Closed richard-viney closed 2 years ago
Makes sense to me. Can you confirm that it results in an empty field when writing out again? Looks to me like it will.
Nice catch, I've pushed a fix for it writing back "null" in the serialisation case. It now writes back the empty string again so the read=>write roundtrip with a "" value in IntegerString/DecimalString now keeps it unchanged, as one would expect (this wasn't the case before this PR, i.e it would be transformed to zero).
Also worth noting that this is a breaking API change and so would need a v0.23 release.
Also worth noting that this is a breaking API change and so would need a v0.23 release.
Note that we use these commit message conventions so this would trigger a new major version, which seems a bit much for this. Maybe this is better seen as a fix, since as you pointed out the behavior was basically incorrect before.
Also worth noting that this is a breaking API change and so would need a v0.23 release.
Note that we use these commit message conventions so this would trigger a new major version, which seems a bit much for this. Maybe this is better seen as a fix, since as you pointed out the behavior was basically incorrect before.
Sounds good to me.
:tada: This PR is included in version 0.23.1 :tada:
The release is available on:
Your semantic-release bot :package::rocket:
This is needed when doing C-FIND with dcmjs-dimse, where "" means 'match everything', not zero.