Relatively new to SAS Enterprise Guide. I'm looking for a way to prompt the user to upload a file during the process flow execution. The file will be used in the next steps of the process flow.
Is this possible? What alternatives are available?
SAS doesn't appear to support 'native' calls to file open dialog boxes, but it does allow you to make calls to the Windows API which provides the common file picker dialog box you see everywhere.
You will need to make use of the sascbtbl filename and calls to modulec in order to do so. I'm putting this answer here as a stub for the moment and will try and expand on it next week.
In the meantime, here's an example of how to use the above to get the current process ID (PID) in windows:
filename sascbtbl "%sysfunc(pathname(work))\sascbtbl.txt" lrecl=500;
data _null_;
file sascbtbl;
infile datalines;
input;
put _infile_;
datalines4;
routine GetCurrentProcessId
minarg=0
maxarg=0
stackpop=called
callseq=byvalue
module=kernel32;
;;;;
run;
%let current_sas_process_id = %sysfunc(modulen(*e,GetCurrentProcessId));
%put ¤t_sas_process_id;
The above is the basic approach you will need to take but instead of call the GetCurrentProcessId class you will probably need to call the FileOpenDialog class or something like that (which is more complicated as it takes parameters).
Related
I have a very strange problem.
I have a standard program with the following piece of code that tries to create a file in response to a previous attempt to open it with OPEN DATASET ... FOR INPUT IN BINARY MODE.
CATCH SYSTEM-EXCEPTIONS dataset_too_many_files = 6
open_dataset_no_authority = 7
open_pipe_no_authority = 8
dataset_no_pipe = 9.
OPEN DATASET filename FOR OUTPUT IN BINARY MODE
MESSAGE msg.
ENDCATCH.
Suprisingly the response to that is sy-subrc = 8 which according to SAP documentation can happen only when OPEN DATASET is used with FILTER.
The message in msg variable has that File could not be opened, which is irrelevant because we are trying to create this file.
Did anybody experienced something like that? I suppose it has something to do with authority to create a file in a given directory on the operating system level but I cannot find any other log or trace to that. The error message and sy-subrc = 8 seem to be actually misleading in this case. Could more pieces of information be seen by activated tracing in ST01?
It turned out that the cause of the problem was in the first place the lack of the directory in which the file should be created. No wonder the system could not create the file in a non-existing folder. The error message is in such a case a misleading one anyway.
Open Dataset Docu:
and
Open datset os additions
Suprisingly the response to that is sy-subrc = 8 which according to SAP documentation can happen only when OPEN DATASET is used with FILTER.
Not exactly what the docu says. Worth another look.
Ie Would add sy-subrc = 8 on the open dataset command means
The operating system could not open the file.
I have a requirement were I need to read ABAP code written by SAP developers. I want to write my own client using Java/Python which can integrate with SAP system and get me the ABAP code.
What I understand that ABAP code is stored in SAP database like HANA, mysql etc. So is there a way which SAP provides where we can read the code like we can do in Git/SVN etc.
I've used RFC calls RPY_FUNCTIONMODULE_READ and RPY_FUNCTIONMODULE_READ_NEW through the perl NWRFC wrapper/library to retreive ABAP code.
You can access tables with below techniques:
Using SAP Connectors via RFC (RFC_READ_TABLE)
Using SOAP Web Service with same function (RFC_READ_TABLE)
Using custom web services with existing functions which are reading report, functions, etc.
You can use both Java or Pyhton for RFC, there is already exits github repo for python.
If you will select reading directly in db table, you need to know structure of saved data. It has own mechanism for OOP objects. Daniel Berlin try to implement binary parser in C++ in sap-reposrc-decompressor project. Never forget this source depended with SAP version.
I think using ADT (ABAP Development Tools) plugin is good for updated systems. There is already Eclipse plugin exists for ADT. ADT not exists in old systems.
If you are planning to use your solution in old system (after 7.01), you can build your own solution with abapGit and custom web services.
NOTE: Keep in mind, report and data elements (variables, tables, types) saved in separate tables. Dynpro objects (screens etc), reports (Smartforms) hard things to decompile.
Before you re-invent a wheel, Take a look at:
ABAPgit. https://docs.abapgit.org/
or the old SAPLink https://wiki.scn.sap.com/wiki/display/ABAP/SAPlink.
If you want JUST the source code, You could expose a very simple rest service/ Endpoint in SAP.
This service would just read the raw code and return it as plain text.
Every abaper could create this for you.
BUT is the raw source only. There is much more to a complete development
and why tools like ABAPGIT exist.
In SICF, create a new endpoint / service.
EG ZCODE_MONKEY with the class below as an example.
Now activate the service.
Call the endpoint
http://server:PORT/zcode_monkey?name=ZCODE_MONKEY
Sample implementation
CLASS zcode_monkey DEFINITION
PUBLIC
CREATE PUBLIC .
PUBLIC SECTION.
INTERFACES: if_http_extension.
ENDCLASS.
CLASS zcode_monkey IMPLEMENTATION.
METHOD if_http_extension~handle_request.
DATA: lo_src type ref to CL_OO_SOURCE,
l_name TYPE string,
l_repname type c length 30,
l_clskey type seoclskey ,
l_source type rswsourcet,
resultcode TYPE string.
FIELD-SYMBOLS: <line> TYPE LINE OF rswsourcet.
l_name = server->request->get_form_field( name = 'NAME' ).
l_clskey = l_name.
l_repname = l_name.
create OBJECT lo_src
EXPORTING
clskey = l_clskey
EXCEPTIONS
class_not_existing = 1
others = 2 .
IF sy-subrc <> 0.
read REPORT l_repname into l_source.
else.
lo_src->read( ).
lo_src->if_oo_clif_source~get_source( IMPORTING source = l_source ).
ENDIF.
LOOP AT l_source ASSIGNING <line>.
CONCATENATE resultCode
cl_abap_char_utilities=>cr_lf
<line>
INTO resultCode RESPECTING BLANKS. " always show respect ;)
ENDLOOP.
SErver->response->set_content_type( content_type = 'text/plain' ).
server->response->set_cdata( EXPORTING data = resultcode ).
server->response->set_status(
EXPORTING
code = 200
reason = 'this is a 3.50 piece of code. Dont ask...its a demo ' ).
ENDMETHOD.
ENDCLASS.
I need to write to a non-VSAM dataset in the mainframe. I know that we need to use the ZFile library to do it and I found how to do it here
I am running my Java batch job in the WebSphere Liberty on zOS. How do I specify the dataset? Can I directly give the DataSet a name like this?
dsnFile = new ZFile("X.Y.Z", "wb,type=record,noseek");
I am able to write it to a text file on the server itself using Java's File Writers but I don't know how to access a mvs dataset.
I am relatively new to the world of zOS and mainframe.
It sounds like you might be asking more generally how to use the ZFile API on WebSphere Liberty on z/OS.
Have you tried something like:
String pdsName = ZFile.getSlashSlashQuotedDSN("X.Y.Z");
ZFile zfile = new ZFile(pdsName , ...options...)
As far as batch-specific use cases, you might obviously have to differentiate between writing to a new file that's created for the first time on an original execution, as opposed to appending to an already-existing one on a restart.
You also might find some useful snipopets in this doctorbatch.io repo, along with the original link you posted.
For reference, I'll copy/paste from the ZFile Javadoc:
ZFile dd = new ZFile("//DD:MYDD", "r");
Opens the DD namee MYDD for reading
ZFile dsn = new ZFile("//'SYS1.HELP(ACCOUNT)'", "rt");
Opens the member ACCOUNT from the PDS SYS1.HELP for reading text records
ZFile dsn = new ZFile("//SEQ", "wb,type=record,recfm=fb,lrecl=80,noseek");
Opens the data set {MVS_USER}.SEQ for sequential binary writing. Note that ",noseek" should be specified with "type=record" if access is sequential, since performance is greatly improved.
One final note, another couple useful ZFile helper methods are: bpxwdyn() and getFullyQualifiedDSN().
I got many scripts. I want to be able to manage them all in 1 in script.
What I want is that the main script will activate a certain script, then when the secondary script is done, it returns a value to the main script. After that, the main script calls another secondary script, etc...
Is there a proper way to do this?
More precise question:
Is it possible to activate a AHK script from another script AHK?
At the moment, to detect that at a secondary script is complete, the way I currently use is that right before the end of the secondary script, I press a combinaison of keys that the main script will detect. And once detected, it will increase a main script variable by one and this will trigger the activation of the next script. Is there a better way to achieve this?
The main script could call the other scripts using RunWait. The scripts could then communicate back before terminating themselves.
The best option for communication would be to use OnMessage.
The following is a working example from the documentation:
; Example: Send a string of any length from one script to another. This is a working example.
; To use it, save and run both of the following scripts then press Win+Space to show an
; InputBox that will prompt you to type in a string.
; Save the following script as "Receiver.ahk" then launch it:
#SingleInstance
OnMessage(0x4a, "Receive_WM_COPYDATA") ; 0x4a is WM_COPYDATA
return
Receive_WM_COPYDATA(wParam, lParam)
{
StringAddress := NumGet(lParam + 2*A_PtrSize) ; Retrieves the CopyDataStruct's lpData member.
CopyOfData := StrGet(StringAddress) ; Copy the string out of the structure.
; Show it with ToolTip vs. MsgBox so we can return in a timely fashion:
ToolTip %A_ScriptName%`nReceived the following string:`n%CopyOfData%
return true ; Returning 1 (true) is the traditional way to acknowledge this message.
}
; Save the following script as "Sender.ahk" then launch it. After that, press the Win+Space hotkey.
TargetScriptTitle = Receiver.ahk ahk_class AutoHotkey
#space:: ; Win+Space hotkey. Press it to show an InputBox for entry of a message string.
InputBox, StringToSend, Send text via WM_COPYDATA, Enter some text to Send:
if ErrorLevel ; User pressed the Cancel button.
return
result := Send_WM_COPYDATA(StringToSend, TargetScriptTitle)
if result = FAIL
MsgBox SendMessage failed. Does the following WinTitle exist?:`n%TargetScriptTitle%
else if result = 0
MsgBox Message sent but the target window responded with 0, which may mean it ignored it.
return
Send_WM_COPYDATA(ByRef StringToSend, ByRef TargetScriptTitle) ; ByRef saves a little memory in this case.
; This function sends the specified string to the specified window and returns the reply.
; The reply is 1 if the target window processed the message, or 0 if it ignored it.
{
VarSetCapacity(CopyDataStruct, 3*A_PtrSize, 0) ; Set up the structure's memory area.
; First set the structure's cbData member to the size of the string, including its zero terminator:
SizeInBytes := (StrLen(StringToSend) + 1) * (A_IsUnicode ? 2 : 1)
NumPut(SizeInBytes, CopyDataStruct, A_PtrSize) ; OS requires that this be done.
NumPut(&StringToSend, CopyDataStruct, 2*A_PtrSize) ; Set lpData to point to the string itself.
Prev_DetectHiddenWindows := A_DetectHiddenWindows
Prev_TitleMatchMode := A_TitleMatchMode
DetectHiddenWindows On
SetTitleMatchMode 2
SendMessage, 0x4a, 0, &CopyDataStruct,, %TargetScriptTitle% ; 0x4a is WM_COPYDATA. Must use Send not Post.
DetectHiddenWindows %Prev_DetectHiddenWindows% ; Restore original setting for the caller.
SetTitleMatchMode %Prev_TitleMatchMode% ; Same.
return ErrorLevel ; Return SendMessage's reply back to our caller.
}
Well, I'm not sure why you'd want to make one script run another one... but here are a few other methods:
Include a script in another one
but, you know you can include a script inside another one, right? That is, you can use another scripts functions in your main script.
Make sure a particular script is loaded
"I got many scripts" too. Sometimes I need to make sure that a particular one is included before I can use it, so I put this at the top:
;make sure core.ahk is loaded since it is required
#include c:\ahk\core.ahk
And you don't have to worry about it getting included more than once (unless you need it) because:
#Include ensures that FileName is included only once, even if multiple inclusions are encountered for it. By contrast, #IncludeAgain allows
multiple inclusions of the same file, while being the same as #Include
in all other respects.
Now, when I include file.ahk in main.ahk, I am assured of no problems using the functions from core.ahk that file.ahk requires. And even if I include core.ahk again in main.ahk it is no worry (unless it contains subroutines instead of just functions - in which case they get run at the point where they were included, so it's best not to put subroutines in your ahk libraries).
Use good ole' RUN on Scripts
Aside from that, you know you can always use the run command to launch an ahk script. You don't have to do all that fancy WM_SENDMESSAGE stuff.
Communicate betweenst scripts using a hidden GUI
Another way for two scripts to communicate between each other is for script #1 to keep open a hidden GUI window that has an edit box and a submit button. This window will never be shown. Now, Script #2 hunts for that message box, uses send to put a string in the edit box, and then control-click to push the submit button. Now script #1 has just received input from script #2. And you don't even have to hunt for the window if you put the windows hwnd value in both scripts (so they already know it ahead of time). This works like a charm.
Tell if a script has completed
If you use ahk's run command, there is an parameter that will give you back the PID of that process (PID = Process ID). You can use this PID to check to see if the script is running, and you can use it to terminate the process.
Also, if you use runwait - the script using that command will pause and wait for the runn-ed process to complete and close before continuing.
theoretically you could also use a file object between the scripts as a sort of stdin/stdout method as when opening a file with the file object you can set it as shared.
You could also set an environment variable and pass the name of the variable to the script ,given that you have setup argument handling in the target script, which then sets the environment variable value on closing. using RunWait and this you could find out what the return result of the script is after it runs.
Lastly, look into using a function as that is probably the "best practice" for what you are trying to do. Since a function can do anything a script can do and you can pass it an array to operate on or with using ByRef on the array param. This would mean that you don't have to write in a bunch of parameters when writing the function and the variables would release memory once the function is complete, automatically. You can even write your functions in a separate file and use #Include to use them in your script.
I hope someone can help. I'm trying to build a subroutine to launch SAS from VBA. I've been doing so with SAS Workspace Manager. As I'm strolling pretty far from my comfort zone I've followed mostly what I found in these two sources. Here's what I came up with so far in VBA:
Public Sub SubmitSasProg(usrid As String, passid As String, path As String, sasprog As String, varinput As String, Optional logreturn)
Dim obWsMgr As New SASWorkspaceManager.WorkspaceManager
Dim obSAS As SAS.Workspace
Dim xmlInfo As String
Set obSAS = obWsMgr.Workspaces.CreateWorkspaceByServer("Local", VisibilityProcess, Nothing, usrid, passid, xmlInfo)
Dim obStoredProcessService As SAS.StoredProcessService
Set obStoredProcessService = obSAS.LanguageService.StoredProcessService
obStoredProcessService.Repository = "file:" & path
obStoredProcessService.Execute sasprog, varinput
If IsMissing(logreturn) Then logreturn = 100000
MsgBox obSAS.LanguageService.FlushLog(logreturn)
End Sub
And I have myself a little SAS program, let's call it "Test.sas":
%let loopTimes=3;
*ProcessBody;
data a;
do x= 1 to &loopTimes;
y=x*x*x;
output;
end;
run;
Now this line will work juste fine:
Call SubmitSasProg("myuserid", "mypassword", "somepath", "Test", "loopTimes=10")
But whenever I try to execute a SAS proceedure that modifies files/libnames etc. I'll get a either a "Invalid operation for this SAS session" or "User does not have access".
Please note that I'm using SAS locally, not on a server.
So I'm guessing that I'm not correctly logged in with my SAS worksession and lack permission. I thought that the userId and password parameters in CreateWorkspaceByServer are supposed to log me in.
So my question would be how to succesfully start the SAS session with my credentials on my local computer and have all the normal access I'd have by opening the windowed environement.
Just to clarify, this SAS process would work fine in the windowed environement:
Data _NULL_;
*x del C:\WINDOWS;
x mkdir C:\Users\Myname\Desktop\NewFolder;
run;
But it would fail with an "Invalid operation for this SAS session" code if started from VBA.
Something similar happens if I try to write SAS datasets.
I've been looking for some time now but most threads are about SAS server sessions.
Any help would be appreciated.
Well, it looks like most of my problems stem from the fact that I'm not an admin on my system.
I have managed to write datasets with the above technique given that I have write privileges on the directory with my account (duuh). Allas, the x command is stubborn and won't work lest I get myself an admin account. So, two choices to «solve» this problem:
1. Get administrator privileges.
or
2. Forget the workspace manager. Use OleObjects.
While less elegant and perhaps more time consuming upon execution (not tested) OleObject will let me use SAS to its full extent. Here is the VBA code:
Dim OleSAS As Object
Set OleSAS = CreateObject("SAS.Application")
OleSAS.Visible = True
OleSAS.Top = 1
OleSAS.Title = "Automation Server"
OleSAS.Wait = True
OleSAS.Submit(yourSAScode)
OleSAS.Quit
Set OleSAS = Nothing
If you want to run a specific process and change some macro variables as with *ProcessBody; just do OleSAS.Submit("%let "& variable_name & "=" & "yourValue") and OleSAS.Submit("%include" & your_program).
Anyway, I'm pretty sad of loosing the Log report return that I had with the Worspace Manager, it was really great for fast debugging.
Hope this was usefull.
I know that previously I've had experienced issues with SAS/Intrnet not having the required security policies setup. I can't guarantee that this will fix it, but it's somewhere to start looking:
Control Panel->Administrative Tools->Local Security Policy
Security Settings->Local Policies->User Rights Assignments->Log on as a batch job
Make sure that the account that SAS will be using to run the jobs has this right.
Also, are you saying that your test.sas program fails because it tries to write a dataset?
Also, you may be experiencing some of the same issues I've had here:
Using SAS and mkdir to create a directory structure in windows