Need to select a directory name.

I need a dialog that will allow me look through my computer and select a directory name only. The file name is automatically generated, I figured this part out.

Thanks for your answers, but I already found how to use a "File Dialog Function" from looking at other postings. But I guess it's not working the way I am expecting.
I attached a constant to the select mode input and set it to "existing dir" mode 3. Then attached a start path control to the start path.
Now when the dialog opens to select a "directory", I can only select a file. The directory information is in the total file path but I really don't want to select on a file. I need to select a directory and have this information show up. I don't want to see a file name in the final selection.
Any more information?

Similar Messages

I guess I need help just using this forum. What happened to the question I just asked? Where do I find the answers? And finally, why do I need to select a new name every time I need information?

The tutorial seems to assume you want to catalog everything on your camera/card. I want to be able to download my day's shooting so I can browse through them on my computer to decide what to save, what to delete, what to catalog, etc. I'm not opposed to change, but so far it seems that Lightroom has its own flow that does not fit how I work. Am I missing something?

Two things to note on this:
When you are in the Import dialog box, you can click Uncheck All and then click on the small box (upper left) for the photos you want to import.
Even easier, Lightroom won't import duplicates. Photos you have already imported will be dim. New photos will be checked. It's a good idea to also turn on "Don't Turn on Suspected Duplicates" on the right side of the dialog, under File Handling.

File Directory Name Select

Hi,
in my application i have a text field item to store the whole name of a file directory in a database table. For example 'C:\Temp\Test\'.
Now i want to make it possible for the user instead of typing manually, select the directory name analog as it's possible to select a file in an item with 'Display As' = 'File Browse ...'
Many Thanks in advance.
Best regards,
Martin

Martin,
You may want to have a look in this thread:
Using the file browser.
and in my demo application:
http://htmldb.oracle.com/pls/otn/f?p=31517:15
Denes Kubicek

FM for browsing the directory name in presentation server

Hi All,
Can somebody give me the FM for browsing the directory name (not the file name) in the presentation server so that the path comes automatically in the parameter.
Thanks in advance.
Regards,
Arun Mohan

Please see the following program. This will give you what you need.
report zrich_0001 .
parameters: p_path type localfile.
at selection-screen on value-request for p_path.
data: path_str type string.
call method cl_gui_frontend_services=>directory_browse
      exporting
         window_title    = 'Select Directory'
      changing
         selected_folder = path_str
      exceptions
         cntl_error = 1.
call method cl_gui_cfw=>flush
       exceptions
          cntl_system_error = 1
          cntl_error        = 2.
p_path = path_str.
Regards,
Rich Heilman

WebCenter Form Recognition WebVerifier directory name {Directory Path} is Invalid

Hi
I'm facing the below issue while trying to open the Web Center Form Verifier using web Verifier.
The WebVerifier directory name {Directory Path} is Invalid. please contact your system administrator.
OS : Window Server 2008 R2
Software: Oracle WebCenter Form recognition 11.1.1.8.0
find the attached error screen shot and help out with the solution.
Thanks
Sanjeev

You are getting this message because the WFR web server is trying to access the project file in a directory that is not valid on that server. I assume that path is correct on the Designer machine but the web server is on a different machine?
Assuming the project file is on a file server, you should share that folder then open the project in Designer via the UNC path to the share. Then, go to the Options menu and select the Users, Groups and Roles... option. On the Users tab of the Project Authentication dialog, ensure that the Allow Database Authentication option is checked, then click the Export to Database... button. That will create an entry in the database for that project pointing to the UNC path instead of the local path, so the web server should then be able to access it (provided you have assigned sufficient permissions on the share).
Alternatively, you could copy the project folder structure to the web server on the same path as it is on the Designer machine. This should work but will cause issues for ongoing administration of the project, as you will need to keep the copy on the web server synchronized with that on the Designer machine. I wouldn't recommend doing this.

Precompiling JSPs changes directory name

I am using the weblogic.appc compiler to precompile JSPs and I noticed that if my source folder name contains a hyphen, eg. my-test, after precompilation the class files are stored under \WEB-INF\classes\jsp_servlet\_my_45_test\ directory.
The "-" in the name is substituted by its ASCII value. Is there a way to stop this from happening other than choosing a different name for the source folder?
Thanks

cranestar wrote:
Oracle db versesion : 11g
Apex version: 4.1.0.00.32
I have created a file browse object on a page. After filling several text boxes and selecting a file the page writes a row into a table
in the database.
When a file is selected with the file browse the name looks like:
C:\OracleApex\PDM-BOM comparison app\datafiles\B-15680-49_D_SAP.CSV
However in the database the name appears as:
F1531709975/B-15680-49_D_SAP.CSV
What happened to the directory name? This causes problems because I need to use the directory and file name in a utl_file
utility.<tt>F1531709975</tt> is produced to provide a unique file identifier in the <tt>apex_application_files</tt> view, and is the session state value of the file browse control that uploaded the file. The actual filename is available in the <tt>filename</tt> column.
As has been noted in many previous threads on this topic: For security/privacy reasons recent versions of browsers by default do not send local file path information from File Browse items to the server, nor expose the file path in the control's JavaScript methods. Firefox, Safari and Chrome only provide the filename. IE6 & IE7 still yield the path in Windows format. IE8+ and Opera have adopted an irritating approach of replacing the path with a wholly imaginary "C:\fakepath\"—and this monstrosity has sadly had to be enshrined in the HTML5 spec...
Changing IE's security config setting "Include local directory path when uploading files" enables the path to be exposed in IE, but unless you're working in an intranet environment where: IE is the only browser used; it's possible to make remote changes to this setting on every desktop; and this won't break/expose anything else, then trying to achieve this is pointless.
For more information see:
http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2009-March/018980.html
http://blogs.msdn.com/ie/archive/2009/03/20/rtm-platform-changes.aspx
http://developers.whatwg.org/number-state.html#file-upload-state

Selecting a directory issue.

I need to select all of the files within a certain directory and place them in an array. Here is the code which I have written so far.
for (int i=0; i<numGroups; i++){
 System.out.print("Enter the path of the directory containing the images for group " + (i+1) + ".\nALL files must be in JPEG format. ");
 JFileChooser chooser = new JFileChooser();
 chooser.setFileSelectionMode(JFileChooser.DIRECTORIES_ONLY);
 int returnVal = chooser.showOpenDialog(null);
 File[] files = chooser.getSelectedFile().listFiles();
 for (int j = 0; j < files.length; j++){
 System.out.println(files);
It looks like it should work, but right before opening the file chooser it spits out this grimy error:
2008-06-25 10:40:07.443 java[7248] CFLog (0): CFMessagePort: bootstrap_register(): failed 1103 (0x44f), port = 0xfb03, name = 'java.ServiceProvider'
See /usr/include/servers/bootstrap_defs.h for the error codes.
2008-06-25 10:40:07.444 java[7248] CFLog (99): CFMessagePortCreateLocal(): failed to name Mach port (java.ServiceProvider)
I have no idea how to even read this error message or how to fix it.
Thanks,
Collin

You're on OS X, correct? Possibly debugging with Eclipse?
I've had strange errors running/debugging with Eclipse on OS X in the past, but not this one. I googled for "CFMessagePort: bootstrap_register()" and come up with lots of stuff saying that it's a harmless error caused when you try to launch two processes with the same name (in this case "java") on the command line. An Apple developer even explicitly says "If your app is crashing, it's for some other reason." So this probably isn't an actual problem for you. It could be how Eclipse is launching and your application (assuming you're running Eclipse...).
aquowf wrote:
Also, the list of files which the program prints is incorrect. It prints out a .DS_Store file five times and does not print out any of the jpeg files which are in the folder.Not sure about this, but please note that when using JFileChooser to select a directory, you must navigate INTO the directory you want to select. Navigating to the parent directory, clicking on the desired directory with the mouse, and clicking "OK" does NOT cut it. In DIRECTORIES_ONLY mode, JFileChooser returns the directory whose contents it is displaying. A file chooser is an unintuitive way to select a directory; if you have the time and motivation, you may want to create a simple directory chooser dialog and use that instead, something like a tree view of all directories in the file system.
As camickr said, create a shoft, self-contained, compilable example program that clearly demonstrates the multiple ".DS_Store" problem and we can help you diagnose what's wrong.

Index.html pages in local sub-directories will not load automatically when sub-directory name entered into address field, followed by a forward-slash

I am in the process of designing a website-on-a-CD. In order to make things easier for my clients, I decided to organize this site into local sub-directories with an index.html page in each one. However, when I try to enter the local sub-directory name, followed by a forward-slash, the index page does not open automatically. Instead, I get a raw directory listing that includes the index.html file, and anything else that is present within it. I would like to know if there is any way to force this page to load automatically within Firefox from a local storage medium as it would load from a web-server.

You need to use file:// as the protocol instead of C:/. The latter may never work (C:\ might).
Where is the main HTML located ?
Easiest if to store the images in a sub folder of the location because you can't go back via ../ beyond that root location for security reasons.
See:
*http://kb.mozillazine.org/Links_to_local_pages_do_not_work

"There was an error opening this document. The filename, directory name, or volume label syntax is i

"There was an error opening this document. The filename, directory name, or volume label syntax is incorrect."
I tried different versions, no avail.
I think it has to do with the password somehow, but I do not know. I can copy the files from the web folder to my desktop and they open fine (after entering the password AGAIN), but I can not open them directly form the web folder - it does not ask for the password again, just pops up the error.
Adobe's support really upset me (insert explicative here) by keeping me on the phone for 20 min, collecting all sorts of information about me (not the problem) with some lady who barely spoke English, just to tell me to go to the website. No surprise they need to charge for tech support, customer service is doing everything but "service". I hate companies that do this. I am thinking of using another PDF reader, I have heard good things about them. Check
http://en.wikipedia.org/wiki/List_of_PDF_software
Has anyone ever been able to find a fix? I can not find anything on this site.

Howdy, Moritz!
Thanks for the suggestion, but this is not an issue of filename and language conflicts, at least in my environment. I highly doubt that's the root cause of this issue.
The PDF files in my tests have all had very simple names, such as 123.pdf and test.pdf, as well as more complicated names with spaces in them. Nothing about the files contains anything outside of the English language, including the content and filenames.
From everything I've been able to determine, this is a WebDAV-related security issue that affects programs like Acrobat Reader. I know for a fact that MS's KB892211 WebDAV patch doesn't create the issue, however MS's update to that 892211 patch DOES create the issue. As I've outlined earlier in this thread, my extensive testing has proven this more than once.
At this point, I haven't been able to accurately uninstall the update to 892211 after the update has been installed. And, as far as I can tell, the update to 892211 comes down from MS as a Office and/or Windows Update component. So, everyone has it, the uninstaller doesn't work, and there's no documentation detailing what registry entries are being changed when the update to 892211 is installed.
My ultimate goal is to bring a test machine to the point just before the update to 892211 goes on, fire up the Registry, Process, and File monitors on my test box, and install the 892211 update. This should give me an accurate picture of what I need to restore in order to rid the machine of the update to 892211.
Until Microsoft officially fixes the issue, I am not totally sure how this will be resolved. I've ceased my work on the issue for now since copying the PDF file from the WebDAV share to the user's desktop gets the user what they wanted in the first place - the ability to open the PDF file!
Maybe, once some other projects slow down a bit, I might get back to creating a solution, but it's going to be a "hack" no matter which way you slice it. I'd much rather MS spend their time creating the solution. If I have to do it, my fee to MS will be quite HIGH! :*)
George B.

SharePoint 2010, Visual Studio 2010, Packaging a solution - The specified path, file name, or both are too long. The fully qualified file name must be less than 260 characters, and the directory name must be less than 248 characters.

Hi,
I have a solution that used to contain one SharePoint 2010 project. The project is named along the following lines:
<Company>.<Product>.SharePoint - let's call it Project1 for future reference. It contains a number of features which have been named according
to their purpose, some are reasonably long and the paths fairly deep. As far as I am concerned we are using sensible namespaces and these reflect our company policy of "doing things properly".
I first encountered the following error message when packaging the aforementioned SharePoint project into a wsp:
"The specified path, file name, or both are too long. The fully qualified file name must be less than 260 characters, and the directory name must be less than 248 characters."
I went through a great deal of pain in trying to rename the project, shorten feature names and namespaces etc... until I got it working. I then went about gradually
renaming everything until eventually I had what I started with, and it all worked. So I was none the wiser...not ideal, but I needed to get on and had tight delivery timelines.
Recently we wanted to add another SharePoint project so that we could move some of our core functinality out into a separate SharePoint solution - e.g. custom workflow
error logging. So we created another project in Visual Studio called:
<Company>.<Product>.SharePoint.<Subsystem> - let's call it Project2 for future reference
And this is when the error has come back and bitten me! The scenario is now as follows:
1. project1 packages and deploys successfully with long feature names and deep paths.
2. project2 does not package and has no features in it at all. The project2 name is 13 characters longer than project1
I am convinced this is a bug with Visual Studio and/or the Package MSBuild target. Why? Let me explain my findings so far:
1. By doing the following I can get project2 to package
In Visual Studio 2010 show all files of project2, delete the obj, bin, pkg, pkgobj folders.
Clean the solution
Shut down Visual Studio 2010
Open Visual Studio 2010
Rebuild the solution
Package the project2
et voila the package is generated!
This demonstrates that the package error message is in fact inaccurate and that it can create the package, it just needs a little help, since Visual Studio seems to
no longer be hanging onto something.
Clearly this is fine for a small time project, but try doing this in an environment where we use Continuous Integration, Unit Testing and automatic deployment of SharePoint
solutions on a Build Server using automated builds.
2. I have created another project3 which has a ludicrously long name, this packages fine and also has no features contained within it.
3. I have looked at the length of the path under the pkg folder for project1 and it is large in comparison to the one that is generated for project2, that is when it
does successfully package using the method outlined in 1. above. This is strange since project1 packages and project2 does not.
4. If I attempt to add project2 to my command line build using MSBuild then it fails to package and when I then open up Visual Studio and attempt to package project2
from the Visual Studio UI then it fails with the path too long error message, until I go through the steps outlined in 1. above to get it to package.
5. DebugView shows nothing useful during the build and packaging of the project.
6. The error seems to occur in
CreateSharePointProjectService target called at line 365 of
Microsoft.VisualStudio.SharePoint.targetsCurrently I am at a loss to work out why this is happening? My next task is to delete
project2 completely and recreate it and introduce it into my Visual Studio solution.
Microsoft, can you confirm whether this is a known issue and whether others have encountered this issue? Is it resolved in a hotfix?
Anybody else, can you confirm whether you have come up with a solution to this issue? When I mean a solution I mean one that does not mean that I have to rename my namespaces,
project etc... and is actually workable in a meaningful Visual Studio solution.

Hi
Yes, I thought I had fixed this my moving my solution from the usual documents to
c:\v2010\projectsOverflow\DetailedProjectTimeline
This builds ok, but when I come to package I get the lovely error:
Error 2 The specified path, file name, or both are too long. The fully qualified file name must be less than 260 characters, and the directory name must be less than 248 characters. C:\VS2010\ProjectsOverflow\DetailedProjectTimeline\VisualDetailedProjectTimelineWebPart\Features\Feature1\Feature1.feature VisualDetailedProjectTimeline
Now, the error seems to be related to
Can anyone suggest what might be causing this. Probably some path in an XML file somewhere. Here is my prime suspect!
<metaData>
<type name="VisualDetailedProjectTimelineWebPart.VisualProjectTimelineWebPart.VisualProjectTimeline, $SharePoint.Project.AssemblyFullName$" />
<importErrorMessage>$Resources:core,ImportErrorMessage;</importErrorMessage>
</metaData>
<data>
<properties>
<property name="Title" type="string">VisualProjectTimelineWebPart</property>
<property name="Description" type="string">My Visual WebPart</property>
</properties>
</data>
</webPart>
</webParts>
.... Unless I can solve this I will have to remove the project and recreate but with simple paths. Tho I will be none the wiser if I come across this again.
Daniel

Certutil -crl problems (the directory name is invalid)

Another problem for you fine experts to consider...2 tier PKI, offline Root 2008 R2, 1 Sub Ent CA in Domain1 (2008 R2) and 1 Sub Ent CA in Domain2 (2012 R2).
SubCA 1 and 2 are configured pretty much identically, however when setting up SubCA 2 I am having issues running the Certutil -CRL command to publish the CRL.
My CDP locations are configured as follows;
65:c:\WIndows\System32\CertSrv\CertEnroll\%3%8%9.crl
79:ldap://CN=%7%8,CN=CDP,CN=Public Key Services,CN=Services,%6%10
6:http://pki.domain2/CertEnrolment/%3%8%9.crl
65:file://\\pki.domain2\CertEnrolment\%3%8%9.crl
I can confirm that the base CRL publishes correctly to the CertEnroll location and LDAP correctly. But it fails trying to publish to the HTTP/File location (which is the same path).
I get the error:
CertUtil: -CRL command FAILED: 0x8007010b (WIN32/HTTP: 267 ERROR_DIRECTORY)
CertUtil: The directory name is invalid
Also the Delta CRL fails on the CertEnroll default directory as well as the file/http path with error;
Active Directory Certificate Services could not publish a Delta CRL for key 0 to the following location: file://\\pki.domain\CertEnrolment\CANAME+.crl.
Operation aborted 0x80004004 (-2147467260 E_ABORT)<o:p></o:p>
I'm pretty certain it's not a permissions issue as I've added Everyone for NTFS/share permissions to test without any change. The install was done with an Enterprise
Admin account but I'm doing all the testing now with a normal admin account (admin in the CA/server but not domain or enterprise admin).<o:p></o:p>
<o:p></o:p>
The File/HTTP location is on the CA itself (I know this is likely not best practise, but needs to be there in the short term) so not sure if the Windows firewall comes into play.
Thanks!

Hi driko,
It's not a best practise to give Everyone NTFS/share permissions!
What I suggest is you
1. Create a dedicated folder f.e. "C:\Repository" on CA and share it only with permissions to specific account (see below)
2. In CA publish CRLs to c:\WIndows\System32\CertSrv\CertEnroll\%3%8%9.crl
only and create a task in task scheduler that will be running on the dedicated account and will copy c:\WIndows\System32\CertSrv\CertEnroll\*.crl
to \\pki.domain2\Repozitory\*.crl
3. Make sure that account that is running this task on CA1 (Domain 1) has enough permissions for Repository share in Domain 2 (try running cmd as this user on
CA1 and copy files manually to \\pki.domain2\Repository\)
4. Map your http://pki.domain/CertEnrolment URL with IIS to physial C:\Repository\ folder path
Did my post help you or make you laugh? Don't forget to click the Helpful vote :) If I answered your question please mark my post as an Answer.

DR within Same Site Using Different Virtual Directory Names

We are deploying Exchange 2013 and have setup a set of 2013 servers in a different building we would like to use for DR.
We have 2 servers in our main datacenter behind a load balancer and 2 servers in a 2nd datacenter within the same DAG. So mailbox databases are already on the servers. However for the CAS role, we are looking at just setting the DR servers to
have different virtual directory names so in a DR situation, we just need to tell people to access owadr.domain.com instead of owa.domain.com. The DR servers would be behind a 2nd load balancer only configured to talk to those two servers.
Both of these datacenters are in the same AD site, and the network vLANS are stretched across each datacenter so DAG replication is already working. The primary serves are already working behind the production load balancer. Just need to make
sure we are going in the right direction for the DR setup. Our public certificate already names the names of owa.domain.com, owadr.domain.com, and autodiscover.domain.com on the cert. We are not really concerned so much with Outlook/ActiveSync
traffic in a DR situation, but would just like OWA to be a temporary solution while the main datacenter recovers. So a different virtual directory name should work right?

Hi,
From your description, I know four servers are in the same site but with different building. So you create two load balancer. And it seems you add the DAG members into Load Balancer. Could you confirm if you use WNLB?
In fact, WNLB can't be used on Exchange servers where mailbox DAGs are also being used because WNLB is incompatible with Windows failover clustering. If you're using an Exchange 2013 DAG and you want to use WNLB, you need to have the Client Access server
role and the Mailbox server role running on separate servers. See more details in the following link:
http://technet.microsoft.com/en-us/library/jj898588(v=exchg.150).aspx.For
For your environment, I have a suggestion: we can deploy two servers with only CAS role installed. One cas in one datacenter. Put them in one Load Balancer. Point OWA URL to this load balancer. So when one datacenter is down, the request will
go to anoter CAS in another datacenter. And the OWA URL is not changed..
But if you want to use two OWA URLs for each datacenter. owa.domain.com and owadr.domain.com. You can set OWA internal URL and external url on CAS servers in DR site to use owadr.domain.com. And ask users to login
https://owadr.domain.com/owa instead of
https://owa.domain.com/owa when the primary datacenter is down.
Sent By
Silver

Get Directory name in Chinese and Chinese chars to Excel

Hi,
Question:1
I use "cl_gui_frontend_services=>directory_browse"
to get the Directory name from the presentation server.
It works good for the directory names in English.
However I need to perform a similar operation when the
user has logged in "Chinese" at OS level,
say Chinese Version of Windows. In this case this FM
doesn't get properly the directory names in chinese.
This might be because the directory name is of type
string. is there any other way to achieve this?
SAP Ver: 4.7
Question:2
I need to download chinese description to an excel file from an internal table. I could able to achieve this when I use Chinese version of windows and log into SAP in 'ZH'.
I agree that to see the chinese characters properly we may need to log in the chinese version of Windows. But I don't understand the reason that why do we need to log into chinese version of SAP? i.e 'ZH'. I have decla=red the description field as character in the internal table declaration and in the FM 'GUI_DOWNLOAD', I specify the file type as 'ASC'. Any clues on this to get this executed being logged in EN version of SAP?
Thanks,
Siva

Hey there KCR875015,
It sounds like like you need to change the language your iPad is set to, but you cannot understand the language it currently is. The following article should help you do that:
iOS: How to change the language when it's set to one you don't understand
http://support.apple.com/kb/ht2371
Thank you for using Apple Support Communities.
All the best,
Sterling

Dynamic File and Directory Name without Mapping

Hello Experts,
We have following requirement:
1) Files will be picked from R/3 AL11 directory and would be placed in corresponding folder in target system.
2) On source side ,there would be only one folder for all types of files(around 20),but on target side,there would be one folder for each kind of file(20 folders)
3) File name should be same on the target side but target directory should be selected based on file name.
I have gone through a number of posts related to similar requirements and hence,sorry for a new post but I am not yet able to find a solution to this.
I could understand,this can be achieved using DynamicConfiguration UDF .
But I have no possibility to have mapping in my scenario.It would just be a pass through scenario.
Can anyone please suggest a solution to this?
Thanks.
Regards,
Shweta

Hello,
Thanks a lot for suggesting solution to this problem.
I could achieve this using following Java mapping:
import com.sap.aii.mapping.api.*;
import java.io.*;
import java.text.*;
import java.util.*;
public class GetDynamicConfiguration implements StreamTransformation {
    private Map param;
    public void setParameter(Map map1) {
        this.param = map1;
    public void execute(InputStream inputstream, OutputStream outputstream) throws StreamTransformationException {
        try {
               AbstractTrace trace = null;
            // a) Set ouput File name
            String directory=null;
               trace = (AbstractTrace)param.get(StreamTransformationConstants.MAPPING_TRACE );
            param.put(DynamicConfigurationKey.create("http://sap.com/xi/XI/Dynamic", StreamTransformationConstants.DYNAMIC_CONFIGURATION), "");
            DynamicConfiguration conf = (DynamicConfiguration) param.get(StreamTransformationConstants.DYNAMIC_CONFIGURATION);
            DynamicConfigurationKey key = DynamicConfigurationKey.create("http://sap.com/xi/XI/System/File", "FileName");
               DynamicConfigurationKey key1 = DynamicConfigurationKey.create("http://sap.com/xi/XI/System/File", "Directory");
            String filename =conf.get(key);
            conf.put(key, filename);
               trace.addInfo("File name is "+filename);
            if(filename.equals("in.txt"))
            directory = "/home/ftpxi/in";
               if(filename.equals("test.txt"))
               directory = "/home/ftpxi/in/test";
               if(filename.equals("shweta27.txt"))
               directory = "/home/ftpxi/in/test";
               trace.addInfo("Directory name is "+directory);
               conf.put(key1, directory);
            // b) Just copy input file to output file
            byte[] b = new byte[inputstream.available()];
            inputstream.read(b);
            outputstream.write(b);
        } catch (Exception exception) {
            exception.printStackTrace();
Thanks again.
Regards,
Shweta

XI File Adapter: Sender and the use of wildcards in the directory name

Hi,
Quick question for you XI guru's. I have a requirement to read files from an ftp server that live in separate directories like:
/dir1/subdir1/in/
/dir1/subdir2/in/
/dir1/subdir3/in/
I want to setup the communication channel so that it does something like:
/dir1/*/in
Normally I would specify the other directories in the advanced tab however there could be potentially 1000's of subdir's so I want to use a wildcard. Is it possible to do this with a directory name?
I have read this link and it seems to only indicate using wildcards in the filename.
http://help.sap.com/saphelp_nw04/helpdata/en/bc/bb79d6061007419a081e58cbeaaf28/frameset.htm

Hi,
I think.. try as below
It is better to use OS script to read the said three directories.
bcz you know three directories. so you can use OS script file and that file can be configured in Ftp adapter.
OS Command from FTP
Please see the below links
/people/michal.krawczyk2/blog/2005/08/17/xi-operation-system-command--error-catching
OS Command on FTP
OS command line script - Need help
FTP - Run OS Command before file processing
/people/sameer.shadab/blog/2005/09/21/executing-unix-shell-script-using-operating-system-command-in-xi - Call UNIX Shell Script
Chilla

Need to select a directory name.

Similar Messages

Maybe you are looking for