TEXT-encoding of 'aa9' and 'aa`' identical

[adrianschlatter]

Encoding the messages 'aa9' and 'aa`' to TEXT gives identical results for both which can't be correct.

toText('aa`', [ /* TEXT */ 239, 31, 0, 0, 32, 9, 29, 47, 1, 33, 57, 9, 44, 64, 1, 43, 90, 2, 64, 95, 1, 69, 122, 9, 83, 127, 2, 96, 255, 1, 0])

=> [239, 89, 190, 254]

toText('aa9', [ /* TEXT */ 239, 31, 0, 0, 32, 9, 29, 47, 1, 33, 57, 9, 44, 64, 1, 43, 90, 2, 64, 95, 1, 69, 122, 9, 83, 127, 2, 96, 255, 1, 0])

=> [239, 89, 190, 254]

[datalog]

First of all thank you for your feedback.

It is true that this function may return the same result for different chunks of data. Please, keep in mind that this is not standalone function, this is just a part of a cascade. Every next step will revaluate data again in order to choose optimum encoding method... This also helps to avoid mentioned collision.

[adrianschlatter]

I agree with you. However, it is apparently not enough to assure correctness. I tried a bit with your nice datamatrix generator tool and found a collision (i.e. same datamatrix) for these strings:

• "aaaaaaaaaaa9"
• "aaaaaaaaaaa`"