You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
While working on #49, I stumbled upon some implementation differences between webforms and JavaRosa, that came about while thinking about "how and why does JR allow a randomization seed from a node with text, is that intentional or accidental, and should Webforms do the same" (so far I only have answers to #1 and maybe #2). Anyway:
The equivalent of a JS Number as used in Webforms is the Java Double. Same thing, same precision, so in theory this should allow identical outcomes on both platforms.
Evidently the JR implementation is much more extensive; and there's talk about "xpath type conversion rules" — which I haven't looked up, but those sound as if they could be relevant for webforms.
Poignantly, JR allows for "123,123" -> 123.123, while Webforms makes this a NaN.
Conversely, Webforms allows "0x12" -> 18, while JR would make this a NaN.
The question is thus: which one (if any) of the two is right? And if JR is right, what'd be the appropriate site for the Webforms implementation of same?
The text was updated successfully, but these errors were encountered:
While working on #49, I stumbled upon some implementation differences between webforms and JavaRosa, that came about while thinking about "how and why does JR allow a randomization seed from a node with text, is that intentional or accidental, and should Webforms do the same" (so far I only have answers to #1 and maybe #2). Anyway:
The equivalent of a JS
Number
as used in Webforms is the JavaDouble
. Same thing, same precision, so in theory this should allow identical outcomes on both platforms.JavaRosa implementation (think of it as a
toDouble
, as that's what it does)The correspondent code site in Webforms might be here
Evidently the JR implementation is much more extensive; and there's talk about "xpath type conversion rules" — which I haven't looked up, but those sound as if they could be relevant for webforms.
Poignantly, JR allows for
"123,123"
->123.123
, while Webforms makes this aNaN
.Conversely, Webforms allows
"0x12"
->18
, while JR would make this aNaN
.The question is thus: which one (if any) of the two is right? And if JR is right, what'd be the appropriate site for the Webforms implementation of same?
The text was updated successfully, but these errors were encountered: