If you’re on a desktop workstation, the simplest solution might be a keyboard wedge. Physically, it’s just a USB barcode scanner. Electronically, it’s a USB device that looks like a keyboard to the operating system, sending the scanned barcode as keystrokes. It doesn’t care what kind of application it’s using, making setup very simple. There are usually control codes you can program into it to configure its nonvolatile RAM. You probably wouldn’t want to send enter after the barcode, for example.
used a cursor to loop through the group and if the logged in user wasn’t part of the group then the error displayed. If you put the username in the WHERE clause of your cursor, you can trade the O(n) performance of a brute force linear search with the O(log n) performance of an index search. You’d just check whether the cursor was %FOUND and stop searching when it is. More importantly, this could make your code smaller and easier to read.
We’ve encountered this, too. The XML export only includes IFSAPP and IFSINFO objects and doesn’t include Oracle system privileges, grants to non-IFS roles, or grants on SYS objects. FND_RUNTIME, for example, is supposed to have the CREATE SESSION privilege. Exporting and reimporting it will basically break the whole application.
I've found using Oracle Loader external tables is easier than using migration jobs. They're far easier to reuse for multiple purposes.You'll need to edit this if you want more than 10 columns or more than 256 characters in your data. You'll also need to cast strings to numbers or dates as needed downstream.Reading from a CSV file sitting in Oracle directory is a lot like reading from a view into that file. -- AS SYS:CREATE OR REPLACE DIRECTORY c_external_csv_dir AS '\\your\server\location';GRANT read, write, execute ON DIRECTORY c_external_csv_dir TO ifsapp;GRANT read, write, execute ON DIRECTORY c_external_csv_dir TO ifssys;-- AS IFSAPP:CREATE TABLE c_external_csv_tab( a1 VARCHAR2(256), a2 VARCHAR2(256), a3 VARCHAR2(256), a4 VARCHAR2(256), a5 VARCHAR2(256), a6 VARCHAR2(256), a7 VARCHAR2(256), a8 VARCHAR2(256), a9 VARCHAR2(256), a10 VARCHAR2(256))ORGANIZATION EXTERNAL( TYPE ORACLE_LOADER DEFAULT DIRECTORY c_external_csv_dir ACCESS PARAMETERS
You might consider making a union view to achieve this on the database. SELECT * FROM view_one WHERE condition = TRUEUNION ALLSELECT * FROM view_two WHERE condition = FALSE
We use broadcast messages in our nonproduction refresh scripts to tell users when the environment was last refreshed from PROD. The bright orange banner also helps them notice they’re not in PROD. (This is made with a call to COMMON_MESSAGES_API.NEW__.)
Try some other values:Use an existing module, such as FNDBAS. (I've never created a module, but I'd imagine this would require an insert into MODULE_TAB.)Use as your LU name ZPurchaseOrder (case sensitive). Then do this in order:Grant your package to IFSSYS.Refresh your dictionary cache.Grant your package to a permission set.Refresh your security cache.
If you’re willing to keep the data resident on the Purchase Order custom field, what benefit are you getting by not retaining the history logging longer on the PurchaseOrder LU? Those bytes need to fit somewhere in your database, and they’re all in the IFSAPP_DATA tablespace, so it’s not like you’re gaining much by keeping them in PURCHASE_ORDER_CFT while purging them from HISTORY_LOG_TAB/HISTORY_LOG_ATTRIBUTE_TAB.For a given set of triggers of the same type, Oracle does not guarantee the order that triggers fire. They added FOLLOWS/PRECEDES options in Oracle Database 11, but IFS Events do not implement this, as far as I know.https://docs.oracle.com/en/database/oracle/oracle-database/21/lnpls/CREATE-TRIGGER-statement.html#:~:text=information%20crossedition%20triggers-,trigger_ordering_clause,-FOLLOWS%20%7C%20PRECEDESIf you get lucky and find they work well together today, that might change tomorrow. If you’re worried about the history log size, I’ve found that the cleanup job that purg
Thank you, this is probably not wise to use an event. Thank you for the input. Yes, and I believe you’re allowed to mark your own response as the best answer.
Bump
This would be impossible without some custom code. An event action runs as the end user, and most end users aren’t going to have permission to deactivate users, even themselves. As a word of warning, a deactivated user is deactivated everywhere and can no longer work without IT intervention to reactivate them, even if they only made a mistake. To put that more bluntly, you can’t make money without working employees. I added some protection to ensure your app owner and other system accounts don’t get locked. I didn’t test this. SET SQLBLANKLINES ONCREATE OR REPLACE PACKAGE c_deactivate_user_api AUTHID DEFINER IS/*module_ CONSTANT VARCHAR2(25) := 'FNDBAS';lu_name_ CONSTANT VARCHAR2(25) := 'CDeactivateUser';These comments get read by DATABASE_SYS. DO NOT REMOVE.*/PROCEDURE init;PROCEDURE deactivate_current_user;END c_deactivate_user_api;/CREATE OR REPLACE PACKAGE BODY c_deactivate_user_api ISPROCEDURE init IS BEGIN NULL; END init;PROCEDURE deactivate_current_userIS attr_ VARCHAR2
I don’t believe .html files can be attached in IFS, but you could attach the file in .txt perhaps. If needed, you can add new file types under EDM Basic → File Types.XML is already a type on our system, but I don’t know if we added it.
No, this isn’t possible yet, and these kinds of limitations are why custom events have been offered as a tailoring option.A custom event need not be complicated. You might create a new permission set called CAN_EDIT_FIELD_X, then set up your event and event action on the _CFT table. BEGIN IF NVL('&OLD:CF$_FIELD_X', 'NULL') != NVL('&NEW:CF$_FIELD_X', 'NULL') THEN IF security_sys.has_user_role_access('CAN_EDIT_FIELD_X') = 'FALSE' THEN error_sys.appl_general('FndBas', 'CANTEDITX: You do not have permission to edit Field X. (If you think you should, ask IT to grant you CAN_EDIT_FIELD_X.)'); END IF; END IF;END;Then you’ll grant CAN_EDIT_FIELD_X to the appropriate users.
Any further thoughts on Contacts, anyone?
Moving upward in an ascending sequence is safe to do. You’ll find your sequences already have holes because sequence IDs that get consumed during a session aren’t reused by another session after an error rollback. Side note:In the critical areas of IFS where sequences can’t have holes, the application uses regular tables to manage the numbers, since those kinds of updates get rolled back on error. (For example, see “Enterprise Identities Series” and “Voucher Series per Voucher Type”.) Messing around with those is far less safe from an accounting or compliance perspective.
I have yet to find a way to perform client-side warnings and confirmations reliably, because Client_Sys() has not been properly hooked into all IFS operations, only some. That means the IFS client is rendered useless, because we need the ability to ask the user questions and display warnings in many business scenarios. If anyone finds a workaround for this issue I hope they will post and let us know! I’m anxiously looking forward to Business Process Automation in IFS Cloud.IFS Cloud 21R1 – Business Process Automation Workflow Engine Overview and Demo | GoToStage.com
when I add a new procedure to it and other procedures within the package are being accessed at the time of compilation and the compile gets stuck You won’t get this issue if you remove the global variables. You will need them there so the dictionary cache gets updated and assigns your package to a module and logical unit, but the process that reads the code doesn’t mind if they’re commented out.Lesson Learned: How to avoid ORA-04068: "existing state of packages has been discarded" | IFS Community Creating a new package is a little bit of a pain. If you follow the example I put here, with the tweak I just mentioned, it includes all the extra steps that are necessary.Is there a way to run Custom APIs as scheduled database tasks ? | IFS Community
We ran into the issue, that the configured “Prospect” Category was lost during this process and all of them are converted to “Customer”, since this was the only setting available in the legacy version of Apps 8. Thanks. We saw that and addressed it by adding the basic data.DECLARE info_ VARCHAR2(32767); objid_ VARCHAR2(32767); objversion_ VARCHAR2(32767); attr_ VARCHAR2(32767);BEGIN BEGIN client_sys.clear_attr(attr_); client_sys.add_to_attr('IDCOD', 'cca.PC', attr_); client_sys.add_to_attr('IDCNA', 'CCA', attr_); vmo_base_code_api.new__(info_, objid_, objversion_, attr_, 'DO'); EXCEPTION WHEN OTHERS THEN dbms_output.put_line('ERROR calling vmo_base_code_api.new__: ' || SQLERRM); END; BEGIN client_sys.clear_attr(attr_); client_sys.add_to_attr('IDCOD', 'cca.PC', attr_); client_sys.add_to_attr('IDLNG', 'en', attr_); client_sys.add_to_attr('CODEDESC', 'Prospect', attr_); client_sys.ad
Are either of these work-arounds acceptable?Create a new queue, then move your critical jobs to it that you don’t want blocked. Increase the parallel degree for your default queue.
Are you running Windows? If so, you can write directly to the MWS without needing to copy the file first.Oracle Database runs as the SYSTEM user by default, but on the network, that means it runs as the corresponding Computer principal in Active Directory. If the server hostname is DBSERVER, then Oracle would run as a user called DBSERVER$ (with a dollar sign on the end). You can grant folder access to that Computer principal to allow Oracle to write to a UNC path. First, you’ll create your share on the middleware server.\\your_middleware_server\share_name Next, you’ll grant that share to the user running Oracle. Here’s what that grant screen might look like in Windows: Finally, you’ll create your directory object in Oracle:CREATE OR REPLACE DIRECTORY your_directory AS '\\your_middleware_server\share_name';GRANT ALL PRIVILEGES ON DIRECTORY your_directory TO ifsapp; Footnote:Oracle recommends you run their software as a regular user instead of this default of using the SYSTEM user. I
If you’re confident your deployment is repeatable, you can export those tables from your nonproduction Apps 10 environment and reimport those two tables to restore the profiles after going live (as well as FNDRR_USER_CLIENT_PROFILE_TAB if you’re changing user base profile assignments).That will get you into trouble if you deploy too many or too few custom fields, too many or too few IFS deliveries, if the users log in with a different language that has a different NLS_SORT setting, if you create new profiles in the meantime (or new users), or if users expect new changes made in 9 after today to carry forward to 10. Otherwise, it’s a pretty quick and stable way to go. We used this strategy for Quick Reports when we upgraded from 8 to 9. We took the whole QUICK_REPORT_TAB table out of the nonproduction environment and dropped that right into the new production system. After applying the deliveries but before dropping in this new version, though, I took a backup. After restoring the new t
This script exports your ACL definitions as executable SQL text. Run this with SQL*Plus as SYS and store the output. From that output, grab just what you need to drop and recreate the offending ACL. Edit your script to change the host name. Run your edited script to recreate what you need.ORACLE-BASE - DBA Scripts: network_acls_ddl.sql Keep in mind that the IFS PL/SQL Access Provider uses an ACL for the database to talk to the application server, so you don’t want to remove everything. You may also have integrations that need these too. -- ------------------------------------------------------------------------------------- File Name : https://oracle-base.com/dba/11g/network_acls_ddl.sql-- Author : Tim Hall-- Description : Displays DDL for all network ACLs.-- Requirements : Access to the DBA views.-- Call Syntax : @network_acls_ddl-- Last Modified: 28-JUL-2017-- -----------------------------------------------------------------------------------SET SERVEROUTPUT ON FORMAT WRAP
Do you need this object connection? If not, does this error go away if you remove it?
What are you hoping to achieve when you re-import them? Is this for backup and restoration purposes? The headers are stored in FNDRR_CLIENT_PROFILE_TAB, where personal profiles have an OWNER and the base profiles don’t. The values are stored in FNDRR_CLIENT_PROFILE_VALUE_TAB.
My working idea so far is to migrate those Sales and Marketing Companies as Customers first, then when the migration is complete, do UPDATE queries against CUSTOMER_INFO_TAB.CUSTOMER_CATEGORY and VMO_COMPANY_TAB.IDCODCCA to demote them back to P:rospects.This feels less invasive than writing directly to the tables to create Person and Customer Contact records. I’m trying to reinvent the fewest wheels.
Already have an account? Login
No account yet? Create an account
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.
Sorry, we're still checking this file's contents to make sure it's safe to download. Please try again in a few minutes.
Sorry, our virus scanner detected that this file isn't safe to download.
