grep/sed/awk - Replace all "$X" in file with new computed value "$X/10"-CodePudding

This is my original file:

#Game No : 1000693
***** 888poker Hand History for Game 1000693 *****
$1/$2 Blinds No Limit Holdem - *** 18 11 2021 10:41:44
Table DD Poker 9 Max (Real Money)
Seat 7 is the button
Total number of players : 9
Seat 1: Monroe ( $2 )
Seat 2: Mpeti ( $2 )
Seat 3: Bowen ( $2 )
Seat 4: Riccardo ( $2 )
Seat 5: Ramman ( $2 )
Seat 6: Reva ( $2 )
Seat 7: Miklos ( $2 )
Seat 9: Samlet ( $2 )
Seat 10: Geralyn ( $2 )
Hamlet posts small blind [$1]
Geralyn posts big blind [$2]
** Dealing down cards **
Monroe calls [$2]
Mpeti folds
Bowen folds
Riccardo folds
Ramman folds
Reva folds
Miklos calls [$2]
Hamlet folds
** Dealing flop ** [ As, 6s, Ah ]
** Dealing turn ** [ 8s ]
** Dealing river ** [ 4h ]
** Summary **
Monroe shows [ Jh, Td ]
Miklos shows [ Tc, Jc ]
Samlet shows [ Th, 8c ]
Geralyn shows [ Js, Ts ]
Geralyn collected [ $17 ]

For all the dollar values "$X" I need to replace them with "$X/10" (Eks. $2/10=0.2) I am able to extract the dollar values and divide them with 10 with this piped command:

grep -o '\$[[:digit:]]*' dollar.txt | sed 's/[^0-9]//g' | awk '{print $1/10}'

0.1
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.1
0.2
0.2
0.2

How can I alter the whole original file (or create a new file) so that the rest of the file consist while all the "$X" values are replaced with "$X/10"?

CodePudding user response：

If you can use perl for this, the solution will be as short as

perl -i -pe 's{\$(\d )}{($1/10)}ge' file

See the online demo. Details:

-i - inline file changing option
s{\$(\d )}{($1/10)}ge - a substitution command, where the RHS (replacement part) is a Perl expression (this is enabled with the e flag), matching all (g) occurrences on each line of a $ symbol and then one or more digits (captured into Group 1, $1), and replacing each match with the result of division of the Group 1 value by 10.

If you want to explicitly forbid to perform substitution of float values (e.g. as in $10.500), you can replace \$(\d ) with \$(\d )(?!\.?\d).

CodePudding user response：

One implementation of this, as a pure awk script (no bash required), might look like:

#!/usr/bin/env awk -f
/[$][[:digit:]] / {
  while (match($0, /[$][[:digit:]] /) > 0) {
    prefix = substr($0, 1, RSTART-1)
    suffix = substr($0, RSTART RLENGTH)
    input = substr($0, RSTART 1, RLENGTH-1)
    output = input / 10
    $0 = prefix output suffix
  }
}
{ print }

If run with ./awkscript <in.txt or awk -f awkscript <in.txt, output is:

#Game No : 1000693
***** 888poker Hand History for Game 1000693 *****
0.1/0.2 Blinds No Limit Holdem - *** 18 11 2021 10:41:44
Table DD Poker 9 Max (Real Money)
Seat 7 is the button
Total number of players : 9
Seat 1: Monroe ( 0.2 )
Seat 2: Mpeti ( 0.2 )
Seat 3: Bowen ( 0.2 )
Seat 4: Riccardo ( 0.2 )
Seat 5: Ramman ( 0.2 )
Seat 6: Reva ( 0.2 )
Seat 7: Miklos ( 0.2 )
Seat 9: Samlet ( 0.2 )
Seat 10: Geralyn ( 0.2 )
Hamlet posts small blind [0.1]
Geralyn posts big blind [0.2]
** Dealing down cards **
Monroe calls [0.2]
Mpeti folds
Bowen folds
Riccardo folds
Ramman folds
Reva folds
Miklos calls [0.2]
Hamlet folds
** Dealing flop ** [ As, 6s, Ah ]
** Dealing turn ** [ 8s ]
** Dealing river ** [ 4h ]
** Summary **
Monroe shows [ Jh, Td ]
Miklos shows [ Tc, Jc ]
Samlet shows [ Th, 8c ]
Geralyn shows [ Js, Ts ]
Geralyn collected [ 1.7 ]

CodePudding user response：

I would use GNU AWK for this task following way, let file.txt content be

#Game No : 1000693
***** 888poker Hand History for Game 1000693 *****
$1/$2 Blinds No Limit Holdem - *** 18 11 2021 10:41:44
Table DD Poker 9 Max (Real Money)
Seat 7 is the button
Total number of players : 9
Seat 1: Monroe ( $2 )
Seat 2: Mpeti ( $2 )
Seat 3: Bowen ( $2 )
Seat 4: Riccardo ( $2 )
Seat 5: Ramman ( $2 )
Seat 6: Reva ( $2 )
Seat 7: Miklos ( $2 )
Seat 9: Samlet ( $2 )
Seat 10: Geralyn ( $2 )
Hamlet posts small blind [$1]
Geralyn posts big blind [$2]
** Dealing down cards **
Monroe calls [$2]
Mpeti folds
Bowen folds
Riccardo folds
Ramman folds
Reva folds
Miklos calls [$2]
Hamlet folds
** Dealing flop ** [ As, 6s, Ah ]
** Dealing turn ** [ 8s ]
** Dealing river ** [ 4h ]
** Summary **
Monroe shows [ Jh, Td ]
Miklos shows [ Tc, Jc ]
Samlet shows [ Th, 8c ]
Geralyn shows [ Js, Ts ]
Geralyn collected [ $17 ]

then

awk 'BEGIN{FPAT="[^0-9]|([0-9]*)";OFS=""}{dollar=0;for(i=1;i<=NF;i =1){if(dollar){$i*=0.1};dollar=($i=="$")};print}' file.txt

output

#Game No : 1000693
***** 888poker Hand History for Game 1000693 *****
$0.1/$0.2 Blinds No Limit Holdem - *** 18 11 2021 10:41:44
Table DD Poker 9 Max (Real Money)
Seat 7 is the button
Total number of players : 9
Seat 1: Monroe ( $0.2 )
Seat 2: Mpeti ( $0.2 )
Seat 3: Bowen ( $0.2 )
Seat 4: Riccardo ( $0.2 )
Seat 5: Ramman ( $0.2 )
Seat 6: Reva ( $0.2 )
Seat 7: Miklos ( $0.2 )
Seat 9: Samlet ( $0.2 )
Seat 10: Geralyn ( $0.2 )
Hamlet posts small blind [$0.1]
Geralyn posts big blind [$0.2]
** Dealing down cards **
Monroe calls [$0.2]
Mpeti folds
Bowen folds
Riccardo folds
Ramman folds
Reva folds
Miklos calls [$0.2]
Hamlet folds
** Dealing flop ** [ As, 6s, Ah ]
** Dealing turn ** [ 8s ]
** Dealing river ** [ 4h ]
** Summary **
Monroe shows [ Jh, Td ]
Miklos shows [ Tc, Jc ]
Samlet shows [ Th, 8c ]
Geralyn shows [ Js, Ts ]
Geralyn collected [ $1.7 ]

Explanation: I inform GNU AWK that field is either (0 or more digits) or single any other character using FPAT. Then I iterate over fields using for loop, firstly I replace field with 0.1 of value if dollar variable holds true, then I set dollar variable to true if field is equal to $. Order is crucial, as this mean first operation is to field immediately after dollar field. After processing all fields in line I print it. I set output field separator (OFS) to empty string, to prevent injecting unwanted characters.

(tested in GNU Awk 5.0.1)

CodePudding user response：

If .2 is sufficient for 0.2, you could just put a dot before the last digit of each dollar value:

sed -E 's/(\$)([0-9]*)([0-9])/\2.\3/g

Otherwise:

sed -E '
    s/(\$[0-9]*)([0-9])/\1.\2/g
    s/(\$)(\.[0-9])/\10\2/g
    s/(\$)([0-9])/\2/g'

Put a dot before the last digit of each dollar value
Add a 0 before the dot if necessary
Remove $
This won't work if the input can contain decimal dollar values already (like $10.00)