Fixed length records

From Rosetta Code
Task
Fixed length records
You are encouraged to solve this task according to the task description, using any language you may know.
Fixed length read/write

Before terminals, computers commonly used punch card readers or paper tape input.

A common format before these devices were superseded by terminal technology was based on the Hollerith code, Hollerith code.

These input devices handled 80 columns per card and had a limited character set, encoded by punching holes in one or more rows of the card for each column.

These devices assumed/demanded a fixed line width of 80 characters, newlines were not required (and could not even be encoded in some systems).

Task

Write a program to read 80 column fixed length records (no newline terminators (but newline characters allowed in the data)) and then write out the reverse of each line as fixed length 80 column records.

Samples here use printable characters, but that is not a given with fixed length data. Filenames used are sample.txt, infile.dat, outfile.dat.

Note: There are no newlines, inputs and outputs are fixed at 80 columns, no more, no less, space padded. Fixed length data is 8 bit complete. NUL bytes of zero are allowed.

These fixed length formats are still in wide use on mainframes, with JCL and with COBOL (which commonly use EBCDIC encoding and not ASCII). Most of the large players in day to day financial transactions know all about fixed length records and the expression logical record length.

Sample data

To create the sample input file, use an editor that supports fixed length records or use a conversion utility. For instance, most GNU/Linux versions of dd support blocking and unblocking records with a conversion byte size.

Line 1...1.........2.........3.........4.........5.........6.........7.........8
Line 2
Line 3
Line 4

Line 6
Line 7
     Indented line 8............................................................
Line 9                                                                 RT MARGIN


prompt$ dd if=sample.txt of=infile.dat cbs=80 conv=block
will create a fixed length record file of 80 bytes given newline delimited text input.
prompt$ dd if=infile.dat cbs=80 conv=unblock
will display a file with 80 byte logical record lengths to standard out as standard text with newlines.


Bonus round

Forth systems often include BLOCK words. A block is 1024 bytes. Source code is stored as 16 lines of 64 characters each (again, no newline character or sequence to mark the end of a line).

Write a program to convert a block file to text (using newlines). Trailing spaces should be excluded from the output.

Also demonstrate how to convert from a normal text file to block form. All lines either truncated or padded to 64 characters with no newline terminators. The last block filled to be exactly 1024 characters by adding blanks if needed. Assume a full range of 8 bit byte values for each character.

The COBOL example uses forth.txt and forth.blk filenames.

COBOL[edit]

      *> Rosetta Code, fixed length records
*> Tectonics:
*> cobc -xj lrecl80.cob
identification division.
program-id. lrecl80.
 
environment division.
configuration section.
repository.
function all intrinsic.
 
input-output section.
file-control.
select infile
assign to infile-name
organization is sequential
file status is infile-status
.
select outfile
assign to outfile-name
organization is sequential
file status is outfile-status
.
 
data division.
file section.
fd infile.
01 input-text pic x(80).
 
fd outfile.
01 output-text pic x(80).
 
working-storage section.
01 infile-name.
05 value "infile.dat".
01 infile-status pic xx.
88 ok-input value '00'.
88 eof-input value '10'.
 
01 outfile-name.
05 value "outfile.dat".
01 outfile-status pic xx.
88 ok-output value '00'.
 
procedure division.
 
open input infile
if not ok-input then
display "error opening input " infile-name upon syserr
goback
end-if
 
open output outfile
if not ok-output
display "error opening output " outfile-name upon syserr
goback
end-if
 
*> read lrecl 80 and write the reverse as lrecl 80
read infile
perform until not ok-input
move function reverse(input-text) to output-text
 
write output-text
if not ok-output then
display "error writing: " output-text upon syserr
end-if
read infile
end-perform
 
close infile outfile
 
*> from fixed length to normal text, outfile is now the input file
open input outfile
if not ok-output then
display "error opening input " outfile-name upon syserr
goback
end-if
 
read outfile
perform until not ok-output
display function trim(output-text trailing)
read outfile
end-perform
 
close outfile
 
goback.
end program lrecl80.

Given a starting file sample.txt of

Line 1...1.........2.........3.........4.........5.........6.........7.........8
Line 2
Line 3
Line 4

Line 6
Line 7
     Indented line 8............................................................
Line 9                                                                 RT MARGIN

And a setup pass of

prompt$ dd if=sample.txt of=infile.dat cbs=80 conv=block
Output:
prompt$ cobc -xj fixed-length.cob
8.........7.........6.........5.........4.........3.........2.........1...1 eniL
                                                                          2 eniL
                                                                          3 eniL
                                                                          4 eniL

                                                                          6 eniL
                                                                          7 eniL
..............................................................8 enil fo tuO
NIGRAM TR                                                                 9 eniL
prompt$ file outfile.dat
outfile.dat: ASCII text, with very long lines, with no line terminators

blocks[edit]

Demonstrate text to Forth source block form.

      *> Rosetta Code fixed length records, text to Forth block
identification division.
program-id. blocking.
 
environment division.
configuration section.
repository.
function all intrinsic.
 
input-output section.
file-control.
select infile
assign to infile-name
organization is line sequential
file status is infile-status
.
select outfile
assign to outfile-name
organization is sequential
file status is outfile-status
.
 
data division.
file section.
fd infile.
01 input-text pic x(64).
 
fd outfile.
01 output-text pic x(64).
 
working-storage section.
01 infile-name.
05 value "forth.txt".
01 infile-status pic xx.
88 ok-input value '00'.
88 eof-input value '10'.
 
01 outfile-name.
05 value "forth.blk".
01 outfile-status pic xx.
88 ok-output value '00'.
 
procedure division.
 
*> read a line, padded to or truncated at 64 as defined in FD
open input infile
if not ok-input then
display "error opening input " infile-name upon syserr
goback
end-if
 
open output outfile
if not ok-output
display "error opening output " outfile-name upon syserr
goback
end-if
 
move 0 to tally
read infile
perform until not ok-input
move input-text to output-text
 
write output-text
if not ok-output then
display "error writing: " output-text upon syserr
end-if
 
add 1 to tally
if tally > 15 then move 0 to tally end-if
 
read infile
end-perform
 
*> Output up to next 1024 byte boundary
if tally > 0 then
compute tally = 16 - tally
move spaces to output-text
 
perform tally times
write output-text
if not ok-output then
display "error writing: " output-text upon syserr
end-if
end-perform
end-if
 
close infile outfile
 
goback.
end program blocking.

Demonstrate Forth source block to text form.

      *> Rosetta Code fixed length records, Forth blocks to text.
identification division.
program-id. unblocking.
 
environment division.
configuration section.
repository.
function all intrinsic.
 
input-output section.
file-control.
select infile
assign to infile-name
organization is sequential
file status is infile-status
.
select outfile
assign to outfile-name
organization is line sequential
file status is outfile-status
.
 
data division.
file section.
fd infile.
01 input-text pic x(64).
 
fd outfile.
01 output-text pic x(64).
 
working-storage section.
01 infile-name.
05 value "forth.blk".
01 infile-status pic xx.
88 ok-input value '00'.
88 eof-input value '10'.
 
01 outfile-name.
05 value "forth.txt".
01 outfile-status pic xx.
88 ok-output value '00'.
 
procedure division.
 
open input infile
if not ok-input then
display "error opening input " trim(infile-name) upon syserr
goback
end-if
 
open output outfile
if not ok-output
display "error opening write " trim(outfile-name) upon syserr
goback
end-if
 
*> read a fixed length line, 64 characters
read infile
perform until not ok-input
move trim(input-text) to output-text
 
write output-text
if not ok-output then
display "error writing: " output-text upon syserr
end-if
read infile
end-perform
 
close infile outfile
 
goback.
end program unblocking.

Delphi[edit]

See Pascal

Free Pascal[edit]

See Pascal

J[edit]

Solution:

Using the 720 byte input file linked to in the Perl6 entry.

   _80 ]\ fread 'flr-infile.dat'   NB. reads the file into a n by 80 array
_80 |.\ fread 'flr-infile.dat' NB. as above but reverses each 80 byte chunk
'flr-outfile.dat' fwrite~ , _80 |.\ fread 'flr-infile.dat' NB. as above but writes result to file (720 bytes)
processFixLenFile=: fwrite~ [: , _80 |.\ fread NB. represent operation as a verb/function

Example Usage:

   'flr-outfile.dat' processFixLenFile 'flr-infile.dat'  NB. returns number of bytes written
720

jq[edit]

To read the raw input and write out raw strings, the command-line options -R (raw input), -s ("slurp"), and -r (raw output) options are necessary:

   jq -Rrs -f program.jq infile.dat

program.jq:

For the sake of generality, we define `cut` with an argument corresponding to the number of columns. To illustrate that an efficient recursive definition is possible, we define an inner helper function, `c`, with arity 0, as currently jq's tail-call optimization is restricted to zero-arity filters:

def cut($n):
def c: if length==0 then empty else .[:$n] , (.[$n:] | c) end;
c;
 
cut(80) | explode | reverse | implode;

Julia[edit]

The program reads from an "infile.dat" file created with dd, as described in the task instructions.

 
function processfixedlengthrecords(infname, blocksize, outfname)
inf = open(infname)
outf = open(outfname, "w")
filedata = [ read(inf, blocksize) for _ in 1:10 ]
 
for line in filedata
s = join([Char(c) for c in line], "")
@assert(length(s) == blocksize)
write(outf, s)
end
end
 
processfixedlengthrecords("infile.dat", 80, "outfile.dat")
 

Go[edit]

package main
 
import (
"fmt"
"log"
"os"
"os/exec"
)
 
func reverseBytes(bytes []byte) {
for i, j := 0, len(bytes)-1; i < j; i, j = i+1, j-1 {
bytes[i], bytes[j] = bytes[j], bytes[i]
}
}
 
func check(err error) {
if err != nil {
log.Fatal(err)
}
}
 
func main() {
in, err := os.Open("infile.dat")
check(err)
defer in.Close()
 
out, err := os.Create("outfile.dat")
check(err)
 
record := make([]byte, 80)
empty := make([]byte, 80)
for {
n, err := in.Read(record)
if err != nil {
if n == 0 {
break // EOF reached
} else {
out.Close()
log.Fatal(err)
}
}
reverseBytes(record)
out.Write(record)
copy(record, empty)
}
out.Close()
 
// Run dd from within program to write output.dat
// to standard output as normal text with newlines.
cmd := exec.Command("dd", "if=outfile.dat", "cbs=80", "conv=unblock")
bytes, err := cmd.Output()
check(err)
fmt.Println(string(bytes))
}
Output:
8.........7.........6.........5.........4.........3.........2.........1...1 eniL
                                                                          2 eniL
                                                                          3 eniL
                                                                          4 eniL

                                                                          6 eniL
                                                                          7 eniL
............................................................8 enil detnednI
NIGRAM TR                                                                 9 eniL

Bonus round

package main
 
import (
"bufio"
"fmt"
"log"
"os"
"strings"
)
 
func check(err error) {
if err != nil {
log.Fatal(err)
}
}
 
func block2text(inputFile, outputFile string) {
in, err := os.Open(inputFile)
check(err)
defer in.Close()
 
out, err := os.Create(outputFile)
check(err)
defer out.Close()
 
line := make([]byte, 64)
empty := make([]byte, 64)
for {
n, err := in.Read(line)
if err != nil {
if n == 0 {
break // EOF reached
} else {
log.Fatal(err)
}
}
str := string(line)
str = strings.TrimRight(str, " \000")
out.WriteString(str + "\n")
copy(line, empty)
}
}
 
func text2block(inputFile, outputFile string) {
in, err := os.Open(inputFile)
check(err)
defer in.Close()
 
out, err := os.Create(outputFile)
check(err)
defer out.Close()
 
scanner := bufio.NewScanner(in)
count := 0
for scanner.Scan() {
str := scanner.Text()
count++
le := len(str)
if le > 64 {
str = str[0:64]
} else if le < 64 {
str = fmt.Sprintf("%-64s", str)
}
out.WriteString(str)
}
if rem := count % 16; rem > 0 {
str := strings.Repeat(" ", (16-rem)*64)
out.WriteString(str)
}
}
 
func main() {
block2text("block.dat", "block.txt")
text2block("block.txt", "block2.dat")
}

M2000 Interpreter[edit]

Fixed length Record[edit]

Form Buffer object (hold a memory block) we can read a byte at offset. So Print Eval(Line80,3) return the 4th byte, and Return Line80, 2:=255,5:=0 set two bytes (unsigned, we place any number and interpreter convert it to byte).

 
Module FixedFile {
Read fixed$
OldLocale=Locale
\\ chr$(string_argument$)
\\ use Locale to convert from Ansi to Utf-16LE
\\ Read Ansi form files also use Locale
Locale 1032
Try ok {
\\ Make the file first
Const Center=2
Font "Courier New"
Bold 0
Italic 0
Def long m, z=1, f
Def text2read$,test3write$
Form 100, 50 ' 100 by 60 characters
Document txt$={Line 1...1.........2.........3.........4.........5.........6.........7.........8
Line 2
Line 3
Line 4
 
Line 6
Line 7
Indented line 8............................................................
Line 9 RT MARGIN
}
\\ use Help Open in M2000 console for details
\\ Method one
Report Center, "Make file"
\\ for WIDE Random \\ for Utf-16
Open fixed$ for Random Exclusive as #f len=80
m=Paragraph(txt$, 0)
z=1
If forward(txt$, m) then
while m, z<10
text2write$=Paragraph$(txt$,(m))
Print format$("Len:{0}, Data: {1}",Len(text2write$),text2write$)
Put #f, text2write$ , z
\\ record number from 1
\\ if number is total records plus one
\\ we append a record
z++
End while
End If
Print "Press any key"
Push Key$ : Drop
Form 80, 40
Report Center, "Method1"
For z=1 to 9
Get #f, text2read$, z
text2read$=StrRev$(text2read$)
Put #f, text2read$, z
Print text2read$
Next z
Close #f
Report Center, "Method2"
\\ Method2
\\ Buffer Clear Line80 ... \\ to clear memory
\\ here we write all bytes so not needed
Buffer Line80 as byte*80
m=filelen(fixed$)
If m mod 80=0 Then
m=1
\\ now Get/Put read write at byte position
\\ we have to use seek to move to byte position
\\ This way used for Binary files
Open fixed$ for Input as #f1
Open fixed$ for Append as #f2
while not eof(#f1)
seek #f1, m
Rem Print seek(#f)
Get #f1, Line80
Return line80,0:=Str$(StrRev$(Chr$(Eval$(line80,0,80))))
seek #f2, m
Put #f2, Line80
seek #f1, m
Get #f1, Line80
Print Chr$(Eval$(line80,0,80))
m+=80
End While
Close #f1
Close #f2
End if
}
\\ use Close with no parameters for close all files if something happen
If error then Close: Print Error$
Locale OldLocale
}
FixedFile "fixed.random"
 
 

Forth Blocks[edit]

 
Form 80,50
Print "Forth's Blocks"
\\ Forth Blocks
Structure Line16 {
a`Line as byte*64
}
NewBlock=lambda Line16 -> {
Buffer a`Block as Line16*16
\\ fill spaces
Return a`Block, 0:=str$(string$(" ",1024))
=a`Block
}
\\ Events are value types, but for closures and groups are reference types
Event Doit {
Read something$
}
Header=Doit
DisplayBlock= Lambda NewBlock, Line16, Doit, Header (Blocks`File$,Block`Number, UseLocale=1033)->{
Page1=NewBlock()
Open Blocks`File$ for input as #f
Seek #f, 1024*(Block`Number-1)+1
Get #f,Page1
Close #f
Document NewDoc$
\\ need to convert from Ansi
Call Event Header, "Block:"+Str$(Block`Number)
oldlocale=locale
locale UseLocale
For i=0 to 15
lineAny$=chr$(Eval$(Page1,i, Len(Line16)))
Call Event Doit, format$("{0::-2} {1}",i,chr$(Eval$(Page1,i, Len(Line16))))
Next i
locale oldlocale
}
Document ForthCode$={( Large letter F)
 : STAR [CHAR] * EMIT ;
 : STARS 0 DO STAR LOOP ;
 : MARGIN CR 30 SPACES ;
 : BLIP MARGIN STAR ;
 : BAR MARGIN 5 STARS ;
 : F BAR BLIP BAR BLIP BLIP CR ;
}
\\ Make Document bigger than 16 lines
\\ doc.par(ForthCode$) return paragraphs (here we have no wrap)
\\ actuall lines per layer can be found from Report (the renderer)
\\ using Reportlines. layer can be the printer page.
ForthCode$=string$(ForthCode$,5)
Print "Make Block"
Page1=NewBlock()
Locale 1033
Blocks`File$="Forth Blocks"
Block`Number=1
\\ Apppend three times same blocks
For Pass=1 to 3
If Doc.Len(ForthCode$)>0 then
For i=1 to Doc.par(ForthCode$)-1
\\ we give order number but Paragraph$ use unique number for paragraphs
\\ if we didn't delete or insert lines, then these two are the same
Print Paragraph$(ForthCode$, Paragraph(ForthCode$,i))
\\ convert to Ansi using Locale
\\ offset from 0, so minus 1
\\ offset 1 is at len(Line16)
\\ Page1(0) is the real address, but here doesn't matter
Return Page1, (i-1) mod 16:=Str$(Paragraph$(ForthCode$, Paragraph(ForthCode$,i)))
if i mod 16=0 then Gosub SaveBlock
Next i
i--
if Not i mod 16=0 then Gosub SaveBlock
End if
Next Pass
\\ now we read from disk
Class DocumentKeeper {
Document Text$
Function AppendLine(aline$) {
\\ right trim$
.Text$<=Mid$(Trim$("*"+aline$),2)+{
}
}
}
Function Disp(aline$) {
Print aline$
}
DocumentKeeper=DocumentKeeper()
Event Doit New Disp()
Event Header New Disp()
For i=1 to Block`Number-1
Call DisplayBlock(Blocks`File$, i)
Print "Press any key"
Push key$ : Drop
Next i
Event Doit Drop Disp()
Event Doit New DocumentKeeper.AppendLine()
Event Header Hold
For i=1 to Block`Number-1
Call DisplayBlock(Blocks`File$, i)
Next i
Report DocumentKeeper.Text$
 
End
SaveBlock:
Print "Save as Number ";Block`Number
If Exist(Blocks`File$) then
\\ check if there is space for this block
If Not filelen(Blocks`File$) div 1024>=Block`Number-1 Then
Error "Wrong Block Number"
End if
Else
Print "not exist"
Open Blocks`File$ for output as #f
Close #f
Wait 100
\\ or Error "Empty File" if we wish only read
If Block`Number<>1 then Error "Wrong Block Number"
End if
Open Blocks`File$ for append as #f
Seek #f, Block`Number*1024-1023 ' so we seek to first byte
Put #f, Page1
Close #f
Block`Number++
Page1=NewBlock()
Return
 

Neko[edit]

/**
fixed length records, in Neko
*/

 
var LRECL = 80
 
var reverse = function(s) {
var len = $ssize(s)
if len < 2 return s
 
var reverse = $smake(len)
var pos = 0
while len > 0 $sset(reverse, pos ++= 1, $sget(s, len -= 1))
return reverse
}
 
var file_open = $loader.loadprim("[email protected]_open", 2)
var file_read = $loader.loadprim("[email protected]e_read", 4)
var file_write = $loader.loadprim("[email protected]_write", 4)
var file_close = $loader.loadprim("[email protected]_close", 1)
 
var input = file_open("infile.dat", "r")
var output = file_open("outfile.dat", "w")
 
var len
var pos = 0
var record = $smake(LRECL)
 
while true {
try {
len = file_read(input, record, pos, LRECL)
if len != LRECL $throw("Invalid read")
 
len = file_write(output, reverse(record), pos, LRECL)
if len != LRECL $throw("Invalid write")
} catch a break;
}
file_close(input)
file_close(output)
Output:
prompt$ dd if=infile.txt cbs=80 conv=block status=none of=infile.dat
prompt$ nekoc fixed-length.neko
prompt$ neko fixed-length
prompt$ dd if=outfile.dat cbs=80 conv=unblock status=none
8.........7.........6.........5.........4.........3.........2.........1...1 eniL
                                                                          2 eniL
                                                                          3 eniL
                                                                          4 eniL

                                                                          6 eniL
                                                                          7 eniL
............................................................8 enil detnednI
NIGRAM TR                                                                 9 eniL

Pascal[edit]

program reverseFixedLines(input, output, stdErr);
const
lineLength = 80;
var
// in Pascal, `string[n]` is virtually an alias for `array[1..n] of char`
line: string[lineLength];
i: integer;
begin
while not eof() do
begin
for i := 1 to lineLength do
begin
read(line[i]);
end;
 
for i := lineLength downto 1 do
begin
write(line[i]);
end;
writeLn();
end;
end.

Perl[edit]

open $in,  '<', 'flr-infile.dat';
open $out, '>', 'flr-outfile.dat';
 
while ($n=sysread($in, $record, 80)) { # read in fixed sized binary chunks
syswrite $out, reverse $record; # write reversed records to file
print reverse($record)."\n" # display reversed records, line-by-line
}
close $out;
Output:
8.........7.........6.........5.........4.........3.........2.........1...1 eniL
                                                                          2 eniL
                                                                          3 eniL
                                                                          4 eniL
                                                                                
                                                                          6 eniL
                                                                          7 eniL
............................................................8 enil detnednI     
NIGRAM TR                                                                 9 eniL

input file and output file

Perl 6[edit]

Link to a copy of the input file used: flr-infile.dat

Essentially the same as task Selective_File_Copy except more boring.

$*OUT = './flr-outfile.dat'.IO.open(:w, :bin) orelse .die; # open a file in binary mode for writing
while my $record = $*IN.read(80) { # read in fixed sized binary chunks
$*OUT.write: $record.=reverse; # write reversed records out to $outfile
$*ERR.say: $record.decode('ASCII'); # display decoded records on STDERR
}
close $*OUT;
Output:
8.........7.........6.........5.........4.........3.........2.........1...1 eniL
                                                                          2 eniL
                                                                          3 eniL
                                                                          4 eniL
                                                                                
                                                                          6 eniL
                                                                          7 eniL
............................................................8 enil detnednI     
NIGRAM TR                                                                 9 eniL

Phix[edit]

You might want to investigate using builtins such as get_text() and read_lines(), but they were never really designed for fixed length record handling and for me, for this task, their limits outweight any advantages. To simplify matters, this creates any files needed on the fly.

constant sample_text = """
Line 1...1.........2.........3.........4.........5.........6.........7.........8
Line 2
Line 3
Line 4
 
Line 6
Line 7
Indented line 8............................................................
Line 9 RT MARGIN"""
 
constant indat = "infile.dat",
outdat = "outfile.dat"
 
if not file_exists(indat) then
object text = get_text("sample.txt")
if text=-1 then text = sample_text end if
sequence s = split(text,'\n')
integer fn = open(indat,"wb")
for i=1 to length(s) do
puts(fn,s[i]&repeat(' ',80-length(s[i])))
end for
close(fn)
printf(1,"%s created (%d bytes)\n",{indat,get_file_size(indat)})
end if
 
function get_block(integer fn, size)
string res = ""
for i=1 to size do
res &= getc(fn)
end for
return res
end function
 
integer fn = open(indat,"rb"),
isize = get_file_size(indat)
puts(1,"reversed:\n")
for i=1 to isize by 80 do
 ?reverse(get_block(fn,80))
end for
close(fn)
 
-- Bonus part, 16*64 (=1024) block handling:
 
procedure put_block(integer fn, sequence lines)
lines &= repeat("",16-length(lines))
for i=1 to length(lines) do
string li = lines[i]
if length(li)>64 then
li = li[1..64]
else
li &= repeat(' ',64-length(li))
end if
puts(fn,li)
end for
end procedure
 
fn = open(outdat,"wb")
put_block(fn,split(sample_text,'\n'))
close(fn)
integer osize = get_file_size(outdat)
printf(1,"\n%s created (%d bytes):\n",{outdat,osize})
fn = open(outdat,"rb")
sequence lines = {}
for i=1 to osize by 64 do
string line = get_block(fn,64)
 ?line
lines = append(lines,trim_tail(line,' '))
end for
lines = trim_tail(lines,{""})
puts(1,"\ntrimmed:\n")
pp(lines,{pp_Nest,1})
{} = delete_file(indat) -- (for consistent 2nd run)
Output:
infile.dat created (720 bytes)
reversed:
"8.........7.........6.........5.........4.........3.........2.........1...1 eniL"
"                                                                          2 eniL"
"                                                                          3 eniL"
"                                                                          4 eniL"
"                                                                                "
"                                                                          6 eniL"
"                                                                          7 eniL"
"............................................................8 enil detnednI     "
"NIGRAM TR                                                                 9 eniL"

outfile.dat created (1024 bytes):
"Line 1...1.........2.........3.........4.........5.........6...."
"Line 2                                                          "
"Line 3                                                          "
"Line 4                                                          "
"                                                                "
"Line 6                                                          "
"Line 7                                                          "
"     Indented line 8............................................"
"Line 9                                                          "
"                                                                "
"                                                                "
"                                                                "
"                                                                "
"                                                                "
"                                                                "
"                                                                "

trimmed:
{"Line 1...1.........2.........3.........4.........5.........6....",
 "Line 2",
 "Line 3",
 "Line 4",
 {},
 "Line 6",
 "Line 7",
 "     Indented line 8............................................",
 "Line 9"}

REXX[edit]

/*REXX pgm reads fixed-length 80 byte records;  reverses each record, displays to term. */
iFID= 'FIXEDLEN.TXT' /*the file's filename (used for input).*/
call charin iFID, 1, 0 /*open the file, point rec pointer to 1*/
/* [+] just to be safe, position file.*/
do j=1 while chars(iFID)>=80 /*read data records of LRECL ≡ eighty. */
@.j= charin(iFID, , 80) /*read a data record of eighty bytes. */
end /*j*/
#= j - 1 /*adjust # of records (J is 1 too high)*/
do k=1 for # /* [+] process all the records read. */
say reverse(@.k) /* reverse a record and write to term. */
end /*k*/ /*stick a fork in it, we're all done. */
output   when using the default input file:
Output:
8.........7.........6.........5.........4.........3.........2.........1...1 eniL
                                                                          2 eniL
                                                                          3 eniL
                                                                          4 eniL

                                                                          6 eniL
                                                                          7 eniL
............................................................8 enil detnednI
NIGRAM TR                                                                 9 eniL

zkl[edit]

Reading from the dd formatted file, which is 720 characters:

Line 1...1.........2.........3.........4.........5.........6.........7.........8Line 2                                                                          Line 3                                                                          Line 4                                                                                                                                                          Line 6                                                                          Line 7                                                                               Indented line 8............................................................Line 9                                                                 RT MARGIN
File("infile.dat","rb")	// could contain nulls and newlines
// Since we are writing to a ASCII terminal, ignore nulls
.walker(3).chunk(80,String).pump(Console.println,"reverse"); // 3-->read chars
Output:
8.........7.........6.........5.........4.........3.........2.........1...1 eniL
                                                                          2 eniL
                                                                          3 eniL
                                                                          4 eniL
                                                                                
                                                                          6 eniL
                                                                          7 eniL
............................................................8 enil detnednI     
NIGRAM TR                                                                 9 eniL

To write to a file (as one big line), preserving nulls:

in,out := File("infile.dat","rb"), File("outfile.dat","wb");
in.walker(0).chunk(80).pump(out,"reverse"); // may contain nulls and newlines
// 0-->read bytes, chunk to list of bytes, reverse and write the bytes
in.close(); out.close();

outfile.dat:

Output:
8.........7.........6.........5.........4.........3.........2.........1...1 eniL                                                                          2 eniL                                                                          3 eniL                                                                          4 eniL                                                                                                                                                          6 eniL                                                                          7 eniL............................................................8 enil detnednI     NIGRAM TR                                                                 9 eniL

Bonus Round

// read block file (as in read a file of blocks) to text
fcn readFourthBlock(inFileNm){
out,f,buf := Sink(String), File(inFileNm,"rb"), Data(1024);
while(f.read(1024,buf,False)){
// read 64 chars from buf, strip ws from right side, repeat
buf.walker(3).chunk(64,String).pump(out,T("strip",1),'+("\n"));
}
f.close(); out.close();
}
// read text file and write as block to file
fcn formatToFourthBlock(inFileNm,outFileNm){
n,blk,in,out := 0, Data(), File(inFileNm,"r"), File(outFileNm,"wb");
foreach line in (in.walker(11)){ // right side stripped
if(not line) blk.write(" "*64);
else blk.write(line.walker().chunk(64,String).pump(String,"%-64s".fmt));
if(blk.len()==1024){ out.write(blk); blk.clear(); }
}
if(blk) out.write(blk, " "*(1024 - blk.len()));
f.close(); out.close();
}