Concept question on these technologies: Xquery, XSU & XSQL

Hi, I wonder if anyone thought about this:
I tried out a few samples on Xquery and that does meet my requirement --
"Convert my relational data into an XML".
Recently however, I learned there are also other technologies like XSU or XSQL(?).
I tend to go a bit crazy when there exists more than one answer to the solution.
Judging from some arcticles about XSU and XSQL, I could accomplish the same goal using one of these technologies.
My questions are:
1) I am not clear whether these 3 products are 'competing' products or complementing to each other.
2) Which product would be better suited ( mostly ease of use from developer's point of view) to convert a relation data onto XML and may stay around longer into the future?
I don't want to pick up a product that would go extinct in a few years.
Any idea?
Sun

We have pretty much the same question.
Of the three technologies... is XSQL a viable long-term option?
Is there a more standards based solution like SQL/XML that is now considered the future?
XSQL appeals to us because we can take our existing SQL queries and incorporate them into an xsql page and we manipulate results with XSLT if needed.
What is the best way to leverage our existing SQL and output XML from our relational db tables?
Thanks in advance for any info/clues.
Jim

Similar Messages

Concept Question about EDI and Error Processing

Hello All,
This is a concept question, I was wondering how others would approach this scenario.
Let's take this scenario. In EDI often the transmission for a purchase order comes in with an invalid material number. Normal error processing is for the EDI team to research the issue (find the correct matnr), edit the idoc and reprocess the Idoc. I would like to move that type of error processing out of the hands of the EDI folks and into the hands of the business users.
How would they correct the material number? It has to be more intuitive than editing an IDoc, it has to be an easy and intuitive user interface. This process has to also work for many types of EDI transmissions both inbound and outbound.
There would be many ways to handle this: create a report with an editable grid that would change the idoc during input and create the order with the corrections, convert the idoc to xml and use simple transformations, use an xsl report...etc. You get the point.
How would you approach this? Before I start the design I want to make sure I've given all the options due consideration...all options, webdynpro, transformations, regex, you name it...all is available, except there is no XI.
Personally I'm leaning towards an editable grid and processing buttons that would post the idoc in background...but I do that type of thing all the time and I may be in a rut and neglecting better design options.
Thanks for your input,
Greg

Paul: So it runs the transaction silently in BDC format, until the error occurs, then opens up in dialogue to allow the user to change the invalid material, and then continue on with processing.
This works when the processing function module uses BDC. But even then I think this is possibly nice from user perspective, but a nightmare from auditing perspective. I.e. correct me if I'm wrong, but I'm pretty sure there's no log indicating that the user changed the material number. Thus for anybody comparing the IDoc contents against the posted document (including change history) there's no trail that shows this change. Of course you can assume that this is what must have happened, but I personally prefer if I can track in the system what happened and have proof for that.
Reddy:
<ol>
<li>it can be run daily basis, which should select all idocs which are in status 51 with message number (..related to wrong material number). report output should inlcude :idoc number-wrong mat no- space for new material to be entered by business against wrong 1. And there should be one button for RUN.</li>
<li>After RUN, the material number should be changed to new 1 in segments and idocs should be reprocessed.</li>
<li>repeat the run until business enter right mat num.</li>
</ol>
Design seems to limited to me (takes care only of one error message). Might work if that's the main pain point and this is the only one the user is dealing with. Otherwise I'd expect pretty soon they start complaining about having to use different tools for the possible errors. I'd keep the report more general, but allow this special form of processing only for a given error message (otherwise it's a normal re-process as triggered for example via BD87).
Also, I assume that when you talk of changing the IDoc you mean that you actually keep an original copy around (like SAP does when you edit an IDoc). Often this is required from an auditing perspective. I'm not sure why you wouldn't want to check the material number before trying to process the IDoc to avoid wasting system resources (but maybe I misunderstood the step).
Anyhow, in theory you could also achieve all of this via workflow. You can add custom columns to the work item overview in the inbox, only issue here is that it doesn't scale well (so issues with larger volumes).

Java Usage Concept Questions

I have a java Command line running program. Which allow input/edit data of clients and store them into a Flat file. The problem is that i have to make a GUI, ( a thin web based client ) and using only HTML and Javascript.
The next step is the db should be frontended by servlet or JSP that interact with thin client to perform the database retrievals.
Now my concept Questions:
1. Is it possible to communicate to a Java Program using HTML and Javascript? If so, any exampler, pointers?
2. I was thinking of using HTML and Javascript to directly modify the Database. ( Since it is a flat file therefore nothing more than a txt file ) But isn't this side stepping the problem?
3. If i could modify the Flat file using HTML and Javascript only. Why would i need JSP and Servlets? And if i could do it with JSP and Servlets only why would i need the think client ( HTML bit )?
If all of above 3 are correct ( conceptialy ), what is the orginal command line java program for?
The last question is why i think i have the concept wrong. Please i am new to web programming so feel free to point out any error.

Hello,
You a little off on a few things. let me explain what each is used for to give you a better idea what each i used for.
HTML - Used to format data. Is static and can't do any File I/O and such
Javascript - Is usaually used for client side validation and some cool effects.
JSP - Is a scripting language that allows you to execute code on the server to create dynamic content and server side stuff like saving to a DB or a file
Servlets - Kinda like JSP but gives you more control over a few things. Usaually used for form processing and Servlet/Applet communication
for you application i would probably embed HTML/JavaScript for layout and validation into a JSP page for your display pages.. then build JSP pages to handle all you business logic and processing.
--Zer0C00L

Have a question on why all these technologies

Okay this may be a dumb question, but I am curious. I see you can write EJB's now looks like hibernate, J2EE, Ruby etc etc to put on the J2EE server. You also have WebDynpro and Java with HTMLB, okay. So are WebDynpro and Java with HTMLB the only applications that are supposed to connect to R3 or are people using hibernate, EJB, Ruby to connect to R3 via JCA and if so are you creating Iviews with this or making full applications to push to the portal. I just see all this noise and was wondering what its function is, as the examples all show putting your own database etc
Cheers,
Devlin
p.s. Any books or links would be greatly appreciated

JCA is a EAI from Sun organisation. It can used to connect to other systems including R3. Refer <a href="http://en.wikipedia.org/wiki/J2EE_Connector_Architecture">Wikipedia</a> which tells you simple deifination of it.For more detail architecture of JCA refer <a href="http://java.sun.com/j2ee/connector/">Sun JCA</a>
This is similar to SAP XI - an EAI which can be used to connect SAP with other systems.

A few questions about bluetooth technology and headsets...

Alright I am new to bluetooth technology and headsets, and want to do the proper research before I buy one.
1)I have seen some have 1.2 and 2.0 technology, so what is the difference and should I only look for one that offers 2.0?
2)Do they all offer a vibrating alert?
2b)Or just those really big kinds that look like hearing aides that have that little box behind your ear?...
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&ssPageName=STRK:MEWAX:IT&item=2002 46049729
3) Do they all have a multi-function button, 3way, mute, volume etc?
3b) Or are some just a single button that only answers and hangs up?
4) Do any of these charge via bluetooth?
5) I've seen some bluetooths have a micro sd chip in them... what is the point of this? Can they be used like a usb storage device?
6) Can you buy those usb to mini-usb adapters for charging and data transfer (not sure what kind of data you would transfer to a headset) separate from having to buy a combo kit that comes with it?

You might try Google and look for answers, these are pretty comprehensive questions regarding a broad sweep of headsets. You might get answers on a few models, or just Apple's, here.
If you click on each of the models in the link I gave you it goes into more detail about each one.

The website Kotaku brings me to a blank white page when i try do a search or click a link to an older post, it says done in the loading bar. Also had an error where I couldn't log in properly while trying to post this question on these forums.

Yesterday I clicked the "link Facebook account" on the website www.kotaku.com.au. Prior to that I can't remember having this issue. Not sure if its related.
My first idea was to use safe mode, which worked. so i disabled my addons in non-safe mode but it still didn't work.
The only thing that solved the problem was clearing my cookies, so for now I have the cookies of kotaku blocked. Obviously not an ideal situation.
I found these error message in console:
Error: Permission denied for <http://network.alluremedia.com.au> to call method Location.toString on <http://www.kotaku.com.au>.
Error: pageTracker is not defined
Source File: javascript:pageTracker._trackPageview('/interaction/rollover');
Line: 1
I was originally going to post this as a website error, however when trying to ask this question using Firefox I was presented with my log in name and a log out button, rather than the Ask a New Question page. Once again, it worked fine when I disabled cookies.
So now I'm thinking it may not be a single site problem.
Can't find many useful topics on the internet, needing some help.

Thanks for the flag J.K. - they contaced me and solved my problem. On this account (which I created just to ask this question), I log in with my email rather than my user name (mtalldud). On the account that I was having trouble with, they told me to use my user ID instead of my email to log in and it worked. This is a bit confusing and seems a bit inconsistent but at least I don't need this temporary account any more to get into this community and find answers!

INCLUDE File Concept Questions

Hello All!
I am extremely new to ABAP development; most of my experience is in Windows Programming.
While building a Local Class Hierarchy in an INCLUDE file, I ran into a few "quirks". Hopefully, answers to these questions will help me understand the inter-op between REPORT files and INCLUDE Files much better as well.
**NOTE** I have searched through SAP Library, this forum (admittedly, I may have missed older threads) and Googled each topic for each of these issues and not found information that satisfactorily answered my questions. I'm hoping the experts here may shed some light on my questions.
My questions are in regard to the linking between a REPORT and an INCLUDE file. When I created the INCLUDE file, I started a new instance of SE38 and went through the option dialogs to establish the file as an INCLUDE. Then, UML in hand, I started coding. Since I am new to ABAP, I run Syntax checks regularly (Finally getting used to those periods!!!). Not suprisingly, I couldn't run a Syntax check on the include. I decided to start another instance of SE38 and open the REPORT file this INCLUDE is being built for (initially). I added the INCLUDE statement and ran a Syntax Check, Success! Ever since that point however, the quirks have been appearing.
For instance, I would like to define Text Elements local to the include file; however, when I bring up the Text Elements Screen (from the INCLUDE file), the Text Elements from the REPORT are loaded in. Is it possible to define Text Elements Local to an INCLUDE file? If not, are INCLUDE files really intended to be used by multiple REPORTs? It would seem not since it looks to be tightly coupled to the REPORT I have it included in; # this could lead to naming conflicts with Text Elements as well #.
I think I may have thought of a reason for the other quirk, but would like to run it by you all to see if I am on the mark. When I have the INCLUDE file open and attempt to activate it, if a syntax error is in the active version of the report due to the include file (i.e. Tables declared more then once), it will warn me before activating. I believe this is due to the fact that SAP will try make sure existing applications will not be harmed by changes to an include file. If this is true, then INCLUDE files must be fairly closely coupled with each REPORT they are included in. Is this correct?
Thank you for your time and consideration.
I look forward to the day I know enough to begin helping each of you as well.
Steven Coolbaugh
Edited by: Steven Coolbaugh on Jan 20, 2009 6:21 PM - Clarifying a Thought - # Delimited.

Hi
U should considere the INCLUDE is just like a box can be inserted in a program (REPORT), the include can't exist without a report (the main program).
The include can be used to arrange the program: several thounsends of abap code lines can be in a program, if you split them in several includes, it'll be easier to read and manage the program.
It can group components of the same type (for example the data declaration, the routine definitions,....) into the same include, in this way it'll be easier to manage the program for the maintenaces in the future.
An include can be used by different programs, so if u need to use certain definations for many program, u can insert them in a include once and then call the include in all program u need: in this way if you update that include you'll update all program using that include.
So you'll need to insert the corrections in the include just once instead of to repeate it for every program.
Anyway all text elements used in include belongs to main program, not to the include, so u need to define it again for every program.
Max

Anyone heard of static Constructor? Pure Concept Question

Hi All,
I guess this is a little obscure but let me ask this anyway.
Has anyone heard about a "static constructor"? No I am not talking about a static block in your code. I am talking about a static method which will be called every time you make a static call on a particular class. Just like the constructor gets called every time you create an Object. The difference will be that the static code will be executed only once while the static constructor will be called every time you call a static method.
My question: Do you see a need, from the experience you have got, about such a construct being a feature of a language? I was talking to some people of a very well known company - who were telling me about this special feature - as they put it - being part of their Proprietary language. They couldn't convince me as to why this "special" feature was required. I seek help from the Community to help me think about this. I tried seeking an answer to this question on the net but to no avail.
I for one, am not going to stop my quest after posting here - but I would be grateful if any of you could help me.
Thanks for all your support.
Best Regards,
Manish

Hi All,
I am overwhelmed by the response. Thanks for spending your time thinking about this.
Some of the questions you asked were...
teknologikl : Did the people say about how is it useful?
javax.pert : No. They tried to tell me things like initializing static objects which I could have as well done in a common method which I call in the start of every static method. Basically they could not convince me about the use of a static constructor. May be they did not know too well why it was put in the first place. Because I was talking to programmers rather than Designers. But I did not want to stop at that. I wanted to tickle my gray cells and yours to find out why would someone put this as part of a language? I agree with most of your posts but instead of telling me that this is not going to be very useful ( which I already thought of), I would appreciate if you can help me think Why would someone have put it at all. You see what I am saying?
rjwr: static factory methods
javax.pert: Yup. I am aware and use static factory methods. But this is a little different. This is called automatically by the VM - just like a normal constructor - whenever a static method is called.
dubwai: This sounds pretty lame to me...
javax.pert: Sorry if I wasted your time. But please look at my response to teknologikl above.
DrClap:
javax.pert: Yup, yup. I thought just the way you are saying. I am actually trying to get in touch with some one who knows more about this proprietary language - maybe one of the designers. I would be glad to share the findings when it comes - if it is worth sharing of course. Thanks for your time though. :)
rvflannery: Security framework.
javax.pert: True. I agree with you about this being a wrapper for security purposes. Maybe this is what they thought of. One reason I see is that because they expose the APIs to third party vendors. These APIs can talk, by extending certain objects, to the sensitive areas of the database. I guess this could be one of the reasons. I will share it once I know more about it.
trejkaz: For debugging.
javax.pert: See my answer to rvflannery above.
Thanks all of you for spending your time with me. I will keep you informed whenever I get something worth sharing.
Thanks again,
Best Regards,
Manish

Forms-launched Reports - General Concept Question

With client/server forms and reports in 6i I launched reports by selecting them directly off a menu (mmx). This brought up the report's (attached) parameter and then the user ran the report.
I'm re-engineering the application in 10gR2 and the documentation I've looked at describes how you need to create a reports object under the report node within an Oracle form. I'm missing a piece of the general concept here. I've got 120 reports that currently each appear as menu items and when selected that launches the report (as described above). Should I be replacing each of those menu item calls to a report with a call to a form that is now the 'jump off' point for each report? i.e.- Do I have to create a form for each of my reports?

Hi,
I have done the similar migaration of reports...
2 Ways for this....One - You can create forms and then call the reports........I hope you can do the changes...
Two - You can add report object to the form in which the menu is attached and add few parameters ( i added five parameters bu these may be different in your case ) to the reports so that you can display the reports parameter form in the browser itsef before running the reports from menus......you can use RP2RRO library for this.

Why do some questions on these forums begin with HT or TS and some numbers?

This is a random question that I've been curious about for a while. Some of the questions' titles on these forums begin with the letters HT and end in random numbers. Some also begin with TS... What does this mean?

I'm sorry, Allan, but those are always preseeded by the Apple symbol as in

Concept Question : Migrate SQL Models

Hi Oracle SQL Data Modeler users.
I have a question. How, if it at all, do you migrate a Database from Version 1 of the model to version 2?
e.g. I've created the tutorial for a library. (Patrons; transactions; books).
Now I've generated the DDL and imported it into Oracle (or any other database).
Now I wish to change the model by adding an ISBN number and changing the domain of Person Name to 50 characters instead of 25.
All of this is fine. But Oracle generates new DDL for complete new tables.
How do I migrate the current data model from Version 1 to Version 2? Is this possible? How do other users manage this problem?
i.e. if the solution is with DDL, I'm looking for alter table statements rather than create table statements
Thanks
Martin
Edited by: mac24nz on Apr 11, 2011 3:07 AM

Hi Martin,
I suggest you look at these messages in the forum:
Generate DDL - change is a new column and I want to generate a alter table
Create 'alter tables ...' etc from Data Modeller
David

Simple java2D concept question

Dear all,
I am trying to write a chess game, but I am new to java2d. I really need some help on the java2d concept. Here is what I plan
1. A Main Class
2. A Chessboard Class
3. A Chess Class
I need to create a chessboard at the begin of the game, then add chess during the game
I don't think the following is right, but have no other idea to do in other way
1st --- use the main class to contain the chessboard and chess objects
2nd --- use JFrame in the chessboard class to contain the chessboard (then paint it in the chessboard class), use JFrame in the chess class to contain the chess image (then paint the chess in the chess class)
3rd --- set the location and the visibility of the chess on the chessboard in the main class
Do I have concept correct?
If so, how to put the chess image on the chessboard?

marco_wu wrote:
Dear all,
I am trying to write a chess game, but I am new to java2d. I really need some help on the java2d concept. Here is what I plan
1. A Main Class
2. A Chessboard Class
3. A Chess Class
1st --- use the main class to contain the chessboard and chess objectsWhat does this mean? If you mean to hold a reference to these then yes, that may be appropriate.
2nd --- use JFrame in the chessboard class to contain the chessboard (then paint it in the chessboard class), use JFrame in the chess class to contain the chess image (then paint the chess in the chess class)
No, do not display directly to a JFrame: use a JPanel.
3rd --- set the location and the visibility of the chess on the chessboard in the main classNo, do not think of your "Main Class" as the do it all area: make this a control area and provide the appropriate steps to get the game setup, but be sure to give an "initialize entity" to allow a new game to start without running through the "Main" again.
You need to have board Tiles--which make up the 2D array of your game board. Each Tile needs to be able to contain a reference to a chess piece. Each piece needs to be a Chess Piece Class. Each piece needs to have it's movements defined there in.
When you initialize you will need to have the board, an array of white chess pieces, and array of black chess pieces.
Remember games like this are basically animations upon request, so plan for that, you need to looking in to SWING graphics using the javax.swing.Timer for animation.

Java application- database related concept question

hi Friends,
I have 2 concept related questions,can someone please help me on how to implement them.I am making a java-based VOTING application.First thing,I am making a generic application,so I
don't know how many contestants would be there in each contest.
So,how can I design my database table??Should I try to create a database table on the fly,depending upon how many contestant are there to participitate??
Secondly,if somebody sends a vote like "contest_name contestant_name",I want to increment the counter of that particular person by one.So,should i have a separate database table,with the
name of all the contestant and their repective counters.What if somebody sends a wrong contestant name,do i need to query the database for that too,or i can store beforehand all the contestant names in the session or somewhere else.........please help
Thanks

Layer 4.

Concept question

Hello All and Happy Friday,
I have a quick question from a newbie. I am running my 11.1.0.7 db on a Redhat5 host in ARCHIVE LOG MODE. I have my archived redos going into two different
locations. u02/oradata/CUST/ and s/22/san/flash_recovery_area/CUST. My s22 is getting kind of full, as a matter of fact, DB control tell me its 85% full. I only have 34 days of archive redos in the folder. Can I take the last 20 days and just copy them into another mountpoint with out disturbing the DB operation? We are doing on-line RMAN backups every night.
Second Question: We enabled this db host approximately on Mar 1st. On Mar 1st, we told RMAN to do a full back (use as base of incremental) and since then, it's been doing incrimentals every night (approx 370mb per night). So with that said, i only have 1 full backup. I don't see any harm in offloading or copying the archive redos to tape or another location.
Appreciat your help..
Rob

ronthe11gdba wrote:
Hello All and Happy Friday,
I have a quick question from a newbie. I am running my 11.1.0.7 db on a Redhat5 host in ARCHIVE LOG MODE. I have my archived redos going into two different
locations. u02/oradata/CUST/ and s/22/san/flash_recovery_area/CUST. My s22 is getting kind of full, as a matter of fact, DB control tell me its 85% full. I only have 34 days of archive redos in the folder. Can I take the last 20 days and just copy them into another mountpoint with out disturbing the DB operation? We are doing on-line RMAN backups every night.
Second Question: We enabled this db host approximately on Mar 1st. On Mar 1st, we told RMAN to do a full back (use as base of incremental) and since then, it's been doing incrimentals every night (approx 370mb per night). So with that said, i only have 1 full backup. I don't see any harm in offloading or copying the archive redos to tape or another location.
Appreciat your help..
RobYou can (should) use rman to back up the archive log files and perform housekeeping on them. Here's what I do:
run {
   backup incremental level 1 cumulative database tag='bkup_incr';
   backup archivelog all not backed up 2 times tag='bkup_arch';
   delete noprompt archivelog all backed up 2 times to device type disk;
run {
delete noprompt obsolete;
crosscheck backup;
delete noprompt expired backup;
crosscheck archivelog all;
delete noprompt expired archivelog all;
}This completely automates all backup and housekeeping of my archivelogs, even with multiple archive destinations.
Note the above is my daily incremental backup. I run that Mon-Sat. On Sun I run a script that is identical expcept the 'incremental level' is zero instead of one. You really should take a periodic full (incremental 0) backup. It will make recovery much easier. Remember, you have to keep all archivelogs (in place or in a backup) back to at least your most recent full backup.

Simply question: Are these two query same?

Question 1:
There is this query from someone: (I'm not sure how Join ON works)
SELECT NULL AS order_id
,line_id
,o.item_id
,o.customer_id
INTO v_line_table
FROM xx_items_v i,
JOIN order_dtl o ON i.item_id = o.item_id
WHERE o.que_id = 3380;
Is this same as below?
SELECT ORDER_ID , line_id , o.item_id, o.customer_id
into v_line_table
FROM xx_items_v i, order_dtl o
where i.item_id = o.item_id (+)
and o.que_id = 3380;
Question 2:
I want to retrieve ORDER_ID information from a 3rd table (order_header.order_id). The key between order_dtl & order_header are que_id. How can I add to the first query above?

No. Your first query will never work it has a comma between the first table and the join key word remove it then...
remove the (+) from you second piece of code and it will be the same as the amended first piece of code, or turn the upper piece of code into an outer join by specifying whether it is a left or a right outer join.
In your case it's a left join:
select [column list]
from table1 i
left join table2 o on i.key = o.key
where o.column = 1234;However, the where clause turns this query back into an inner join becuase the right side of the join must exist. You can either allow o.column to be null in the where clause, or move the predicate up into the join. The two folowing queries are equivalent in their results:
select [column list]
from table1 i
left join table2 o on i.key = o.key
where o.column = 1234
or o.column is null;
select [column list]
from table1 i
left join table2 o on i.key = o.key and o.column = 1234;Message was edited by:
Sentinel

Concept question on these technologies: Xquery, XSU & XSQL

Similar Messages

Maybe you are looking for