Modifying a DynaSet - OO4O and Wizards(Archived)

I need to be able to modify a DynaSet and then, once needed, commit the changes to the Database. Anyone know how? Thanks, Jeremy 

This is quite straight forward with OO4O. In fact, in-place editing is one of the best features.
It's all in the help file, check out ODynaset::Edit, ODynaset::Update & ODatabase::CommitTrans.
Regards
Adrian
LBS Limited, UK

Related

Key iteration via cursor

Hi all,
While looking at the documentation, I have been unable to find a way to iterate over the keys of a database without loading the values. Did I miss something or is this not possible? If not possible, are there any plans to change this?
The reason I ask is that BDB JE is used as the default back-end for Voldemort (http://project-voldemort.com/) and for certain admin functions, it's useful to be able to work out which nodes have which keys. It would be nice to be able to do this without loading the values.
Best,
Ismael 
Ismael, no it is not possible with JE today. It will be possible in a future release, but only if ReadUncommitted is used.
--mark                                                                                                                                                                                                                                                                            
Thanks Mark, that's good enough.
Ismael

Data Grid, how to avoid unintentional data manipulation

Hi,
Is there a option to avoid unintentional writing in the data grid ?
And if not, please implement such a toggle button for Security reasons.
Andre 
Hi,
Is there a option to avoid unintentional writing in the data grid ?1. Not using it. (use a select in the worksheet instead)
2. Revoking insert/update/delete privileges.
3. Voting for the "read only connection" request at the SQL Developer Exchange, or adding it as a new request if you think it's not related.
K. 
Ensure you have autocommit off. If you update a row and navigate away from the data grid you'll be prompted to commit your changes.
Regards
Sue 
... of course... and remember: additionally you can revert any unintentional changes issuing a rollback.
K. 
Dear friends,
thank you, but I can accept these advices at most as a work around and not as a solution (especially that with revoking IUD seems to be a joke).
Please take a short look at SQLNavigator.
That's the way it has to work.
Please make this an ER.
Andre
Message was edited by:
andreml
Message was edited by:
andreml 
As matt said, this is already requested. Please go and vote for it on exchange.
Eric 
Hi Eric,
I'm not really sure if "read only connection" means the same thing as "read only data grid". If this is one and the same in meaning so excuse me please and let me know about that.
I will then vote immediately for that.
Thank you in advance!
Andre 
If you don't like that feature request or don't feel it specific enough for your needs, add a new one. The forums are NOT for feature requests and sitting here on a user forum whining about it won't help you any.
Eric 
Someone correct me if this will not work.....
If you open a connection you could issue this:
set transaction read only
in the worksheet for that connection and it will apply to everything done on that connection. This should prohibit insert/update/deletes. Here's what I get when I try to edit the grid after doing this:
UPDATE "DEPT" SET LOC = 'asdfasdf' WHERE ROWID = 'AAASB2AAEAAADCcAAB' AND ORA_ROWSCN = '16185885'
One error saving changes to table "DEPT":
Row 2: ORA-01456: may not perform insert/delete/update operation inside a READ ONLY transaction
Then to edit data again you'd have to issue this:
set transaction read write;
-kris 
kris,
please see my message from Mar 2, 2007 11:33 AM
There is nothing to say in addition. But one.
SQL Developer is a GUI and it should be able to use it this way.
When using SQLplus "set transaction read only" is Ok with me, but not in a GUI Tool.
#EricH: You never saw me whining. I want to to promote the tool. And you?
Regards
Andre
Message was edited by:
andreml 
Andre,
I was simply pointing out the only way to do it currently as there is no way to accomplish what you are asked in the tool. So any solution today would have to be based on the server side. If your after an "help me not make a mistake" solution this transaction is the only current thing to do and maybe we could offer some simple UI to issue the statement. If you after a readonly access for a user, then that user should only have select access to a table and this is not something that should be enforced by any client tool.
If this ranks high on the xchange, we'll definitely take a look at it.
I've been meaning to blog on how you can add XXX to the context menus. So I took this and I just did a blog on how you can add this yourself:
http://krisrice.blogspot.com/2007/03/add-ability-to-make-connection-read.html
-kris 
kris ,
thank you for the blog advise.
Nevertheless I have to contest your statement.
The data grid is not equal to a connection.
I can have more then one data grids open - every one of them with another table or even n times the same table.
I could be connected as the schema owner and nevertheless I want so be sure that nothing can go wrong while I browse the data via the grid.
That means in consequence that the grid supports me by the ability to be read only.
Many tools use this functionality. Why not SQL Developer?
Be aware you have the great chance to start from the scratch with SQL Developer.
I know that things takes and needs time. And I know it is impossible to get all the useful and requested stuff at ones.
But I really hope that you have studied some of the market leading tools.
Please don't make things just different from them - make good things good as well and if possible even better.
As far as I know the Oracle crew, I’m very sure that you will get it done.
And I do hope that every single squabble from my site will help a little bit to accomplishment this goal.
Thank you very much
and warm regard
Andre 
I could be connected as the schema owner and nevertheless I want so be sure that nothing can go wrong while I browse the data via the grid.Come on! What can go wrong here?
Sorry to put it this way, but it's like you've got a bit of a inferiority complex.
If you can't trust yourself a) making changes, and b) committing, you must think you've got a split personality ;-P
Many tools use this functionality. Why not SQL Developer? As stated before, if this EVER gets into sqldev, it'll be because you've asked it on the Exchange, and other users vote for it.
Keeping insisting here IS whining, or do you think at the end Kris will say: "OK, you win, tomorrow I'll send you a patch"?
I know that things takes and needs timeNow you're making sense ;-)
Thanks for making suggestions, amongst all the users we CAN make a difference. Everyone here wants this to be the coolest tool, and you DO have the chance of making your contribution...
Regards,
K. 
Hello K.
thank you for interesting in and care about my mental health.
I'm alive and kicking.
But if I follow your logic the developers of TOAD must have a real big (mental) problem.
They are obviously that kind of paranoid that the say:
“Our data grid is protected always against data modification as long as the user does not explicitly put the rowid to the column list to undoubtedly make clear that they want to manipulate some data.”
I asked them if this is really needed, because in SQL Navigator there is just a simple button to toggle between RW and R only (What I definitely prefer) .
And the said YES – to avoid unexpected / accidentally changes.
Do you find that still strange?
If yes these guys must be more ill then the two of as can imagine.
But I’m very sure that they are alive and kicking too.
By the way - one of my questions remained unanswered:
Did you ever analysed other tools to pick the best feature of them in order to implement they into your tool?
If not, believe me you should. It is worth the effort.
Otherwise it is quite possible that you will take an unfounded amount of time till you will have reached a equivalent level.
Or do you want to be behind the others for eternity?
If you had the task to develop an new car would it ever come into your mind to build it with without a backward gear and simply to wait until someone requests something like that.
Some (useful) standard has been set before you came up.
What about a little less self satisfaction and a little bit more respect.
That’s my advice for you (also) beyond the sqldev tool .
Regards
Andre 
We read your posts, we listen to your views, we watch the market, we have a "competitive" machine... ;-)
We do want to build a product the customer base needs, something that can be used and useful, we do not necessarily want replicate what is there and so try to listen to all points of view.
Some of the concerns voiced on this forum, related to making the client tool be the enforcer of rules, are that the client tool is not the only entry point to the database. I understand accidental changes, but I'd hope my bank has more security in place in its database to avoid an "oops, I seem to have updated this accounts balance by mistake"
This request is different from the read only user. Please log it on the Exchange as a new request and you can use this forum to ask others to vote on it. We have a good list of feature requests for our future releases. We can certainly look at this.
Sue
(PM)

Import not working

Hi,
I imported through the BI Discoverer tool a business area and a workbook (.eex files). When I import, I choose to refresh the object and only take ownership if the original owner can't be found. I get the message back the import was successful. However, when I go through the front end (like a user would) and look at the file, the old report is there, not the new one I just imported that the BI tool said went fine.
Any ideas? We've tried deleting the old report, etc. and it didn't work.
Thanks,
Victoria 
Hi Victoria
First of all let me reassure you that the Import functionality of Discoverer does work. Therefore if the workbook is not being imported then there must be an issue with either the name or identifier of the workbook being imported.
Have you tried looking inside the EUL Owner's list of workbooks to see if it is there? If it is then there is something different in the name.
By the way did you try importing the workbook using the BY NAME feature as opposed to BY IDENTIFIER? I ask because almost certainly the identifiers will not be the same whereas the name should be.
Best wishes
Michael 
Hi,
No, that wasn't the problem. The problem is I have to run the eul5_id.sql script that changes the keys. If you import to a place where the sequence numbers/primary keys are the same, the import doesn't work. I don't understand why Oracle would not allow you to write over a Discoverer report and automatically change the numbers for you. People modify reports all the time, there should be an option to allow you to overwrite reports.
Victoria 
Hi Victoria
Why must you run the eul5_id script? I've done imports before and they have always worked. Perhaps I'm missing something which is quite possible :-)
One idea that comes to mind is deleting old reports before importing the new ones. Have you tried this?
Best wishes
Michael 
Yes, on my first post, deleting the file and reinstalling the file was one of the things we tried. I've not usually had problems with imports, but it seems for those reports that have been put in before, this happens.
I don't know why this works this way. I was posting here so that someone with more understanding could tell us why this happens. 
Hi Victoria
So deleting and importing works? Interesting.
How about if you import using NAME and not identifier? Does this fail too?
If it does then it looks like you will have to use the delete method but please make sure you raise this with Oracle support. By the way, which version of Discoverer are you using?
Best wishes
Michael 
When the report was deleted and then I tried to import it, it still gave me the same problem of not updating the report. So no, it didn't work.
I haven't tried importing using the name instead of identifier. I've only been taught how to import using 1 particular way, we're not supposed to do anything else. The Oracle consultant who started all this up taught us one way to import, and nothing else.
I don't even know if we have any Oracle support. The Oracle BI Discoverer version I have on my PC is 10.1.2.1 and the client version is 10.1.2.48.18. Thanks for all your comments - appreciate the help.
Victoria 
Hi Victoria
Oh dear, it sounds like you may have not had the best advice. Within Discoverer there's nearly always more than one way of doing something. When you import matching on identifiers in different databases you may well have a hard time matching up those identifiers, unless that is you take control of them yourself which is so not possible with worksheets.
In all Discoverer systems there is only one sequence number generator as you will see if you run this SQL:
SELECT SEQ_NAME FROM EUL5_SEQUENCES;
which will more than likely return EUL5_ID_SEQ.
You can then take that value and insert it into a statement like this:
SELECT EUL5_ID_SEQ.NEXTVAL FROM DUAL;
which will tell you the next sequence number that will be used.
I can almost guarantee that if you run this on both EULs that the answer will be different which means that someone along the way the same objects have gotten different identifiers.
Your best bet for getting this to work is to match on names, especially for workbooks. Give it a try and let me know what happens.
It really sounds like you may need to get some proper Discoverer training / mentoring in how to use it effectively. There are many good training organizations around, ours being one of them, and I'm not trying to tout for business. However, should you wish to ask for help just drop me an email.
Best wishes
Michael

Schema compare tool seems to do nothing - at least not to compare schemas.

Hello,
(Version: complete zip-download from today)
the Schema compare tool seems to do nothing - at least not to compare schemas.
I can choose a source an a target schema, but when I press the action button (in german the descripton is "Anwenden") nothing happens.
For every single click on that button a windows appears for the fraction of a second with the header text "Performing Diff".
And that's all ...
Regards
Andre
**************
hmm...,
OK, seems I have to reword my first posting:
Did anyone ever managed the Schema-Diff-Tool to compare Schema(s)/Object(s) in the past?
Regards
Andre
Message was edited by:
andreml 
Dear dev-team,
would you like rather to spread the cover of silence about this topic.
At least this I want to know about.
A simple Yes or No would be sufficient.
Thank you.
Regards
Andre 
The feature might or might not work. Some users have no problems with it, but lots of them do.
As the feature is not complete yet (the team plans fixing it in the future), I suggest you don't use it. You definitely don't want to compare dev with stage.
Regards,
K. 
Hello K.,
so it comes in the end that I have neither a "yes" nor a "no", but a "perhaps" :-)
I'm sorry, but I cannot understand the sense/meaning of your last phrase/proposition.
Would you please reword/rephrase this for me please?
Thank you.
Andre 
As it's a "perhaps", you better not use it in real-life situations. Definitely not for comparing your development database with the one your end users are using, because you might forget to create a critical component because it didn't show up in the comparison.
K. 
Hi,
what a great dryly humour. Deliciously!.
But hey, since for me the tool simply does nothing there is no danger that it can demolish something somehow somewhere.
This is at least a one positiv aspect , isn't it?
You should without fail continue in having and making fun!
Thank you very much!
Andre 
Andre,
If you do not check the check box for individual objects on the Source tab, but just check the high level "TABLES" checkbox, then nothing is selected. Which is why no report runs. There is nothing to compare. This is a known bug. This means that if you only check the high level check boxes and then move to the destination tab, nothing is effectively selected.
The work around is to expand the nodes. Then, by just by expanding the nodes, all the objects are selected. You can then move to the Destination tab and select the schema you wish to compare. To see the syntax, select the Apply button on the Results tab.
We are aware the Schema Diff is not fully functional and have many a debate about removing it from the product until it's ready. But there are those who are using it, and so it remains. We will be redoing the UI and feature support for our next release, together with Schema Copy. (See the updated Statement of Direction for more)
Regards
Sue 
Hello Sue,
that sounds pretty good ! Much better than the other post.
If anyone knows about the limits (or bugs) it's possible to drive round.
Thank you very much.
Regards
Andre

Business rules in HTML DB

Dear developer of HTML DB
HTML DB enables to represent many "Business Rules" in a so good format that I can show the edit pages of my application to business people as well.
But I do not want to give them the possibility to edit my application. So I can not give them the role "Developer". I want to give the role "Reviewer".
Is It possible in my workspace ?
Thank You 
this is not possible in the current release of htmldb. improvements along these lines are underway. please consider exporting/importing your "apps for review" into a separate workspace where your business rule reviewers can see your app w/o risking your production copy.
regards,
raj 
Thank You
It is very sad. This solution is not the best.
But I hope that this problem can easily solved in the next release. It is very important to work together with business people.
Andres 
Dear developer of HTML DB
We have so many questions as "HTML DB" is a "black box". We do not see the code, but we can see the repository.
We can change/add rules in this repository.
So, as we can see the tables and triggers of the schema "FLOW%" then we can add some additional triggers.
To solve this problem of business people
1. we can define them as developers of HTML DB
2. and restrict their updates in a trigger of the table
WWV_FLOWS (as you update this table)
as you update this table everywhere
update wwv_flows set
last_updated_on = sysdate,
last_updated_by = wwv_flow.g_user
We can add a new trigger with the following rule
if user not in ('A','B','C') then
raise_application_error(-20100,'Do not change !');
end if;
What do you think about this solution ?
I think that it solves the problem today ?
Andres 
Andres,
Although your suggestion may be a solution to your problem, your install of HTML DB would no longer be supported.
Sergio 
Dear Sergio
We are the partners. We together solve problems of our customers.
Sorry. I did not asked You question about technical support. I asked question about technical solution.
Is this solution technically correct ? Does it solve the problem ?
We do not know the architecture of HTML DB as it is not documented for us. So we need some advice. We are obsoletely not interested to make any big changes in the schema FLOW%. The system is great. But we need today a little bit more. If You can not solve the problem then we must do something.
If I did something wrong here then please write a e-mail to andres#iesoft.ee . I try to be correct in the future.
Thank You
Andres

Categories

Resources