Hello Neil,
thanks a lot for your quick reply.
My final goal is to extract unitarily all documents (with their metadata) from a specific Application Group with ARSDOC command or ODWEK java library. In order to re-import them in another system.
I have also found the table sa2 which identifies (unitary document by document) each document loaded in OnDemand. In this table, there is a LOAD_ID attribute which looks to be a concatenation of AppGrpID-PRINID-SECNID-DOCNAME-Date-Date-AppID. But how to be sure this is a unique ID (in order to re-associate a physical file to its metadata) ?
By using ODWEK, when I call ODFolder.search, I get a list of ODHit and each of them is identified (I don't know where and how) by a m_docid identifier which is an alpha-numeric ~200-characters string (probably unique and non ambiguous) :
Do you know how ODWEK gets this m_docid identifier (somewhere in the DB2 database ? Or in live built ?) ?
Because, it is this same m_docid identifier which is used to download the PDF file from server :
Thanks for your clarification
------------------------------
Manuel PAITREAULT
------------------------------
Original Message:
Sent: Wed June 05, 2024 01:25 PM
From: Neil Parrott
Subject: CMOD DocumentID
Hello Manuel,
Re: I suppose there is a link between its DocId and the "physical" PDF file stored on the server FileSystem ?
No. The Unique ID Field is optional. The Unique ID field contains a 36-byte unique identification string for each row in the database. It does not provide a physical link to the file system storage.
Regards
------------------------------
Neil Parrott
Senior Product Manager
IBM Content Services
IBM Software
------------------------------
Original Message:
Sent: Tue June 04, 2024 03:52 AM
From: Manuel PAITREAULT
Subject: CMOD DocumentID
Hello everybody,
I come back on this "old" ticket as I'm interested in.
I'm using ODWEK ODApi 10.5.0.4 to search and get documents from OnDemand.
Indeed, I get an (probable) unique ID from the document with com.ibm.edms.od.ODHit.getDocId() (on 218 characters of alphanumeric string m_docid attribute), and I can get the "physical" PDF document with com.ibm.edms.od.ODHit.retrieve (com.ibm.edms.od. ArsWWWInterface.apiRetrieve) thanks to this DocId.
I suppose there is a link between its DocId and the "physical" PDF file stored on the server FileSystem ? But I don't manage to find it in the DB2 database of OnDemand:
select agid,table_name from ARSSEG where agid=17260#AGID TABLE_NAME---------- ------------------17260 FKH2
But no track of this ID in the database :
select * from FKH2 where APPID="BLABLAARC"#BDATE IDATE APPID CCUST CAGEN DOC_NAME DOC_OFF DOC_LEN COMP_OFF COMP_LEN ANNOT COMP_TYPE RESOURCE PRI_NID SEC_NID------ ----------- --------------- --------- ----- ----------- ----------- ----------- ----------- ----------- ------------- --------------------- ----------- ------- ------- -------------------------------18213 1576603980 BLABLAARC REF123-5 001 976FAAA 0 27289 0 27289 x'55' x'44' 0 2 0
Do you know where I can find it ?
Thanks in advance
Regards
------------------------------
Manuel PAITREAULT
Original Message:
Sent: Mon November 02, 2020 11:13 AM
From: Bud Paton
Subject: CMOD DocumentID
Stanislav, CMOD V10.1 added support for Unique Document ID.
Generates a unique identifier for every row loaded into OnDemand
Configurable per application group
String field of length 36 bytes
Example:
ac9223b3-f741-4608-ba92-749fda48d866
If defined, then automatically generated during load
If multiple rows to same document, each row has its own unique identifier
Previously with ERM – such rows could not participate in ERM
Only applies to new data loaded
------------------------------
Bud Paton