Hi Christian! unfortunately there are runtime dependencies, since generated (un)marshallers rely on (the most efficient, IMHO) 3rd party streaming APIs... but what is good is that everything is organized in fine modules, so users include just the part are interested on (i.e. there could be user interested on having the JSON config only) and 3rd part dependencies are included as transitive dependencies.
OH, forgot to mention that I already applied ASF headers/license, in case you want to include it. All the best, -Simo http://people.apache.org/~simonetripodi/ http://twitter.com/simonetripodi On Wed, Jun 12, 2013 at 12:27 PM, Christian Grobmeier <[email protected]> wrote: > Hi, > > I like this approach. I also think its worth mentioning that we don't have > additional runtime dependencies. All code generation happens at build time, > to my knowledge. > > Christian > > On Wed, Jun 12, 2013 at 11:40 AM, Simone Tripodi > <[email protected]> wrote: >> Hi all mates, >> >> I recently got involved in the Modello project[1], adding some useful >> stuff such as JSON and YAML (un)marshallers generation and took some >> existing configurations as sample to test the quality of new >> marshallers... and "accidentally" took also the Log4j2 configuration. >> >> So, I tracked my progresses in a private project that I just shared on >> GitHub[2] in the case some of you would like to have a look and, since >> Log4j2 is still under beta version, I wonder if there is enough space >> to discuss the inclusion of that configuration (has to be completed, >> I'm sure I missed some elements) in the current trunk development, >> these are the added value with the new approach: >> >> * One single model definition and Java POJOS/XSD/XML/JSON/YAML >> support generated with 0 effort; >> >> * Users have APIs to describe and generate Log4j configurations in >> different formats; >> >> * All parsers/serializers are based on Streaming APIs (jackson-core >> for JSON, snake-yaml for YAML); >> >> * XML format[3] differs a little to current format when defining >> properties, but it is a little less verbose; >> >> * lighter, less verbose JSON notation[4]; >> >> * new YAML support[5]; >> >> Last, but not least, have a look at the generated APIs[6]! >> >> WDYT? >> Many thanks in advance, all the best! >> -Simo >> >> [1] http://modello.codehaus.org/index.html >> [2] https://github.com/simonetripodi/log4j2-config/ >> [3] >> https://github.com/simonetripodi/log4j2-config/blob/master/tests/src/test/resources/org/apache/logging/log4j/config/io/log4j.xml >> [4] >> https://github.com/simonetripodi/log4j2-config/blob/master/tests/src/test/resources/org/apache/logging/log4j/config/io/log4j.json >> [5] >> https://github.com/simonetripodi/log4j2-config/blob/master/tests/src/test/resources/org/apache/logging/log4j/config/io/log4j.yaml >> [6] >> https://github.com/simonetripodi/log4j2-config/blob/master/tests/src/test/java/org/apache/logging/log4j/config/io/AbstractConfigurationTestCase.java >> >> http://people.apache.org/~simonetripodi/ >> http://twitter.com/simonetripodi >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: [email protected] >> For additional commands, e-mail: [email protected] >> > > > > -- > http://www.grobmeier.de > https://www.timeandbill.de > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
