Messages

46

Report Indexing / Re: Changing indexed report to non index report

« on: September 01, 2020, 05:41:24 AM »

Could you use the full report browse function? You could keep indexing and loading the report as you do now, but give the users Full Report Browse permission in the folder. They will then see a View Full Report button when that folder is open in the client. Select one document from the Document List, click on the View Full Report button. The first document in the report opens, they can then search within the document as normal. They will see a message to continue searching thru all segments of the report. Click on Continue All and the search string is found, no matter how many documents / segments have to be retrieved. Another plus for this technique is that it works with reports that are already loaded.

47

Report Indexing / Re: Is _*FINDONCE*_ Supported with the PDF Indexer

« on: July 28, 2020, 06:17:29 AM »

The find once function is unique to the OS400 Indexer.
I don't know of a method to accomplish the same function with the PDF Indexer.

48

Report Indexing / Re: ACIF lineData Arabic

« on: July 15, 2020, 05:46:05 AM »

Code page 850 is for Western European languages. It looks like code page 864 or 1256 would be correct for ASCII Arabic data.

49

Report Indexing / Re: PDF Indexer

« on: June 22, 2020, 05:46:47 AM »

To expand on what Justin said:

Mixing levels of PDF Indexer and Content Manager OnDemand Server is definitely not tested and you definitely won't get any help from support if you have problems.

50

Report Indexing / Re: Multiple PDF Page Piece Dictionary indexes that reference the same document

« on: June 15, 2020, 11:58:02 AM »

See the Content Manager OnDemand Newsletter for 3Q 2018 for more information on page piece dictionaries.
I think the tip titled 'Multiple Page-Piece Dictionary rows' will answer your question.
https://www.ibm.com/support/pages/node/628001

51

iSeries / Re: Retrieving spool files

« on: June 09, 2020, 06:44:19 AM »

There is no built in method to retrieve spooled file data from the server and store it in a flat (text only) file. When data is retrieved using arsdoc get or RTVDOCOND, it is assumed that the data will either be reloaded into another server or viewed using a client appropriate for the data type.

If you retrieve the data to a spooled file, you can copy it to a stream file using CPYSPLF, for example:
CPYSPLF FILE(CKS) TOFILE(*TOSTMF) JOB(006123/DBRYANT/QPRTJOB) SPLNBR(54) TOSTMF('/home/dbryant/cks.txt')
The stream file will be encoded in EBCDIC (the CCSID of your job).

If you need the stream file in ASCII, these steps will work (set the CCSID to whatever value you require). (Note that Notepad would still not display the file correctly for me, even when encoded in ASCII.)
QSH CMD('touch -C 819 /home/dbryant/cks2.txt)
CPYSPLF FILE(CKS) TOFILE(*TOSTMF) JOB(006123/DBRYANT/QPRTJOB) SPLNBR(54) TOSTMF('/home/dbryant/cks2.txt') STMFOPT(*REPLACE)

You can also convert to PDF using CPYSPLF, for example:
CPYSPLF FILE(CKS) TOFILE(*TOSTMF) JOB(006123/DBRYANT/QPRTJOB) SPLNBR(54) TOSTMF('/home/dbryant/cks.pdf') WSCST(*PDF)

More information on the using CPYSPLF to a stream file can be found at:
https://www.ibm.com/support/pages/new-cpysplf-function-copy-splfs-stream-files-ifs

In addition to arsdoc print and PRTRPTOND, there is also a PRTDOCOND command, which has more flexibility than PRTRPTOND.

52

iSeries / Re: Retrieving spool files

« on: June 08, 2020, 01:50:22 PM »

There are two methods to retrieve data stored in Content Manager OnDemand for i and put that data into an IFS file.

1) arsdoc get
2) RTVDOCOND (an IBM i user friendly front end to arsdoc get that covers all the popular arsdoc options)

Are you planning to reload the data into another system or to send it to some type of client application?

53

Tips and Tricks / Re: Is there a way to find out the load date for a specific document in ondmeand?

« on: June 05, 2020, 07:53:07 AM »

I've switched to an AG table that will result in multiple rows being returned. I'm basically getting the same error as you, with everything I try ending in the error 'Result of SELECT more than one row.' I'll keep looking at it and see if I can find a solution.

54

Tips and Tricks / Re: Is there a way to find out the load date for a specific document in ondmeand?

« on: June 04, 2020, 07:31:34 AM »

You can also find that information from the system log, although that will probably be a lot more work. You probably have multiple system log tables, SL2, SL3, etc, and maybe made a change from the old 'time stamp' to the new timestamp data type at some point. The old 'time stamp' column is TIME_STAMP and the new timestamp column is TIME_STAMP_DT. Note the leading wildcard % on the SELECT DISTINCT statement as the load id is embedded in the message text.

The SQL I used with the SL2 table and TIME_STAMP_DT is:

SELECT TIME_STAMP_DT, MSG_TEXT FROM ONDENU.SL2 WHERE MSG_TEXT LIKE
(SELECT DISTINCT '%93322-' || PRI_NID || '-'|| SEC_NID || '-' || 
LEFT(DOC_NAME, LENGTH(TRIM(DOC_NAME))-1) || '%'                   
FROM ONDENU.LNJD1  WHERE RPT_DATE = 0) AND MSG_NUM = 87           

And the output is:

TIME_STAMP_DT               MSG_TEXT                                                                                               
2020-03-31-13.48.48.087472  Application Group Load: Name(TEST-LOAN) LoadId(93322-2-0-1FAA-19940110000000-19940110000000-93347) File(/tmp/...

55

Tips and Tricks / Re: Is there a way to find out the load date for a specific document in ondmeand?

« on: June 04, 2020, 06:34:20 AM »

The system load facility was introduced in 2010 at server level 8.4.1. However, on MP and z/OS, it was optional. Also optional was using timestamps. The SA2 table also contains a column, LOAD_TIME, which I assume is used if you don't turn on timestamps. My example is from OnDemand for i, where use of the system load facility, and the use of timestamps, was automatic, the administrator did not have to make a decision to turn them on.

56

Tips and Tricks / Re: Is there a way to find out the load date for a specific document in ondmeand?

« on: June 03, 2020, 07:54:17 AM »

Here's an SQL that works for me. I'm sure some adaptation will be needed for your circumstances. Because you know the AGID Name, UGA, you can look up the AGID from the ARSAG table. In my example, my AGID Name is LNJD and my AGID is 93322. My instance name is ONDENU. The DOC_NAME column in the AG table requires the last character to be removed so that doc name 1FAAA becomes 1FAA, which is the correct name for the load ID.

SELECT LOAD_TIME_DT, LOAD_ID FROM ONDENU.SA2 WHERE LOAD_ID LIKE
(SELECT DISTINCT '93322-' || PRI_NID || '-'|| SEC_NID || '-' ||
LEFT(DOC_NAME, LENGTH(TRIM(DOC_NAME))-1) || '%'               
FROM ONDENU.LNJD1  WHERE RPT_DATE = 0)                         

The output in my example is:
LOAD_TIME_DT                LOAD_ID                                           
2020-03-31-13.48.48.087472  93322-2-0-1FAA-19940110000000-19940110000000-93347

I hope this helps.

57

Windows Client / Re: CMOD Client 10.1 on Win7 upgrading to Win10 64-bit

« on: May 13, 2020, 10:41:16 AM »

You can run both the 32-bit and 64-bit Administrator clients (and Windows end-user clients) on the same PC. So you can use the 32-bit Administrator when working with PDF (or 100 percent of the time if you choose to).

58

Windows Client / Re: Destination of the copy to file funtion

« on: May 06, 2020, 07:17:23 AM »

The destination is specified in the registry at HKEY_CURRENT_USER\Software\IBM\OnDemand32\Client\Preferences, key name COPY_PAGES_PATH.

59

iSeries / Re: retrieving sys logs from system i

« on: March 23, 2020, 06:47:37 AM »

You can also use the native IBM i command, Query Document (QRYDOCOND). It is an easier to use front end to arsdoc query and is available at v7.2 and higher. See the Content Manager OnDemand newsletter for 2Q2016 and 3Q2017. Links can be found at https://www.ibm.com/support/pages/node/628001

60

MP Server / Re: Issue with CMOD admin client, application and pdf index

« on: November 07, 2019, 09:42:05 AM »

As I understand it, the problem is that Acrobat is still a 32-bit application. But I'm definitely not the expert in this area.