Advices & rules for package hierarchy

Hi,
I developing a new project and I'll use packages for the first time. (My company doesn't have rules for package)
I would like to know the gereral rules for creating a good hierarchy.
i'm thinking of something like:
com.mycompany.myproject.form
com.mycompany.myproject.dialog
com.mycompany.myproject.sql
com.mycompany.myproject.dataobjects
com.mycompany.myproject.table
com.mycompany.myproject.tree
Should I use also packages like com.mycompany.table for generals stuff like table models that could be of use for others projects ?
Is this a good idea ? Do you have any suggestions ?

I am not aware of any hard and fast rules for packaging. There are some things I keep in mind when I setup hierarchies though:
1) The "standard" (more common than standard) is your businesses internet address reversed. So for instance, if you are www.foo.com, your packages would start out with "com.foo". Some people leave off the "com" part as well, so you would have "foo". I don't like this however as more and more companies reserve "*.net" and "*.com" and "*.org" etc addresses. Leave the top level domain on there. This is mostly just for namespace sake, so you don't get two pieces of code from two different companies with the same name.
2) Is this the only product your company produces? OR will this be the only product your company produces? If the answer is no, you may want to put the product name next, so now we have "com.foo.product1". This will allow you to build nice, clash safe hierarchies.
3) Most products are broken down into a number of projects or modules. Sometimes its a good idea to divide the package structure with this information. So now we have "com.foo.product1.module1", "com.foo.product1.module2" etc.
4) If its an enterprise application (multitiered, J2EE, etc), I personally like to divide the code into tiers. It makes building easier and packaging for deployment much more intuitive. You might have web tier code, database tier code, application tier code, code common to all tiers etc. So now we have "com.foo.product1.module1.app" and "com.foo.product1.module1.web", etc. If its not an enterprise multi-tiered application, skip this step.
5) Now within the hierarchy we have built up, its pretty much up to you to logically divide your code.
6) It is now possible to "push" code up to certain levels to make it available to other projects/products/tiers. If you develop a nice data structure that you think might be useful to other products your company is developing, move the data structure up to an appropriate level to make it available to all products. For instance, "com.foo.structures.AwesomeStructure", instead of "com.foo.product1.structures.AwesomeStructure".
7) Products and projects should not import outside of their "scope". This prevents cross project dependency and cross product dependency. If another project or product needs to import something outside its scope then the code should be moved to a more common place to make it available.
Some will call this overly anal, but it has worked well for me for a few years. Hopefully others can provide you with more tips.
Hope this helps.

Similar Messages

How correct rule for Pcon calc for different hierarchy withsame Ent andICP?

Hi! My question is below, after problem and describing of current situation
The problem:
Calculation of percentage of consolidation is runnig only for fixed hierarchy by consolidation rule
The situation:
I've got three hierarchys for my Entity's in system: one is base and two others are alternative.
#root ConsBase
ConsBase BaseChild
BaseChild BaseChild1
BaseChild BaseChild2
#root ConsAlt1
ConsAlt1 BaseChild1
ConsAlt1 BaseChild2
#root ConsAlt2
ConsAlt2 BaseChild1
ConsAlt2 BaseChild2
And I have different value for Pcon for each member of hierarchys:
BaseChild.BaseChild1, pcon = 30%
BaseChild.BaseChild2, pcon = 30%
ConsAlt1.BaseChild1, pcon = 50%
ConsAlt1.BaseChild2, pcon = 50%
ConsAlt2.BaseChild1, pcon = 70%
ConsAlt2.BaseChild2, pcon = 70%
Because of the different pcon for same entity members I've wrote a rule for correct consolidation proccess:
Pieces of elimination rule:
'After checking possibility for elimination starts percentage of consolidation rule for Entitys and ICPs:
If CanEliminate Then
dPCon = HS.Node.PCon("")
'ICP parent define (restriction: consolidate only hierarchy ConsAlt1)
call GetParent(strICPParent, strICP)
'calculate Pcon for ICP
dPConICP = HS.Node.PCon("E#"&strICPParent&"."&strICP)
'allow for portion of all parents
call GetParent(strICPSecondParent, strICPParent)
dPConICP = dPConICP * HS.Node.PCon("E#"&strICPSecondParent&"."&strICPParent)
'Parent rule define for entitys and icps
Sub GetParent(strParent, strEntity)
+For each EntParent in HS.Entity.List("ConsAlt1",[Parents])+ '!!! fixed hierarchy ConsAlt1 !!!!
If (HS.PARENT.ISCHILD(EntParent,strEntity) = TRUE) Then
strParent = EntParent
End If
Next
End Sub
I want have a process of consolidation with correct calculation of percentage of consolidation for Entity, ICP and not in one fixed hierarchy (ConsAlt1) - but in ConsBase, ConsAlt2 without making corrections in rules to fix Entity. How can I do it in my situation?
Edited by: 806947 on 02.11.2010 3:58

Hi DST
This is a great effort and gesture. thank you on behalf of all the newbies.
PJ

Transfer rules for Functional Location hierarchy

Hi
I am trying to install Transfer rules for 0FUNC_LOC_HIER from BI Content.
But i could not find any .
Where do i can find T/Rs or Is there any procedure for that ?
Regards
sateesh

Hi Sateesh,
I am not sure if the hierarchy datasource 0FUNC_LOC_HIER supports the transfer methog PSA,
if it only supports IDoc then its will not have transfer rules that you can maintain. You can
check in table ROOSOURCE in the source system (field transfer method) to see if the datasource supports
PSA transfer methd. If it does then you could create your own transfer rules.
Best Regards,
Des

Standard Smartform for label printing for packages - pls advice

Hi All,
I need help in changing a print program ie Standard Smartform for label printing for packages ?
Standard Driver program: SDPACKDR
Please provide me with the detailed steps for editing the Std Smartform according to any requirement.
Points will be rewarded promptly.
Thanks.

hi,
1. The smartform name is SD_PACK_ETIK.
2.Just copy the standard program into Zprogram and make chanes and assign in NACE with the form.
3.But,in smartform lable printing is not possible.
4.you can print barcodes.
Regards,
Kumar(Reward if helpful).

How to structure my package hierarchy?

As an inexperienced Java programmer, I am having a hard time understanding how to structure packages hierarchically.
Let's say that I am developing a client/server AddressBook application. So automatically, the following package hierarchy comes to my mind:
(company/org name is ACME for example)
com.acme.addressbook
com.acme.addressbook.protocol
The above two packages form the reusable core of the AddressBook system and include classes such as AddressBook, AddressBookEntry, etc. and related helper classes. Additionally, the protocol sub-package contains classes that pertain to the C/S network protocol. And both of these two packages are packaged up in a jar: AcmeAddressbook.jar
com.acme.addressbook.client
The above is the package for the Client application: AcmeAddressBookClient.jar
com.acme.addressbook.server
The above is the package for the Server application: AcmeAddressBookServer.jar
My questions are: does such a package structure and hierarchy make sense? Or are there any glaring mistakes that I have made?
Is it acceptable to keep the client and server applications into their own separate packages?
Also, is there any recommended limit for nested packages i.e. how deep should packages be nested within the top-level AddressBook package?
I am struggling to understand good package design and package structuring conventions.
Thanks any for any info, critique, advice, etc.

Lets the two client and server packages are respectively :
com.acme.addressbook.client
com.acme.addressbook.server
Yes, such a package structure and hierarchy make sense!
If you are having common utility classes and many library jars inside the package addressbook.library, you can just refer in client as well as in the server package.
If you are going to create separate project for client and for server, you want to maintain multiple copies, which further results in the problems due to redundant code. ie., whenever you changed any library files, you want to update in both places.
You can create two jars as you mentioned as,
AcmeAddressBookClient.jar
AcmeAddressBookServer.jar
Before building jars you just want to set the executable Main class appropriate for Client and Server before making the client and server jars respectively.
There is no limit for the nested package, if you need to group meaningfully, you can make as many nested package as you want!
For your reference :
http://articles.techrepublic.com.com/5100-22-1046677.html

Essbase Build Dimension using load rules - Jagged Recursive hierarchy

Hi,
I am building a dimension from data that has the following parent/child relationship
PARENT CHILD
A B
C D
C E
C F
A C
Load dimension results in the following hierarchy:-
A
|__B
|__C
|_F
C
|_D
|_E
I would like the hierachy to be:-
A
|__B
|__C
|_F
|_D
|_E
Is there a way to achieve this? Is is to do with the order of rows in the data file?
Thanks
Dinesh

Hi,
I agree with the advice already posted about allowing moves in your dim build settings on the dimension in question. I just want to help clarify the other comment about how your file is sorted. As long as you are building a main rollup only (i.e., no shared member, alternate hierarchies in the same input file), the Essbase parent/child build method doesn't require any particular record order (as long as the pairings are correct in and of themselves). Essbase p/c method is powerful for ragged hierarchy creation and also able to "figure out" the intended hierarchy.
Now if you are including shared member relationships in the file, just be sure to have those records AFTER all records for the main rollup (logically as if you were simply using a second input file for the same rule). They don't have to be sorted within the rollup either.
BTW, the other strength of the p/c method is the ability to handle upper level shares (i.e., when your shared members are referencing members that are above level 0 in their stored (main hierarchy) positions.
Darrell

How can I activate the transfer rules for the ODS updating a data target.

We are on BW 3.5 and I'm loading data into the 0FIGL_O10 ODS and then uploading the data into the cube 0FIGL_C10. The data loads just fine to the ODS but when I try to 'update the data target' I get a date & time stamp' error on the info-package transfer rules.
I then Replicate the datasource 80FIGL_O01.
I must then 'activate' the transfer rules.
However I cannot get the transfer rules for 80FIGL_O10 in CHANGE MODE to activate them.
How can I activate the transfer rules for the ODS updating a data target.
The error text is as follows:
DataSource 80FIGL_O10 has to be replicated (time stamp, see long text)
Message no. R3016
Diagnosis
DataSource 80FIGL_O10 does not have the same status as the source system in the Business Information Warehouse.
The time stamp in the source system is 02/15/2007 10:42:33.
The time stamp in the BW system is 11/07/2006 13:11:54.
System response
The load process has been terminated.
Procedure
Copy the DataSource again and then activate the transfer rules that belong to it. You have to activate the transfer rules in every case, even if they are still active after the DataSource has been copied.
Thanks for your assistance.
Denny

Hi Dennis,
Try, using Business Content to activate your data source
hope this will help you
How activate business content?
http://help.sap.com/saphelp_nw04/helpdata/en/80/1a66d5e07211d2acb80000e829fbfe/frameset.htm

AT&T has rules for the iPhone 3G S launch - Lines allowed to form starting at 7am

AT&T has rules for iPhone 3G S launch - Lines allowed to form starting at 7am
Perhaps AT&T has learned some from the troubles with iPhone past launches and they are going to try to make the experience better by settings some rules this time around. While AT&T is calling these “rules” guidelines, we think it really is nothing more than an effort to make a success of what the company hopes is one of the biggest tech event launches of the year.
AT&T stores will receive inventory in time for the actual launch on June 19th. AT&T is already accepting pre-orders online as well as in stores. This change is key this time around because those that choose not to place a pre-order could be left out in the cold. From our understanding, pre-order customers will be given priority.
The way that it will work is when the AT&T store opens, customers in line will then be split into two lines, with those already having a pre-order placed in one line and those who don’t have a pre-order placed in the other line. Pre-order customers will be able to start receiving their new iPhone 3G S starting shortly after 7 A.M., while those without a pre-order will have to wait until the store’s official regular opening time. Upon the actual regular opening time of the store, the two lines will then be merged into one line again. Apple retail stores will also be an option as they will open at 8 A.M. on launch day, but only those that pre-order from the Apple retail store will actually be able to pick up their phones early.
As for the pricing, current AT&T customers that have had a contract with AT&T for at least 18 months will be able to get the new 16GB model for $199 or a 32GB model for $299. If you happen to be a current AT&T iPhone owner, but have not gone 18 months on your current contract, you will be able to pick up a new iPhone 3G S but will have to pay between $399 to $599, depending on how much time is left on your current contract and whether you want the 16GB or 32GB model. From what we understand, you will have to agree to another 2-year contract to get the iPhone 3G S this time around.
The only actual possible loophole in this deal is to cancel your current AT&T contract and pay the $175 early termination fee and then wait three months. After the three months have lapsed, you can return to AT&T and then get new customer pricing, which would save you $225, but you have to endure a three-month wait.
While the pricing and the options might make some AT&T customers unhappy, it is what it is. The bottom line is that the iPhone is still so hot, AT&T knows that people will pay what it takes to sport the new version if they are a big iPhone fan. According to our sources, AT&T should have enough inventory to more than cover the pre-orders, but beyond the pre-orders it is difficult to say if they will be able to meet the demand on launch day for additional units. Our advice is that if you want to have one on launch day, you had better get your pre-order in soon.
Source:
http://www.fudzilla.com/content/view/14181/1/

I believe if you are a current AT&T Customer and wish to ADD a new phone to your account you will be able to get the new 16GB model for $199 or a 32GB model for $299. I went through
http://www.apple.com/iphone/buy/
and selected "Add a new line to my existing account" and was shown pricing as $199 for 16 GB and $299 for 32 GB.
Hard to see this as any benefit unless you need an additional line on an existing account, which I happen to (I bought a new "additional line" 3G just before the 3GS was announced, and will fit into the "return" window)

Is there any Specific rule for using "IS" and "AS" keyword?

Hi,
i am confused about the use of these two keywords "IS" and "AS".It seems to me they are used randomly anywhere in procedure,package or triggers etc..
Is there any rule for there usages , or only way is to remember the actual syntax whenever they are used ?

They are generally interchangeable, so you can use whichever you choose.
In the English language we say that "some object IS" and "do an action AS" so, for example...
in a package if we define a type it reads better as:
TYPE mytype IS TABLE OF NUMBER;because we are saying that this type called mytype IS a table of numbers.
but if we were to create the type on the database we would use e.g.
CREATE OR REPLACE TYPE mytype AS RECORD (x number, y number);because we are saying that we are creating (action) a type called mytype AS a record.
Just a slight difference in grammatical meaning, but Oracle isn't bothered if you choose to use the other one in most cases.

Appl. Hierarchy and Package Hierarchy - what is the Best practice ?

Hi!
We plan to create a Package Hierarchy in our system, but then come to think of the Application Hierachy is built the way we think of building the Package Hierarchy. We planned to build a Main package called fex. HR and below a structure package called HR-PA and below that ordinary packages within that area. But should we rather create our own application hierarchy for this?
Or should we assign the packages to Sap standard application Hierarchy in the HR area ?
What is the Best practice in this area ? Have any tried to create their own hierarchies ?
Regards, Tine

Hi!
You can create your own Application Hierarchy as well .
From ABAP Workbench select Overview->Applic. hierarchy->SAP or Customer
This is what SAP says about this :
The application component hierarchy is a method of splitting up the SAP System from a logical or business point of view. Packages are a method of modularizing the system from a technical point of view. This technical modularization can, but need not, match the logical division of the system. Assignments should be made however between these two views.
Regards, Tine

JDeveloper 10g CVS package hierarchy gone after check out?

Hello!
We are using JDeveloper 9.0.5.2 (Build 1618) together with WinCvs 1.3.17.2.
We successfully imported with JDev a project structure as
client/moduleA/ui;
client/moduleA/db;
client/moduleA/ctr;
client/moduleB/ui;
client/moduleB/db;
client/moduleB/ctr;
into my repository - connecting with a pserver.
If an other pserver-User tries to check out the module CVSRoot, the package hierarchy - as described above - is gone.
All files are mixed up in flat packages ,,Miscellaneous Packages'' and ,,Miscellaneous Files''.
The created file structure is ok (the package hierarchy remained). But as mentioned, the Application Navigator just shows the flat Miscellaneous file structure.
For check out with the ,,Check Out from CVS Wizard'', in step 4 of 4, the options
,,prune empty folders'' ,
,,checkout into Target Folder'' ,
,,Do Not Shorten Pathnames'',
were checked.
What do we have to do, to see in the Application navigator - after a check-out - the imported tree of packages, instead of a flat structure?
Thank you in advance for your answer!

Hi,
Was the project imported to CVS at the same time as the package structure?
Do the project source paths match the filesystem? For example, does the java source path match the root of the java source? Jdev normally puts files in a flat structure if the files are not in the right file system folder for their java package.
If changing the project path does not fix this can you tell me what technologies your using, such as BC4J, struts, Tag libs, etc.. Then I can try and reproduce this here.
Many thanks,
Geoff
JDev SCM Team
-------------

Rules for iMovie/FCE/iDVD

Reading through many posts seems I find some general rules of thumb not found elsewhere. Thanks in advance for all comments.
For serious video editing work and creation an external firewire HDD is needed, mainly to avoid dropped frames in either FCE or iMovie (iM). The new 20" iMac and external HDD (with 2 firewire connections, for the input device (DV camcorder or DV/SVHS deck) and to the iMac) should be able to run either FCE/iM.
I just did a 45 min project in iM with no noted dropped frames but likely would in much longer projects if I don't get the external drive?
Audio bit rate. Advice I read for iM is that without 16 bit audio the audio gets out of sync in long projects--does FCE also need 16 instead of 12? Not knowing this need, I've recorded a number of projects (with a DV deck and a DV camcorder) in 12--can I capture to the iMac/external HDD in 16 if I'm playing from a source tape that's 12?
Chapter Markers. Seems there are "Rules for Chapter Markers" that I won't repeat here but suspect they apply to both FCE/iM when used with iDVD. Do those rules apply when you use DVDStudio instead of iDVD? One rule I haven't seen yet is that one should likely not use the apply chapter markers option in iDVD as it could apply a marker that violates the "rules" and cause strange results in the DVD when played, right?
Any other rules or concerns I've missed--perhaps always max out on RAM even if its rather costly? What other basic tips or secrets should I know for editing with these products?
Dale

"..The new 20" iMac and external HDD (with 2 firewire connections, for the input device (DV camcorder or DV/SVHS deck) and to the iMac) should be able to run either FCE/iM.."
This may not be true if you're trying to connect a Canon camcorder to the Mac at the same time as an external hard disc: Canons like to have the FireWire circuit all to themselves, and so you may need to do it in two steps with a Canon ..capture to the internal hard disc, then unplug the Canon, plug in the external disc and transfer the movie clips across to the external, and then use that for editing. This doesn't affect Sonys, which happily work with other FW devices connected to the system. I don't know about Panasonics and other brands (..I haven't tried my Panny with an external disc connected, and am in too much of a hurry at the moment to try it. Maybe later today..)
"..can I capture to the iMac/external HDD in 16 if I'm playing from a source tape that's 12?.."
No ..FireWire's a digital connection system, so what goes in at one end (..12-bit source tape..) is exactly what's captured at the other end (..iMac/external HDD in 16..)
One way to convert would be to play your source video into another camcorder through analogue (RCA) connections, with the 2nd camcorder set to record in 16-bit ..or use some other device which accepts an analogue signal IN, and provides a digital signal OUT. That'll recapture the sound in 16-bit, and it can then be transferred into your Mac in 16-bit. But the video may be degraded a little, and it's rather long-winded.
It may be that, once imported as 12-bit, GarageBand can convert 12-bit audio into 16-bit ..I don't know: I'm not that familiar with GarageBand, sorry..
Just try testing your 12-bit audio in iMovie or FCE first ..some people have no difficulties with it ..others do. (..That may depend on whether you're using only your original 12-bit audio, or if you're also trying to mix it with 16-bit audio within the movie project..)
However, note that - once again - Canons are quirky. Because of the way that their own particular internal 12-bit timing runs, Canon-shot 12-bit tapes may give the audio drifting out-of-sync a little on single clips which are longer than 20 minutes. This shouldn't be apparent on ordinary short clips, but if you've shot, say, a school play or concert as a single take on a Canon in 12-bit audio mode, the audio may slowly go out of sync after 20 mins, and would need correcting in iMovie or FCE.
Can't comment on Chapter Markers in DVD Studio vs. iDVD ..I've never had any trouble with Chapter Markers.
"..always max out on RAM.." ..more RAM usually helps; the most noticeable improvement is when jumping up from, say, 256MB to anything above that. The next jump, from 512MB to 1GB is also useful, but doesn't provide such a spectacular improvement as that first increase. More RAM generally means that your Mac has less need to swap things from RAM to "virtual memory" on your hard disc, so less "disc thrashing" as material's repeatedly swapped back and forth, and greater speed of access if you're not constantly reading to and writing from the hard disc.
Fitting a second internal disc, in Macs which can take it, and putting your movie project on that ..or on an external disc, as you suggest.. also relieves "disc thrashing" and any instances of the OS trying to get access to the hard disc at the same time as the movie program wants access.
To stop any delays in accessing an external hard disc, as soon as you've reformatted it to 'Mac OS Extended' (..using Disk Utility..) then open System Prefs, and the 'Spotlight' preference, and then click the 'Privacy' tab, and drag the external disc's icon into the big square space which is labelled "Prevent Spotlight from searching these locations". That stops Spotlight indexing every video clip which you import onto the external disc, which can slow down disc access.
The cup-of-tea icon is a good thing to click on every now and again, in both iMovie and FCE. That lets you stop editing for a while, have tea and biscuits, and come back refreshed, with new ideas..!

Bug: package hierarchy is displayed 'flat' in Jdeveloper navigator

Hi,
suppose I have the following package (in accordance with java's guidelines for package naming):
package nl.ibuildings.myapp.submodule;
If I add a UML Class Diagram to submodule, it's displayed correctly in the navigator:
- nl
- ibuildings
- myapp
+ submodule
But in the 'other' items in the navigator (where the object entities appear) the hierarchy is wrong:
+ nl (with no contents other then a bc4j.xcfg and a nl.xml file)
+ nl.ibuildings (same 2 files)
+ nl.ibuildings.myapp
+ nl.ibuildings.myapp.submodule
Why isn't this structure in treeform?
Greetings,
Ivo

Hi,
Whether or not packages are shown hierarchically is an option in the navigator. Select the "Sources" node (or any node below it and use the navigator toolbar:
The modelers section of the navigator currently does not support all the modes that the main part of the navigator does. This is being tracked as an ER.
Thanks,
Brian
JDev Team

Business Package - Hierarchy pop up permissions

Hi,
I am using item details iview of standrad business package in portal to create/edit master record.
When user with System Administrator role try to create/edit a master record, the pop up for selecting hierarchy field value displays the hierarchy in tree structure.
But, other users who are assigned to this iview are not able to see the hierarchy in the pop up! it is showing a portal run time error.
I have solved this problem once by modifying the permission of the item details iveiw. But now i forgot.!
Can you please help me to fix this problem?
Thanks and best regards,
Arun prabhu S

Hi Gaurav,
Thanks for your quick response!
Yes you are correct! it is due to permission
"Access denied (Object(s): com.sap.portal.system/security/com.sap.pct.mdm.appl.masteriviews/low_safety)"
How to set the permission for other users?
Thanks and best regards,
Arun prabhu S

GRIR - Criteria set in OB74 - Additional rule for automatic clearing

Hi,
WIth reference to the subject of this post, I would like to have advice on the commonly used criteria for the GRIR GL account.
My GL master for this GRIR settings: sort key is 010: Purchase order no., which is EBELN (Tcode: OB16).
Currently the settings at OB74 - Additional rule for automatic clearing for my GRIR are as follows:
DMBTR, Amt.in loc.cur.
ZUONR, Assignment
EBELN, Purchasing Document
My issue is due to the criteria of DMBTR, Amt.in loc.cur, I am not able to clear the GRIR GL using F.13, and after I remove the criteria: DMBTR, Amt.in loc.cur, I manage to clear all my GR and IR entries which have 0 balance in amount in local currency.
KIndly advise.
P/S: DO I need to have the following settings, additionally?
Define Tolerance Groups for G/L Accounts
Define Tolerance Groups for Employees
Assign Users to Tolerance Groups
Create Accounts for Clearing Differences
Thanks in advance.

Thanks.
But my problem is as follows:
My issue is due to the criteria of DMBTR, Amt.in loc.cur, I am not able to clear the GRIR GL using F.13, and after I remove the criteria: DMBTR, Amt.in loc.cur, I manage to clear all my GR and IR entries which have 0 balance in amount in local currency.
Do you have any idea?

Advices & rules for package hierarchy

Similar Messages

Maybe you are looking for