Good to hear Vern;

So you can basically boil it down to something like :

with xmldata as (
select path_name , xml2clob( xmlparse (document
get_xml_file(path_name))) payload
from table (
qsys2.ifs_object_statistics(
start_path_name => '/tmp/',
subtree_directories => 'NO',
object_type_list => '*STMF'
)
)
where lower(path_name) like '%cust%.xml'
)
select path_name ,payload
from xmldata
where payload like '%ID%'
with ur;



On Tue, Aug 16, 2022 at 9:57 AM Vern Hamberg via MIDRANGE-L <
midrange-l@xxxxxxxxxxxxxxxxxx> wrote:

Hi Niels

All this is well and good but not needed for what I am doing. We already
use XMLTABLE in out programs to consume the information and put into our
application tables.

One of the main things I am interested in is to get a list of the XMLs
that contain certain kind of information - it could be about a service
done in the field, a status message from the field, or something else.
The XML has a top-level element like <Service> to identify these. The
name of the XMLs are GUIDs - not a clue about __what__ the content is
about. (In actual use, we get a list of GUIDs for each type, I believe -
I am not the person doing most of this development - so we know what
they are the first time we see them. But once they have arrived, I don't
know what type they are from the name.

So I am interested in a list of file names that are service information.
I am also interested to pull a certain value within the XML. I can see
that XMLTABLE could be useful there, but I'm not interested here in a
full shredding of the content into tables.

So using locate and substring works very well, both for the column list
and for the WHERE clause.

Again, this is a kind of ad hoc thing - I have a custom "example" in RSS
that I can use, setting values of a couple global variables, and off I go!!

I do like what you present, it is just more than I need, for now, anyhow!

Vern

On 8/16/2022 2:33 AM, Niels Liisberg wrote:
Here is my take on that:

You can pasthe the following directly into ACS - SQL script and play with
it:

___________________________

-- Lets say your XML is a representation of a simple table with rows like
this one:Examples
Select *
from qiws.qcustcdt;

-- Now lets save that table ( id and name) as a XML file to the IFS so we
have something to play with:
call ifs_write_utf8 (
path_name =>'/tmp/mycust.xml',
line => xml2clob((
Select
xmlelement( name "CUSTOMERS" ,
xmlagg(
xmlelement( name "CUSTOMER" ,
xmlelement( name "ID" , cusnum),
xmlelement( name "NAME" , trim(LSTNAM))
)
)
)
from qiws.qcustcdt
)) ,
overwrite => 'REPLACE',
end_of_line => 'NONE'
);

-- Now to your real question:
-- First it you need to look into the XML ( any XML) -you can use this
approach:
-- note the "with ur" that is "uncommitted read" - you need commit since
XML is streams
values (
xmlparse (document get_xml_file('/tmp/mycust.xml'))
) with ur;


-- You could be tempted to simply use POS or INSTR in the above to find
a
specific value.
-- However since SQL supports the access to the XML as table we can use
this much more robust way to do it:
Select *
from xmltable(
'/CUSTOMERS/CUSTOMER'
passing xmlparse (document get_xml_file('/tmp/mycust.xml'))
columns
id bigint path 'ID',
name varchar(32) path 'NAME'
)
with ur;

-- So now we have the XML data as tabular info. But you need to
-- to read it from a directory of XML files.
-- Here you can combine the CTE list of XML files
-- and join/pass each file
with XML_files as (
select path_name
from table (
qsys2.ifs_object_statistics(
start_path_name => '/tmp/',
subtree_directories => 'NO',
object_type_list => '*STMF'
)
)
where lower(path_name) like '%cust%.xml'
),
customer_from_XML_files as (
select *
from XML_files , xmltable(
'/CUSTOMERS/CUSTOMER'
passing xmlparse (
document get_xml_file(path_name)
)
columns
id bigint path 'ID',
name varchar(32) path 'NAME'
)
)
select *
from customer_from_XML_files
with ur;

-- Voila !! XML files as rows you do all the SQL magic on you can imagine






On Sun, Aug 14, 2022 at 10:59 AM Birgitta Hauser <Hauser@xxxxxxxxxxxxxxx

wrote:

Vern,

You had to use the CROSS JOIN and LATERAL because you used the IFS_READ
in
a way that it returned multiple rows.
I used Get_Clob_From_File(Path_Name) which returns a LOB-Locator on the
IFS file. LOB locators can be used like character columns. So I used it
in
the WHERE Condition in composition with the LIKE predicate.
If you only want to know in which file the data is located, using
GET_CLOB_FROM_FILE is the easiest way (no need of any CROSS JOIN or
LATERAL).
If you exactly want to see the row and the content of the row, where you
find the string, your solution is the better one.

Mit freundlichen Grüßen / Best regards

Birgitta Hauser
Modernization – Education – Consulting on IBM i


"Shoot for the moon, even if you miss, you'll land among the stars."
(Les
Brown)
"If you think education is expensive, try ignorance." (Derek Bok)
"What is worse than training your staff and losing them? Not training
them
and keeping them!"
„Train people well enough so they can leave, treat them well enough so
they don't want to.“ (Richard Branson)


-----Original Message-----
From: MIDRANGE-L <midrange-l-bounces@xxxxxxxxxxxxxxxxxx> On Behalf Of
Vern Hamberg via MIDRANGE-L
Sent: Samstag, 13. August 2022 18:17
To: midrange-l@xxxxxxxxxxxxxxxxxx
Cc: Vern Hamberg <vhamberg@xxxxxxxxxxxxxxx>
Subject: Re: Find string in multiple it's files using sql services

Thanks, Birgitta! I finally got a solution, similar to yours, might be
the
first real use I've made of LATERAL.

The problem was, I got a message from Ifs_Read that said
'/YourFolder/Dir2/Dir3' was not valid for the operation - well,
obviously,
because it is a directory!! I did not have the named parameter
Object_Type_List - now I do. Here's what I came up with using CROSS JOIN
LATERAL with Ifs_Read - and a couple global variables.

This doesn't need to be run under commitment control, although that
isn't
usually a problem, as I recall. Performance, in any case, is probably
very
dependent on how many streamed files there are and their size.

-- category: Custom
-- description: Find IFS files for 2 specific values create or replace
variable typeofxml varchar(50); set typeofxml = '<Service>'; create or
replace variable testvalue varchar(32); set testvalue = 'DGDO5778';

SELECT path_name,
line_number,
line
FROM TABLE (
qsys2.ifs_object_statistics(start_path_name =>
'/Lvl1/Lvl2/Guids', OBJECT_TYPE_LIST => '*STMF')
) a
CROSS JOIN
LATERAL (
SELECT line_number,
line
FROM TABLE (
qsys2.ifs_read(
path_name => a.path_name,
end_of_line =>
'ANY', maximum_line_length => default,
ignore_errors => 'NO')
)
) b
WHERE LOCATE(UPPER(testvalue), UPPER(line)) > 0 and
LOCATE(UPPER(typeofxml), UPPER(line)) > 0;

I do so appreciate the efforts of the database team at IBM for giving us
these really useful tools, these SQL services. Kudos, Scott!! and all!

Cheers
Vern

On 8/13/2022 4:09 AM, Birgitta Hauser wrote:
Try something like this:
Select Path_Name, Get_Clob_From_File(Path_Name) "IFS File Content"
From Table (Qsys2.Ifs_Object_Statistics(Start_Path_Name =>
'/YourFolder/Dir2/Dir3',
Subtree_Directories =>
'YES',
Object_Type_List =>
'*ALLSTMF')) x
Where Right(Trim(Path_Name), 5) = '.xml'
and Get_Clob_From_File(Path_Name) like '%Whatever you are
looking for%'
Order By Path_Name ;

Note: GET_CLOB_FROM_FILE has to be run under Commitment Control

Mit freundlichen Grüßen / Best regards

Birgitta Hauser
Modernization – Education – Consulting on IBM i


"Shoot for the moon, even if you miss, you'll land among the stars."
(Les Brown) "If you think education is expensive, try ignorance."
(Derek Bok) "What is worse than training your staff and losing them?
Not
training them and keeping them!"
„Train people well enough so they can leave, treat them well enough so
they don't want to.“ (Richard Branson)


-----Original Message-----
From: MIDRANGE-L <midrange-l-bounces@xxxxxxxxxxxxxxxxxx> On Behalf Of
Vern Hamberg via MIDRANGE-L
Sent: Samstag, 13. August 2022 00:55
To: Midrange Systems Technical Discussion
<midrange-l@xxxxxxxxxxxxxxxxxx>
Cc: Vern Hamberg <vhamberg@xxxxxxxxxxxxxxx>
Subject: Re: Find string in multiple it's files using sql services


I know how to use grep and all, this is not another chapter in that
unpublished rag, for my purposes an sql solution serves me more
directly, I
wanted to avoid a command line solution.
Cheers
Vern


-----Original Message-----

From: Jack <midrange-l@xxxxxxxxxxxxxxxxxx>
To: Midrange <midrange-l@xxxxxxxxxxxxxxxxxx>
Cc: Jack <jwoehr@xxxxxxxxxxxxxxxxxxxxxxxx>
Date: Friday, 12 August 2022 4:10 PM CDT
Subject: Re: Find string in multiple it's files using sql services

On Fri, Aug 12, 2022 at 10:38 AM Roger Harman
<roger.harman@xxxxxxxxxxx>
wrote:

As opposed to using "grep"?

The only sensible way to do it, and if you are on IBM i and not
planning to off-platform, it's half-past time to learn grep sed tr
etc. and your basic pipeline tools.

*Another chapter in the continuing saga of "Open Source Is Not An
Option, It's A Necessity".*


-----Original Message-----
From: MIDRANGE-L <midrange-l-bounces@xxxxxxxxxxxxxxxxxx> On Behalf Of
Vern Hamberg via MIDRANGE-L
Sent: Friday, August 12, 2022 8:15 AM
To: Midrange Systems Technical Discussion
<midrange-l@xxxxxxxxxxxxxxxxxx>
Cc: Vern Hamberg <vhamberg@xxxxxxxxxxxxxxx>
Subject: Find string in multiple it's files using sql services


I have XML files in a folder, I want to see which ones have a certain
value in them, such as an account code. I tried ifs_object_statistics
with start_path_name ==> set, joined to ifs_read, the path_name from
ifs_object_statistics in path_name ==>.

--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing
list To post a message email: MIDRANGE-L@xxxxxxxxxxxxxxxxxx To
subscribe,
unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives at
https://archive.midrange.com/midrange-l.

Please contact support@xxxxxxxxxxxxxxxxxxxx for any subscription
related
questions.

Help support midrange.com by shopping at amazon.com with our affiliate
link: https://amazon.midrange.com

--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing
list
To post a message email: MIDRANGE-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/midrange-l.

Please contact support@xxxxxxxxxxxxxxxxxxxx for any subscription
related
questions.

Help support midrange.com by shopping at amazon.com with our affiliate
link: https://amazon.midrange.com


--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/midrange-l.

Please contact support@xxxxxxxxxxxxxxxxxxxx for any subscription related
questions.

Help support midrange.com by shopping at amazon.com with our affiliate
link: https://amazon.midrange.com


As an Amazon Associate we earn from qualifying purchases.

This thread ...

Follow-Ups:
Replies:

Follow On AppleNews
Return to Archive home page | Return to MIDRANGE.COM home page

This mailing list archive is Copyright 1997-2024 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].

Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.