ﺘﻭﻓﺭ ﺤﻠﻴﻥ ﻟﻬﺫﻩ ﺍﻟﻤﺸﻜﻠﺔ:

1- :National Character Set  ﺍﻟﺫﻱ ﻗﺩﻡ ﻓﻲ ﺍﻹﺼﺩﺍﺭ 8 Oracle Database ، ﻭﻫﻲ Character Set ﺜﺎﻨﻴﺔ ﻴﺘﻡ ﺍﺨﺘﻴﺎﺭﻫﺎ ﻟﺤﻅﺔ ﺇﻨﺸﺎﺀ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ ﻭﺘﺴﺘﺨﺩﻡ ﻟﺘﺨﺯﻴﻥ ﺍﻟﺒﻴﺎﻨﺎﺕ

ﻓﻲ ﺍﻷﻋﻤﺩﺓ ﻤﻥ ﺍﻟﻨﻭﻉ NCLOB and NCHAR and NVARCHAR2، ﻓﻠﻭ ﺘﻭﻗﻊ ﻤﺩﻴﺭ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ ﺒﺄﻥ ﺃﻏﻠﺏ ﺒﻴﺎﻨﺎﺘﻪ ﺴﺘﻜﻭﻥ ﺒﺎﻟﻠﻐﺔ ﺍﻹﻨﺠﻠﻴﺯﻴﺔ ﻭﺍﻟﺒﻌﺽ ﺒﺎﻟﻠﻐﺔ ﺍﻟﻴﺎﺒﺎﻨﻴﺔ ﻓﺴﻭﻑ ﻴﺨﺘﺎﺭ

Western European Character Set  ﻟلDatabase Character Set  ﻭﺴﻭﻑ ﻴﺨﺘﺎﺭ Kanji Character Set  ﻙNational Character Set.

  2- Unicode: ﻟﻜﻥ ﻓﻲ ﺍﻹﺼﺩﺍﺭ Oracle 9i ﺘﻐﻴﺭ ﺍﻟﺴﻴﻨﺎﺭﻴﻭ ﻗﻠﻴﻼﹰ ﺤﻴﺙ ﺃﺼﺒﺢ ﻤﻥ ﺍﻟﻀﺭﻭﺭﻱ ﺃﻥ ﻴﻜﻭﻥ ﺍلNational Character Set  ﻤﻥ ﺍﻟﻨﻭﻉ Unicode ﻭﻫﻭ Character Set ﻋﺎﻟﻤﻲ

ﻴﺴﺘﻁﻴﻊ ﺘﻤﺜﻴل ﺠﻤﻴﻊ ﺍﻟﺤﺭﻭﻑ ﺍﻟﺘﻲ ﻴﻤﻜﻥ ﺍﺴﺘﺨﺩﺍﻤﻬﺎ ﻓﻲ ﺃﻱ ﻜﻤﺒﻴﻭﺘﺭ. ﻭﻫﻨﺎﻙ ﻨﻭﻋﺎﻥ ﻤﻥ ﺍلUnicodes ﻴﺴﺘﻁﻴﻌﺎﻥ ﺃﻥ ﻴﺩﻋﻤﺎﻥ الNationa Character Set:

Fixed-Width, Two-Byte Character Set :AL16UTF16 .

Variable-Width Character Set :UTF8 .

ﻭﺍﻻﺨﺘﻴﺎﺭ ﺍﻷﻓﻀل ﺒﻴﻥ ﻫﺫﻴﻥ ﺍﻟﺨﻴﺎﺭﻴﻥ ﻴﺩﻭﺭ ﺤﻭل ﺍﻷﺩﺍﺀ ﻭﻜﻔﺎﺀﺓ ﺍﻟﺘﺨﺯﻴﻥ.

ﻜﻼﹰ ﻤﻥ ﺍلDatabase Character Set  ﻭﺍلNational Character Set  ﻴﺘﻡ ﺍﺨﺘﻴﺎﺭﻫﻤﺎ ﻟﺤﻅﺔ ﺇﻨﺸﺎﺀ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ.

ﺘﻐﻴﻴﺭ ﺍلCharacter Set:

ﻗﺩ ﻴﺤﺘﺎﺝ ﻤﺩﻴﺭ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ ﺃﺤﻴﺎﻨﹰﺎ ﻟﺘﻐﻴﻴﺭ ﺍلCharacter Set ﻟﻌﺩﺓ ﺃﺴﺒﺎﺏ، ﻋﻠﻲ ﺴﻴﺒل ﺍﻟﻤﺜﺎل ﻟﻨﻔﺘﺭﺽ ﺃﻥ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ ﺘﻡ ﺇﻨﺸﺎﺅﻫﺎ ﺒﺎﻟﻘﻴﻤﺔ ﺍﻻﻓﺘﺭﺍﻀﻴﺔ USVASCII،

ﻻﺤﻘﺎﹰ ﺘﺒﻴﻥ ﻟﻤﺩﻴﺭ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ ﺃﻨﻪ ﺒﺤﺎﺠﺔ ﻟﺘﺨﺯﻴﻥ ﺤﺭﻭﻑ ﻟﻴﺴﺕ ﻤﻀﻤﻨﻪ ﻓﻲ ﻫﺫﻩ ﺍلCharacter Set، ﻤﺜﻼﹰ ﺍﺴﻡ ﻓﺭﻨﺴﻲ،

ﻗﺒل ﺍﻹﺼﺩﺍﺭ 9i ﻟﻡ ﻴﻜﻥ ﻤﻥ ﺍﻟﻤﻤﻜﻥ ﺘﻐﻴﻴﺭ ﺍلCharacter Set ﻭﻟﻜﻥ ﻤﻥ ﺍﻹﺼﺩﺍﺭ 9i ﻓﺼﺎﻋﺩﺍﹰ ﺃﺼﺒﺤﺕ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ ﺘﺩﻋﻡ ﺘﻐﻴﻴﺭ ﺍلCharacter Set ﻭﻟﻜﻥ ﻟﻴﺱ ﻫﻨﺎﻙ ﻀﻤﺎﻥ

ﻟﻨﺠﺎﺡ ﻫﺫﻩ ﺍﻟﻌﻤﻠﻴﺔ، ﻭﻤﻥ ﻤﻬﺎﻡ ﻤﺩﻴﺭ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ ﺍﻟﺘﺄﻜﺩ ﻤﻥ ﺃﻥ ﻋﻤﻠﻴﺔ ﺘﺤﻭﻴل ﺍلSet Character ﻟﻥ ﺘﺘﻠﻑ ﺍﻟﺒﻴﺎﻨﺎﺕ. ﺍﻟﻤﺸﻜﻠﺔ ﺒﺒﺴﺎﻁﺔ ﺃﻥ ﻋﻤل ﺘﺤﻭﻴل ﺍلCharacter Set

ﻗﺩ ﻻ ﺘﺴﺘﻁﻴﻊ ﺼﻴﺎﻏﺔ ﺍﻟﺒﻴﺎﻨﺎﺕ ﺍﻟﺤﺎﻟﻴﺔ ﻓﻲ ﺍلDatafile، ﻋﻠﻲ ﺴﺒﻴل ﺍﻟﻤﺜﺎل ﻟﻭ ﻗﺎﻡ ﻤﺩﻴﺭ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ ﺒﺘﻐﻴﻴﺭ ﺍلCharacter Set ﻤﻥ Western European ﺇﻟﻲ Eastern European

ﻓﺈﻥ ﺍﻟﻌﺩﻴﺩ ﻤﻥ ﺍﻟﺒﻴﺎﻨﺎﺕ ﺍﻟﻤﺸﺎﻋﺔ ﻓﻲ ﺃﻭﺭﺒﺎ ﺍﻟﻐﺭﺒﻴﺔ ﺴﺘﻅﻬﺭ ﺒﻨﺘﺎﺌﺞ ﻜﺎﺭﺜﻴﺔ.

ﺘﻭﻓﺭ ﺃﻭﺭﻜل ﺃﺩﺍﺘﺎﻥ ﺘﺴﺎﻋﺩﺍﻥ ﻓﻲ ﺘﻘﺭﻴﺭ ﺇﻤﻜﺎﻨﻴﺔ ﺘﻐﻴﻴﺭ ﺍلCharacter Set:

1- Database Character Set Scanner ﻭﻫﻲ ﺃﺩﺍﺓ ﺘﻘﻭﻡ ﺒﺎﻟﺘﻨﻔﻴﺫ ﺒﺸﻜل ﻤﺴﺘﻘل ﺤﻴﺙ ﺘﺘﺼل ﺒﻘﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ ﻭﺘﺼل ﻟلDatafiles ﻭﺘﻘﻭﻡ ﺒﺈﺼﺩﺍﺭ ﺘﻘﺭﻴﺭ ﺒﺎﻟﻤﺸﺎﻜل ﺍﻟﻤﺤﺘﻤﻠﺔ

(csscan.exe on Windows & Unix on csscan) .

ﻫﺫﺍ ﺍﻷﻤﺭ ﻴﻘﻭﻡ ﺒﺎﻻﺘﺼﺎل ﺒﻘﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ ﻋﺒﺭ ﺍﻟﻤﺴﺘﺨﺩﻡ SYSTEM ﻭﻴﻘﻭﻡ ﺒﻤﺴﺢ ﻤﻠﻔﺎﺕ ﺍلData Files ﻭﺍﻟﺘﺄﻜﺩ ﻫل ﻋﻤﻠﻴﺔ ﺍﻟﺘﺤﻭﻴل ﻟلUTF8 ﻗﺩ ﺘﺴﺒﺏ ﺃﻱ ﻤﺸﺎﻜل.

ﻭﺍﻟﻤﺸﻜﻠﺔ ﻓﻲ ﺍﻟﺘﺤﻭﻴل ﻟلUTF8 ﻫﻲ ﺃﻥ ﺍﻟﺤﺭﻭﻑ ﺍﻟﺘﻲ ﻜﺎﻥ ﻴﺘﻡ ﺘﺭﻤﻴﺯﻫﺎ ﻓﻲ ﺍلCharacter Set ﺍﻷﺼﻠﻴﺔ ﺏByte One ﺴﻭﻑ ﺘﺘﻁﻠﺏ Byte Two ﻓﻲ ﺍلUTF8،

ﻟﺫﻟﻙ ﻓﺎﻥ ﺍﻟﺒﻴﺎﻨﺎﺕ ﻗﺩ ﻻ ﺘﺼﻠﺢ ﻓﻲ ﺍﻟﻌﻤﻭﺩ ﺒﻌﺩ ﺍﻟﺘﻐﻴﻴﺭ.

ﺍلDatabase Character Set Scanner ﺴﺘﻨﺘﺞ ﺘﻘﺭﻴﺭﹰﺍ ﺸﺎﻤﻼﹰ ﻟﻜل ﺼﻑ ﺴﻴﺤﺩﺙ ﻓﻴﻪ ﻤﺸﻜﻠﺔ ﻓﻲ ﺍلSet Character ﺍﻟﺠﺩﻴﺩ، ﺜﻡ ﻴﺠﺏ ﺃﻥ ﺘﺘﺨﺫ ﺍﻹﺠﺭﺍﺀﺍﺕ ﺍﻟﻤﻨﺎﺴﺒﺔ

ﻟﻌﻼﺝ ﺍﻟﻤﺸﺎﻜل ﻗﺒل ﺍﻟﺘﺤﻭﻴل ﺇﻥ ﻜﺎﻥ ﺫﻟﻙ ﻤﻤﻜﻨﺎﹰ.

2- The Language and Character Set File Scanner ﻫﺫﻩ ﺍﻷﺩﺍﺓ ﻤﻥ ﺸﺄﻨﻬﺎ ﻤﺤﺎﻭﻟﺔ ﺍﻟﺘﻌﺭﻑ ﻋﻠﻲ ﺍﻟﻠﻐﺔ ﻭﺍلCharacter Set ﺍﻟﻤﺴﺘﺨﺩﻤﺔ ﻓﻲ ﻤﻠﻑ ﻨﺼﻲ،

ﻓﻬﻲ ﺘﻌﻤل ﻋﻠﻲ ﻨﺼﻲ ﻋﺎﺩﻱ ﻓﻘﻁ، ﻫﺫﻩ ﺍﻷﺩﺍﺓ ﻤﻔﻴﺩﺓ ﺠﺩﺍﹰ ﺇﺫﺍ ﻜﺎﻨﺕ ﻟﺩﻴﻙ ﺒﻴﺎﻨﺎﺕ ﺘﺭﻴﺩ ﺘﺤﻤﻴﻠﻬﺎ ﻟﻘﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ ﻭﻻ ﺘﺩﺭﻱ ﻤﺎ ﻫﻲ ﺍﻟﺒﻴﺎﻨﺎﺕ، ﺘﻘﻭﻡ ﺍﻷﺩﺍﺓ

ﺒﻔﺤﺹ ﺍﻟﻤﻠﻑ ﻭﺘﺨﻤﻴﻥ ﺍﻟﻠﻐﺔ ﻭﺍلCharacter Set ﻟﻠﺒﻴﺎﻨﺎﺕ. ﺒﻌﺩ ﺍﻟﺘﺄﻜﺩ ﻤﻥ ﺇﻤﻜﺎﻨﻴﺔ ﺘﻐﻴﻴﺭ ﺍلCharacter Set ﺩﻭﻥ ﺃﻀﺭﺍﺭ ﻴﻤﻜﻨﻙ ﺘﻨﻔﻴﺫ ﺍﻷﻤﺭ

ﻜﻤﺎ ﻴﻤﻜﻨﻙ ﺘﻐﻴﻴﺭ ﺍلNational Character Set  ﺒﻨﻔﺱ ﺍﻟﻁﺭﻴﻘﺔ Alter database national character set ، ﻭﻟﻜﻥ ﻟﻴﺱ ﻫﻨﺎﻙ ﻀﻤﺎﻥ ﺒﻌﺩﻡ ﻭﺠﻭﺩ ﻤﺸﺎﻜل ﻭﻫﺫﻩ ﻤﻬﻤﺔ ﻤﺩﻴﺭ ﻗﺎﻋﺩﺓ ﺍﻟﺒﻴﺎﻨﺎﺕ.

----------------------------