Fix for handling of alphabetic repetition separator

[hoffrocket]

Problem

Valid X12 ISA headers can sometimes include an alpha-numeric character at element 11. In that case, the value should be ignored in favor of the default repetition separator "^"

We've encountered real world messages that have a V at ISA 11 fail to parse with the following error message on later events. I'm Assuming that a V was encountered elsewhere in the segments and the parser assumed that it was a control character.

io.xlate.edi.internal.stream.tokenization.EDIException: EDIE003 - Invalid processing state; INVALID (previous: TAG_1); input: 'V' in segment AMT at position 25
	at io.xlate.edi.internal.stream.tokenization.Lexer.error(Lexer.java:432)

Besides the spec, there's supporting evidence for this here: https://www.ibm.com/support/pages/isa-11-repetition-separator-letter-u-and-u-character-data-it-does-not-fail-x12-deenvelope

Fix

This change checks will elementRepeater to the default if the header value is alphabetic

[MikeEdgar]

@hoffrocket - thanks for bringing this up and for opening the PR. I have a couple suggestions around disabling the repeater rather than defaulting it. Please have a look and let me know what you think.

[MikeEdgar]

Thinking a bit more about this, what about adding this method to CharacterSet and calling instead of !Character.isAlphabetic? This will take into account other delimiters and user configuration.

    public boolean isValidDelimiter(int character) {
        switch (getClass(character)) {
        case CONTROL:
            return !extraneousIgnored;
        case WHITESPACE:
        case OTHER:
            return true;
        default:
            return false;
        }
    }

[coveralls]

Pull Request Test Coverage Report for Build 1002774612

• 4 of 5 (80.0%) changed or added relevant lines in 2 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage decreased (-0.02%) to 98.057%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
src/main/java/io/xlate/edi/internal/stream/tokenization/CharacterSet.java	3	4	75.0%

Totals
Change from base Build 976119375:	-0.02%
Covered Lines:	4795
Relevant Lines:	4890

---

💛 - Coveralls