| Time |
S |
Nick |
Message |
| 02:42 |
|
|
Bombe joined ##javaee |
| 03:41 |
|
|
fedorstudio joined ##javaee |
| 04:38 |
|
|
mocrunsthecity joined ##javaee |
| 04:46 |
|
|
philbot joined ##javaee |
| 04:46 |
|
|
Topic for ##javaee is now Core Java (Java SE) AND Java Enterprise Edition (Java EE) discussion | ##javaee-offtopic for all kinds of non-tech chat| logs at http://irclog.greptilian.com/javaee/today |
| 10:43 |
|
|
philbot joined ##javaee |
| 10:43 |
|
|
Topic for ##javaee is now Core Java (Java SE) AND Java Enterprise Edition (Java EE) discussion | ##javaee-offtopic for all kinds of non-tech chat| logs at http://irclog.greptilian.com/javaee/today |
| 10:46 |
|
|
javaeebot joined ##javaee |
| 11:08 |
|
|
BooleanByte joined ##javaee |
| 11:32 |
|
|
javaeebot joined ##javaee |
| 12:00 |
|
firebird1 |
i'm usiing Date |
| 12:00 |
|
firebird1 |
no day light saving |
| 12:00 |
|
firebird1 |
no manipulation |
| 12:00 |
|
firebird1 |
safe ? |
| 12:18 |
|
|
alexas joined ##javaee |
| 12:19 |
|
|
rarduin joined ##javaee |
| 12:47 |
|
|
pdurbin joined ##javaee |
| 13:07 |
|
|
weyer joined ##javaee |
| 13:52 |
|
|
tommmied joined ##javaee |
| 14:21 |
|
|
philbot joined ##javaee |
| 14:21 |
|
|
Topic for ##javaee is now Core Java (Java SE) AND Java Enterprise Edition (Java EE) discussion | ##javaee-offtopic for all kinds of non-tech chat| logs at http://irclog.greptilian.com/javaee/today |
| 14:32 |
|
|
Naros joined ##javaee |
| 14:42 |
|
|
kobain joined ##javaee |
| 15:07 |
|
fabioportieri |
http://hackertyper.net/ :> |
| 15:18 |
|
firebird1 |
seriously i'm not jsf flow |
| 15:18 |
|
firebird1 |
i'm sooo difficult jsf :P |
| 15:33 |
|
|
balo joined ##javaee |
| 15:34 |
|
firebird1 |
difference between viewScoped and RequestScoped ? |
| 15:35 |
|
fabioportieri |
http://stackoverflow.com/questions/6025998/difference-between-view-and-request-scope-in-managed-beans |
| 15:37 |
|
fabioportieri |
if you don't need to persist values across http request you should go with requestScoped |
| 16:09 |
|
|
javaeebot` joined ##javaee |
| 16:18 |
|
|
sajjadg joined ##javaee |
| 16:18 |
|
|
syncsys joined ##javaee |
| 16:25 |
|
|
philbot joined ##javaee |
| 16:25 |
|
|
Topic for ##javaee is now Core Java (Java SE) AND Java Enterprise Edition (Java EE) discussion | ##javaee-offtopic for all kinds of non-tech chat| logs at http://irclog.greptilian.com/javaee/today |
| 16:38 |
|
|
neuro_sys joined ##javaee |
| 16:38 |
|
|
neuro_sys joined ##javaee |
| 16:47 |
|
sess |
anyone got experience making a wildcard definition with tiles? |
| 16:47 |
|
sess |
i just keep getting infinite loops.. |
| 17:02 |
|
whartung |
I've never used tiles, I just use raw tagfiles |
| 17:07 |
|
sess |
it seems very nifty if i can just get the wildcards to work |
| 17:07 |
|
sess |
so i get free definitions with the same name as the jsp path |
| 17:08 |
|
whartung |
yea, i'm not even sure of the functionality. |
| 17:09 |
|
sess |
it's just an extra layer between the controller and the jsp |
| 17:09 |
|
sess |
that allows for templating |
| 17:09 |
|
sess |
since pure jsp is bullocks |
| 17:11 |
|
whartung |
tag files make tempting trivial |
| 17:11 |
|
whartung |
templating... |
| 17:11 |
|
sess |
it is not possible to do proper templating with tag files |
| 17:11 |
|
sess |
it only allows for composition |
| 17:12 |
|
sess |
you still have to copypaste that include to every single jsp |
| 17:12 |
|
sess |
and <head> is a bitch with jsp:includes |
| 17:13 |
|
whartung |
define "proper templating" |
| 17:50 |
|
sfisque |
i think he means tiles supports proper inheritence. you can define a decoration or composition as inheriting a parent one, and everything works fine without having to doctor up the inheriting templates/tiles/fragments |
| 17:54 |
|
|
BooleanByte joined ##javaee |
| 17:57 |
|
whartung |
ok |
| 18:16 |
|
sess |
well not the inheritence part per say, but rather that the contet JSPs need not be aware of menus, head parts, footer etc |
| 18:18 |
|
sfisque |
you can do that with "include" in jsp's. the real power of tiles is you dont have to explicitly re-include everything if you need several "parallel" templates. just inherit one, and make the small mods |
| 18:21 |
|
whartung |
you can do that with tag files, you just need to wrap the parent tag |
| 18:22 |
|
syncsys |
is there anything wrong with this method? any comments? using spring and hibernate with JPA http://pastie.org/8651314 |
| 18:23 |
|
syncsys |
I mean ofcourse, its not good |
| 18:23 |
|
whartung |
method looks fine to me syncsys |
| 18:24 |
|
whartung |
(besides the println before the empty list check...) |
| 18:24 |
|
syncsys |
line 8 and 10 also? |
| 18:24 |
|
syncsys |
ah. |
| 18:25 |
|
whartung |
yea, line 8 has the potential to go boomski |
| 18:26 |
|
sess |
sfisque: not really, include means each jsp has to know all other parts |
| 18:26 |
|
sess |
and explicitly include them |
| 18:26 |
|
sfisque |
i didnt say it did it well. just that you "could" do it with raw jsp. tiles makes it more elegant |
| 18:27 |
|
sess |
well or not, it is exactly the opposite of "contet JSPs need not be aware of menus, head parts, footer etc" |
| 18:27 |
|
whartung |
so where does the composition happen in tiles, in a config file? |
| 18:28 |
|
sess |
yeah |
| 18:28 |
|
sfisque |
combination of configuration and design |
| 18:28 |
|
sess |
im upgrading a really old system that uses frames |
| 18:28 |
|
sfisque |
it is similar to how jsf does it |
| 18:28 |
|
sess |
so i slap on tiles, and just remove a few parts from the jsps |
| 18:28 |
|
sfisque |
nice. frames == t3h debil |
| 18:28 |
|
sess |
that's the theory anyways, if i can get the wildcards to work |
| 18:28 |
|
sess |
because im NOT making definitions for every single jsp |
| 18:29 |
|
sess |
convention over configuration any day |
| 18:29 |
|
sfisque |
well, be careful. once you start using knockouts " {x} |
| 18:29 |
|
syncsys |
whartung, how about this (with changed highlighted lines ) http://pastie.org/8651326#2,9-12 |
| 18:29 |
|
sfisque |
" for directory fragments, your design can get fragile |
| 18:29 |
|
sess |
knockouts? |
| 18:30 |
|
whartung |
looks fine to my syncsys |
| 18:30 |
|
sess |
syncsys: your last one was better |
| 18:31 |
|
sess |
an exception should be thrown if it's not found |
| 18:31 |
|
sess |
just move the get(0) to after the check |
| 18:31 |
|
syncsys |
added email.tolowercase() http://pastie.org/8651332#1-3,9-13 |
| 18:32 |
|
syncsys |
sess, hm |
| 18:32 |
|
sess |
to make it even snazier |
| 18:32 |
|
sess |
use getSingleReault |
| 18:32 |
|
sess |
throw exception automatically if 0 results |
| 18:32 |
|
syncsys |
sess, my last one was better? but this one is ok too? |
| 18:32 |
|
sess |
result* |
| 18:32 |
|
sess |
I'd prefer an exception over return null |
| 18:33 |
|
sess |
but both are fine technically |
| 18:33 |
|
sfisque |
and for the love of the baby java, get rid of the system.out.println's. they are not guaranteed to work in a contained environment. use a Logger. |
| 18:33 |
|
syncsys |
sess, sfisque whartung one other problem is. entityManager will never return a null for getResultList() |
| 18:33 |
|
sfisque |
and they're a performance bottleneck. println is synchronized |
| 18:34 |
|
sess |
syncsys: no it wont, so you wont have to check that either |
| 18:34 |
|
sess |
i dont see how it is a problem |
| 18:34 |
|
sess |
also use a typed query |
| 18:35 |
|
sess |
syncsys: http://pastie.org/8651342 |
| 18:35 |
|
sess |
this is fine tbh |
| 18:35 |
|
sess |
catch NoResultException in the calling method |
| 18:35 |
|
sess |
and fail the login there |
| 18:36 |
|
sfisque |
swap the println for LOGGER.info( "blah blah blah" ); |
| 18:36 |
|
sfisque |
and define LOGGER at the top of the class |
| 18:36 |
|
sfisque |
private static final Logger LOGGER; |
| 18:37 |
|
sfisque |
well, only final if you instantiate it in place |
| 18:37 |
|
sfisque |
otherwise just static and initialize lazy or aggressive as you see fit |
| 18:38 |
|
sfisque |
if you're working in NB, eclipse, or idea, there's no reason for a println. you can wire up the debugger and step through the code as desired |
| 18:38 |
|
sfisque |
this isnt 1973 |
| 18:58 |
|
|
neuro_sys joined ##javaee |
| 19:11 |
|
pdurbin |
bummer. I used to be able to use curl to POST some JSON to create an object (and persist it to the database) but our object model changed and now I'm getting this: javax.servlet.ServletException: Exception [EclipseLink-44] (Eclipse Persistence Services - 2.5.0.v20130507-3faac2b): org.eclipse.persistence.exceptions.DescriptorException Exception Description: Missing class indicator field from database row |
| 19:11 |
|
pdurbin |
[UnmarshalRecordImpl()]. Descriptor: XMLDescriptor(edu.harvard.iq.dataverse.DvObjectContainer --> []) |
| 19:12 |
|
pdurbin |
anybody know how I can get this working again? |
| 19:13 |
|
whartung |
you using Entity Inheritance pdurbin ? |
| 19:16 |
|
pdurbin |
um. I'm not sure of the details of how the object model changed. I'll hear about it tomorrow |
| 19:17 |
|
pdurbin |
here's my method (newlines removed) that used to work: @POST public String add(Dataset dataset) { datasetService.save(dataset); return "dataset " + dataset.getId() + " created/updated\n;"} |
| 19:17 |
|
pdurbin |
that method takes a Dataset as an argument but in practice I was sending it JSON |
| 19:18 |
|
whartung |
just sounds like that there is some inheritance going on and the classifier column is missing (so JPA can tell what class the row of data really is) |
| 19:18 |
|
whartung |
that's the way I read the error |
| 19:19 |
|
pdurbin |
whartung: yeah, I guess that makes sense... not sure how to fix it though |
| 19:19 |
|
whartung |
add the column and put the correct value in it :) |
| 19:19 |
|
pdurbin |
here's some example JSON I was sending that was working: {"author":"Dr. Bird","citationDate":"2003-12-11","description":"bird study 1","distributor":"For the Birds","id":"2","owner":{"id":"2"},"title":"bird dataset"} |
| 19:20 |
|
whartung |
what's the schema in the db? |
| 19:23 |
|
* pdurbin |
downloads http://schemaspy.sourceforge.net |
| 19:24 |
|
whartung |
you don't have a sql listener like psql handy? |
| 19:24 |
|
pdurbin |
I have psql. I thought I'd show rather than tell :) |
| 19:25 |
|
whartung |
you're just comparing the columns in the table to the ones in your entity |
| 19:33 |
|
pdurbin |
whartung: I heard you but I haven't been working on the schema. here it is: http://dvn-5.hmdc.harvard.edu/tmp-schemaspy-tmp3/dvn-schemaspy-4861940/ |
| 19:37 |
|
pdurbin |
"Exception Description: Missing class indicator field from database row [UnmarshalRecordImpl()]. Descriptor: XMLDescriptor(edu.harvard.iq.dataverse.DvObjectContainer --> [])" and DvObjectContainer extends DvObject so I guess I should be looking at http://dvn-5.hmdc.harvard.edu/tmp-schemaspy-tmp3/dvn-schemaspy-4861940/tables/dvobject.html |
| 19:39 |
|
sfisque |
pdurbin is there a complementary "read" operation. read a row of the same data type, and see what columns you're missing |
| 19:42 |
|
pdurbin |
sfisque: yes, actually... and what's new in the JSON hash that's returned is a key called "@type" ... it's the only key with an @ in from of it... the others have no @ such as "title" etc. |
| 19:44 |
|
sfisque |
intriguing. i've not seen that convention in a java->json object serialization |
| 19:45 |
|
sfisque |
any idea what the @ denotes? |
| 19:46 |
|
pdurbin |
it seems to be the type of the object. so for us... dataverses vs. datasets vs. files |
| 19:46 |
|
pdurbin |
previously I was able to POST all three kinds to create/update and persist them |
| 19:46 |
|
pdurbin |
I'm using moxy in glassfish, as I mentioned here: http://irclog.greptilian.com/javaee/2013-12-05 |
| 19:48 |
|
pdurbin |
the @POST method here got me started: https://github.com/javaee-samples/javaee7-samples/blob/master/jaxrs/moxy/src/main/java/org/javaee7/jaxrs/moxy/MyResource.java |
| 19:52 |
|
pdurbin |
whartung: no inheritance going on in that example |
| 20:18 |
|
syncsys |
any one knows why in spring security in CustomUserDetailsService implements UserDetailsService{ it makes dbUser.getPassword().toLowerCase()? |
| 20:20 |
|
syncsys |
http://krams915.blogspot.com/2010/12/spring-security-mvc-integration_18.html e.g line 54 |
| 20:35 |
|
semiosis |
syncsys: no, that seems strange to me. i coerce *usernames* to lowercase, but not passwords |
| 20:35 |
|
syncsys |
i coerce ? |
| 20:35 |
|
semiosis |
username.toLowerCase() |
| 20:35 |
|
syncsys |
ya but line 54 clearly is doing dbUser.getPassword().toLowerCase(), |
| 20:36 |
|
syncsys |
I think it should be to username. not password |
| 20:36 |
|
semiosis |
i agree |
| 20:36 |
|
syncsys |
thanks |
| 20:36 |
|
semiosis |
but you'd have to ask the author of that article, i guess |
| 20:36 |
|
syncsys |
hm |
| 20:51 |
|
semiosis |
pdurbin: sfisque: http://www.w3.org/TR/2014/REC-json-ld-20140116/ |
| 20:51 |
|
semiosis |
this is new |
| 20:51 |
|
semiosis |
http://www.w3.org/TR/2014/REC-json-ld-20140116/#typed-values |
| 20:52 |
|
semiosis |
JSON-LD 1.0 - A JSON-based Serialization for Linked Data - W3C Recommendation 16 January 2014 |
| 20:52 |
|
semiosis |
4 days ago |
| 20:52 |
|
semiosis |
sounds related, and the timing is right |
| 20:55 |
|
semiosis |
pdurbin: are you running new code against an existing test/dev database on your machine? you probably need to update your local db schema |
| 21:10 |
|
sfisque |
ah, so the @…. attributes are "magic" attributes for json. nifty |
| 21:14 |
|
semiosis |
that's my guess |
| 21:15 |
|
semiosis |
json-ld specifically |
| 21:23 |
|
|
mocrunsthecity joined ##javaee |
| 21:51 |
|
|
tommmied left ##javaee |
| 22:08 |
|
pdurbin |
semiosis: I dropped my old database and started fresh |
| 22:09 |
|
semiosis |
did that fix it? |
| 22:09 |
|
pdurbin |
nope |
| 22:11 |
|
|
weyer joined ##javaee |
| 22:14 |
|
pdurbin |
maybe I need the @XmlAccessorType and/or @XmlType stuff mentioned at http://stackoverflow.com/questions/9875378/descriptorexception-missing-class-indicator-field-from-database-row-exception/9881317#9881317 |
| 22:14 |
|
pdurbin |
(even though I'm trying to use JSON rather than XML) |
| 22:15 |
|
whartung |
depending on how you're doing it, JSON and JAXB can be using the same underlying infrastructure. |
| 22:21 |
|
pdurbin |
hmm. ok |
| 22:29 |
|
semiosis |
pdurbin: sounds to me like there's a mismatch between your db schema & your domain entities |
| 22:30 |
|
|
Quest- joined ##javaee |
| 22:30 |
|
pdurbin |
semiosis: but if I drop my db and let all the tables get recreated when I deploy my app... there should be no mismatch, right? |
| 22:30 |
|
semiosis |
what creates them? |
| 22:31 |
|
pdurbin |
magic |
| 22:31 |
|
semiosis |
magic should work |
| 22:31 |
|
pdurbin |
jpa, I guess |
| 22:31 |
|
pdurbin |
I have no idea how this stuff works |
| 22:31 |
|
semiosis |
try those Xml annotations |
| 22:32 |
|
semiosis |
plenty of json stuff uses jaxb |
| 22:32 |
|
semiosis |
as whartung pointed out |
| 22:33 |
|
pdurbin |
trying. many things competing for my attention. as usual |
| 22:49 |
|
pdurbin |
interesting. I'm playing with @type not as a key not at the first level but at the second level (within the "owner") key, and I'm making progress, maybe... a different exception at least: javax.persistence.NonUniqueResultException |
| 22:50 |
|
semiosis |
pdurbin: are you sure you're using the DB you think you're using? |
| 22:51 |
|
pdurbin |
lolz |
| 22:52 |
|
pdurbin |
it's like it's trying to create the owner object... which I don't want to do... I'm trying to specify who the owner is without creating it |
| 22:52 |
|
pdurbin |
since the owner already exists |
| 22:55 |
|
semiosis |
pdurbin: i think you need to pull the existing item from the db via jpa |
| 22:57 |
|
pdurbin |
huh. I got something to work. weird. maybe a bug somewhere else in our code |
| 23:06 |
|
pdurbin |
well... I got something to be persisted... using @type nested inside my JSON, but the resulting object is all broken and weird |
| 23:07 |
|
pdurbin |
here's the JSON I used: {"affiliation":"For the Trees","alias":"chestnuts","contactEmail":"chestnuts trees.com","description":"chestnuts etc","id":"9","name":"Chestnuts","owner":{"@type":"dataverse","alias":"tmpAlias","name":"tmpName","contactEmail":"tmptmp.com","id":"4"}} |
| 23:08 |
|
pdurbin |
anyway, no more DescriptorException... and I never put any of the XML annotations in... so maybe it's just a matter of me figuring out what JSON to feed it to get a proper object persisted |
| 23:32 |
|
sfisque |
yeah, i'd guess they use the same reflection framework and just load a different "de/encoder" for the actual read/write phase |
| 23:49 |
|
pdurbin |
yeah |
