Re^2: Need your help in a pattern based text manipulation algorithm

Hello ,

Thanks a lot for your time. To add to my post, my a_copy.txt is something like this:

sjdsajdsd kjfksls'fsaf

sandlksadnksalndop
djsndsan;sa


sdsakdslkdsa

sakdsakdsa
.....................
.................


NAME|VALUE = a

ksadksad
dewidewjrw
djsdnsalkd


dskdsa;dkjsa
....................
.............................
TASK|VALUE = copy
dsakdmsald;sadsa

dsalkdsldk'dls


................................


dskmdsldsdsda
CAPS|VALUE = 34
dksdmlsajdsajdsa
..............................
dsdksld;sads

PKG_TYPE|VALUE = minimal
dsmds.dsa.d
lsdksadjsldjsdjpos
...........................
............................

..................................
jndlkjsandlksandsndsa
jdnsakjdnkjs;dsa

kdjslkdjsa;lkds

STACK|VALUE = 4
dfdskjff;sdf
.............................
........................

SHIP|VALUE = junk
kdskdsakjdl'sa
jdsajd;lksahdsakda


..........................

dsjads


.........................
[download]

And input file is user based, so its not confined to only the content you see in my example. It can have multiple values under "NAME|VALUE =" and "TASK|VALUE =" . I have to read these two keywords from the input file, and open the correct (NAME|VALUE)_(TASK|VALUE).txt. In my example, the first one happens to be "a_copy.txt". Second one happens to be "z_cut.test" and third one is "c_paste.txt". There can be many more like this based on the user input. So, I have to grep for the lines that follow "NAME|VALUE = a" and "TASK|VALUE = copy" in the input file, and replace those lines in the .txt file For example, in this case, a_copy.txt, I have to replace the lines "CAPS|VALUE = 34" with "CAPS|VALUE = 0" and "PKG_TYPE|VALUE = minimal" with "PKG_TYPE|VALUE = premium". In the c_paste.txt I have to replace "STACK|VALUE = 65" with "STACK|VALUE = 2" and "SHIP|VALUE = medium" with "SHIP|VALUE = lowtier" and so on. And you are right, since z_cut.txt file doesnt exist, the code has to throw an error in the end giving a list of all the non existent files and exit. The exit should happen only after the code has successfully run on the existing files. For example, post processing , a_copy.txt should look like:

sjdsajdsd kjfksls'fsaf

sandlksadnksalndop
djsndsan;sa


sdsakdslkdsa

sakdsakdsa
.....................
.................


NAME|VALUE = a

ksadksad
dewidewjrw
djsdnsalkd


dskdsa;dkjsa
....................
.............................
TASK|VALUE = copy
dsakdmsald;sadsa

dsalkdsldk'dls


................................


dskmdsldsdsda
CAPS|VALUE = 0
dksdmlsajdsajdsa
..............................
dsdksld;sads

PKG_TYPE|VALUE = premium
dsmds.dsa.d
lsdksadjsldjsdjpos
...........................
............................

..................................
jndlkjsandlksandsndsa
jdnsakjdnkjs;dsa

kdjslkdjsa;lkds

STACK|VALUE = 4
dfdskjff;sdf
.............................
........................

SHIP|VALUE = junk
kdskdsakjdl'sa
jdsajd;lksahdsakda


..........................

dsjads


.........................
[download]

Comment on Re^2: Need your help in a pattern based text manipulation algorithm Select or Download Code

Replies are listed 'Best First'.
Re^3: Need your help in a pattern based text manipulation algorithm by AnomalousMonk (Archbishop) on May 18, 2020 at 18:19 UTC
It's a few days since I posted and I don't know if you're still looking for support, but I thought I'd follow up and close the loop. I'd like to mention that the example input/output file pair posted here seems to contain much irrelevant data and thus to be overlong. The goal in composing such example files is to include all that is necessary (including, in this case, a bit of general text that is not subject to alteration) and little else. Please see Short, Self-Contained, Correct Example. In any event, here's a version of the full script previously posted, updated to reflect my current understanding of your requirements based upon this. As before, it is only minimally tested and I have made no attempt to provide a GUI wrapper. And if you're using a version of Perl prior to 5.10, a fix to the regex that uses the `\K` operator can easily be made. Read more... (4 kB) PerlMonks exists to provide support and assistance to Perl users at all levels of expertise and is not, in general, a free, on-line code writing service. You say that you are a Perl novice, so I have provided a fair amount of code that, in other circumstances, I would have expected you to have contributed to substantially. I have no hesitation about providing help to you, but in future if you have any qustions, please provide the code with which you are working (or at least a reference to it), and please try to provide short, pertinent example files for development and testing. Give a man a fish: `<%-{-{-{-<`	[reply] [d/l] [select]

Replies are listed 'Best First'.

Re^3: Need your help in a pattern based text manipulation algorithm
by AnomalousMonk (Archbishop) on May 18, 2020 at 18:19 UTC

It's a few days since I posted and I don't know if you're still looking for support, but I thought I'd follow up and close the loop. I'd like to mention that the example input/output file pair posted here seems to contain much irrelevant data and thus to be overlong. The goal in composing such example files is to include all that is necessary (including, in this case, a bit of general text that is not subject to alteration) and little else. Please see Short, Self-Contained, Correct Example.

In any event, here's a version of the full script previously posted, updated to reflect my current understanding of your requirements based upon this. As before, it is only minimally tested and I have made no attempt to provide a GUI wrapper. And if you're using a version of Perl prior to 5.10, a fix to the regex that uses the \K operator can easily be made.