HV-1999 new CNPJ alphanumeric format #1365
Conversation
| @@ -31,7 +31,7 @@ | |||
| * | |||
| * @author George Gastaldi | |||
| */ | |||
| @Pattern(regexp = "([0-9]{2}[.]?[0-9]{3}[.]?[0-9]{3}[/]?[0-9]{4}[-]?[0-9]{2})") | |||
| @Pattern(regexp = "([0-9A-Z]{2}[.]?[0-9A-Z]{3}[.]?[0-9A-Z]{3}[/]?[0-9A-Z]{4}[-]?[0-9]{2})") | |||
There was a problem hiding this comment.
Since this pattern now depends on the CNPJ's version/format, I'd suggest moving it into the CNPJ validator itself and picking the pattern that suits the format.
Doing so would also mean that @ReportAsSingleViolation can be removed.
| * @return Whether upper case letters characters in the validated input should be allowed ({@code true}) or result in a | ||
| * validation error ({@code false}). | ||
| */ | ||
| boolean alphanumeric() default false; |
There was a problem hiding this comment.
I'd suggest limit the changes to CNPJ constraint only
| @@ -45,6 +45,8 @@ | |||
|
|
|||
| Class<? extends Payload>[] payload() default { }; | |||
|
|
|||
| boolean alphanumeric() default false; | |||
There was a problem hiding this comment.
Wouldn't an enum here be more future-proof option? something like
| boolean alphanumeric() default false; | |
| FormatRevision formatRevision() default FormatRevision.V1; |
enum FormatRevision {
// Obviously, the constant names have to be more meaningful
// and connected to the versions of current and new formats.
V1, V2;
}This way, if another version is published, it can be added without any new parameters on the constraint annotation...
| protected void initialize(int startIndex, int endIndex, int checkDigitIndex, boolean ignoreNonValidCharacters) { | ||
| this.startIndex = startIndex; | ||
| this.endIndex = endIndex; | ||
| this.checkDigitIndex = checkDigitIndex; | ||
| this.ignoreNonDigitCharacters = ignoreNonDigitCharacters; | ||
| this.ignoreNonValidCharacters = ignoreNonValidCharacters; | ||
| this.alphanumeric = false; | ||
|
|
||
| this.validateOptions(); | ||
| } |
There was a problem hiding this comment.
maybe just call
initialize(startIndex, endIndex, checkDigitIndex, ignoreNonValidCharacters, false)and also if you'll end up still needing to modify the ModCheckBase consider using this overloaded version of initialize in places where alphanumeric is not needed so that less classes are modified.
| return Character.digit( value, DEC_RADIX ); | ||
| if ( alphanumeric ) { | ||
| if ( (value >= '0' && value <= '9' ) || (value >= 'A' && value <= 'Z') ) { | ||
| return value - BASE_CHAR_INDEX; |
There was a problem hiding this comment.
This way of converting seems somewhat specific to CNPJ, I'd suggest creating an extended version of Mod11CheckValidator in CNPJValidator and overriding extractDigit in it.
|
Thanks for the work and for bringing up the fact that there's an upcoming change in the format. I've addressed the review suggestions and aligned things more closely to how HV does things on top of your initial patch in this PR #1386 . I am closing this one. Thanks again! |
https://hibernate.atlassian.net/browse/HV-1999
Added an option to allow for gradual adoption.