index of string from beginning of line vs beginning of file

index of string from beginning of line vs beginning of file

Post by Jesse B » Sat, 27 Mar 2010 01:19:12


I am trying to write a basic script to implement "silent comments"
In the example anything from "input.txt" that is enclosed between "--/"
and "/--" will not be output to "output.text" It appears that the
problem I am having is that the opening string is indexed from beginning
of file, whereas the closing string is indexed from beginning of line.
So I would like to figure out why.
Also when using Ruby1.9 there is an error message about using "each"
with a string that I need to find a workaround to.
any help is greatly appreciated. thanks in advance.
here is the code:

infile = IO.readlines('input.txt','').to_s
outfile = File.new("output.txt", "w")
begins = "--/"
ends = "/--"
start_ss = infile.index(begins)
end_ss = infile.index(ends)
infile[start_ss, end_ss] = ""

infile.each {
|i|
outfile.write i
}
puts start_ss
puts end_ss
outfile.close()
--
Posted via http://www.yqcomputer.com/
 
 
 

index of string from beginning of line vs beginning of file

Post by Jes Gabrie » Sat, 27 Mar 2010 01:34:16


Check: http://www.yqcomputer.com/ #M000771

It says: If passed two Fixnum objects, returns a substring starting at
the offset given by the first, and a length given by the second.

So the second parameter should be the length of the text, not the
index of the token. Try this:

infile[start_ss, (end_ss - start_ss)] #untested


It's also better to use the block form of File.open to ensure proper
closing, even after an exception.

You could also try something like this:

infile = File.read("input.txt")
File.open("output.txt", "w") do |outfile|
outfile.puts infile.split(%r{--/.*?/--}m)
end

Take into account that neither this nor the above solution with
indexes supports nested comments.

Jesus.

 
 
 

index of string from beginning of line vs beginning of file

Post by Jesse B » Sat, 27 Mar 2010 05:09:40

Thank you Jesus,
Your solution is much more elegant.
However it leaves a blank line where my string was. any idea how to
close that gap?

i.e.

opening text
--/text to delete/--
closing text

now renders as:

opening text

closing text

and what I am going for is:

opening text
closing text

thanks again for all of your help!
--
Posted via http://www.yqcomputer.com/
 
 
 

index of string from beginning of line vs beginning of file

Post by Jes Gabrie » Sat, 27 Mar 2010 07:35:33


infile = File.read("input.txt")
File.open("output.txt", "w") do |outfile|
outfile.puts infile.split(%r{\n?--/.*?/--}m).join
end

Two changes:

- I added an optional \n at the beggining of the matched area to cover
that case, although there might be other corner cases.
- I joined the array before writing it, because puts with an array
prints a newline between each element.

See if this works for you and test any other edge case regarding
carriage returns. You can print what the split returns to see what you
get in each case:

infile = File.read("input.txt")
File.open("output.txt", "w") do |outfile|
splitted = infile.split(%r{\n?--/.*?/--}m)
puts splitted.inspect
outfile.puts splitted.join
end

Jesus.
 
 
 

index of string from beginning of line vs beginning of file

Post by Jesse B » Sun, 28 Mar 2010 00:04:21

Thank you again for your help.

I am finding that spaces before the "comment" cause it not to work.

this works as expected:
opening text
--/text to delete/--
closing text

however this (one space before "comment")
opening text
--/text to delete/--
closing text

outputs a newline

opening text

closing text

The command line output for the different cases looks like:
C:\Ruby\bin\file_write>ruby jesus.rb
["opening text", "\nclosing text\n\n\n"]

C:\Ruby\bin\file_write>ruby jesus.rb
["opening text\n ", "\nclosing text\n\n\n"]

as you can see, there is the \n after opening text. The three \n at the
end were present in the input.txt, so no problem.

thank you for your time.




--
Posted via http://www.yqcomputer.com/
 
 
 

index of string from beginning of line vs beginning of file

Post by Robert Kle » Sun, 28 Mar 2010 01:00:58

2010/3/25 Jes Gabriel y Gal < XXXX@XXXXX.COM >:


If you slurp in the whole file anyway then you can do

content = File.read "input.txt"
content.gsub! %r{--/.*?/--}m, ''

File.open "output.txt", "w" do |out|
out.write content
end

Kind regards

robert

--
remember.guy do |as, often| as.you_can - without end
http://www.yqcomputer.com/
 
 
 

index of string from beginning of line vs beginning of file

Post by Jesse B » Sun, 28 Mar 2010 01:17:34

Thank You Robert,
I tried running this and unfortunately it shows the line break (newline)
with or without the leading space before the "comment".

So I think Jesus is unto something with the "joined the array before
writing it because puts with an array prints a newline between each
element."

Now I just need to figure out how to have it not do the newline when
there are spaces before the "comment"
thanks for your time.




--
Posted via http://www.yqcomputer.com/
 
 
 

index of string from beginning of line vs beginning of file

Post by Jes Gabrie » Sun, 28 Mar 2010 05:48:18


I think there's no solution to that. I mean, the requirement is to
remove what's inside the comments, and that leaves a line with a
single space. Both my solution and Robert's do exactly that (I think
Robert's solution is the better one).

What you are asking now is for an additional condition, that it might
be that if after removing the comments there's an empty line (a line
with only spaces), that should be removed too.

content = File.read "input.txt"
content.gsub! %r{--/.*?/--}m, ''
content.gsub! %r{\A\s+\z}, '' # untested and i think you might to
remove a \n in this case too, give it a try.

File.open "output.txt", "w" do |out|
out.write content
end


Or maybe the condition is to remove all the whitespace and \n that
surround the comments.

content = File.read "input.txt"
content.gsub! %r{[\n\s]*--/.*?/--[\n\s]*}m, '' #untested also, but
you get the idea.

File.open "output.txt", "w" do |out|
out.write content
end

Hope this helps,

Jesus.
 
 
 

index of string from beginning of line vs beginning of file

Post by Robert Kle » Sun, 28 Mar 2010 07:01:29


A bit more to play with

content = File.read "input.txt"
content.gsub! %r{^[ \t]*--/(?:.(?!/--))*.?/--\s*?$\s}m, ''
content.gsub! %r{[ \t]*--/(?:.(?!/--))*.?/--[ \t]*}m, ' '

File.open "output.txt", "w" do |out|
out.write content
end


I used this for testing:


robert@fussel:~$ cat input.txt
1. before after
X1.1
before --/ comment
comment /-- after
X1.2
2. before
X2.1
before --/ comment
comment /--
X2.2
3. after
X3.1
--/ comment
comment /-- after
X3.2
4. -
X4.1
--/ comment
comment /--
X4.2
5. end
robert@fussel:~$

;-)

Cheers

robert

--
remember.guy do |as, often| as.you_can - without end
http://www.yqcomputer.com/
 
 
 

index of string from beginning of line vs beginning of file

Post by Jesse B » Mon, 29 Mar 2010 01:04:27

thanks robert, thats perfect!

--
Posted via http://www.yqcomputer.com/