Rick Hillegas (JIRA) wrote:
[ http://issues.apache.org/jira/browse/DERBY-1079?page=all ]
Rick Hillegas updated DERBY-1079:
---------------------------------
Attachment: bug1079_split.diff
Attaching bug1079_split.diff. This patch splits the public api javadoc into two separate
sets: one for JDBC3 applications and one for JDBC4 applications. The patch also bolts a
webpage on top of these two sets. The webpage explains when to use each of these sets.
This addresses the problems raised on derby-dev in the email thread titled "Javadoc
lies".
The J2ME/CDC DataSource (EmbeddedSimpleDataSource) goes into the JDBC3 set of
javadoc. If someone has the itch, they can add an additional J2ME/CDC
DataSource to implement the JDBC4 DataSource api. This additional DataSource
can then go into the JDBC4 set.
I have one comment (on Java naming):
In tools/javadoc/publishedapi_index.html, instead of saying
<li><a href="jdbc3/index.html">JDBC2/JDBC3 Public API</a> - Consult
this javadoc if your application runs on
JDK 1.3, JDK 1.4, or J2SE 5.0. Also consult this javadoc if your
application runs with the CDC/Foundation Profile(JSR-169).</li>
<li><a href="jdbc4/index.html">JDBC4 Public API</a> - Consult this
javadoc if your application runs on J2SE 6.0.</li>
we should say
<li><a href="jdbc3/index.html">JDBC2/JDBC3 Public API</a> - Consult
this javadoc if your application runs on
J2SE 1.3, J2SE 1.4, or J2SE 5.0. Also consult this javadoc if your
application runs with the CDC/Foundation Profile (JSR-169).</li>
<li><a href="jdbc4/index.html">JDBC4 Public API</a> - Consult this
javadoc if your application runs on Java SE 6.</li>
Two reasons:
a) Because one does not need a JDK (or SDK) to run Derby
b) We should (IMHO) strive to comply with current Java naming and
versioning [1, 2] in official Derby documentation.
--
John
[1]
http://java.sun.com/developer/technicalArticles/JavaOne2005/naming.html
[2]
http://java.sun.com/javase/downloads/index.html
