On Thu, Aug 11, 2011 at 5:22 PM, David Buchmann <[email protected]> wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > hi, > > implementing the php port of jcr as closely to the java spec as the > language differences permit, i have a question about the javadoc of > Property.setValue(Value value) > > the javadoc [1] tells that "If the property type is constrained, then a > best-effort conversion is attempted." > however, the jcr 2.0 specification defines in 3.6.4 an exact list of > what can be converted into which types and when to throw the > ValueFormatException. > the best-effort conversion suggerates that the implementation might > convert more of the cases or "just does something" like convert string > "hello" to integer 1.
no, that's not the intention. best-effort in this case means that a conversion is attempted but it's not guaranteed to succeed. > however, such behaviour would result in > non-portable client code because what works with one implementation > works not with an other. > > should the setValue method javadoc read "...then a conversion *according > to jcr spec paragraph 3.6.4* is attempted." ? i agree that the javadoc could be more specific WRT the supported value conversions. could you please file a jira issue [0]? thanks stefan [0] http://java.net/jira/browse/JSR_333 > > cheers,david > > [1] > http://www.day.com/maven/javax.jcr/javadocs/jcr-2.0/javax/jcr/Property.html#setValue%28javax.jcr.Value%29 > [2] > http://www.day.com/specs/jcr/2.0/3_Repository_Model.html#3.6.4%20Property%20Type%20Conversion > - -- > Liip AG // Agile Web Development // T +41 26 422 25 11 > CH-1700 Fribourg // PGP 0xA581808B // www.liip.ch > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.10 (GNU/Linux) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ > > iEYEARECAAYFAk5D87wACgkQqBnXnqWBgIuhbgCeO+yezmf+eqM1xEG6t/8GbhBR > kbwAnAqYx5JtA6xf/HrzspFE/iwqxirB > =OJyi > -----END PGP SIGNATURE----- >
