I'm using free Pascal 2.6.4 and I created this code. It's a program that asks for number which represents line in file. Everything works except one thing. When I want to display 1. line it stops with "exitcode 217". Why?
Program FileTruncate;
uses
SysUtils;
label znova;
const
filename = 'C:\Users\KVIKY\Desktop\Pascal\Projects\FileHandling\test.txt';
var
myfile: text;
line: string;
counter:integer;
position:double;
begin
znova:
Writeln('Zadaj cislo riadku: ');
Readln(position);
if position=0 then exit;
if position>26 then exit;
Assign(myfile, filename);
Reset(myfile);
counter:=0;
Repeat
inc(counter);
readln(myfile);
until counter = position-1;
readln(myfile, line);
Close(myfile);
writeln(line);
Writeln('Stlacte enter pre pokracovanie.');
Writeln('Zadajte 0 pre ukoncenie programu.');
readln;
goto znova;
end.
Because when position is 1, your repeat-until condition will never be met (because your counter=1 and position-1 is zero, so counter = position-1 will never happen).
counter represents the line before the target, so...
Instead, you could initialize differently:
counter := -1
or, better, change your loop to a while-do:
while counter < position-1 do
because you have a readln within the loop
Related
I keep wanting to do something like this:
my $block := {
state $n = 0;
say $n++;
last if $n > 3;
};
loop $block;
Or even:
$block.loop;
I'm not expecting that this is possible but it would sure be cool if it was.
How would I find out where a particular routine comes from?
$ perl6
To exit type 'exit' or '^D'
> &loop.^name
===SORRY!=== Error while compiling:
Undeclared routine:
loop used at line 1
Making $block.loop work, is rather easy and could live in module land:
use MONKEY;
augment class Block {
method loop($self:) {
Nil while $self()
}
}
my $a = { print "a" };
$a.loop # aaaaaaaaaaaaaaaaaaa (with apologies to Knorkator)
Making loop $block work would be rather more involved, as this would involve changes to the action handling of the Perl 6 grammar.
Using what is already in Perl 6, you can use Seq.from-loop in sink context.
(Note that the REPL doesn't put the last statement on a line into sink context)
my $block := {
state $n = 0;
say $n++;
last if $n > 3;
}
Seq.from-loop: $block;
Seq.from-loop: {say $++}, {$++ <= 3};
I was trying to understand how we are generating verilog code out of "for" loop in chisel.
Generally verilog code used to unroll body as many time as loop progress but here in chisel it's only unrolling it once.
val io = new Bundle {
val a = UInt(INPUT, 2)
val output = UInt(OUTPUT, 2)
}
io.output := UInt(0)
for(j <- 0 to 4){
io.output := io.a
}
Corresponding verilog code for the above program is :
module LutSimpleALU(
input [1:0] io_a,
output[1:0] io_output
);
assign io_output = io_a;
endmodule
it would be very helpful if someone can tell how for loop is working.
Your for loop is doing the same thing for each iteration. You aren't using the "j" iterator variable anywhere, so this is what it expands to:
io.output := io.a
io.output := io.a
io.output := io.a
io.output := io.a
The semantics here is that the last writer wins, so the last statement of "io.output = io.a" would be the final value. In fact, the previous three statements mean nothing, so they would be pruned from the graph.
I' m reading large process output data in Lazarus using the TProcess and the suggestions from this freepascal wiki page.
The wiki page suggests to create a loop to read the process output data like this:
// ... If you want to read output from an external process, this is the code you should adapt for production use.
while True do
begin
MemStream.SetSize(BytesRead + 2024); // make sure we have room
NumBytes := OurProcess.Output.Read((MemStream.Memory + BytesRead)^, READ_BYTES);
if NumBytes > 0
then begin
Inc(BytesRead, NumBytes);
Write('.') //Output progress to screen.
end else
BREAK // Program has finished execution.
end;
// "Then read the MemStream to do your job"
The wiki page also mentions that the calling program should read from the output pipe to prevent it from getting full.
So, how much data makes the output pipe full?
Why we should use a MemStream (TMemoryStream) and not directly read from OurProcess.Output stream (using the bytesAvailable, etc) in the above loop?
I'm reading 80MB of wav data from a process and I have noticed that both MemStream and OurProcess.Output streams have the same amount of data! The memory usage gets doubled. So the suggested method from the wiki cannot be considered as efficient or optimized. Or there is something I'm missing?
Afaik output/input streams are a stream form of a pipe, not memory streams. The values you see are retrieved from the OS handle, not from memory allocated to the FPC app per se.
It is just like you can ask for the .size of a file on disk without reading the whole file.
procedure RunExternalAppInMemo(DosApp:String;AMemo:TMemo);
const READ_BYTES = 2048;
var
aProcess: TProcess; //TProcess is crossplatform is best way
MemStream: TMemoryStream;
NumBytes: LongInt;
BytesRead: LongInt;
Lines: TStringList;
begin
// A temp Memorystream is used to buffer the output
MemStream := TMemoryStream.Create;
Lines :=TStringList.Create;
BytesRead := 0;
aProcess := TProcess.Create(nil);
aProcess.CommandLine := DosApp;
aprocess.ShowWindow := swoHIDE;
AProcess.Options := AProcess.Options + [poUsePipes];
aProcess.Execute;
while aProcess.Running do
begin
// make sure we have room
MemStream.SetSize(BytesRead + READ_BYTES);
// try reading it
NumBytes := aProcess.Output.Read((MemStream.Memory + BytesRead)^, READ_BYTES);
if NumBytes > 0 // All read() calls will block, except the final one.
then Inc(BytesRead, NumBytes)
else
BREAK // Program has finished execution.
end;
MemStream.SetSize(BytesRead);
Lines.LoadFromStream(MemStream);
AMemo.lines.AddStrings(Lines);
aProcess.Free;
Lines.Free;
MemStream.Free;
end;
I was dealing with this problem today, I've modified Georgescu answer, as I wanted Memo to display output stream on the fly
procedure RunExternalAppInMemo(DosApp:String;AMemo:TMemo);
const READ_BYTES = 2048;
var
aProcess: TProcess; //TProcess is crossplatform is best way
NumBytes: LongInt;
Buffer: array of byte;
begin
// set the size of your buffer
SetLength(Buffer,READ_BYTES);
aProcess := TProcess.Create(nil);
aProcess.CommandLine := DosApp;
aprocess.ShowWindow := swoHIDE;
AProcess.Options := AProcess.Options + [poUsePipes];
aProcess.Execute;
while aProcess.Running do
begin
// try reading it
NumBytes := aProcess.Output.Read(Buffer[0], length(buffer)*sizeof(byte)); // I usually do it that way, so I can change Buffer size on if needed
AProcess.Suspend; //I have no experience with pipes, but it seems way I won loose eny output?
if NumBytes > 0 then // All read() calls will block, except the final one.
begin
AMemo.Lines.Add(Pchar(Buffer);
application.ProcessMessages;
AProcess.Resume;
end
else
BREAK; // Program has finished execution.
end;
setlength(Buffer,0);
aProcess.Free;
end;
I have a SREC file which is a simple text file and I want to read it line by line in verilog. How can I do that?
The following reads through a file, 1 line per clock cycle: expected data format is one decimal number per line.
integer data_file ; // file handler
integer scan_file ; // file handler
logic signed [21:0] captured_data;
`define NULL 0
initial begin
data_file = $fopen("data_file.dat", "r");
if (data_file == `NULL) begin
$display("data_file handle was NULL");
$finish;
end
end
always #(posedge clk) begin
scan_file = $fscanf(data_file, "%d\n", captured_data);
if (!$feof(data_file)) begin
//use captured_data as you would any other wire or reg value;
end
end
Thank you for the solution.
I modified it just a little to use 2 .txt file containing 32 HEX numbers on each row and found some difficulties on the way since I didn't understand what each line of code did. My findings were the following.
Just vars and regs declaration
////////I'm using inputs.txt and outputs.txt to read both lines at the same time
module Decryption_Top_Testbench;
////////TEXT DOC variables
integer file_outputs ; // var to see if file exists
integer scan_outputs ; // captured text handler
integer file_inputs ; // var to see if file exists
integer scan_inputs ; // captured text handler
//TXT
reg [127:0] captured_outputs; ///Actual text obtained from outputs.txt lines
reg [127:0] captured_inputs; ///Actual text obtained from inputs.txt lines
Opening both files
initial
begin
// TEXT FILE outputs///////////////////////
file_outputs = $fopen("C:/outputs.txt", "r"); //Opening text file
//you should use the full path if you don't want to get in the trouble
//of using environment vars
if (file_outputs == 0) begin // If outputs file is not found
$display("data_file handle was NULL"); //simulation monitor command
$finish;
end
// TEXT FILE inputs///////////////////////
file_inputs = $fopen("C:/inputs.txt", "r"); //Opening text file (inputs)
if (file_inputs == 0) begin //If inputs file is not found
$display("data_file handle was NULL");
$finish;
end
end
At this part, I will read line by line in HEX format and store it in "captured_outputs" register and "captured_inputs" register.
///Since I'm using it just to simulate I'm not interested on a clock pulse,
/// I want it to happen all at the same time with whatever comes first
always #(* )
begin
if (!$feof(file_outputs))
begin
///!$feof means if not reaching the end of file
///file_outputs is always returning a different number other than "0" if the doc
///has not ended. When reaching "0" it means the doc is over.
///Since both of my docs are the same length I'm only validating one of them
///but if you have different lenghts you should verify each doc you're reading
///
scan_inputs = $fscanf(file_inputs, "%h\n", captured_inputs); //Inputs Line text
scan_outputs = $fscanf(file_outputs, "%h\n", captured_outputs); //Outputs line text
$display ("Line :[inputs: %h _ outputs: %h ]" captured_inputs, captured_outputs);
// Displaying each line at the simulation monitor
///$fscanf means formatted text, $scanf would read text ignoring the format
/// %h\n means it should expect HEX numbers and the end of line character, that means
/// the line is over, but if you want to use a diff criteria
/// you can replace \n to whatever you may need
end
else
begin
$finish;
$fclose(file_outputs); //Closing files just in case to prevent wasting memory
$fclose(file_inputs);
end
end
I just wanted to contribute with something anybody who's starting to code in Verilog could understand and attach this great feature to his/her project.
Enjoy!
Hello I'm a beginning programming Student and I am practicing using loops to validate input. Unfortunately, the loop works but skips the inner loop entirely... I get now error message or prompt...
Here is my code: [I BORROWED IT FROM AN ANSWER ON THIS SITE ABOUT VALIDATING INPUT SO I COULD TEST IT.]
import java.util.Scanner;
public class ValidationTest
{
public static void main(String[] args)
{
Scanner sc = new Scanner(System.in);
int number;
do {
System.out.println("Please enter a positive number!");
while (!sc.hasNextInt())
{
System.out.println("That's not a number!");
sc.next(); // this is important!
}
number = sc.nextInt();
} while (number <= 0);
System.out.println("Thank you! Got " + number);
}
}
The inner loop :
while (!sc.hasNextInt())
{
System.out.println("That's not a number!");
sc.next(); // this is important!
}
number = sc.nextInt();
only check if your input is not a number, if you input -123, the function !sc.hasNextInt() is false so it'll skip the loop, if you want to check the number is negative, add this check after the assign number value:
if(number <= 0 ){
System.out.println("The number is negative!");
}
You don't have to make another loop for checking the number is negative or not because of the first loop had it, and do...while loop will make sure you have to run the loop at least one time
The while loop will be skipped if the condition (in parentheses after while) is false when the while loop is first executed.
You can use a debugger to step through the code to see what is going on.
If you do not know how to use a debugger, stop everything you are doing and do some Google searches or whatever you need to do to find out how to use one.
your int number is NaN (not a number). Try setting it to -1, so you can enter the first loop.
In second loop sc didn't ever scan for input it's only initialized.