Gracile noticed that citeproc-js currently tests “true” with the
“is-numeric” and “is-uncertain-date” conditions when the tested
variable(s) are empty:
Gracile, Sebastian, and I all think that the result should be “false”
instead. Unless anybody objects within the next few days I’ll modify
the CSL specification to require the latter behavior.
Rintze
Gracile noticed that citeproc-js currently tests “true” with the
“is-numeric” and “is-uncertain-date” conditions when the tested
variable(s) are empty:Problems with details of Lettres et sciences humaines style repository - Zotero Forums
Gracile, Sebastian, and I all think that the result should be “false”
instead. Unless anybody objects within the next few days I’ll modify
the CSL specification to require the latter behavior.Rintze
Ouch. I agree, that certainly shouldn’t happen.
Frank
Gracile noticed that citeproc-js currently tests “true” with the
“is-numeric” and “is-uncertain-date” conditions when the tested
variable(s) are empty:Problems with details of Lettres et sciences humaines style repository - Zotero Forums
Gracile, Sebastian, and I all think that the result should be “false”
instead. Unless anybody objects within the next few days I’ll modify
the CSL specification to require the latter behavior.Rintze
As far as I can tell, is-uncertain-date with an empty date variable is
returning false. I may be getting tangled up in double-negatives, but
that would be correct, right?
is-numeric is definitely behaving badly, though.
Frank
I only mentioned is-uncertain-date because it should behave analogous to
is-numeric, and if the spec is changed, both should be addressed.