Question on the use of hash tables

I have created a extract program that extract data from the bkpf and the bseg tables. I am extracting a lot of data which is needed for the auditors and depending on the selection criteria (company code and date range), this extract takes quite awhile to run. I had heard from another developer working in a different project that hash tables are used when dealing with a lot of data. I am not that familiar with hash tables and was wondering if the hash table approach would help with the processing time of my process.
thanks in advance for the help

this is only part of the code but this is the part when the selects and the writing of the file are. let me know if I have to post the entire program.
FORM f_get_data .
SELECT * INTO TABLE wt_bkpf
    FROM bkpf
    WHERE bukrs IN s_bukrs
    AND   belnr IN s_belnr
    AND   blart IN s_blart
    AND   bldat IN s_bldat
    AND   budat IN s_budat
    AND   bstat IN s_bstat.
IF sy-dbcnt IS INITIAL.
    MESSAGE i208(00) WITH text-001.
    STOP.
ENDIF.
SORT wt_bkpf BY bukrs belnr gjahr.
SELECT mandt bukrs belnr buzei buzid bschl koart shkzg dmbtr
         wrbtr pswbt sgtxt kostl saknr hkont dmbe2
    INTO TABLE wt_bseg
    FROM bseg
    FOR ALL ENTRIES IN wt_bkpf
    WHERE bukrs EQ wt_bkpf-bukrs
    AND   belnr EQ wt_bkpf-belnr.
ENDFORM.                    " f_get_data
FORM f_split_data .
DATA wlv_index LIKE sy-tabix.
DESCRIBE TABLE wt_bkpf LINES wv_index.
wlv_index     = 0.
wv_item_index = 1.
WHILE wlv_index LT wv_index.
    ADD 1 TO wlv_index.
    CLEAR wt_bkpf.
    READ TABLE wt_bkpf INDEX wlv_index.
    IF NOT sy-subrc IS INITIAL. EXIT. ENDIF.
    LOOP AT wt_bseg FROM wv_item_index
      WHERE bukrs EQ wt_bkpf-bukrs
      AND   belnr EQ wt_bkpf-belnr.
      wv_item_index = sy-tabix + 1.
      move wt_bkpf-bukrs to ws_bseg_hold-bukrs.
      move wt_bkpf-belnr to ws_bseg_hold-belnr.
      move wt_bkpf-gjahr to ws_bseg_hold-gjahr.
      move wt_bkpf-blart to ws_bseg_hold-blart.
      move wt_bkpf-bldat to ws_bseg_hold-bldat.
      move wt_bkpf-budat to ws_bseg_hold-budat.
      move wt_bkpf-monat to ws_bseg_hold-monat.
      move wt_bkpf-cpudt to ws_bseg_hold-cpudt.
      move wt_bkpf-cputm to ws_bseg_hold-cputm.
      move wt_bkpf-usnam to ws_bseg_hold-usnam.
      move wt_bkpf-tcode to ws_bseg_hold-tcode.
      move wt_bkpf-xblnr to ws_bseg_hold-xblnr.
      move wt_bkpf-bktxt to ws_bseg_hold-bktxt.
      move wt_bkpf-waers to ws_bseg_hold-waers.
      move wt_bkpf-bstat to ws_bseg_hold-bstat.
      move wt_bkpf-ausbk to ws_bseg_hold-ausbk.
      move wt_bseg-mandt to ws_bseg_hold-mandt.
      move wt_bseg-buzei to ws_bseg_hold-buzei.
      move wt_bseg-buzid to ws_bseg_hold-buzid.
      move wt_bseg-bschl to ws_bseg_hold-bschl.
      move wt_bseg-koart to ws_bseg_hold-koart.
      move wt_bseg-shkzg to ws_bseg_hold-shkzg.
      move wt_bseg-dmbtr to ws_bseg_hold-dmbtr.
      move wt_bseg-wrbtr to ws_bseg_hold-wrbtr.
      move wt_bseg-pswbt to ws_bseg_hold-pswbt.
      move wt_bseg-sgtxt to ws_bseg_hold-sgtxt.
      move wt_bseg-kostl to ws_bseg_hold-kostl.
      move wt_bseg-saknr to ws_bseg_hold-saknr.
      move wt_bseg-hkont to ws_bseg_hold-hkont.
      move wt_bseg-dmbe2 to ws_bseg_hold-dmbe2.
      APPEND ws_bseg_hold TO wt_bseg_output.
    ENDLOOP.
ENDWHILE.
ENDFORM.                    " f_split_data

Similar Messages

WHAT IS THE USE OF RETURN TABLE IN BW

Hi Gurus,
I would like to know what is the use of return table in bw.. can any one give me the perfect answer....
Ramesh

Hii
Imagine a situation where 1 record coming into a cube needs to be split into multiple records. Lets say you get values total values for sector in the PSA. Now you may want to partition that value based on the various sectors existing. In that case you can use a return table and split a single record into multiple records.
return table option is available in update rules
go thru these links
Return table
Return table
Message was edited by:
Sheeba Bhaskaran

What is the use of setup tables in LO extraction

Hi Friends,
What is the use of setup tables in LO extraction, what is the functionality of setup tables
Thanks in advance

hi,
to prepare data to be extracted to bw,
more details take a look Roberto's weblog
/people/sap.user72/blog/2004/12/16/logistic-cockpit-delta-mechanism--episode-one-v3-update-the-145serializer146
/people/sap.user72/blog/2004/12/23/logistic-cockpit-delta-mechanism--episode-two-v3-update-when-some-problems-can-occur
/people/sap.user72/blog/2005/01/19/logistic-cockpit-delta-mechanism--episode-three-the-new-update-methods
/people/sap.user72/blog/2005/04/19/logistic-cockpit-a-new-deal-overshadowed-by-the-old-fashioned-lis
/people/sap.user72/blog/2005/02/14/logistic-cockpit--when-you-need-more--first-option-enhance-it
hope this helps.

How to use one hash table inside another hash table

Hi everyone,
Any example of hash table inside another hash table.
Can one here help me how to write one hash table inside another with repeating keys for the first hash table.
Thanks,
kanty.

Do you mean you want the 'value' entries in a hash table to themselves be hash tables? Easy but this often indicates a design flaw.
Hashtable<String,<Hashtable<String,Value>> fred = new Hashtable<String,<Hashtable<String,Value>> ();But what do you mean by "with repeating keys for the first hash table"?
Edited by: sabre150 on Jul 2, 2010 10:11 PM
Looks like you have already handled the declaration side in your other thread. I suspect you should be writing your own beans that hold the information and these beans would then be stored in a Map. The problem I have is that your description is too vague so I can't be certain.

A Case Study Question in the Use of Stills with Final Cut Pro

Believe it or not, for the last week I have been reading endlessly about pixels, resolution, aspect ratio, Photoshop resizing, the bloody battle over the usefulness of the term "dpi" in video discussions, and generally about what one should do in order to use still images in a film edited in Final Cut Pro. And yet, though I have a PhD from an Ivy League university, and can understand a lot of things, I am sorry to say I am still at a loss.
Much of the discussions of photos on forums like this one are either quite technical or broadly philosophical. But my questions are completely practical and specific. I am not a photographer or a printer, nor am I an experienced editor. I am just a fairly smart person trying to use photos in his film. So I am going to ask some specific questions and hope for specific answers, without any philosophical waffling or technical mumbo jumbo.
I have one of those HP all in one printer/fax/scanner machines. I am told this device, while not fancy, is perfectly adequate to the task. Some of my photos I scan with this device from prints. Some of them I get from other people. (I have stopped downloading photos from the web, as the quality is nearly always poor.) I have Photoshop Essentials (not Photoshop). I am making a 90-minute HDV documentary, 16x9. The doc is mostly HDV footage with a still here and there. I want to be able to move on the photos (zoom, pan, etc.). I want, obviously, for them to look good. My sense is that the work flow is: (a) scan the photo (or obtain from some source as a digital file); alter the photo in some way in Photoshop; and (b) import the photo into your FCP project.
Question #1: If I have a 6x8 print, what settings should I scan it in with -- is it possible there really is no difference between scanning in it at 72dpi and 300dpi? Is it really all about the dimensions of the image? Isn't a photo whose dimensions are enlarged in Photoshop to make it available for panning and zooming in my sequence going to look fuzzier and more pixelated if it was originally scanned at 72dpi than at 300dpi? Also: Does it matter if I save it as a jpeg or tiff? Bottom line: What settings do I use in my scanning software when I scan a photo for this purpose?
Question #2: Once I have the image file on my hard drive, I suppose I now have to load it into Photoshop Elements and do something to it. What is it exactly I am using this program for? I'm going to take a wild guess. I use PS Elements to (1) crop the photo so that it has a 16x9 aspect ratio, and (2) resize it so that I can do pans and other Ken Burnsy moves on it in FCP. Bottom line: What exactly do I do in PSE to the photo and what settings do I use to both (a) make it available for moves and (b) not injure the quality of the photo by increasing its dimensions?
(In passing, I will say that I have done experiments that show that if I increase a photo's dimensions to, say, 4000 pixels by 2000 pixels, the photo's quality definitely and obviously suffers. I don't understand how a picture can be increased in size without losing sharpness. So, what am I doing wrong? Is this a function of it having been scanned in improperly to begin with?)
Question #3: If I am obtaining a photo from an outside source -- say, a professional photographer -- what are the specifications I ask for in both a hard print and a photo file? If the former, what dimensions do I ask for? If the latter, do I specify resolution and dimensions and file type?
These are my questions. At this point, I really don't want to know WHY any of this is the way it is. I just want someone to say: "Do this; do this; and then do this."
Thank you very much.

Studio X is absolutely correct; while DPI has no meaning in the video world, it often has significant meaning in scanning (and, of course, printing). If your scanning program only allows image size adjustments in DPI, then you'll have to adapt to that ... keeping in mind that the scanner's DPI setting directly relates to pixel resolution. And it is pixel resolution (not DPI) that will make all the difference in the world for images used in video. You can test this yourself by scanning an image at 1000 x 1200 @ 72 DPI and again at 1000 x 1200 at 300 DPI. They will look identical on TV or projection as long as the pixel resolution remains the same.
As an example, my Epson scanning software allows me to adjust image size by DPI or pixel resolution. Wanting to maintain the correct aspect ratio for the image area that I'm scanning, I adjust the DPI setting which in turn adjusts the pixel resolution automatically. But it doesn't work the other way around; if I adjust the pixel resolution settings manually, the DPI setting does not change. Your software may vary.
As Jim Cookman suggested, a scan setting of 300 DPI works well for most images ranging from 3 x 5 to 4 x 6 (typical snapshot size) in a DV Sequence. I may scan smaller images at 400 or 600 DPI and larger images (8 x 10 and up) at 150 DPI. Other size images fall in between those numbers. But I'm constantly watching the pixel resolution numbers to make sure I'm getting what I'll need to support the animation I intend to apply to the images once in FCP. Also bear in mind that I don't always scan the entire image. I only scan the area of an image that I intend to use in my program.
So the answer is that it all depends on your scanner's software and how far you plan to push into an image once in FCP. Do a few experiments ... you're a bright fellow so it shouldn't take long for you to find the formula that suits your purposes.
I haven't edited an HD Sequence using images yet, so I cannot offer definitive pixel resolutions for you there.
On to your specific questions:
1. See above. There is no one correct answer It all depends on what you intend to do with the image once it's in FCP. If you only plan to push in about 10%, you won't need much higher resolution than the equivalent of your FCP Sequence resolution. If you plan to push in further, you'll need a higher resolution to support that move.
2. I've been doing these for many years (SD Sequences only) and I've never loaded them into Photoshop first. I import the entire folder into FCP and drop it on the Timeline so I can see what the images look like on TV; unless your program is for CD-ROM or web delivery, that's where it counts. When you place them on the Timeline, FCP will automatically scale to your Sequence settings and pixel aspect ratio. If I see that some of them need adjustment using an external editor such as Photoshop, then and only then do I use that program. And by the way, if you have a vertical image and a 16 x 9 aspect ratio in FCP, how do you propose to crop it to fit? Perhaps you meant another term instead of "crop."
3. If it's a print, it won't matter what size they supply because you'll control the image size when scanning. If it's a file ... once again, the size (pixel resolution) depends on what you intend to do with the image once in FCP.
I really don't want to know WHY any of this is the way it is. I just want someone to say: "Do this; do this; and then do this."<<</div>
With that attitude, how did you ever earn a PhD from an Ivy League university? But if that is truly all you want, you might consider hiring an experienced editor to do it for you.
-DH

Questions concerning the use of Adobe Interactive Forms

Hi All,
I have two questions that I would to see if I could get an answer to.
1) Has or can one use an Adobe Interactive form on a Tablet PC to allow a user to draw, say a small diagram or map, and have that image then stored in an R/3 database?
2) I may already know the answer to this, but want to verify it anyway. Can Adobe Interactive forms be used to provide data to other applications outside of SAP using NetWeavers J2EE or .Net API?
Any help is greatly appreciated.
Thank you,
Mark

Hello!
1) As far as I know (and can see in Adobe Designer opened at my PC right now) there are no standard controls
in the library that can be used to add drawing (ink) area to the form. Maybe external are available, but unlikely.
Even more, googling shoes that Adobe itself is not eager to support ink in their Acrobat products... Pity but so.
So my answer for the first question is "NO".
2) That depends on what you mean on "providing data to other applications outside of SAP using NetWeaver J2EE and .NET".
If you are using NetWeaver J2EE than you are still using SAP.
If you want to install SAP's ADS server on J2EE server from another vendor, than it is not a good idea:
First - it should be merely legal to use SAP software like this.
Second - Adobe's standalone forms solution (Link: [Adobe LiveCycle Forms ES|http://www.adobe.com/products/livecycle/]) should be cheaper.
On the other hand, if you create a simple form in LifeCycle Designer not inside SAP, but launched standalone and
set this form to send data to let's say XML-service, than there is no problem to use any other server application / platform.
It's just have to be a web-service like: support HTTP-requests with XML inside.
In the last case still the question persists "why use SAP in this case?"
Just few ERP 2005 licenses will cost more, than entire Adobe product line, so maybe you have to change the platform?

What is the use of bridge table in obiee

If any body replay me use of bridge table with scenario
which situation you can use it.

An example of a bridge table I use frequently is w_gl_linkage_information_g . Pasted a sample query below:
SELECT APF.*
FROM w_gl_linkage_information_g GLI
,w_gl_other_f GLO
,w_ap_xact_f APF
WHERE GLI.journal_line_integration_id = GLO.integration_id
AND GLI.source_distribution_id = APF.acct_doc_id
AND GLO.etl_proc_wid = APF.etl_proc_wid;
AND GLO.integration_id = '4969904~2';

Some questions about the use of the relay

I connect Relay coil to the AC power source,but the switch always act.Is there any suitable relay for ac power ?
and when I replaced time delay relay in simulation, there was an error. I don't know how to get the right parameter setting ?
please help.
Attachments:
time_delay relay.png ‏5 KB

Hi there,
Thank you for your question. There are two ways to do this, for your origianl circuit you can connect the time delay relay in this way:
Set the frequency of the source to 1Hz and simulate, then you can see the NC and NO switch back and forth.
Second method: Instead of using a AC relay, we can use an AC coil which can be found at Electro_Mechanical-->COILS_RELAYS-->ENERGIZING_COIL_AC
In addition, we can add a normally closed (NC) contact or a normally open(NO) contact spicified for this AC coil(say K1)
Just type RELAY_CONTACT_NC or RELAY_CONTACT_NO to place the components and then place them under K1.
Change the Controlling device reference as shown in the picture below,
Now the AC relay can be constructed in this way.
For time delay, we can use TIME_DELAY_NC and TIME_DELAY_NO contact
Hope these two methods help!:-)
Best,
Chen_T
National Instruments

Question about the use of asdoc

I want to use asdoc to compile the flex SDK documentation myself. The reason is I need it in another format, and for a specific SDK (4.5.1).
I simply cannot get it to run. Out of the box, what should the command line look like if I'm in the bin the bin folder? Everything I try just keeps throwing errors.

1002099 wrote:
+"When an application completes, you need to close any connections that you have created. Failure to close a connection can cause resources not to be released by the JMS provider. Closing a connection also closes its sessions and their message producers and message consumers.+
+queueConnection.close();+
+topicConnection.close();+
+Before your application can consume messages, you must call the connection's start method; for details, see Section 3.5, "Message Consumers." If you want to stop message delivery temporarily without closing the connection, you call the stop method."+
Should I close the connection as the last paragraph say?. If so, I don't know how to do it without lose the service, because if I call "close" method on the connection and immediatelyi start it, i will lose all my consumers and producers for a little time.
You should close a connection when you have no further use for it as this will release resources.
You must start a connection if you want to receive messages from it.
Most applications don't use stop(). However if your application needs to temporarily pause delivery of messages (to the application) for some reason, call stop(). When your application wants delivery of messages (to the application) to restart, call start() again.
Nigel

Question about the use of JMS connections

Good Morning,
I'm a new user of JMS technology. I'm having some problems with the performance of my application. Specifically, i'm developing a 24x7 application. We are using MessageListener for all consumers.
Doing performance and continuity tests I detected memory leaks in certain JMS classes. After passing a profiler tool, I saw a large amount of ObjectMessage and Message instances. We are not closing the session, MessageListeners, MessageConsumers or Connetion instances because we want to reuse them all the time, and we can't close them because we lose the high availability service.
Reading, the JMS tutorial, I'm saw this paragraph in the "Connection" section:
+"When an application completes, you need to close any connections that you have created. Failure to close a connection can cause resources not to be released by the JMS provider. Closing a connection also closes its sessions and their message producers and message consumers.+
+queueConnection.close();+
+topicConnection.close();+
+Before your application can consume messages, you must call the connection's start method; for details, see Section 3.5, "Message Consumers." If you want to stop message delivery temporarily without closing the connection, you call the stop method."+
Should I close the connection as the last paragraph say?. If so, I don't know how to do it without lose the service, because if I call "close" method on the connection and immediatelyi start it, i will lose all my consumers and producers for a little time.
Any suggestion?
Thank you in advance,
Regards

1002099 wrote:
+"When an application completes, you need to close any connections that you have created. Failure to close a connection can cause resources not to be released by the JMS provider. Closing a connection also closes its sessions and their message producers and message consumers.+
+queueConnection.close();+
+topicConnection.close();+
+Before your application can consume messages, you must call the connection's start method; for details, see Section 3.5, "Message Consumers." If you want to stop message delivery temporarily without closing the connection, you call the stop method."+
Should I close the connection as the last paragraph say?. If so, I don't know how to do it without lose the service, because if I call "close" method on the connection and immediatelyi start it, i will lose all my consumers and producers for a little time.
You should close a connection when you have no further use for it as this will release resources.
You must start a connection if you want to receive messages from it.
Most applications don't use stop(). However if your application needs to temporarily pause delivery of messages (to the application) for some reason, call stop(). When your application wants delivery of messages (to the application) to restart, call start() again.
Nigel

Question about the use of the logger

I am trying to use the logger but it seems that unless i create a location named: "myview.class", i will not be able to see it? is this correct.
This code is generated automatically:
<b>private static final com.sap.tc.logging.Location logger =
com.sap.tc.logging.Location.getLocation(myView.class);</b>
And i can not change it, so if i dont have the location i can not use it.
<b>logger.errorT("myMessage"); - this will not be seen.</b>
The only way i could see my log messages was if i did this:
com.sap.tc.logging.Location.getLocation("ExistingOldLocation").errorT("mymessage");

Hi Erez,
You have to create a location like this. E.g. I have class
com.dbeisert.Hello that has logger.
public class Hello{
private static final com.sap.tc.logging.Location logger =
com.sap.tc.logging.Location.getLocation(Hello.class);
If you create a location with the name of "com.dbeisert"
everything in this package and each subpackage will be logged to this location. So just create location with the package prefix of your application and logging will work for each class under this package. It is basically like Log4J.
rgds
David

Question regarding the use of Cisco Aironet 1140 series

We are looking at deploying some of the Cisco Aironet Access Points as standalone. We are essentially setting up a hotspot. They will mainly be in conference rooms at each of our field offices so visitors attending our meetings can have internet access. Do clients have to use Cisco wireless adapters in order to connect
to any of Cisco Aironet Access Points or can they use whatever wireless card they have as long as it supports the bands available?
Thx in advance for any advice given.

Any product which displays the Wi-Fi Certified logo, including everything from Cisco, should interoperate with any other Wi-Fi Certified gear.

Question regarding the use of Built-in Bridge SIP call setup to recording device

We have an application that uses Built-in Bridge (BiB) to setup a SIP call to our recording application. Recently the topology of our network was changed and for some reason the RTP for the call is not being sent. What appears to be happenning is that the SIP messages are sent from the UCCM and once the call is established we get an end call event.
We are not sure how to troubleshoot our network to determine the root cause. Any pointers will be appreciated.
Thanks.
-Arun

Hi
Did you ever figure this out? I am having an issue with recording calls to the PSTN from 9971 SIP phones. Station to station calls seem fine, and 7945 phones running SCCP seem fine to the PSTN.
Thanks,
Aaron

A question regarding the use of multiple CSS templates

Hello,
I use Dreamweaver CS3. I tried to find information about this online, but I cannot seem to find anything at all. I was wondering if it is possible to use a completely differently designed template for a separate part of a website. From a marketing perspective, I understand that this concept may be impractical, because all websites should have consistently arranged information. However, the website I wish to change is not my own small business in any way. It is the website of the company that I work for, and it is a world-renown organization. Many university students access this website everyday. It was recently brought to my attention, that a proposed "brand-new" blogging section is going to be included in our website, and it was to have a different layout. We wish to use the same domain name, and simply change the style of the other part of the website.
I believe what one must do, is make sure that the .dwt (template) file points to a different CSS file, than the original one. Is this correct? The only reason that I have not tried this is because I am not sure what risks will be involved. It is also impossible to make a new page "from template", and then change the tinted uneditable text, so that it points to a different CSS file. Maybe you make a new CSS file and ensure it is in the same directory as the standard CSS file..? And when you click File-->New, you select the new template?
Any help or advice would be greatly appreciated! Thank you to everyone who contributes or even reads this. Thanks again for coping with the many details of this message.
Kind regards,
Mr. V

I think you already know the answer to this. You need to have two unique Templates and two external stylesheets. For this discussion let's call them Main.dwt and Main.css - for the main site sections. Sub.dwt and Sub.css for the subsite. You can copy and paste the code from one file into a new blank page, then make required changes to it and Save As "Sub..."
Link Main.dwt to Main.css. Link Sub.dwt to Sub.css.
To create a new page from Template, Click File > New > Templates Tab and select one.
Nancy O.
Alt-Web Design & Publishing
Web | Graphics | Print | Media Specialists
www.alt-web.com/
www.twitter.com/altweb
www.alt-web.blogspot.com

Use of internal table with hearerline in ABAP OO

Hi,
I have a very basic question regarding the use of Internal table with headerline in ABAP OO.
I consider the concept of Internal table with header as one of the most features of ABAP, which is not there in any of the other porgramming languages.
I accept that OO's concept is one of the most powerful and a very good concept. We should also implement OO's concept in ABAP.
But my concern is why in the process of moving to OO's the concept of Internal table with headerline is no more used.
Can any one tell me the main reason for this. Is there any technical reason for this. By this i mean anything to do with memory or effeciency.
Thanx,
Srinivas

Hi Srinivas,
here is something more which i found about the same
follow this link
https://sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/5ac31178-0701-0010-469a-b4d7fa2721ca
and search for header line.
it says:
Tables with header lines not allowed
Only tables without header lines can be declared in ABAP Objects.
Error message in ABAP Objects if the following syntax is used:
DATA itab TYPE LIKE TABLE OF ... WITH HEADER LINE.
Correct syntax:
DATA: itab TYPE LIKE TABLE OF ... ,
wa LIKE LINE OF itab.
Reason:
It depends on the statement whether the body or header line of a table is accessed. The table name should identify the table uniquely. Programs are easier to read. Tables with header lines do not improve performance.
i hope this will help you.
regards,
Kinshuk
PS mark helpful answers

Question on the use of hash tables

Similar Messages

Maybe you are looking for