Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.lang.java.programmer > #25482 > unrolled thread

Re: XStream vs XMLGregorianCalendar

Started byLew <noone@lewscanon.com>
First post2011-01-30 13:37 -0500
Last post2011-01-30 13:37 -0500
Articles 1 — 1 participant

Back to article view | Back to comp.lang.java.programmer

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: XStream vs XMLGregorianCalendar Lew <noone@lewscanon.com> - 2011-01-30 13:37 -0500

#25482 — Re: XStream vs XMLGregorianCalendar

FromLew <noone@lewscanon.com>
Date2011-01-30 13:37 -0500
SubjectRe: XStream vs XMLGregorianCalendar
Message-ID<ii4b65$qvr$1@news.albasani.net>
Lew wrote:
> (Another question is whether to version-control the generated .java files or
> depend on the build to create them each time. A schema and its associated JAXB
> classes make a nice JAR.) [1]

[1] Too often, we programmers disregard deployment and operations issues. 
It's a good exercise to figure out where to put your JAXB stuff in your 
version-control repository, write and run the build scripts (Ant, for example) 
that work off a clean export from that repository, accounting for code used on 
both ends of the exchange.  Then deploy that yourself on two instances - one 
for each end - of whatever the target environments are (desktop app, 
smartphone app, Java EE app, ...).  Integrate that with a 
continuous-integration platform like Hudson.  For extra credit, put the 
database on a different host.

That exercise tends to make you a fan of framework reduction - too many 
programmers take all the latest Springs and Ajax of outrageous configuration 
without regard for architecture, emergent interactions or transitive version 
dependencies.  One outcome is to undo all the type safety of the Java 
language, substituting instead a mish-mosh of references by name whose 
incompatibilities come forward only at runtime.

Injection has its place, but most of the time, what's wrong with just calling 
a durn method?

-- 
Lew
Ceci n'est pas une fenĂȘtre.
.___________.
|###] | [###|
|##/  | *\##|
|#/ * |   \#|
|#----|----#|
||    |  * ||
|o *  |    o|
|_____|_____|
|===========|

[toc] | [standalone]


Back to top | Article view | comp.lang.java.programmer


csiph-web