XML value need to find and replace

spatil20 · 10-16-2017, 12:03 PM

i have XML file ,in the file some of the line start with <SQL*> and </SQL>,so i need find X column in between <SQL*> and </SQL> and need to replace with trim(x) in the XML file.

I/0
<SQL><SelectStatement modified=\\'1\\' type=\\'string\\'><![CDATA[
SEL A.COLUMN_NMAE1,
A.FIND_REPLCAE
FROM TABLE_NAME
\><Tables collapsed=\\'1\\'></Tables><Parameters collapsed=\\'1\\'></Parameters><Columns collapsed=\\'1\\'></Columns></SelectStatement><SQL>

<SQL><SelectStatement modified='1' type='string'><![CDATA[ SEL A.COLUMN_NMAE1, $D)\(A) A.FIND_REPLCAE FROM TABLE_NAME]]><Tables collapsed='1'></Tables><Parameters collapsed='1'></Parameters><Columns collapsed='1'></Columns></SelectStatement></SQL>

O/P

<SQL><SelectStatement modified=\\'1\\' type=\\'string\\'><![CDATA[
SEL A.COLUMN_NMAE1,
Trim(A.FIND_REPLCAE)
FROM TABLE_NAME
\><Tables collapsed=\\'1\\'></Tables><Parameters collapsed=\\'1\\'></Parameters><Columns collapsed=\\'1\\'></Columns></SelectStatement><SQL>

<SQL><SelectStatement modified='1' type='string'><![CDATA[ SEL A.COLUMN_NMAE1, \(D)\(A) Trim(A.FIND_REPLCAE) FROM TABLE_NAME]]><Tables collapsed='1'></Tables><Parameters collapsed='1'></Parameters><Columns collapsed='1'></Columns></SelectStatement></SQL>

#!/bin/bash

for a in `sed -n '/<SQL>/,/<\/SQL>/H; /<SQL>/h; /\/SQL/{x;s/<SQL>\(.*[^\n]$\n*<\/SQL>/\1/p;}' t.txt|grep x |cut -d',' -f 1`

do
sed -n '/<SQL>/,/<\/SQL>/H; /<SQL>/h; /\/SQL/{x;s/<SQL>$.*[^\n]$\n*<\/SQL>/\1/p;}' t.txt |sed -i 's/$a/trim($a)/g'

done

ondoho · 10-16-2017, 01:49 PM

there's no question.

also, please use code tags for code.

tofino_surfer · 10-16-2017, 04:31 PM

In addition to there being no actual question there appear to be major typos and poor unreadable formatting.

SEL A.COLUMN_NMAE1,
A.FIND_REPLCAE

Should this be

SEL A.COLUMN_NAME1,
A.FIND_REPLACE

Also please put a space before and after each pipe | symbol for readability and use indenting.

Code:

#!/bin/bash

for a in `sed -n '/<SQL>/,/<\/SQL>/H; /<SQL>/h; /\/SQL/{x;s/<SQL>\(.*[^\n]\)\n*<\/SQL>/\1/p;}' t.txt | grep x | cut -d',' -f 1`
do
   sed -n '/<SQL>/,/<\/SQL>/H; /<SQL>/h; /\/SQL/{x;s/<SQL>\(.*[^\n]\)\n*<\/SQL>/\1/p;}' t.txt | sed -i 's/$a/trim($a)/g'
done

spatil20 · 10-17-2017, 12:33 AM

Simple requirement need to find the string A.FIND_REPLCAE column in <SQL>,</SQL> and replcae with trim(A.FIND_REPLCAE) in <SQL> </SQL> in XML.txt file

Input(evey time line numbers will change)
this is @12th line <SQL><SelectStatement modified=\\'1\\' type=\\'string\\'><![CDATA[
SEL A.COLUMN_NMAE1,
A.FIND_REPLCAE
FROM TABLE_NAME
\><Tables collapsed=\\'1\\'></Tables><Parameters collapsed=\\'1\\'></Parameters><Columns collapsed=\\'1\\'></Columns></SelectStatement><SQL>

this is @150 line <SQL><SelectStatement modified='1' type='string'><![CDATA[ SEL A.COLUMN_NMAE1, \(D)\(A) A.FIND_REPLCAE FROM TABLE_NAME]]><Tables collapsed='1'></Tables><Parameters collapsed='1'></Parameters><Columns collapsed='1'></Columns></SelectStatement></SQL>

my output:-
<SQL><SelectStatement modified=\\'1\\' type=\\'string\\'><![CDATA[
SEL A.COLUMN_NMAE1,
Trim(A.FIND_REPLCAE)
FROM TABLE_NAME
\><Tables collapsed=\\'1\\'></Tables><Parameters collapsed=\\'1\\'></Parameters><Columns collapsed=\\'1\\'></Columns></SelectStatement><SQL>

<SQL><SelectStatement modified='1' type='string'><![CDATA[ SEL A.COLUMN_NMAE1, \(D)\(A) Trim(A.FIND_REPLCAE) FROM TABLE_NAME]]><Tables collapsed='1'></Tables><Parameters collapsed='1'></Parameters><Columns collapsed='1'></Columns></SelectStatement></SQL>

syg00 · 10-17-2017, 01:23 AM

Removed - posted across threads.

ondoho · 10-17-2017, 01:59 AM

you need to use code tags for code.
==============================

Quote:

Originally Posted by spatil20

need to find the string A.FIND_REPLCAE column in <SQL>,</SQL> and replcae with trim(A.FIND_REPLCAE) in <SQL> </SQL> in XML.txt file

and why is the code snippet from post #1 not sufficient? show us.

but really i suspect that sed might not be the right tool for the job...
see here: https://www.linuxquestions.org/quest...on-4175615769/

TB0ne · 10-17-2017, 06:28 AM

Quote:

Originally Posted by spatil20

Simple requirement need to find the string A.FIND_REPLCAE column in <SQL>,</SQL> and replcae with trim(A.FIND_REPLCAE) in <SQL> </SQL> in XML.txt file

Input(evey time line numbers will change)
this is @12th line <SQL><SelectStatement modified=\\'1\\' type=\\'string\\'><![CDATA[
SEL A.COLUMN_NMAE1,
A.FIND_REPLCAE
FROM TABLE_NAME
\><Tables collapsed=\\'1\\'></Tables><Parameters collapsed=\\'1\\'></Parameters><Columns collapsed=\\'1\\'></Columns></SelectStatement><SQL>

this is @150 line <SQL><SelectStatement modified='1' type='string'><![CDATA[ SEL A.COLUMN_NMAE1, \(D)\(A) A.FIND_REPLCAE FROM TABLE_NAME]]><Tables collapsed='1'></Tables><Parameters collapsed='1'></Parameters><Columns collapsed='1'></Columns></SelectStatement></SQL>

my output:-
<SQL><SelectStatement modified=\\'1\\' type=\\'string\\'><![CDATA[
SEL A.COLUMN_NMAE1,
Trim(A.FIND_REPLCAE)
FROM TABLE_NAME
\><Tables collapsed=\\'1\\'></Tables><Parameters collapsed=\\'1\\'></Parameters><Columns collapsed=\\'1\\'></Columns></SelectStatement><SQL>

<SQL><SelectStatement modified='1' type='string'><![CDATA[ SEL A.COLUMN_NMAE1, \(D)\(A) Trim(A.FIND_REPLCAE) FROM TABLE_NAME]]><Tables collapsed='1'></Tables><Parameters collapsed='1'></Parameters><Columns collapsed='1'></Columns></SelectStatement></SQL>

Yes, parsing XML is fairly simple...which is why there are many Perl/CPAN modules specifically written to parse it. Have you examined any?
https://metacpan.org/search?q=xml

We're happy to help if you're stuck, but you've not posted anything you've written/done/tried on your own thus far. We aren't going to write your programs for you.