Routine to validate US State Codes

[bapajju]

Hello,

In a routine I want to validate if the code is one of the valid US State codes . If yes then return 1 else return 0. I have used the below mentioned routine code to get the result. But I do not think this is an efficient and compact code. Can some one help on this.


Ans =' '
If Arg1 = 'AR' or Arg1 = 'AZ' or Arg1 = 'CA' or Arg1 = 'CO' or Arg1 = 'CT'or Arg1 = 'FL' or Arg1 = 'GA' or Arg1 = 'IA' or Arg1 = 'IL' or Arg1 = 'IN'or Arg1 = 'KS' or Arg1 = 'LA' or Arg1 = 'MA' or Arg1 = 'MI' or Arg1 = 'MN' or Arg1 = 'MO' or Arg1 = 'MS' or Arg1 = 'NE' or Arg1 = 'NH' or Arg1 = 'NJ' or Arg1 = 'NM' or Arg1 = 'NV' or Arg1 = 'OH' or Arg1 = 'OK' or Arg1 = 'OR' or Arg1 = 'PA' or Arg1 = 'PR' or Arg1 = 'IN' or Arg1 = 'TN' or Arg1 = 'TX' or Arg1 = 'UT' or Arg1 = 'VA' or Arg1 = 'WA' or Arg1 = 'WI' or Arg1 = 'WV'or Arg1 = 'DF' or Arg1 = 'MS' or Arg1 = 'NE' or Arg1 = 'NH' or Arg1 = 'NJ' or Arg1 = 'NM' or Arg1 = 'NV' or Arg1 = 'OH' or Arg1 = 'OK' or Arg1 = 'OR'
Then
Ans=1
End
Else
Ans=0
End

[chulett]

Load them into a hashed file.

[kcbland]

Put all of your state codes into a comma separated string, then just use the INDEX BASIC function if you can guarantee that Arg1 will always be 2 characters. For added performance, initialize States variable once and use a labeled COMMON to hold the variable. INDEX with a 1 in the third argument means stop on the first occurrence and return the character position where found. If not found, it returns 0.

Code: Select all

States = "AA,BB,CC,DD,EE,FF"

FOUND=INDEX(States, Arg1,1)

[chulett]

Or that.

[bapajju]

Thanks Kenneth. it works. But lets say rather than getting the Character Number of the first occurance I need only 0 and 1. 1 for Found and 0 for not found. Is there any shortcut way to this.

Put all of your state codes into a comma separated string, then just use the INDEX BASIC function if you can guarantee that Arg1 will always be 2 characters. For added performance, initialize States variable once and use a labeled COMMON to hold the variable. INDEX with a 1 in the third argument means stop on the first occurrence and return the character position where found. If not found, it returns 0.

Code: Select all

States = "AA,BB,CC,DD,EE,FF"

FOUND=INDEX(States, Arg1,1)

[chulett]

You've got what you need - a zero is not found and a non-zero is found.

[sun rays]

Use the hash file to look-up, and in the transformer check if the look-up failed or suceeded, by using ISNULL() and assign a zero or 1 based on the result.

[kcbland]

By using COMMON to hold the comma separated list, the list is built once. After that, for short text strings, say under 10K length, the INDEX command is the fastest method. By guaranteeing 2 char values, you don't worry about false hits on a substring match. If the list was large, or the chance of a substring match, then the hash lookup would be a safer and better solution.

As for help on the INDEX command, look in your DS BASIC manual on the start button to learn about it. Craig's correct, non-zero values will be treated as @TRUE if used in a boolean expression.

[ArndW]

In this case all the States are 2 digits and don't contain substrings of each other so the quick INDEX() will work well. If you were looking up a country list which might have ...USA,UK,UKMENISTAN... {doesn't exist, but I couldn't think of a better example} then it would make sense to change the comma delimiter to @FM and use the FIND() function or the LOCATE() function.

[bapajju]

Thanks everyone. It works perfect with INDEX(). Thanks a lot for the prompt help.

In this case all the States are 2 digits and don't contain substrings of each other so the quick INDEX() will work well. If you were looking up a country list which might have ...USA,UK,UKMENISTAN... {doesn't exist, but I couldn't think of a better example} then it would make sense to change the comma delimiter to @FM and use the FIND() function or the LOCATE() function.