(access key for SAP standard modifications needed for test data of this question)
Introduction:
I want to find a certain type of modification to a certain SAP standard repository object (IDES test dataset).
The modification is located in the include of a function group (it is listed under System-defined Include-files in the functionpool of the function group).
So far, I found the table SMODILOG as a central list of modifications (Log of Customer Modifications to Dev. Env. Objects).
Test Data:
My test data is function group V07A, that has e.g. the include LV07A014 (Part of the LV07ANNN include). This was modified by inserting stuff in its source code (one needs an access key in order to be allowed to do this) such as:
*{ INSERT IDSEXAMPLE 1
* this is a comment, which was added
*} INSERT
Goal:
Subsequently I want to find a table where all modifications like this are listed. I want to find the place of modification, i.e. the object type and program id of the object that was modified.
2 Questions:
I realize that the SAP standard include of a function group has a different object type and program ID than the top-include and uxx-includes.
Whereas the latter are of type PROG and prgmid R3TR (found in object catalog entry), the LV07A014 has an object catalog entry identical to the function group that it belongs to, namely R3TR FUGR. This is already peculiar to me. This seems to me as if the resolution to the sub-level (include level) is missing.
In addition, the modification to LV07A014 is listed in table SMODILOG as having the object type (field sub_type) REPS (the pgmid is not included in SMODILOG). I would expect PROG, as for the other inclueds (LTOP, LUXX).
-> Why is there a difference of object types, programids between L_TOP, L_UXX on the one hand and L_NNN Inclues on the other? (Or am I mistaken?)
-> Where can I find information of all SAP standard modifications in my system and the true object type,pgmid belonging to these modified objects (and not the function group that the modified object belongs to, this resolution does not suffice)?
There is no table or something similar where all modifications are listed. The table E071 is a good source to check pgmid and object type.
Why there is a difference of object types, program IDs between L_TOP, L_UXX on the one hand and L_NNN Includes on the other remains a mystery. SAP...
Why do you need a table? For what? Is it purely academic question or connected with real life tasks?
Have you ever tried SE95 transaction? It lists all modifications that were done in system disregard of object type and name. Yes, and function groups too. They are easily searchable by hierarchy
If it is FUGR include that was modified, then it will be listed in the node Outside of modularization units
Finally RTFM, bro...
Related
I want to see the difference of Delivery Class 'A' and 'C'. C for data entered only by the customer, but how can I see it on the code?
I created two tables of type 'A' and 'C'. I add data with ABAP code. I thought I couldn't add data to the table I created with C, but they both work the same.
For A Type:
DATA wa_ogr LIKE ZSGT_DELIVCLS2.
wa_ogr-ogrenci_no = 1.
wa_ogr-ogrenci_adi = 'Seher'.
INSERT ZSGT_DELIVCLS2 FROM wa_ogr.
For C Type:
DATA wa_ogr LIKE ZSGT_DELIVERYCLS.
wa_ogr2-ogrenci_no = 1.
wa_ogr2-ogrenci_adi = 'Seher'.
INSERT ZSGT_DELIVERYCLS FROM wa_ogr2.
Datas get trouble-free when I check with debugging.
Is there a live demo where I can see the working logic of C? Can you describe Delivery Class C better?
Tables with delivery class C are not "customer" tables, they are "customizing" tables. "Customizing" is SAPspeak for configuration settings. They are supposed to contain system-wide or client-wide settings which are supposed to be set in the development system and then get transported into the production system using a customizing transport. But if that's actually the case or not depends on what setting you choose when generating a maintenance dialog with transaction SE54. It's possible to have customizing tables which are supposed to be set in the production system directly without a transport request.
Tables with delivery class A are supposed to contain application data. Data which is created and updated by applications as part of their every day routine business processes. There should usually be no reason to transport that data (although you can do that by manually adding the table name and keys to a transport request). Those applications can be SAP standard applications, customer-developed applications or both.
There is also the delivery class L which is supposed to be used for short-living temporary data as well as the classes G, E, S and W which should only be used by SAP on tables they created.
But from the perspective of an ABAP program, there is no difference between these settings. Any ABAP keywords which read or write database tables work the same way regardless of delivery class.
But there are some SAP standard tools which treat these tables differently. One important one are client copies:
Data in delivery class C tables will always be copied.
Data in delivery class A tables is only copied when desired (it's a setting in the copy profile). You switch it off to create an empty client with all the settings of an existing client or to synchronize customizing settings between two existing clients without overwriting any of the application data. You switch it on if you want to create a copy of your application data, for example if you want a backup or want to perform a destructive test on real data.
Data in delivery class L tables doesn't get copied.
For more information on delivery classes, check the documentation.
I am having the following case:
There is entity 'Master_Entity'. This entity has properties as name, type, duration etc. There are other two type of entities 'Entity' and 'Sub-Entity'. There are identical as the 'Master_Entity' (They have absolutely the same properties).
At the end the 'Master_Entity' should hold a collection of 'Entity' and 'Entity' should hold a collection of 'Sub-Entity'. The tricky part is that records of type 'Entity' can be part of different 'Master_Entity' (same for 'Sub-Entity'), but they can have different values for duration for example. How can achieve such modularity?
Here I came up with, but it's not quite do the work. May you guys help me with this.
Edit: Imagine this as some sort of a work tracker. For example you have a 'Create PHP App' (Master entity). This entity contains duration of how long it will take to finish this job. In addition it contains a entity 'Writing Code' (Entity) and this one can be divided to 'Writing Http Client' (Sub-Entity) which has duration property which is specific for this job.
On other side you might have other job: 'Create an Java App' (Master Entity) which will contain the same 'Writing Code' entity, but with duration which will have different value, because of the context of the Application you are building.
I want to have a single record 'Writing Code', but the duration value that it have should be different for every job it's assigned to. How can achieve that with creating a minimum duplicating records of type 'Entity'?
It sounds like something like these 3 tables will work for you:
Entity
* Id
* Name
* Type
EntityGroup
* Id
* Name
* ParentEntityGroupId
* ParentEntityId
EntityRelationship
* Id
* EntityId
* ParentEntityId
* EntityGroupId
With this structure, you can have an Entity be a member of a Group, or a solo Child of another Entity. You can also have a Group be a Child of an Entity, or even a Child of another Group. Without knowing specifics of your data, it's hard to know what you might need, but this should get you started.
From what you have said, it seems that you don't need EAV at all because you don't have different properties for each item just different values. And thus you should not be using it.
What you need is a combination of lookup tables and then tables that address the actual tracking history of the work. This is because this is time sensitive data. The tasks at the time the projects was created may be substantially different than the tasks associated with that task group two years from now, but you need to record the tasks at the the time of creation). Note that this is not denormalizing, it is creating a picture of data in time. The real duration always goes to teh project not ever to the Task. In the task, you can have a suggested duration to use as a starting point. I used a similar design (with far more fields of course) to design a database for building sales proposals for technical-hardware related projects. The real key here is to recognize what data needs to be stored as a point time time and what is lookup data used to build the final project data. If someone adds a new task to the "Create a Java App" group, you don't want to change details about projects already completed or in work, only new projects.
So you need:
Task group
Task Group ID
Task Group Name
Task
TaskID
Task Name
SuggestedDuration (can be null if you have tasks that are always different
but filled in for tasks that usually have a similar duration)
Task_Taskgroup
TaskID
TaskGroupID
Project
ProjectID
ProjectName
TaskGroupID
ProjectTask (should be filled in automatically when the task group is
chosen for the project)
ProjectID
Task ID
EstimatedDuration (fills in the default value, but can be changed
by the person creating the work project)
ActualDuration (Field in after the task is done, can be used by an
analyst to create more reflective task default duration values)
Of course each of these tables may have other fields depending on the need.
I am accessing flight reservation application built in SAP.
The application has a section on catering which contains: BC_MEAL, BC_MEALT, BC_STARTER, BC_MAINCOURSE, BC_DESSERT.
However, there are no such tables prefixed with BC_.
The tables are SMEAL, SMEALT, SSTARTER, SMACOURSE, SDESSERT instead.
Why is this discripency due to? How does SAP manage to convert application names into table names.
You're looking at the Data Modeler (SD11) and trying to compare it to the Data Dictionary / ABAP Dictionary (SE11). The actual table names are assigned to the entities explicitly:
expand BC_FLIGHT
double-click on BC_SFLIGHT
Button Dict. (?)
--> This screen should show the tables and/or views used to represent the entity.
It is worth noting that for many applications, no explicit data model exists (which is why I personally never bothered with the Data Modeler - a tool like this is virtually useless unless everyone else uses it as well).
I am building a data warehouse for the company's (which I am working for) core ERP application, for a particular client.
Most of the data in the source database, which is related to hierarchies in the data warehouse are in columns as shown below:
But traditionally the model to store dimension data according to my knowledge is as:
I could pivot the data and fit them in the model shown above. But the issue comes when a user introduces a new hierarchy value. Say for instance the user in the future decides to define a new level called Product Sub Category. Then my entire data warehouse model will collapse without a way to accommodate the new hierarchy level defined.
Do let me know a way to overcome this situation.
I hope my answer is clear enough. Just let me know if further details are needed.
Well, nothing should collapse -- the ETL should extract and load the data as always.
Here are a few options to consider:
Simply add one more column for the new hierarchy to the dimProduct.
Try using hierarchy helper table.
Consider adding path string attribute to the dimProduct.
Hi does any one know how to check if a given DB object (Table/View/SP/Function) is used inside Oracle.
For example to check if the table "A" is used in any SP/Function or View definitions. I am trying to cleanup unused objects in the database.
I tried the query select * from all_source WHERE TEXT like '%A%' (A is the table name). Do you thing it is safe to assume it is not being used if it does not return any results?
From this ASKTOM question:
You'll have to enable auditing and then come back in 3 months to see.
We don't track this information by default -- also, even with auditing, it may be very
possible to have an object that is INDIRECTLY accessed (eg: via a foreign key for
example) that won't show up.
You can try USER_DEPENDENCIES but that won't tell you about objects referenced by code in
client apps or via dynamic sql
There's code in the thread for checking ALL_SOURCE, but it's highlighted that this isn't a silver bullet.