On Dec 10, 2007, at 12:32 PM, mark harwood wrote:
I don't know that we have ever checked in IDE settings
GWT development is much easier with the IDE and there is a fair
amount of manual setup required without the settings to run the
"hosted" development environment. Hosted development is the key
productivity benefit and allows debugging in Java (rather than
building, deploying then having to debug Javascript). GWT provides
Eclipse project generators to get started and they do not target
other IDEs e.g. NetBeans because they claim those IDEs typically do
a good job of importing eclipse project settings (can't vouch for
this myself, not having tried).
Aren't they user specific at some point
No, I have taken care to ensure these IDE setting files have all
directory names etc replaced with variables - in the same way ANT
build files use properties to avoid machine-specifics.
Right, you have done this, but that doesn't guarantee that the next
committer who comes along will necessarily be on top of it. That
being said, I don't care too much about it. I use IntelliJ and it has
GWT support in it (although I haven't actually used it) Mostly, I
just don't want to see some proliferation of IDE files in various
places throughout the project.
In that case, then, maybe jetty should be packaged somewhere else
outside of WebLuke?
Yes, I thought that. I tried putting the only other current webapp,
"luceneweb.war" under Jetty but it failed to do anything of interest
"out of the box" because it requires an index to be built first. We
could extend that app to include web-based screens to create and
populate an index but I suspect that rapidly puts us on a
development path heading towards Solr or SearchBlox.
I think it would be reasonable to have a script/batch file that
created an index and then fired up the Lucene demo and WebLuke. Very
simple and by all means nowhere near the level of what Solr or any
other vendor provides. The pieces are all pretty much there, just
need a script around it.
Also, should this be in 2.3?
Might be an idea to let it bed-down a little first. I'm not happy
with the (lack of) security at present and wouldn't want naive users
complaining of vulnerabilities introduced by its deployment.
I think we should wait for 2.3 to come out, either that or mark it as
experimental and put notes about the known issues in a conspicuous
place, such as a README.
-Grant
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
