| Time |
S |
Nick |
Message |
| 00:23 |
|
|
sfisque1 joined ##javaee |
| 00:53 |
|
|
playalpha left ##javaee |
| 00:59 |
|
pdurbin |
SoniEx2: please try to stay on topic :) |
| 01:19 |
|
|
SoniEx2|2 joined ##javaee |
| 01:44 |
|
SoniEx2 |
sfisque: you forgot to add ##GenderQueer to your autojoin list... |
| 01:46 |
|
sfisque |
lol, you are correct |
| 01:47 |
|
sfisque |
fixed |
| 01:48 |
|
sfisque |
so non-java enterprise question. i'm thinking i want to move from PK based ssh access to some token based access. are there any opensource token auth setups? like kernel module for the server and a CLI app to generate tokens on the client side? |
| 01:49 |
|
sfisque |
or soemthing open source that woudl integrate with kerberos |
| 01:50 |
|
pdurbin |
sfisque: https://github.com/fasrc/openauth |
| 01:51 |
|
sfisque |
Oauth was compromised late last year, i thought |
| 01:52 |
|
pdurbin |
was it? oh dear. openauth is different though |
| 01:52 |
|
pdurbin |
I used to help support it a bit: http://rc.fas.harvard.edu/openauth |
| 01:52 |
|
sfisque |
what is diff between openauth and oauth? i was under impression they were same thing... |
| 01:54 |
|
pdurbin |
nope |
| 01:54 |
|
pdurbin |
sfisque: here, this is related to openauth but is java and a little more relevant, I hope: https://github.com/mclamp/JAuth |
| 01:54 |
|
sfisque |
ok. i'll take a look at openauth and jauth |
| 01:55 |
|
pdurbin |
sfisque: cool. here's the server side component you'd need: http://code.google.com/p/google-authenticator/ |
| 01:56 |
|
sfisque |
nice and has an android token generator |
| 02:10 |
|
pdurbin |
sfisque: JAuth is a desktop token generator |
| 04:19 |
|
|
SoniEx2|2 joined ##javaee |
| 04:31 |
|
|
Guest45387 joined ##javaee |
| 07:18 |
|
|
SoniEx2|2 joined ##javaee |
| 10:18 |
|
|
SoniEx2|2 joined ##javaee |
| 10:42 |
|
|
Guest45387 joined ##javaee |
| 11:19 |
|
|
SoniEx2|2 joined ##javaee |
| 12:04 |
|
|
Guest45387 joined ##javaee |
| 12:11 |
|
|
Guest45387 joined ##javaee |
| 12:17 |
|
|
Guest45387 joined ##javaee |
| 12:18 |
|
|
SoniEx2|2 joined ##javaee |
| 13:19 |
|
|
SoniEx2|2 joined ##javaee |
| 13:23 |
|
|
Quest joined ##javaee |
| 13:33 |
|
|
Guest58349 joined ##javaee |
| 13:38 |
|
|
Guest58349 joined ##javaee |
| 13:39 |
|
|
Naros joined ##javaee |
| 14:18 |
|
|
SoniEx2|2 joined ##javaee |
| 15:15 |
|
|
kobain joined ##javaee |
| 15:19 |
|
|
SoniEx2|2 joined ##javaee |
| 16:42 |
|
|
sfisque joined ##javaee |
| 16:46 |
|
Naros |
sfisque: are you familiar with Hibernate at all ? |
| 16:47 |
|
Naros |
IIRC, the use of @OrderBy from the JPA specification requires object properties only ( e.g. you can't specify something like "LEFT(REPLICATE(N' ',40) + orderByField,40)" ) |
| 16:47 |
|
Naros |
but I'm trying to find out whether Hibernate extends JPA by allowing this. |
| 16:48 |
|
Naros |
problem is if I have a NVARCHAR(40) column that contains say the order numbers which is the sort key, order 1000 would show up before 999 in the query because JPA's order by using the DB ordering which does character-by-character comparison, regardless of length. |
| 16:49 |
|
Naros |
1 before 9, thus kicks 1000 before 999. |
| 16:49 |
|
sfisque |
you're not going to get number sorting with an nvarchar. if you're storing numbers why isnt it Number()? |
| 16:50 |
|
Naros |
because it can be alphanumeric. |
| 16:50 |
|
sfisque |
or Decimal |
| 16:50 |
|
Naros |
PO numbers can contain dashes, and other non-numeric values. |
| 16:50 |
|
sfisque |
you'll need to do an in memory sort then |
| 16:50 |
|
sfisque |
and have a custom comparitor |
| 16:50 |
|
Naros |
In memory comparison is not an option. |
| 16:50 |
|
sfisque |
or use a stored procedure |
| 16:50 |
|
Naros |
that isn't an option either. |
| 16:51 |
|
sfisque |
u using oracle |
| 16:51 |
|
sfisque |
? |
| 16:51 |
|
Naros |
No SQL Server. |
| 16:51 |
|
Naros |
2008 |
| 16:51 |
|
Naros |
I don't recall this being an issue on Oracle 8i IIRC. |
| 16:51 |
|
sfisque |
you can create a new column that is "derived" and maybe have a custom index on it? |
| 16:52 |
|
sfisque |
i know in oracle you can specify SQL in the index. not sure if sqls supports that |
| 16:52 |
|
Naros |
But we were able to apply some data manipulation when we were on Oracle 8i because we controlled all the input data. Under this version of the application, we no longer control the input data and it can vary from source system to source system. |
| 16:52 |
|
sfisque |
arbitrary data mean you need some custom sort mech then |
| 16:52 |
|
Naros |
aye, why I thought about using something like LEFT(REPLICATE(N' ',40) + field,40) to give every value a length of 40 |
| 16:53 |
|
Naros |
so comparison would work for both numeric & non-numeric values. |
| 16:53 |
|
Naros |
but keeping sorting at the DB tier where it's far more efficient. |
| 16:53 |
|
sfisque |
aye but i don't know sqls that well. most of my knowledge is ora and mysql with a dash of postgres |
| 16:53 |
|
Naros |
aye, but my question is more about JPA versus Hibernate. |
| 16:54 |
|
Naros |
JPA's order by requires field properties without any functions, yes? |
| 16:54 |
|
sfisque |
afaik yes |
| 16:54 |
|
Naros |
And they haven't (up to 2.0) given any alternative to this? |
| 16:54 |
|
Naros |
I know Hibernate 4.3 is going to be JPA2.1 compliant but we aren't there yet. |
| 16:55 |
|
Naros |
Still at 4.2x |
| 16:56 |
|
sfisque |
only way i see is the options i listed previously |
| 16:56 |
|
sfisque |
for now |
| 16:56 |
|
sfisque |
unless there is a custom annotation you can apply (hibernate specific) |
| 16:57 |
|
Naros |
Aye, iirc I think hibernate's customized @OrderBy is basically a sql fragment append concept. |
| 16:57 |
|
sfisque |
there's your solution then |
| 16:57 |
|
sfisque |
as long as you can absorb a proprietary annotation |
| 16:57 |
|
Naros |
Just wasn't sure if there was a JPA one before I dived into using custom Hibernate ones. |
| 16:58 |
|
Naros |
IIRC Oracle has an RPAD, LPAD functions, yes? |
| 16:59 |
|
sfisque |
yes |
| 16:59 |
|
Naros |
Contemplating whether it might be worthwhile to use a custom sql fragment that can be database agnostic in some way. |
| 16:59 |
|
Naros |
if it means writing my own annotation |
| 16:59 |
|
sfisque |
i wouldn't worry about that. if you change repos, there will be refactoring. we're going through that now with sql -> ora conversion |
| 17:00 |
|
Naros |
true enough, but I do want to try and minimize that |
| 17:00 |
|
Naros |
if possible |
| 17:00 |
|
sfisque |
aye. always good to consider. but i wouldnt "waste" too much effort because you will end up refactoring if y ou change dbms |
| 17:01 |
|
Naros |
True |
| 17:01 |
|
Naros |
ty. |
| 17:18 |
|
|
SoniEx2|2 joined ##javaee |
| 17:48 |
|
Naros |
sfisque: is it generally accepted in the domain model with class cycles since sometimes bidirectional relationships are helpful ? |
| 17:48 |
|
Naros |
package cyclic dependency is obviously bad |
| 17:57 |
|
whartung |
of course there will be cycles |
| 17:57 |
|
whartung |
dependents almost always will have references to their parents. |
| 17:58 |
|
Naros |
Aye, but outside domain entities, it's generally not acceptable either, correct? |
| 17:58 |
|
whartung |
It's either necessary or it's not, it's not really an "optional"/"acceptable"/"unacceptable" thing |
| 17:59 |
|
whartung |
you see it all the time in common data structures, things like linked lists that don't have a PREV link, that's pretty routine, same with tree. |
| 17:59 |
|
whartung |
trees |
| 18:00 |
|
whartung |
but its really up to the use case, the resources used/desired, and programmer convenience/quality of life that affects whether a reference is necessary/desired or not. |
| 18:01 |
|
Naros |
Gotcha. |
| 18:03 |
|
sfisque |
generally you want to avoid them, but if your schema has circularity, the models will too |
| 18:04 |
|
sfisque |
aye whartung but bidirectional connection isnt a true circularity |
| 18:04 |
|
sfisque |
a <-> b isnt the same as a -> b -> c -> a |
| 18:05 |
|
whartung |
true, but a lot of serializers are bone stupid enough to think it does anyway :) |
| 18:05 |
|
sfisque |
:-P |
| 18:05 |
|
sfisque |
machines… stupid machines... |
| 18:05 |
|
Naros |
lol |
| 18:05 |
|
sfisque |
do what i want, not what i tell you… damn it! |
| 18:06 |
|
Naros |
If it were that easy. |
| 18:25 |
|
whartung |
I think I'm going to make an effort to work back to Emacs instead of my IDE…just use the IDE for debugging. |
| 18:33 |
|
sfisque |
brave brave man |
| 23:23 |
|
|
philbot joined ##javaee |
| 23:23 |
|
|
Topic for ##javaee is now Java (general & all kinds of) and Java Enterprise Edition (Java EE) discussion. | logs at http://irclog.greptilian.com/javaee/today. We are a bit more friendly, you may keep silence as a response but you have no right to insult/abuse a person in any case. |
