Shell Programming and Scripting

Hello!

I have quite a bit of map data that I have to edit. I originally had a DOS script that would reverse x1, y1 coordinates in order to change the direction of a particular segment in a map file. It worked wonderfully and all was well, but my bossman told me that there is a boatload of nodes that make up the segments I have flipped. I was not made aware of this when working with DOS.

I have flipped the segment x1, y1 - x2, y2 >> x2, y2 - x1, y1 in order to change the direction of the segments contained in the .asc file I am working with. But!! Now I have a file that contains all of the node data for each segment that exists on the map file in question. The x and y values are now correct and have been flipped so that the streets are going in the correct direction, but now the node values are backward in regards to how the streets are numerically valued(x / y). (this is Edulog data if anyone is curious).

What I need to be able to do is essentially flip the node data (like I did with the segment data)
My task is to reverse the direction of the segments(which I did in my DOS script) and also reverse the incrementation of the nodes contained within each segment. Each node has its own X1, Y1 data that increments in the direction that the segment is going in. (Like addresses going up a street, a way to represent that you are going up / down the street)

I would like to run a script that would reverse the direction (#) of the nodes:

Example of what the data looks like - The data is read from left to right.
(a more thorough example is pasted below)

Code:

23732 N 23732 N 3 Y (1) 1035678 406785 (2) 1035676 406814 (3) 1035668 406858

After the script is run the data would (ideally) look like:

Code:

23732 N 23732 N 3 Y (3) 1035668 406858 (2) 1035676 406814 (1) 1035678 406785

I think the easiest way to accomplish this is to leave the node x,y values alone and just reverse the order in which they are numbered (the # values in wrapped in ()) like I have shown above.

So far, the most node points that a segment has had is 105 Nodes, 105 X1, Y1 pairs that I would like to reverse. In the future, some files may have more than 105 X, Y coordinate values, or, more than 105 nodes).

The approach I have in mind is to have a script that will create 105 empty variable objects at execution I'm having trouble accounting for the possibility of more than 105 potential node values to process

Example of the data I am working with:

Code:

40 Y 40 N 1 N (1) 961884 641632
41 Y 41 N 1 N (1) 967487 627129
42 Y 42 N 1 N (1) 967424 627104
44 Y 44 N 1 N (1) 977911 620540
46 Y 46 N 2 N (1) 979073 620398 (2) 978884 620434
47 Y 47 N 1 N (1) 977997 620602
48 Y 48 N 4 N (1) 979093 620314 (2) 979004 620332 (3) 978913 620350 (4) 978640 620401
56 Y 56 N 1 N (1) 979284 568834
57 Y 57 N 5 N (1) 979494 568276 (2) 979231 568622 (3) 979210 568652 (4) 978921 569034

I have attached an example file of the ShapeData

Like so, for example? Give this a try:

Code:

awk '{for(i=1; i<=6; i++) print $i; for(i=NF-2; i>=7; i-=3) print $i, $(i+1), $(i+2); printf RS}' ORS=" " file

You can use sed too.
If the data don't contain _

Code:

sed 's/(/_/g;:A;s/\([^_]*\)\(.*\)_\([^_]*\)/\1(\3 \2/;tA;s/  / /g;s/  / /g;s/ $//' infile

Thank you so much for your reply! Also happy new year!
I am attempting to get this particular line of code to work but every time I run it with Command Prompt (after installing awk on this machine) it spits out a blank file. I have name script "nodesTwo" and in command prompt (in order to invoke the script) I am typing the following (Where ShapeData.txt is the file I am running the script on and ShapeDataFixed.txt is the intended output file)

Code:

awk nodesTwo ShapeData.txt>ShapeDataFixed.txt

Also!
In the line of code that you so kindly provided, there is an 'f' following the word 'print' just before RS}'
Was this a typo or is this required? is this telling the script to print file?
Thanks again! Your knowledge and experience is very much appreciated

I also tried to place ShapeData.txt>ShapeDataFixed.txt into the awk script itself and run the script but the command prompt just hangs and seems to be doing nothing...
Your beautiful example in practice (EX

Code:

awk '{for(i=1; i<=6; i++) print $i; for(i=NF-2; i>=7; i-=3) print $i, $(i+1), $(i+2); printf RS}' ORS=" " ShapeData.txt>ShapeDataFixed.txt

When I attempt to run this entire line (instead of just invoking the script inside the command prompt, I am given a (the system cannot find the file specified) despite it being inside the same directory of the currently executing script?
So sorry for the basic vibes I'm giving off! I've never used awk before this!
So any and all help is so so appreciated Thank you for your time and patience!

---------- Post updated at 08:56 AM ---------- Previous update was at 08:53 AM ----------

This is excellent! So appreciative to have another approach to this issue Linux Thank you for your suggestion and example! Both are gorgeous!
I try and run the script on a specific file in the directory of the executing script and it is giving me a "line 2: unterminated 's' command" I'm having trouble locating where the particular s command is! so many /'s and \'s!

Also!
Maybe I am not executing the script correctly? I hope I am!
I am running it with

sed -f nodes.sed

where 'nodes.sed' is the name of the file I would like the script to execute on (this file is in the same directory as the script)
Any and all help is appreciated!

What is your OS and version?

Hello!
OS: Windows 7
Version: 6.1

In that case save this as "script.awk" or a name of your choice:

Code:

BEGIN {
  ORS=FS
}
{
  for(i=1; i<=6; i++)
    print $i
  for(i=NF-2; i>=7; i-=3)
    print $i, $(i+1), $(i+2)
  printf RS
}

And execute like this:

Code:

awk -f script.awk file > outfile