Batch command to find a replace string with line b

Question

I have data that is mostly organized in a way that I can convert and import into a spreadsheet. But certain lines have carriage returns and text that my current batch file won't use.

Good Data:

Pipers Cove × 2   $25.00
Pipers Cove Petite × 2    $25.00
Pipers Cove Plus × 2  $25.00
Nordic Club × 2   $25.00
Whiteout × 1  $12.50

Bad Data:

Pipers Cove Kids × 2
Size:
Large - ages 10 to 12
$20.00
Pipers Cove Kids × 2
Size:
Medium - ages 6 to 8
$20.00
Pipers Cove Kids × 2
Size:
Small - ages 2 to 4
$20.00

I need to remove the 2 lines starting with Size, Small, Medium, or Large and have the dollar amount follow the quantity number so my batch file can convert it to a CSV file and so on.

Answer 1

Although you did not show any own efforts, I decided to provide a solution as the task at hand appears not that trivial to me.

The following script -- let us call it clean-up-text-file.bat -- ignores only lines that begin with the words you specified. Any other lines are appended to the previous one until a $ sign is encountered, in which case a new ine is started. With this method, no lines can get lost unintentionally.

@echo off
setlocal EnableExtensions DisableDelayedExpansion

rem // Define constants here:
set _WORDS="Size","Small","Medium","Large"

for %%F in (%*) do (
    set "COLL=" & set "FILE=%%~F"
    for /F delims^=^ eol^= %%L in ('type "%%~F" ^& ^> "%%~F" rem/') do (
        set "LINE=%%L"
        (echo("%%L" | > nul find "$") && (
            setlocal EnableDelayedExpansion
            >> "!FILE!" echo(!COLL!!LINE!
            endlocal
            set "COLL="
        ) || (
            set "FLAG="
            for %%K in (%_WORDS%) do (
                (echo("%%L" | > nul findstr /I /R /B /C:^^^"\"%%~K\>") && (
                    set "FLAG=#"
                )
            )
            if not defined FLAG (
                setlocal EnableDelayedExpansion
                rem // The following line contains a TAB character!
                for /F "delims=" %%E in (^""!COLL!!LINE!    "^") do (
                    endlocal
                    set "COLL=%%~E"
                )
            )
        )
    )
)

endlocal
exit /B

To use the script, provide your text file(s) as (a) command line argument(s):

clean-up-text-file.bat "good.txt" "bad.txt"

Every specified file is modified directly, so take care when testing!

Answer 2

@ECHO OFF
SETLOCAL
SET "sourcedir=U:\sourcedir"
SET "destdir=U:\destdir"
SET "filename1=%sourcedir%\q40953616.txt"
SET "outfile=%destdir%\outfile.txt"
SET "part1="
(
FOR /f "usebackqdelims=" %%i IN ("%filename1%") DO (
 ECHO %%i|FIND "$" >NUL
 IF ERRORLEVEL 1 (
  REM $ not found - set part1 on first such line
  IF NOT DEFINED part1 SET "part1=%%i"
  ) ELSE (
  REM $ found - see whether at start or not
  FOR /f "tokens=1*delims=$" %%a IN ("%%i") DO (
   IF "%%b"=="" (
    REM at start - combine and output and reset part1
    CALL ECHO %%part1%% %%i
    SET "part1="
   ) ELSE (
    ECHO %%i
   )
  )
 )
)
)>"%outfile%"

GOTO :EOF

You would need to change the settings of sourcedir and destdir to suit your circumstances.

I used a file named q40953616.txt containing your data for my testing.

Produces the file defined as %outfile%

Scan each line of the file. If the line does not contain $ then save the first such line in part1.
Otherwise, tokenise the line. If there is only 1 token, then the $ is at the start of the line, so it needs to be output combined with part1
Otherwise, just regurgitate the line.