Hi,
I need to search for a multiple line pattern and remove it
the pattern is search for
(ln number) <TABLE name=*>
and if 3 lines below that the line is
(ln number) </TABLE>
Then remove those 4 lines.
Thank you
Try:
perl -0pe 's/\(ln number\) <TABLE name=\*>.*?\(ln number\) <\/TABLE>//s' file
Sorry that didn't work I will try and be more clearer.
the (ln number) bit is the line number no brackets. There are also bits in the text which has <TABLE name=...> but when it isn't followed 3 lines later by </TABLE> I want to keep it i only want the lines deleted which follow this exact pattern
number <TABLE name=*>
number <ROW>
number </ROW>
number </TABLE>
Thank you for all the help 
Can you provide exact sample data?
perl -0pe 's/\d+ <TABLE name=\*>.*?\d+ <\/TABLE>//s' file
want to remove
112 <TABLE name="something>
113 <ROW>
123 </ROW>
124 </TABLE>
125 <TABLE name="somethingelse>
126 <ROW>
129 </ROW>
130 </TABLE>
want to keep
493 <TABLE name="Headers_11" number="15">
494 <ROW>
495 <dest_semi_11><![CDATA[destination]]></dest_semi_11>
496 <num_semi_11><![CDATA[number sent]]></num_semi_11>
497 <cost_semi_11><![CDATA[cost]]></cost_semi_11>
498 </ROW>
499 </TABLE>
perl -0pe 's/\d+\s+<TABLE name=[^>]*>\n\d+\s+<ROW>\n\d+\s+<\/ROW>\n\d+\s+<\/TABLE>\n//g' file
Unfortunately that isn't working 
I tried cutting out the line numbers and then doing it but it still doesn't remove anything from the script
It is working for the data that you provided:
[root@rhel2 ~]# cat file
112 <TABLE name="something>
113 <ROW>
123 </ROW>
124 </TABLE>
125 <TABLE name="somethingelse>
126 <ROW>
129 </ROW>
130 </TABLE>
493 <TABLE name="Headers_11" number="15">
494 <ROW>
495 <dest_semi_11><![CDATA[destination]]></dest_semi_11>
496 <num_semi_11><![CDATA[number sent]]></num_semi_11>
497 <cost_semi_11><![CDATA[cost]]></cost_semi_11>
498 </ROW>
499 </TABLE>
[root@rhel2 ~]# perl -0pe 's/\d+\s+<TABLE name=[^>]*>\n\d+\s+<ROW>\n\d+\s+<\/ROW>\n\d+\s+<\/TABLE>\n//g' file
493 <TABLE name="Headers_11" number="15">
494 <ROW>
495 <dest_semi_11><![CDATA[destination]]></dest_semi_11>
496 <num_semi_11><![CDATA[number sent]]></num_semi_11>
497 <cost_semi_11><![CDATA[cost]]></cost_semi_11>
498 </ROW>
499 </TABLE>
Can you post output of:
cat -Te sample_file
I have just created a file with just that sample in it and the output I got was correct like you said it
however, when I use it on the larger file it doesn't seem to work :s
the line numbers go into the thousands I dont know if that makes a difference?
---------- Post updated at 01:02 PM ---------- Previous update was at 11:49 AM ----------
maybe if I give a better example with the full range of code it would be easier?
2823 <TABLE name="TotalEventSemiSum_13>
2824 </TABLE>
2827 <TABLE name="TotalEventSemiSum_22>
2828 </TABLE>
2831 <TABLE name="TotalEventSemiSum_57>
2832 <ROW>
2837 </ROW>
2838 </TABLE>
2841 <TABLE name="TotalEventSemiSum_58>
2842 <ROW>
2843 <EventSemiSumTotal_58><![CDATA[�20.40]]></EventSemiSumTotal_58>
2844 <EventSemiSumDescription_58><![CDATA[Roaming text messages non-EU]]></EventSemiSumDescription_58>
2845 <EventSemiSumTotalSent_58><![CDATA[51]]></EventSemiSumTotalSent_58>
2846 <source_58><![CDATA[07775884968]]></source_58>
2847 </ROW>
2848 </TABLE>
2851 <TABLE name="TotalEventSemiSum_16>
2852 <ROW>
2857 </ROW>
2858 </TABLE>
output needed
2841 <TABLE name="TotalEventSemiSum_58>
2842 <ROW>
2843 <EventSemiSumTotal_58><![CDATA[�20.40]]></EventSemiSumTotal_58>
2844 <EventSemiSumDescription_58><![CDATA[Roaming text messages non-EU]]></EventSemiSumDescription_58>
2845 <EventSemiSumTotalSent_58><![CDATA[51]]></EventSemiSumTotalSent_58>
2846 <source_58><![CDATA[07775884968]]></source_58>
2847 </ROW>
2848 </TABLE>
does that help?
Thank you for your continued support you are all very helpful on here
Try:
perl -0pe 's/\s+\d+\s+<TABLE name=[^>]*>\n(\s+\d+\s+<ROW>\n\s+\d+\s+<\/ROW>\n)?\s+\d+\s+<\/TABLE>//g' file
I hope you can see now that simplifying sample data not always help getting the solution 
Many thanks for all your help!
---------- Post updated at 03:12 PM ---------- Previous update was at 02:57 PM ----------
Could you let me know just one more thing, which part of that code indicates the fact that the page number precedes the text? I need to change it so the works when there is no line number and the first letter of each line is then
example
<TABLE name="something>
<ROW>
</ROW>
</TABLE>
<TABLE name="somethingelse>
<ROW>
</ROW>
</TABLE>
<TABLE name="Headers_11" number="15">
<ROW>
<dest_semi_11><![CDATA[destination]]></dest_semi_11>
<num_semi_11><![CDATA[number sent]]></num_semi_11>
<cost_semi_11><![CDATA[cost]]></cost_semi_11>
</ROW>
</TABLE>
Thanks
Try:
perl -0pe 's/\s*(\d+\s+)?<TABLE name=[^>]*>\n(\s*(\d+\s+)?<ROW>\n\s*(\d+\s+)?<\/ROW>\n)?\s*(\d+\s+)?<\/TABLE>//g' file
that didn't work sorry its needs to do the exact same thing as before but now the file no long has line numbers on it
P.S I mean that the input file has no line numbers on it, not that I want them removing when it goes to output
[root@rhel2 ~]# cat file
<TABLE name="TotalEventSemiSum_13>
</TABLE>
<TABLE name="TotalEventSemiSum_22>
</TABLE>
<TABLE name="TotalEventSemiSum_57>
<ROW>
</ROW>
</TABLE>
<TABLE name="TotalEventSemiSum_58>
<ROW>
<EventSemiSumTotal_58><![CDATA[�20.40]]></EventSemiSumTotal_58>
<EventSemiSumDescription_58><![CDATA[Roaming text messages non-EU]]></EventSemiSumDescription_58>
<EventSemiSumTotalSent_58><![CDATA[51]]></EventSemiSumTotalSent_58>
<source_58><![CDATA[07775884968]]></source_58>
</ROW>
</TABLE>
<TABLE name="TotalEventSemiSum_16>
<ROW>
</ROW>
</TABLE>
[root@rhel2 ~]# perl -0pe 's/\s*(\d+\s+)?<TABLE name=[^>]*>\n(\s*(\d+\s+)?<ROW>\n\s*(\d+\s+)?<\/ROW>\n)?\s*(\d+\s+)?<\/TABLE>//g' file
<TABLE name="TotalEventSemiSum_58>
<ROW>
<EventSemiSumTotal_58><![CDATA[�20.40]]></EventSemiSumTotal_58>
<EventSemiSumDescription_58><![CDATA[Roaming text messages non-EU]]></EventSemiSumDescription_58>
<EventSemiSumTotalSent_58><![CDATA[51]]></EventSemiSumTotalSent_58>
<source_58><![CDATA[07775884968]]></source_58>
</ROW>
</TABLE>
The problems I were having were that the at the end of each line there was a ^M which I hadn't noticed until I opened the file by chance in vi editor. Have now solved my problems.
Thanks alot for your continued patience with me