Wildcard in flat file

Hi All,
I need to load a group of flat files into BW and am hoping I can use one infopackage to do it. I want to use the Routine functionality on the "External Data" tab in the infopackage.
I want all files that begin with the word 'pricebook'. The files will be saved on our bw server in the tmp directory.
So far in the routine I have:
p_filename = "/tmp/pricebook*.txt".
The load keeps failing with error:
Error 1 when loading external data
Message no. RSAR234
Diagnosis
Error number 1 occurred when loading external data:
1. Error when reading the file (access rights, file name, ...)
2. File size or number of records does not correspond to the data in the control file
3. Error when generating the IDoc
4. File contains invalid data (errors with an arithmetic operation or data conversion)
I know my access is ok, because when I load the files individually from the bw server, they all load ok.
I'm thinking I cannot use a * as my wildcard character in my routine, but I'm not sure.
Any suggestions?
Thanks

Hello Charla,
You can not use wildcards in the p_filename. This simply would not work because you are expected to place a valid filename there (not wildcards). Here's what you can do though:
Create an ABAP program that will check the filenames in your source directory. Once it has the list of filenames that needs to be loaded, it can then call function modules BAPI_IPAK_CREATE, BAPI_IPAK_CHANGE to change the filename specified in the infopack. It can then use FM BAPI_IPAK_START to schedule the load.
For details and documentation of what function modules (or BAPIs) are available to manage InfoPackages, you can go to transaction BAPI, browse the folder SAP Business Warehouse->Warehouse Management->Infopackage.
Hope this helps

Similar Messages

Golden Gate for flat file

hi,
I have tried with GoldenGate for Oracle/ non-Oracle databases. Now, I am trying for flat file.
What i have done so far:
1. I have downloaded Oracle "GoldenGate Application Adapters 11.1.1.0.0 for JMS and Flat File Media Pack"
2. Kept it on the same machine where Database and GG manager process exists. Port for GG mgr process 7809, flat file 7816
3. Following doc GG flat file administrators guide Page 9 --> configuration
4. Extract process on GG manager process_
edit params FFE711*
extract ffe711
userid ggs@bidb, password ggs12345
discardfile E:\GoldenGate11gMediaPack\V26071-01\dirrpt\EXTFF.dsc, purge
rmthost 10.180.182.77, mgrport 7816
rmtfile E:\GoldenGate11gMediaPack\V26071-01\dirdat\ffremote, purge, megabytes 5
add extract FFE711, EXTTRAILSOURCE ./dirdat/oo*
add rmttrail ./dirdat/pp, extract FFE711, megabytes 20*
start extract FFE711*
view report ffe711*
Oracle GoldenGate Capture for Oracle
Version 11.1.1.1 OGGCORE_11.1.1_PLATFORMS_110421.2040
Windows (optimized), Oracle 11g on Apr 22 2011 03:28:23
Copyright (C) 1995, 2011, Oracle and/or its affiliates. All rights reserved.
Starting at 2011-11-07 18:24:19
Operating System Version:
Microsoft Windows XP Professional, on x86
Version 5.1 (Build 2600: Service Pack 2)
Process id: 4628
Description:
** Running with the following parameters **
extract ffe711
userid ggs@bidb, password ********
discardfile E:\GoldenGate11gMediaPack\V26071-01\dirrpt\EXTFF.dsc, purge
rmthost 10.180.182.77, mgrport 7816
rmtfile E:\GoldenGate11gMediaPack\V26071-01\dirdat\ffremote, purge, megabytes 5
CACHEMGR virtual memory values (may have been adjusted)
CACHEBUFFERSIZE: 64K
CACHESIZE: 1G
CACHEBUFFERSIZE (soft max): 4M
CACHEPAGEOUTSIZE (normal): 4M
PROCESS VM AVAIL FROM OS (min): 1.77G
CACHESIZEMAX (strict force to disk): 1.57G
Database Version:
Oracle Database 11g Enterprise Edition Release 11.1.0.7.0 - Production
PL/SQL Release 11.1.0.7.0 - Production
CORE 11.1.0.7.0 Production
TNS for 32-bit Windows: Version 11.1.0.7.0 - Production
NLSRTL Version 11.1.0.7.0 - Production
Database Language and Character Set:
NLS_LANG environment variable specified has invalid format, default value will b
e used.
NLS_LANG environment variable not set, using default value AMERICAN_AMERICA.US7A
SCII.
NLS_LANGUAGE = "AMERICAN"
NLS_TERRITORY = "AMERICA"
NLS_CHARACTERSET = "AL32UTF8"
Warning: your NLS_LANG setting does not match database server language setting.
Please refer to user manual for more information.
2011-11-07 18:24:25 INFO OGG-01226 Socket buffer size set to 27985 (flush s
ize 27985).
2011-11-07 18:24:25 INFO OGG-01052 No recovery is required for target file
E:\GoldenGate11gMediaPack\V26071-01\dirdat\ffremote, at RBA 0 (file not opened).
2011-11-07 18:24:25 INFO OGG-01478 Output file E:\GoldenGate11gMediaPack\V2
6071-01\dirdat\ffremote is using format RELEASE 10.4/11.1.
** Run Time Messages **
5. on Flat file GGSCI prompt-->_*
edit params FFR711*
extract ffr711
CUSEREXIT E:\GoldenGate11gMediaPack\GGFlatFile\V22262-01\flatfilewriter.dll CUSEREXIT passthru includeupdatebefores, params "E:\GoldenGate11gMediaPack\GGFlatFile\V22262-01\sample-dirprm\ffwriter.properties"
SOURCEDEFS E:\GoldenGate11gMediaPack\V26071-01\dirdef\vikstkFF.def
table ggs.vikstk;
add extract ffr711, exttrailsource ./dirdat/pp*
start extract ffr711*
view report ffr711*
Oracle GoldenGate Capture
Version 11.1.1.0.0 Build 078
Windows (optimized), Generic on Jul 28 2010 19:05:07
Copyright (C) 1995, 2010, Oracle and/or its affiliates. All rights reserved.
Starting at 2011-11-07 18:21:31
Operating System Version:
Microsoft Windows XP Professional, on x86
Version 5.1 (Build 2600: Service Pack 2)
Process id: 5008
Description:
** Running with the following parameters **
extract ffr711
CUSEREXIT E:\GoldenGate11gMediaPack\GGFlatFile\V22262-01\flatfilewriter.dll CUSE
REXIT passthru includeupdatebefores, params "E:\GoldenGate11gMediaPack\GGFlatFil
e\V22262-01\sample-dirprm\ffwriter.properties"
E:\GoldenGate11gMediaPack\GGFlatFile\V22262-01\ggs_Windows_x86_Generic_32bit_v11
_1_1_0_0_078\extract.exe running with user exit library E:\GoldenGate11gMediaPac
k\GGFlatFile\V22262-01\flatfilewriter.dll, compatiblity level (2) is current.
SOURCEDEFS E:\GoldenGate11gMediaPack\V26071-01\dirdef\vikstkFF.def
table ggs.vikstk;
CACHEMGR virtual memory values (may have been adjusted)
CACHEBUFFERSIZE: 64K
CACHESIZE: 1G
CACHEBUFFERSIZE (soft max): 4M
CACHEPAGEOUTSIZE (normal): 4M
PROCESS VM AVAIL FROM OS (min): 1.87G
CACHESIZEMAX (strict force to disk): 1.64G
Started Oracle GoldenGate for Flat File
Version 11.1.1.0.0
** Run Time Messages **
Problem I am facing_
I am not sure where to find the generated flat file,
even the reports are showing there is no data at manager process
I am expecting replicat instead of extract at Flatfile FFR711.prm
I have done this much what to do give me some pointers.....
Thanks,
Vikas

Ok, I haven't run your example, but here are some suggestions.
Vikas Panwar wrote:
extract ffe711
userid ggs@bidb, password ggs12345
discardfile E:\GoldenGate11gMediaPack\V26071-01\dirrpt\EXTFF.dsc, purge
rmthost 10.180.182.77, mgrport 7816
rmtfile E:\GoldenGate11gMediaPack\V26071-01\dirdat\ffremote, purge, megabytes 5
ggsci> add extract FFE711, EXTTRAILSOURCE ./dirdat/oo
ggsci> add rmttrail ./dirdat/pp, extract FFE711, megabytes 20
ggsci> start extract FFE711
You of course need data captured from somewhere to test with. You could capture changes directly from a database and write those to a trail, and use that as a source for the flat-file writer; or, if you have existing trail data, you can just use that (I often test with old trails, with known data).
In your example, you are using a data pump that is doing nothing more than pumping trails to a remote host. That's fine, if that's what you want to do. (It's actually quite common in real implementations.) But if you want to actually capture changes from the database, then change "add extract ... extTrailSource" to be "add extract ... tranlog". I'll assume you want to use the simple data pump to send trail data to the remote host. And I will assume that some other database capture process is creating the trail dirdat/oo
Also... with your pump "FFE711", you can create either a local or remote trial, that's fine. But don't use a rmtfile (or extfile). You should create a trail, either a "rmttrail" or "exttrail". The flat-file adapter will read that (binary) trail, and generate text files. Trails automatically roll-over, the "extfile/rmtfile" do not (but they do have the same internal GG binary log format). (You can use a 'maxfiles' to force them to rollover, but that's beside the point.)
Also, <ul>
<li> don't forget your "table" statements... or else no data will be processed!! You can wildcard tables, but not schemata.
<li> there is no reason that anything would be discarded in a pump.
<li> although a matter of choice, I don't see why people use absolute paths for reports and discard files. Full paths to data and def files make sense if they are on the SAN/NAS, but then I'd use symlinks from dirdat to the storage directory (on Unix/Linux)
<li> both windows and unix can use forward "/" slashes. Makes examples platform-independent (another reason for relative paths)
<li> your trails really should be much larger than 5MB for better performance (e.g,. 100MB)
<li> you probably should use a source-defs file, intead of a dblogin for metadata. Trail data is by its very nature historical, and using "userid...password" in the prm file inherently gets metadata from "right now". The file-writer doesn't handle DDL changes automatically.
</ul>
So you should have something more like:
Vikas Panwar wrote:
extract ffe711
sourcedefs dirdef/vikstkFF.def
rmthost 10.180.182.77, mgrport 7816
rmttrail dirdat/ff, purge, megabytes 100
table myschema.*;
table myschema2.*;
table ggs.*;For the file-writer pump:
+5. on Flat file GGSCI prompt+
extract ffr711
CUSEREXIT flatfilewriter.dll CUSEREXIT passthru includeupdatebefores, params dirprm\ffwriter.properties
SOURCEDEFS dirdef/vikstkFF.def
table myschema.*;
table ggs.*;
ggsci> add extract ffr711, exttrailsource ./dirdat/pp
ggsci> start extract ffr711
Again, use relative paths when possible (the flatfilewriter.dll is expected to be found in the GG install directory). Put the ffwriter.properties file into dirprm, just as a best-practice. In this file, ffwriter.properties, is where you define your output directory and output files. Again, make sure you have a "table" statement in there for each schema in your trails.
Problem I am facing_
I am not sure where to find the generated flat file,
even the reports are showing there is no data at manager process
I am expecting replicat instead of extract at Flatfile FFR711.prm
I have done this much what to do give me some pointers.....The generated files are defined in the ffwriter.properties file. Search for "rootdir" property, e.g.,
goldengate.flatfilewriter.writers=csvwriter
csvwriter.files.formatstring=output_%d_%010n
csvwriter.files.data.rootdir=dirout
csvwriter.files.control.ext=_data.control
csvwriter.files.control.rootdir=dirout
...The main problem you have is: (1) use rmttrail, not rmtfile, and (2) don't forget the "table" statement, even in a pump.
Also, for the flat-file adapter, it does run in just a "extract" data pump; no "replicat" is ever used. The replicats inherently are tied to a target database; the file-writer doesn't have any database functionality.
Hope it helps,
-m

Loading several flat files

I want to load several flat files each week with the same rules file. The files are for instance week12.txt, week13.txt etc. The files are all in the same directory, and need to be loaded each time. Is it possible to use a wildcard with ESSCMD, or MaxL to create a simple loadscript? Or does anybody know another way to load several files?

If you are using a Win OS, you could use a VBS to write a script. Call the VBS in a DOS batch file, then execute the newly created script. Note, In this example, I didn't fully build the IMPORT command. Additionally, I would add some form of error handling and logging.Const sDataPath = "c:\Files"Dim fso, oFolder, oFiles, oFile, oScriptSet fso = CreateObject("Scripting.FileSystemObject")Set oFolder = fso.GetFolder(sDataPath)Set oFiles = oFolder.FilesSet oScript = fso.CreateTextFile("c:\script.scr" ,2, False)oScript.WriteLine "LOGIN " & CHR(34) & "ASPEN" & CHR(34) & " " & CHR(34) & "ADMIN" & CHR(34) & " " & CHR(34) & "PASSWORD" & CHR(34) & ";"oScript.WriteLine "SELECT " & CHR(34) & "SAMPLE" & CHR(34) & " " & CHR(34) & "SAMPLE" & CHR(34) & ";"For Each oFile In oFiles oScript.WriteLine " IMPORT 3 " & CHR(34) & oFile.Name & CHR(34) & ";"NextoScript.WriteLine "LOGOUT;"oScript.WriteLine "EXIT;"oScript.CloseSet oFolder = NothingSet oFiles = NothingSet oScript = NothingSet fso = Nothing

Refreshing the data in a flat file

Hi
I am working on data integration.
I need to fetch data from Oracle data base and then write it to a flat file.
It is working fine now,but for the next fetch I don't need the old data to remain there in the file.The data should get refreshed and only the newly fetched data should be present.
After the data is written to the flat file can i rename the file?
I need the format of the file to be 'File_yyyyMMDDHHmmss.txt'.
My final question is how should I FTP this to the target?
Please help me on this as soon as possible since this is needed in an urgent part of the delivery.

All you ask is achievable:
1) The IKM SQL to file has a TRUNCATE option, which will if set to YES, will start from a clean file.
2) You could rename the file after writing it, but why not just write it with that name? If you set the resource name to be a variable, (e.g. #MyProj.MyFilename), and be sure to set the variable in the package before executing your interface, you should be able to get the file you want. Otherwise, you can use the OdiFileMove tool in your package to rename the file.
To set the name of the variable you can use a query on the database (if you are using Oracle, something like SELECT 'File_'||TOCHAR(SYSDATE) from DUAL.)
3) ODI has ftp classes built in- you can find the doc under doc\webhelp\en\ref_jython\jyt_ex_ftp.htm
Hope this helps
Message was edited by:
CTS

Loading multiple flat file at a time.

hi experts,
I am having 15 flat files with same data structure.so how do i load all the fileswith out creating 15 info packges .
( say all the files are on the desktop.)
I had seen the option that ABAP code an be written in info package in extraction tab.
Can any one share me the abap code
Regards
Laxman.

You can write dynamic code using ABAP if your number of flat files are not fixed...but if you have to 15 flat files always just creating multiple infopackages.
For code, you have to store file names/name pattern in some table and reading it at run time. You can use function modules starting with BAPI_IPAK* to create/change/start template IP like BAPI_IPAK_START. You can set IP parameters at run time using FM RSBATCH_MAINTAIN_PAR_SETTINGS.
Kamaljeet

InfoSpoke Flat File Extract to Logical Filename

I'm trying to extract data from an ODS to a flat file. So far, I've found that the InfoSpoke must write to the application server for large data volume. Also, in order for the InfoSpoke to transport properly, I must use logical filenames. I've attempted to implement the custom class and append structure as defined in the SAP document "How To... Extract Data with OPEN HUB to a Customer Defined Logical Filename". I'm getting an error when attempting to import the included transports (custom class code). It appears to be a syntax error. Has anyone encountered this, and, if so, how did you fix it?

Hello.
I'm getting a syntax error also. I did not import the transport, but applied the notes thru the appendix. When I modified the method "GET_OBJECT_REF_INT" in class CL_RSB_DEST as below, I get a syntax error on the "create object" statement.
    when rsbo_c_desttype_int-file_applsrv.
*{   REPLACE        &$&$&$&$                                          1
*\      data: l_r_file_applsrv type ref to cl_rsb_file_applsrv.
      data: l_r_file_applsrv type ref to zcl_rsb_file_logical.
*}   REPLACE
      create object l_r_file_applsrv
        exporting i_dest    = n_dest
                  i_objvers = i_objvers
Class CL_RSB_DEST,Method GET_OBJECT_REF_INT
The obligatory parameter "I_S_VDEST" had no value assigned to it.

Dynamic file name from input payload (RFC 2 flat file)

Hi,
I have an RFC to flat file scenario. The output flat file has not an XML structure, it's just a plain text file generated with abap mapping.
In my source interface (RFC), I have a field called <FILENAME>, I want to use the value of that field to create the target file using dynamic file name. But if in variable substitution I use payload:ZRFC_NAME,1,FILENAME,1 It doesn't work because the dynamic variable substitution try to access to output payload, not the source one...
What can I do?

Hi Marshal,
You can add a extra node to your target strucutre like
FileName- Node
--FileName - Element.
do the mapping from the field filename of RFC to FileName field in u r target strucure. And use this field path at Refrence in variable subtituion.
In the Content converison add the Name & Values as below
FileName.fileldNames -- FileName
FileName.fieldFixedLengths -- 0
FileName.fixedLengthTooShortHandling -- Cut
So the extra field in u r target structure would not populate in u r target text file.
Cheers
Veera

Using wildcards (*) in sender file adapter - FTP type

Hi guys!
Dooes it work using wildcards in sender file adapter (FTP type(!) (filesystem obviously work))? I tried it and it failed. it works only for exact name..
I read some articles about FTP and the result is, that ftp work always only with 1 file, so I'm wondering, if this is possible.
Thanks for info!
Olian

Thanks for all replies..
I know of course, that * can be used, I use it in many scenarios too. But on a FileSystem. It is not working if the sender type is FTP.
*, ., *.dat, ... nothing like that works.. Just exact file name.
I am able to pick up file, if I specify it's exact name, so there should be no problem with permissions..
Please, I'd appreciate one reply with comfirmation: yes, we are using asterisk (*) , we access source files via FTP and they are processed.
Is there anybody with this experience, that it works?
Thank you!
Olian

Non English Charaters in Flat file(Note pad)

Dear Friends,
I have a flat file from presentation server which needs to be uploaded and do some validations on that file based on the data existing.
I have some Russian language words in that file. E.g: "ОЩИ В ПОДГ". If this word exists in the file i need to get some other data from a database table. But when trying to check for the word, these characters are appearing as junk characters like ####&^^.
I tried logging into RU language also, but still not working. And also tried changing the file as type Unicode, ASCII and also not worked.
Please let me know how to get identify these words to see the exact russian words.
Thanks for your help.
Srinivas.

Hello,
Is your file .txt file.If yes,while saving it gives a pop up for file name.There you have option for encoding, there select encoding as
Unicode and try to upload.
It worked for me.I tried with the data provided by you.
data:begin of it_tab occurs 0,
      val type string,
     end of it_tab.
CALL FUNCTION 'GUI_UPLOAD'
EXPORTING
    filename                      = 'C:\Documents and Settings\xxx\Desktop\SDN1.txt'
FILETYPE                      = 'ASC'
HAS_FIELD_SEPARATOR           = ' '
HEADER_LENGTH                 = 0
READ_BY_LINE                  = 'X'
DAT_MODE                      = ' '
CODEPAGE                      = ' '
IGNORE_CERR                   = ABAP_TRUE
REPLACEMENT                   = '#'
CHECK_BOM                     = ' '
VIRUS_SCAN_PROFILE            = VIRUS_SCAN_PROFILE
NO_AUTH_CHECK                 = ' '
IMPORTING
FILELENGTH                    = FILELENGTH
HEADER                        = HEADER
TABLES
    data_tab                      = it_tab
EXCEPTIONS
   FILE_OPEN_ERROR               = 1
   FILE_READ_ERROR               = 2
   NO_BATCH                      = 3
   GUI_REFUSE_FILETRANSFER       = 4
   INVALID_TYPE                  = 5
   NO_AUTHORITY                  = 6
   UNKNOWN_ERROR                 = 7
   BAD_DATA_FORMAT               = 8
   HEADER_NOT_ALLOWED            = 9
   SEPARATOR_NOT_ALLOWED         = 10
   HEADER_TOO_LONG               = 11
   UNKNOWN_DP_ERROR              = 12
   ACCESS_DENIED                 = 13
   DP_OUT_OF_MEMORY              = 14
   DISK_FULL                     = 15
   DP_TIMEOUT                    = 16
   OTHERS                        = 17
IF sy-subrc <> 0.
MESSAGE ID SY-MSGID TYPE SY-MSGTY NUMBER SY-MSGNO
        WITH SY-MSGV1 SY-MSGV2 SY-MSGV3 SY-MSGV4.
ENDIF.
Thanks.
Ramya.

Error while uploading data from a flat file to the hierarchy

Hi guys,
after i upload data from a flat file to the hierarchy, i get a error message "Please select a valid info object" am loading data using PSA, having activated all external chars still get the problem..some help on this please..
regards
Sri

there is o relation of infoobject name in flat file and infoobjet name at BW side.
please check with the object in the BW and their lengths and type of the object and check your flat file weather u have the same type there,
now check the sequence of the objects in the transfer rules and activate them.
there u go.

How can we search data in flat files

I need to search for a particular list of account numbers whose data is stored in a flat file. Each line represents data for one account number. I also need to put all data for the searched account number into a new flat file.
Can anyone help me on this??
Thanks in advance....

I m very new to java programming and dont know what
to write in the code i.e are there any available
methods to do so..or do we need to write a search
logic..That's correct you have to implemetn the logic, but it's pretty easy. Just a loop which reads each line from the file, and looks for the information.
Kaj

How can we append data in existing flat file

how can we append data to a existing flat file (Text file).

just fill the itab from data which suppose to append to text file.than read text file from GUI_UPLOAD and loop on itab and with the use of read statements you may append the data.

Can we load data in variable screen from flat file?

Hello all,
One of my user once asked me she needs to run a report for some employees, i said you can just type that in employee varaible input box.
She said I need to run the report and I need it for 120 employees, she has the employee numbers in excel sheet.
What is the best way of doing this if any ?
Thanks

This can be done from a flat file, which has list of all values.
If you are using Bex 3.5, then we have an icon in variable screen bottom right to upload values, where we can specify the path of the file.
Naveen.a

HUGE amount of data in flat file every day to external system

Hello,
i have to develop several programs to export all tables data in a flat file for external system ( EG. WEB).
I have some worries like if is possible by SAP export all KNA1 data that contains a lot of data in a flat file using the extraction:
SELECT * FROM KNA1 ITO TABLE TB_KNA1.
I need some advices about these kind of huge extractions.
I also have to extract from some tables, only the data changes, new record, and deleted records; to do this I thought of developing a program that every day extract all data from MARA and save the extraction in a custom table like MARA; the next day when the programs runs compare all data of the new extraction with the old extraction in the ZMARA table to understand the datachanges, new records or deleted record.. It's a righ approach? Can have problems with performance? Do you now other methods?
Thanks a lot!
Bye

you should not have a problem with this simple approach, transferring each row to the output file rather than reading all data into an internal table first:
open dataset <file> ...
select * from kna1 into wa_kna1
transfer wa_kna1 to <file>
endselect
close dataset <file>
Thomas

What is the diff b/w flat file and legacy system?

Hi everyone,
when v say v r working on scenario FILE to FILE? which format of file r v usually working on? and what is the diff b/w flat file and legacy system?
thanx

Hi,
<i>when v say v r working on scenario FILE to FILE? which format of file r v usually working on?</i>
>>>Many a times it will be a Flat file with CSV format,Tab delimited format, fixedlength fields.
<i>what is the diff b/w flat file and legacy system?</i>
>>>We can not differeniate like this..
Flat file may come from any systems, it may be live system or legacy system.
Legacy system- is something like old, or past one. If you talk about SAP , then older versions of SAP can be called as a legacy system.
So it may be a file system, or any system which is old version but it doesnot mean it is not in use,
Regards,
Moorthy

Wildcard in flat file

Similar Messages

Maybe you are looking for