Custom build extractor

Hello Gurus
I am working on Custom Build Extractor through functional module. I have executed the functional module and I got the re cords required.. In RSO2 , I have included functional module and extract structure and activated that. But in rsa3 , if I am giving data records per call as 1000 and display extr call as 1 I am getting same amount of records what I want but If I give data records per call as 1000 and display extr call as 10 ,I am getting around 10 times what the records I want. I dont know what may be the reason. Can any one say where I am going wrong.
Functional module code : FUNCTION ZFI_COSS_COSP.
""Local Interface:
*" IMPORTING
*"     VALUE(I_ISOURCE) TYPE SBIWA_S_INTERFACE-ISOURCE OPTIONAL
*"     VALUE(I_REQUNR) TYPE SBIWA_S_INTERFACE-REQUNR
*"     VALUE(I_MAXSIZE) TYPE SBIWA_S_INTERFACE-MAXSIZE DEFAULT 1000
*"     VALUE(I_INITFLAG) TYPE SBIWA_S_INTERFACE-INITFLAG OPTIONAL
*"     VALUE(I_UPDMODE) TYPE SBIWA_S_INTERFACE-UPDMODE OPTIONAL
*"     VALUE(I_DATAPAKID) TYPE SBIWA_S_INTERFACE-DATAPAKID DEFAULT
*"       50000
*"     VALUE(I_RLOGSYS) TYPE SRSC_S_INTERFACE-RLOGSYS DEFAULT SPACE
*" TABLES
*"      I_T_SELECT TYPE SBIWA_T_SELECT OPTIONAL
*"      I_T_FIELDS TYPE SBIWA_T_FIELDS OPTIONAL
*"      E_T_DATA STRUCTURE ZCOSS_COSP OPTIONAL
*" EXCEPTIONS
*"      NO_MORE_DATA
*"      ERROR_PASSED_TO_MESS_HANDLER
Tables : coss, cosp.
select objnr gjahr kstar vrgng beknz bukrs   geber    wog001    wog002    wog003   wog004 wog005
             wog006 wog007 wog008 wog009 wog010 wog011 wog012 wog013
             from coss into table E_T_DATA.
select objnr gjahr kstar vrgng   beknz   bukrs    geber
             wog001   wog002   wog003   wog004    wog005   wog006
             wog007   wog008   wog009   wog010    wog011    wog012
             wog013 from cosp appending table E_T_DATA.
ENDFUNCTION.
Extract structure = zcoss_cosp.
If I need to do any modification in the code plz let me know where and what I have to modify.
Thanks
Sam

Hi,
I think you need to change the code with passing buffer size into the code. You can insert following three line in your code before selection of the data.
S_S_IF-REQUNR    = I_REQUNR.
S_S_IF-DSOURCE   = I_DSOURCE.
S_S_IF-MAXSIZE   = I_MAXSIZE.
Check and define all import,export,changing and table parameters for function module.
Hope it helps.
Regards,
Aditya

Similar Messages

Difference between Customer Generated Extractors and Generic Extractors

Hi all,
Can anyone please tell me the difference between Customer Generated Extractors and Generic Extractors. I know that for all modules Generic extraction is possible by creating a datasource(RSO2) on top of a table, View or SAP Query in SE11. What are customer generated extractors like LIS, CO/PA and FI/SL.
What kind of extraction is used for FI/AR and FI/AP. To my understanding we have BW Content Extractors. Can anyone explain the extraction method and delta extraction for AR and AP.
Thanks,
Sabrina.

Hi,
Customer generated extractors are running on top of statistics data collection process in R/3. They extract data from statistical tables already available in R/3 and also use the mechanism already provided in R/3 to capture the delta.
For example, PO data goes to statistical table S012 (it is an LIS table) and you can build a generated extractor on top of it. The difference between such extractor and a generic one is the fact that these extractors (eg LIS) use the same mechanism of capturing delta that the statistics table was doing (for example data that goes into S012 is also captured by these generated extractors). Some system tables and structures are generated automatically to support this functionality, you do not have to develop anything for this.
Generic extractors on the other hand simply let you pull data from table/view/query etc. You have to develop code/functionality to support delta capturing for such extractors. If you built a generic extractor on top of S012 (it is a statistical table - LIS), you can pull its data to BW, but you do not have the delta capturing functionality in such case, you will have to code it yourself.
As a thumb rule, you go with the standard/generated extractors, and use generic extractors only if it is necessary to be so.
FI/AR and FI/AP extractors are part of business content (I think) and support delta functionality. To get detailed info on these, you can also look at help documentation -
http://help.sap.com/saphelp_nw04/helpdata/en/ee/cd143c5db89b00e10000000a114084/frameset.htm
cheers,

AP payment custom build program

I have a requirement to create an AP payment file format specific to a Bank (Oracle 11i). The bank accepts payment file in 2 different formats.
Format X: to make payment in its own country (Netherlands). i.e. payment for invoices where supplier bank is in Netherlands.
Format Y: to make payment in all other counties (except Netherlands).
Standard Oracle Payment SELECT/BUILD program selects Invoices for all countries (for the param given: pay batch, currency..)
If I create two payment FORMAT programs; one to pick up the Invoices for payment in Netherlands and another to pickup invoices for payment in all other counties; will it be fine ?
OR
I have to create a custom BUILD program to build a payment batch considering counties of supplier (and payment FORMAT program to create the flat file)
Help Appreciated !!
Thanks
SM
Edited by: user13294561 on 27-Sep-2010 00:22

Define 2 concurrent program to format batch records and create payment file
(or a single program with technical handling of batch type)
Define 2 automatic payment format program
Sample
NL Foreign Payments Program
NL Domestic Payments Program
Define 2 electronic payment formats,
Sample
NL Foreign Payments
NL Domestic Payments
Select each of your payment format program in this setup
Go to Internal Bank Account Setup
Payables Document
Create 2 payables document
EFT DOMESTIC,
EFT FOREIGN
Payment Format,
Select the format programs
Create Batch
Hope this helps
Bilal SARIOZ

How to update apps in a ZTE Open (with FFOS 2.0 custom build)?

I have a ZTE Open I regularly update trough custom builds. In particular, I updated it to the 2.0 version by:
BRANCH=v2.0 VARIANT=user ./config.sh inari
BRANCH=v2.0 VARIANT=user ./build.sh
The operation has apparently been successful, and by the test apps being missing I guess that I really built and flashed the user variant.
Now, I do not expect system updates to work, as I heard that the inari update channel has been shut down.
Though, I would expect it to at least automatically update the third-part apps from the Market store. This does not seem to happen: I have outdated apps (Loqui IM v3.1, in particular) but the "Check now" button from the Settings has no effect (it says "Checking for updates", but nothing else happens).
Moreover, I do not remember of having ever received an update with previous build, so I guess it may be not a problem with this particular build.
Is there something I can do about it?

Are other ZTE Open users with a custom build experiencing the same problem? Do you receive any app update at all?
I can't tell whether there is a problem in my build, in some setting or in the device.

Custom delta extractor: All data deleted in source table in R/3

Hi everyone,
I have made a custom delta extractor from R/3 to a BW system. The setup is the following:
The source table in R/3 holds a timestamp, which is used for the delta. The data is afterwards loaded to a DSO in the BW system. The extractor works as expected with delta capability. Furthermore if I delete a record in the source table, this is not transmitted to the DSO, which is also as expected.
The issue is this however: If we delete all data in the source table, then on the next load there is a request showing 1 record transfered to the DSO. This request does, however, not show up in the PSA, and afterwards all data fields in the DSO is set to initial.
Does anyone know why this happens?
Thank you in advance.
Philip R. Jarnhus

Hi Philip,
As you have used generic extractor I am not sure how the ROCANCEL will work but you can check the below link for more information,
[0RECORDMODE;
Regards,
Durgesh.

15-inch MacBook Pro w/ Retina Custom Build Delay

On 11/25 I ordered a custom build 15-inch MacBook Pro with Retina display 2.6GHz and 1TB PCIe-based Flash Storage.
Item still has not been shipped any idea as to the hold up? Is it the Flash storage?
Be intereseted in hearing from others about their backorder situation.
Thanks

Thanks nbar, wish I could it was ordered as a lease through my company's supplier SHI and their 3rd party apple company, twice removed from me. SHI says they have no way of giving me the apple order number which I find hard to believe. All the other apple accessories have arrived.

Error during Custom Build - error PRJ0019: A tool returned an error code from "Performing Custom Build Step"

Hi All,
I was trying to develop C++ web service using GShop with Visual Studio 2005.I refered a tutorial :-
http://guruce.com/blogpost/hosting-webservices-on-windows-embedded-compact-windows-ce-using-gsoap .
AS per the tutorial i proceeded. But, i stuck-up when making a custom build with Helloworld.wsdl.Here is the output :
1>------ Rebuild All started: Project: HelloWorldWebService, Configuration: Debug Pocket PC 2003 (ARMV4) ------
1>Deleting intermediate and output files for project 'HelloWorldWebService', configuration 'Debug|Pocket PC 2003 (ARMV4)'
1>Performing Custom Build Step
1>'D:\Test' is not recognized as an internal or external command,
1>operable program or batch file.
1>Project : error PRJ0019: A tool returned an error code from "Performing Custom Build Step"
1>Build log was saved at "file://d:\Test Code\HelloWorldWebService\HelloWorldWebService\Pocket PC 2003 (ARMV4)\Debug\BuildLog.htm"
1>HelloWorldWebService - 1 error(s), 0 warning(s)
========== Rebuild All: 0 succeeded, 1 failed, 0 skipped ==========
As given in the tutorial i kept my custom build command line as : -
$(SolutionDir)\gsoap-2.8\gsoap\bin\win32\wsdl2h.exe -s $(InputPath) -o $(ProjectDir)$(InputName).h
and outpouts : $(InputName).h
I kept GSHOP in the project filder :- D:\Test Code\HelloWorldWebService
Kindly help.
Jyotiranjan

Hi Jyotiranjan,
I’m glad to hear that you got it working.
Thank you for sharing your solutions
experience here. It will be very beneficial for other community members who
have similar questions.
Best regards,
Lucy
Lucy Liu [MSFT]
MSDN Community Support | Feedback to us
Get or Request Code Sample from Microsoft
Please remember to mark the replies as answers if they help and unmark them if they provide no help.

"Performing Custom Build Step"-Reg.

Dear All,
I am the new person in Plug-In creation, now i start the "Basic-Dialog" program and compile in VC++ software, but they shows the below error:
'odfrc' is not recognized as an internal or external command,
operable program or batch file.
Project : error PRJ0019: A tool returned an error code from "Performing Custom Build Step"
So please any one can help me!....
Regards
T.R.Harihara SudhaN

Dear Charles Anderson
Many Thanks for the informations and useful guide,
and please help me!
I added the below as you told paths:
"Tools | Options. Go to Projects, then VC++ Directories. In the "Show Directories for" dropdown list, select "Executable Files". Add the directory containing odfrc"...
First I clarify to you... I am using the MicroSoft Visual Studio. NET 2003, using the VC++ compiler...
I am adding the ODFRC file in the executable files, But same error shown in the Output Window. So please give me the solutions.
I am waiting for your solutions.
Thanks & Regards
T.R.Harihara SudhaN

WebDynpro accessing custom-build EJB applications

Hi there,
I have a query regarding WebDynpro accessing custom-build applications deployed on web-logic server and J2EE servers.Following is the scenario.
There is central authentication mechanism deployed in J2EE server. GUI users (in EP) have to be authenticated through this application(EJB).
In EP, this remote call would happen through the WebDynpro. After successful authentication, calls will be sent to another Weblogic server.
Too many remote calls. How about the scalability & performance factors with WebDynPro?

HI,
Webdynpro and EP are almost same except coustomisation coming to picture in EP, why do you want webdynpro to EP to web logic, Just crate your UI in dynpro convert them to Iviews in EP through EP call wenlogic server/J2ee server i hope there will not be any hindarence to scalability or perfomance.
when it comes to remote calls after all your going to call weblogic through EP but dynpro UI will be as an Iview
I hope it helps you
in case if you need further details let me know
Regards
RK

Language Support: Custom Build/Debug actions

Hi!
I'm trying to develop a custom language service, but can't find a way how to register custom build action instead of default C# compiler. I followed this two (Language Service,
Basic Project System) walkthroughs, but unfortunately couldn't find any more additional and up to date resources.
I am pretty confident that of the problems lies in the line of code bellow, but don't know how to correctly change it to register custom tool or ideally method call on build action.
<Import Project="$(MSBuildBinPath)\Microsoft.CSharp.targets" />
If someone could please point me in the right direction.
Thanks!

If your custom language doesn't use the C# or VB.NET compilers but your own compiler then you should provide your own MSBuild .targets and tasks.
MSBuild has a learning curve so I recommend to read some book or documentation. Then, examine the file Microsoft.Common.CurrentVersion.targets (C:\Program Files (x86)\MSBuild\12.0\Bin folder)
* My new blog about VSX: http://www.visualstudioextensibility.com * Twitter: https://twitter.com/VSExtensibility * MZ-Tools productivity extension for Visual Studio: http://www.mztools.com.

MSVC V5.0 IDE and CINs - custom build ??

Hi everyone,
does anyone know how to solve the "custom build" problem under NT?
When I try to build the LVS-file Visual stops at "performing custom
build step" with an error.
I've read the CIN reference manual, NI's knowledge base article
1F3ARTNV, MS's knowledge base article Q191495 and I've tried copying
various files to directories without long file names (just think about
it... Windows NT can't handle long filenames...).
I'd like to avoid postprocessing the DLL manually (so sad, so sad)!
TIA
Bernhard
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.

Ray;
For some strange reason sometimes the custom step cause a lot of troubles when implemented in VC++. I created a small VI just to implement the custom step. You can downloaded from here:
http://www.visecurity.com/tools.shtml
Look for the VI called CIN Custom Step
I hope this helps.
Regards;
Enrique
www.vartortech.com

Reload source file after custom build step

Hello,
I use a source code generator that creates c-code (i.e. hello.c) from some kind of definition file.
I want the generator to create hello.c on each build, so i have decided to put a batch (that calls the generator) into the "custom build step" of my project.
This way the generator is called before each build.
Unfortunatly CVI does not reload the changed hello.c after it has been modified/overwritten by the batch... the build is done with the previous version of hello.c. Not until the build is finished CVI recognizes that hello.c has been modified and prompts "Update Window from Disk", then I have to run a second build in order to get the changes of hello.c applied.
How can I accomplish to have only build process that first calls the generator and then compiles the project (with modified hello.c)?
Thanks

The problem happens because you have the file open in CVI. If a file is opened in CVI, then CVI uses the contents that are available in the opened window and not the contents on disk. Also, once CVI starts a build it does not monitor changes to the file on disk - consider what should happen if the file on disk changes while it is being compiled. So while the custom build step does overwrite the file on disk, CVI uses the contents that are in the source window for the compilation. To avoid this problem, make sure the file is not opened in CVI when you build - this will make CVI load and compile the file from disk after the custom build step overwrites it.

CS5 custom build with GTX 285 for long-form 4k projects

I've been researching nonstop for several days, and it's time to ask for your expert opinions. I would appreciate any help.
I'm ordering a new system for Production Premium CS5, and mostly I will use Premiere Pro CS5 to edit long-form projects (huge timelines with thousands of clips) of Red 4k footage. It's a feature film with only a few F/X shots, but we'd like to do as much F/X, sound mixing, and color correction in CS5 as possible.
Here is what I'm considering:
CUSTOM BUILD from CYBERPOWERPC (about $1950 including items not listed):
OS: Windows 7 Professional x64
CPU: Intel® Core™ i7-930 2.80 GHz 8M Intel Smart Cache LGA1366
HDD: 1TB SATA-II 3.0Gb/s 16MB Cache 7200RPM HDD (Single Hard Drive)
MOTHERBOARD: * (3-Way SLI Support) GigaByte GA-X58A-UD5 Intel X58 Chipset SLI/CrossFireX Ultra Durable™3 Mainboard DDR3/1600 ATX Mainboard w/7.1 Dobly Audio,eSATA,Dual GbLAN, USB3.0, 2 x SATA-III RAID,IEEE1394a,4 Gen2 PCIe,2 PCIe X1 &1 PCI [+132]
MEMORY: 12GB (2GBx6) DDR3/1600MHz Triple Channel Memory Module [+219] (Kingston HyperX)
SOUND: Creative Labs SB X-Fi Xtreme Audio 24-BIT PCI Sound Card [+48]
VIDEO: Nvidia GeForce GTX 285 (I will have to take out their card and install this myself for GPU acceleration)
1. Do you see any poor or strange decisions?
2. I liked that mobo because of the chipset's possibilities and because it has 1394 for old-style DV capture. Plus it has 8 SATA 2.0's, 2 SATA 3.0's, 2. eSATA 2.0's, and 2 USB 3.0's for huge/fast storage options in the future. Is there some better alternative if I'm keen to have at least one of each of these formats?
3. I'll clone their 1TB OS drive onto a smaller 250GB drive to use as the OS drive, then use their 1TB drive as part of a 2TB RAID 0 video storage. Will using the mobo's raid controller or some type of virtual RAID (like I did in Xp64) be fine, or do I need a separate RAID controller (like I had in the late 90's).
4. What do you think about that processor and RAM for what I'm doing? Any reason to prefer 2 Quad-Core's (if dual processors are even compatible with CS5)? Please let me know if the current configuration will lag with huge timelines with thousands of 4k clips.
5. Maybe I don't need the Creative sound card and could just use the mobo's sound?
6. Will it be simple for me to remove their graphics card and install an Nvidia GeForce GTX 285 myself?
7. The Nvidia GeForce GTX 285 seems like a good fit for GPU acceleration for me since I'm not expecting to have 4+ layers of videos, though I will be doing multiple effects on 1 or 2 layers. Would having a Quadro FX give any advantage over the GTX 285 for huge timelines with thousands of 4k clips? Or any advantages for any other use?
8. The GTX 285 has 2 DVI's as its only connectors (plus something that looks like S-video). Can I use one DVI for my workstation and one DVI for monitoring video in realtime on an HDMI/DVI-equipped LCD TV or professional HD video monitor. Any problems with viewing/outputting my 4k video in HD in realtime? What is the best setup to view your video on an external monitor? Back in the CS2 days I used a 1394 through a camcorder into an analog professional monitor, but that's not going to be HD, plus I heard there's a sync problem these days.
Thanks for your advice.

Thanks for your advice, everyone!
I'm gaining a lot of knowledge, but I have some followups so I'll break it down into subcategories and hopefully you can chime in on one or more topics.
To give more info about my daily use and purpose:
--smoothly edit huge amounts of RED 4k footage/clips and view the video externally in as high a resolution as possible (2k or minimum of 1k).
---no reason to have more than 3 layers visible at a time.
--the vast majority of the time only 1 video layer (with video effects) will be visible, with perhaps a handful of fades/dissolves during the entire 2-hour feature.
--I don't mind if it takes slightly more time rendering DPX files at the end, but I don't want any hiccups and never want to render during day-to-day editing and monitoring.
GRAPHICS CARD and VIDEO MONITORS
1. I'm pretty much set on the GTX 285, but I wondered if for my purposes I'd get any significant benefit from the Quadro FX 5800 (+$2000-$2600) or the GTX 480 (+$100)? Either in speed of playback or in output to external video monitors. The 5800 has 4GB RAM versus the GTX 285's 1GB, and the GTX 480 has 1.5 GB. Does this RAM matter for what I'm doing or are the cores much more important? Also, if I chose the GTX 480 then I'd have to pray that everything would be stable with a MPE/CUDA/GPU-acceleration hack, or hope Adobe comes through with full support of that card.
2. What is the 3rd output on the GTX 285? There are two DVI outputs and then something that looks like S-Video. Is it S-video out? Or an HD audio out that travels alongside the DVI output to an external video monitor?
3. Does it slow down the system to send the video at 1k or 2k to an external video monitor since you'd be using both DVI outputs simultaneously?
RED 4k PLAYBACK and RAM vs CPU
1. For day-to-day editing, I'm fine scaling to 1/4 resolution if necessary and monitoring video externally at 1k or 2k. Would my current system already achieve that smoothly?
2. From what I've gathered here and elsewhere, for better 2k or 4k playback it might be better to spend extra on CPU rather than RAM. I could spend
--$1000 more for 24GB RAM (instead of 12GB)
--$290 more for an i7 975 Extreme 3.33 GHz 8M Cache
--$686 more for an i7 980x Extreme 3.33 GHz 12M cache.
Given my needs and the current total system cost of $1950, which of these 3 would you choose? Or should I stay with the cheaper i7 930 2.8 GHz and overclock (which I've never done before).
ECBowen:
It sounds like you've used a system similar to what I'm buying to playback RED 4k on a timeline in full, 1/2, and 1/4 scale. I'd greatly appreciate if you could clarify YES/NO for which are possible to have SMOOTH PLAYBACK with your 980x/12GB RAM/GTX 285 system. I know it's a lot to ask, but it would make me feel so much better about this big purchase:
-full scale 4k of 1 layer with 0 video effects
-full scale 4k of 3 layers with 0 video effects
-full scale 4k of 1 layer with 3 video effects
-full scale 4k of 3 layers with 3 video effects (NO, right?)
-1/2 scale of 1 layer with 0 video effects
-1/2 scale of 3 layers with 0 video effects
-1/2 scale of 1 layer with 3 video effects
-1/2 scale of 3 layers with 3 video effects
-1/4 scale of 3 layers with 3 video effects
RAID CONTROLLER
1. I won't use the virtual/software controller. The GigaByte UD5 says it has RAID 0, 1, 5, 10. Does that mean it has an onboard hardware controller that I can configure in Windows 7 or using some software supplied with the mobo?

DFF in custom build reports

How can we implement dff`s features in custom build reports?
Does it works same way as the oracle bulid-in reports?

This is all documented in the Oracle Applications Developers Guide in the chapter 'Coding Oracle reports concurrent programs'. Basically you will need to call 'FND SRWINIT' in the before report trigger and 'FND SRWEXIT' in the after report trigger. There are also user exits for the flexfield processing.
Alternatively, you can just identofy the column names you require for the flexfield columns and build them into your reports in the same way as any other values.
Edited by: mike_p on Oct 14, 2009 2:10 PM

How to update with a custom build without losing all the data?

I have a ZTE Open I regularly update with custom builds. Every time I do this, I lose all of my data on the phone, such as contacts, installed apps from the Market store, settings (including wireless and email settings).
Recovering all of this data and settings every time is extremely annoying. Is there a way not to lose them in the process, or to save them and repristining them afterwards?

Thanks, Ralph!
I tried the backup_restore_profile.sh tool and apparently it worked great: I backed up my data, flashed my old build on the device (which erased all of my data) and then restored everithing, including installed apps and settings.
I will try this with a new FFOS version (like 2.1) before declaring the problem completely solved.
EDIT: I'm not going to try flashing FFOS 2.1 on the ZTE Open anymore, since I'm now a proud owner of a Flame. The backup_restore_profile.sh tool seemed to work pretty well though, so I will mark the question as solved without further tests.

Custom build extractor

Similar Messages

Maybe you are looking for