Tcl file READ command…

[Tom Rioux]

I’m on 5.6 AIX….

I haven’t had to use this command very much and I can’t seem to find it documented anywhere, but here is what I’m looking for.   I have a new line file and I want to ignore the first few header lines in the file.  Is there a way that I can specify to start reading on a certain line?

Thanks….

Tom Rioux

Topic

[Max Drown (Infor)]

With tcl read I don’t think you can specify to read in a file from x to y. I think you can read a certain number of bites, but not specify where to start. Charlie will know if there’s a way to do this.

When I’m working with files, I normally will slurp the entire file into a variable and then work with the file from there.

An example:

Code:

set fileName [lindex $argv 0]
# Usage: test.tcl file.txt

set f [open $fileName r]
fconfigure $f -translation binary
set msg [read $f]
close $f

# You whole file is now in $msg

foreach line [split $msg n] {
   # Skip blank lines
   # if {$line == “”} {continue}
}

I’m sure there are other ways, even “better” ways to this, but this has worked for me in the past.

-- Max Drown (Infor)

[Jim Kosloskey]

Tom,

I opewn the file getting a file handle.

Then I use a while loop checking for eof

      while ![eof $filehandle]{

Inside the while loop I use gets to get each line. I also check for eof and break inside the while (wearing suspenders and a belt is really safe 😆 ).

You could then keep a counter and as long as the counter is less than what you want continue inside the while. Once you are at the lines you want, start processing.

The read command won’t do what you want as reads in the entire file or just so many characters.

email: jim.kosloskey@jim-kosloskey.com 29+ years Cloverleaf, 59 years IT - old fart.

[Max Drown (Infor)]

Here’s an example from the web of what Jim is referring to:

set f [open /home/keith/.profile]
while {[gets $f line] >= 0} {
[code]set f [open /home/keith/.profile]
while {[gets $f line] >= 0} {

-- Max Drown (Infor)

[Tom Rioux]

Thanks guys for all your help.  The counter was the way to go.  Here is what got it to work for me:

           set fd [open $msgdir/$filename r+]

           set ctr 1

           foreach line [split [read $fd] n] {

             if [expr $ctr > 8] {

               lappend contents $line

             }

             incr ctr

           }

           close $fd

Tom

[Charlie Bursell]

The simplest method to read a file is:

set fd [open file]

while {[gets $fd rec] >= 0} {

     process the record

}

close $fd

As for reading parts of a file refer to the seek and tell commands

 

tell channelId  returns the current file offset

seek channelId offset ?origin?”   moves to a specific offset

[Russ Ross]

Another handy alerternative at the command line to say output starting at line 100 to the end of the file is to use the tail command at the UNIX prompt as follows:

tail +100 myFile

Russ Ross
RussRoss318@gmail.com

[Keith McLeod]

Another command that I like when reading one line at a time is

for_file line {

do something with the line

}

Like other for loops you can throw in a counter if you wish….

[Max Drown (Infor)]

Russ Ross wrote:

Another handy alerternative at the command line to say output starting at line 100 to the end of the file is to use the tail command at the UNIX prompt as follows:

tail +100 myFile

Good one!

in Tcl it would be something like set msg [exec tail +100 $fileName]

-- Max Drown (Infor)

[Author]

Replies