DataMystic

Hi guys,
I have a simple text files with say 100 lines of data (varies in size each time)
say line 5 starts with lineA: yadda yadda....
and line 40 starts with lineB: yadda yadda...
I want to delete all of lines 6 through 39 inclusive..
just leaving lines 1-5 and 40-100 in the file...
now the next file may have different data and line numbers so a simple line count won't cut it. will always be lineA: and lineB: to identify the lines positions though!
How can I trim this using TextPipe..
thanks in advance
AJ

Just use perl Pattern:

(lineA: [^\r\n]*?).*(lineB: )

Replace with

$1$$2

I have a txt file with the following sample..

details: ssssssss
whatever...
Fault: xxxxxx
xxxxxxxxxxx
xxxxxxxxxx
<<yadda yadda>>
yyyy
yyy
<<yippee yippee>>
yyyyyyyyyy
yy
<< more yadda>>
<< even more>>
zzzzzzzzz
zzzzzzzz
zzzz
Details:
more whatever...
more...
Fault:
next lot of txt of the same format!
all I want to keep is:

Details: ssssss
whatever...
Fault: xxxxxx
xxxxxxxxxxx
xxxxxxxxxx
zzzzzzzzz
zzzzzzzz
zzzz

Details: ssssss
whatever...
Fault: xx
xxxxxxxxxxxxxxxxxx
xxxxxx
xxx
zz
zzzzzzz

Details: ssssss
whatever...
Fault: xxxxxxxx
and so on...

So to get rid of everything in between, want to do the same for each subsequent Details: and following data...
there's sometimes several loads of yyyy data seperated with
<< yadda>> lines but there is always a << yadda>> the line preceeding the zzzzzz data which may be 1 or several lins of txt.
xxxxx may be 1 or several lines of txt too but always starts on the Fault: line and has a <<yadda yadda>> on the line directly after the last line of xxxx (also yadda yadda is different in each <<>>)

What I can see.. is to:
locate line with Fault: on it.
KEEP that line and subsequent lines(if any) down to the first <<
and
locate line with Details: on it.
and keep every line above it up to the first line with a >> on it.(going upwards that is)
delete all lines between this lot

hope this is not too cryptic to follow..
thanks
AJ

Please show exactly what the output should be.

all I want to keep is:

Details: ssssss
whatever...
Fault: xxxxxx
xxxxxxxxxxx
xxxxxxxxxx
zzzzzzzzz
zzzzzzzz
zzzz

Details: ssssss
whatever...
Fault: xx
xxxxxxxxxxxxxxxxxx
xxxxxx
xxx
zz
zzzzzzz

Details: ssssss
whatever...
Fault: xxxxxxxx
and so on...

Here tiz...
Sample Original file:

Details: BC884164 33600036 19/03/07 09:19 223181
Serial unit: dggherhertyere
rtyrtyrtytyyrrty
Resolution: frhgrtyhrtyerthyrethyer
eryrtyretyertyertyeryt
Fault: ghdfhdfhgdfhghdhdhdfhgdhdfhgdfhdfh
<<19/03/2007 ST 11:39 fgfdgdf>>
fgsdfjghofdghothperth
okfghoithir
<<19/03/2007 ST 11:40 fgfdgdf>>
rthretherherht
rterrtrtyeryt
<<20/03/2007 ST 12:39 fgfdgdf>>
Required last work done on this call
Maybe 1 or several lines long though?

Details: AC884164 33600136 19/03/07 09:19 223182
Serial unit: dggherhertyere
rtyrtyrtytyyrrty
Resolution: frhgrtyhrtyerthyrethyer
...
and so on for dozens of logs.

call starts at Details: and finishes before the next Details: but I need to capture the lines after the last <<xxx>> of the ticket..
there's varying lines of detail and lots of other guff that I have already stripped out, just down to this detail now but can't work out how to strip out the unrequired lines.
desired output for this example is below:


Details: BC884164 33600036 19/03/07 09:19 223181
Fault: ghdfhdfhgdfhghdhdhdfhgdhdfhgdfhdfh
Required last work done on this call
Maybe 1 or several lines long though?

Details: AC884164 33600136 19/03/07 09:19 223182
Fault: 2222222222hjhgjhg2jh22h2jh22
Required last work done on this call
Maybe 1 or several lines long though?
3 in this ticket!!
...
and so on for dozens of logs.

So far I have stripped out lots of other fields and put 'Details:' at the bottom of the file as I can see that I need a starting and finishing delimiter
this would be the LAST '>>' of the ticket and 'Details:' of the following 1.
problem is there are varying numbers of <<xxxxxxx>> in each ticket so pinpointing the last one is going to be tricky eh.
hope this is clearer?
thanks
AJ

Below illustrates the general technique. You will need your thinking cap on!

1. Find EasyPattern:
[ capture( 'Details: ', 1+ not cr or lf ), cr, lf, 1+ chars, capture('Fault:') ]
Replace with
\x00$1
$2\x01

This inserts \x00 and \x01 as markers which need to be removed later.


2. Now we need to mark the last << >> section after each heading:

EasyPattern:
[ '>>', longest 1+ not ascii(00) ]
Replace with
\x02$0

3. Now remove the preceding unwanted << >>:

Find EasyPattern:
[ ascii(01), 1+ not ascii(02) ]
Replace with nothing.

Sample output...
Fault Notes : archiving problems with main system
<<19/03/2007 ST 11:19 TZ 11:19 ab99665>>
contact caller
<<19/03/2007 ST 11:39 TZ 4WA 09:39 ld76340>>

second process seems to have put the control char after the first occurrence of >> and not the last.
anyway...
maybe...
If we could have some code to do just this..
locate "Fault Notes:"
locate "blank line"

now
I want to keep the line with "Fault Notes:" on it and subsequent lines down to the first line starting with a "<<"
I also want to keep all lines from the last line ending with a ">>" up to the blank line.

eg:
Fault Details:
terte
erert
Fault Notes: dfdfff
xxxxxxx
<<grethert>>
rthr
rtr
<<dfg>>
<<fgt>>
yyyyyyy
yyyyyyy

Fault Details:

result is

Fault Details:
terte
erert
Fault Notes: dfdfff
xxxxxxx
yyyyyyy
yyyyyyy

Fault Details: next ticket here...

This needs to be done for each ticket in the txt file.
is this easier to do maybe?
AJ

Have you purchased? If not, we can provide help - please drop us an email.