[ 
https://issues.apache.org/jira/browse/DERBY-7097?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17266929#comment-17266929
 ] 

Richard N. Hillegas commented on DERBY-7097:
--------------------------------------------

Attaching derby-7097-01-ab-jpms-related-website-changes.diff. This patch adds a 
pointer to the 10.15.1.3 release notes to the seed text for future releases of 
Derby. People upgrading from 10.14 or earlier may need to be pointed to those 
release notes for context on the classpath change.


Touches the following file:

{noformat}
M       java/build/org/apache/derbyBuild/ReleaseNotesGenerator.java
{noformat}


> Update documentation to allow users to properly use EmbeddedDriver
> ------------------------------------------------------------------
>
>                 Key: DERBY-7097
>                 URL: https://issues.apache.org/jira/browse/DERBY-7097
>             Project: Derby
>          Issue Type: Bug
>          Components: Documentation
>    Affects Versions: 10.15.2.0
>            Reporter: Daniel Le Berre
>            Priority: Major
>         Attachments: derby-7097-01-aa-jpms-related-website-changes.diff, 
> derby-7097-01-ab-jpms-related-website-changes.diff, 
> derby-7097-02-aa-releaseSummary.diff
>
>
> In earlier version of Derby, and as reported in the documentation, the 
> EmbeddedDriver class was in derby.jar.
> As such, it was quite easy to deploy a webapp with an embedded derby 
> database: it was sufficient to just add a derby.jar file to the project 
> library.
> In current releases of derby (10.15.2.0), the EmbeddedDriver class is no 
> longer in the derby.jar file but in the derbytools.jar file.
> ```
> $ for i in `ls *.jar`; do echo $i ; jar tf $i | grep Driver ; done
>  derby.jar
>  org/apache/derby/iapi/jdbc/AutoloadedDriver$1.class
>  org/apache/derby/iapi/jdbc/AutoloadedDriver.class
>  org/apache/derby/iapi/jdbc/Driver42.class
>  org/apache/derby/iapi/jdbc/InternalDriver$1.class
>  org/apache/derby/iapi/jdbc/InternalDriver$2.class
>  org/apache/derby/iapi/jdbc/InternalDriver$3.class
>  org/apache/derby/iapi/jdbc/InternalDriver$4.class
>  org/apache/derby/iapi/jdbc/InternalDriver$5.class
>  org/apache/derby/iapi/jdbc/InternalDriver$DaemonThreadFactory.class
>  org/apache/derby/iapi/jdbc/InternalDriver$LoginCallable.class
>  org/apache/derby/iapi/jdbc/InternalDriver.class
>  META-INF/services/java.sql.Driver
>  derbyLocale_cs.jar
>  derbyLocale_de_DE.jar
>  derbyLocale_es.jar
>  derbyLocale_fr.jar
>  derbyLocale_hu.jar
>  derbyLocale_it.jar
>  derbyLocale_ja_JP.jar
>  derbyLocale_ko_KR.jar
>  derbyLocale_pl.jar
>  derbyLocale_pt_BR.jar
>  derbyLocale_ru.jar
>  derbyLocale_zh_CN.jar
>  derbyLocale_zh_TW.jar
>  derbyclient.jar
>  org/apache/derby/client/ClientAutoloadedDriver.class
>  META-INF/services/java.sql.Driver
>  derbynet.jar
>  derbyoptionaltools.jar
>  derbyrun.jar
>  derbyshared.jar
>  derbytools.jar
>  org/apache/derby/jdbc/ClientDriver.class
>  org/apache/derby/jdbc/ClientDriver40.class
>  org/apache/derby/jdbc/EmbeddedDriver.class
> ```
> As such, most of the tutorials found on the internet about "how to use derby 
> in embedded mode" are just wrong because they simply mention derby.jar as a 
> dependency.
> Worst, derby own documentation is not up to date: as such, I had no way to 
> understand why new releases of this tool that I have been using for more than 
> a decade in the classroom suddenly did not work anymore.
> The explanation is finally simple: I just wonder how such impacting decision 
> could be done without proper documentation.
> I am also surprised to be the first one reporting this, since the problem 
> exists since at least a year.
> There are two possible fix to this issue:
>  * move back the EmbeddedDriver class to derby.jar (my favorite option I 
> would say), but I guess there is a good reason for moving those classes to 
> derbytools.jar
>  * update the documentation on Derby's web site, with a quite visible alert 
> about this change.
>  
> I have been a pretty happy user of derby for years, and will certainly be in 
> the future. Thanks for that great tool. However, that breaking change has 
> been particularly annoying, which is the reason of that bug report.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to