[lnkForumImage]
TotalShareware - Download Free Software
Confronta i prezzi di migliaia di prodotti.
Asp Forum
 Home | Login | Register | Search 


 

Forums >

comp.lang.ruby

Gsub!("\n","\n"

Simon Tan

1/20/2009 12:19:00 AM

Hi all, I'm not sure why but I assumed gsub! would allow me to replace
\n with line feeds or \r for carriage returns.

How do I gsub! and replace everything with a nonvisible character like
newline? I was using this results.gsub!("\n","\n") but obviously that
does not work :( also tried results.gsub!("\n",10.chr). No go.

Kinda of frustrating for something so simple. I guess I could always do
some recursion. Thanks.
--
Posted via http://www.ruby-....
9 Answers

Rob Biedenharn

1/20/2009 12:51:00 AM

0

On Jan 19, 2009, at 7:19 PM, Simon Tan wrote:

> Hi all, I'm not sure why but I assumed gsub! would allow me to replace
> \n with line feeds or \r for carriage returns.
>
> How do I gsub! and replace everything with a nonvisible character like
> newline? I was using this results.gsub!("\n","\n") but obviously that
> does not work :( also tried results.gsub!("\n",10.chr). No go.
>
> Kinda of frustrating for something so simple. I guess I could
> always do
> some recursion. Thanks.
> --


Are you trying to turn this:

Hello,
World

Into something like:

Hello,\nWorld

If so, you need to change "\n" (1 character string) into "\\\\n" (3
character string, two escaped backslashs and the letter 'n'). The
replacement string will undergo two different interpretations: as a
string literal "\\\\n" becomes \\n which when interpreted as a
replacement for a regexp undergoes another round (so that things like
\1 and \2 can reference groups in the regexp) and becomes \n

Yes, this is tricky, but it's because there are two levels of escaping
going on and both use the backslash.

-Rob

Rob Biedenharn http://agileconsult...
Rob@AgileConsultingLLC.com

Simon Tan

1/20/2009 1:17:00 AM

0

haha actually, the other way around :) I have a string(I believe...I'm
using Hpricot to retreive web pages and it likes to add slashes to
everything). For example, I want to change "Hello, \nWorld" by changing
\n into a newline. So the output should be like:
Hello
World


I did try your suggestion, but I get "Hello\\nWorld". Will keep playing
with the slashes.

Rob Biedenharn wrote:
> On Jan 19, 2009, at 7:19 PM, Simon Tan wrote:
>
>> --
> Are you trying to turn this:
>
> Hello,
> World
>
> Into something like:
>
> Hello,\nWorld
>
> If so, you need to change "\n" (1 character string) into "\\\\n" (3
> character string, two escaped backslashs and the letter 'n'). The
> replacement string will undergo two different interpretations: as a
> string literal "\\\\n" becomes \\n which when interpreted as a
> replacement for a regexp undergoes another round (so that things like
> \1 and \2 can reference groups in the regexp) and becomes \n
>
> Yes, this is tricky, but it's because there are two levels of escaping
> going on and both use the backslash.
>
> -Rob
>
> Rob Biedenharn http://agileconsult...
> Rob@AgileConsultingLLC.com

--
Posted via http://www.ruby-....

Rob Biedenharn

1/20/2009 1:44:00 AM

0

Are you seeing something like:
irb> look = 'here is a " and I\'ll put a \' here, too'
=> "here is a \" and I'll put a ' here, too"
irb> puts look
here is a " and I'll put a ' here, too
=> nil

The rules for what needs to be escaped depend on the kind of quotes
used.

Are you seeing output that is shown in double quotes to indicate that
it is a string?

-Rob

On Jan 19, 2009, at 8:16 PM, Simon Tan wrote:

> haha actually, the other way around :) I have a string(I
> believe...I'm
> using Hpricot to retreive web pages and it likes to add slashes to
> everything). For example, I want to change "Hello, \nWorld" by
> changing
> \n into a newline. So the output should be like:
> Hello
> World
>
>
> I did try your suggestion, but I get "Hello\\nWorld". Will keep
> playing
> with the slashes.
>
> Rob Biedenharn wrote:
>> On Jan 19, 2009, at 7:19 PM, Simon Tan wrote:
>>
>>> --
>> Are you trying to turn this:
>>
>> Hello,
>> World
>>
>> Into something like:
>>
>> Hello,\nWorld
>>
>> If so, you need to change "\n" (1 character string) into "\\\\n" (3
>> character string, two escaped backslashs and the letter 'n'). The
>> replacement string will undergo two different interpretations: as a
>> string literal "\\\\n" becomes \\n which when interpreted as a
>> replacement for a regexp undergoes another round (so that things like
>> \1 and \2 can reference groups in the regexp) and becomes \n
>>
>> Yes, this is tricky, but it's because there are two levels of
>> escaping
>> going on and both use the backslash.
>>
>> -Rob
>>
>> Rob Biedenharn http://agileconsult...
>> Rob@AgileConsultingLLC.com
>
> --
> Posted via http://www.ruby-....
>

Rob Biedenharn http://agileconsult...
Rob@AgileConsultingLLC.com
+1 513-295-4739
Skype: rob.biedenharn

Bertram Scharpf

1/20/2009 4:04:00 AM

0

Hi,

Am Dienstag, 20. Jan 2009, 09:19:02 +0900 schrieb Simon Tan:
> Hi all, I'm not sure why but I assumed gsub! would allow me to replace
> \n with line feeds or \r for carriage returns.

str.gsub! "\\n", "\n"
str.gsub! "\\r", "\r"

This is a security risk:

eval %Q("#{str}")

Maybe this does:

str = "hello\\nbye"
str.gsub! /\\(\w)/ do |m| eval '"\\' + $1 + '"' end

Okay, \ escaping is a weird matter.

You know that gsub! returns nil in case nothing was changed?

Bertram


--
Bertram Scharpf
Stuttgart, Deutschland/Germany
http://www.bertram-...

Tammo Tjarks

1/20/2009 7:46:00 AM

0

Simon Tan wrote:
If you want to substitute
Hello\nWorld by
Hello
World
you should try to work with single quotes:
a = 'Hello\nWorld'

Simon Tan wrote:
=> "Hello\\nWorld"
irb(main):002:0> puts a
Hello\nWorld
=> nil
irb(main):003:0> a.gsub!('\n',"\n")
=> "Hello\nWorld"
irb(main):004:0> puts a
Hello
World

> Hi all, I'm not sure why but I assumed gsub! would allow me to replace
> \n with line feeds or \r for carriage returns.
>
> How do I gsub! and replace everything with a nonvisible character like
> newline? I was using this results.gsub!("\n","\n") but obviously that
> does not work :( also tried results.gsub!("\n",10.chr). No go.
>
> Kinda of frustrating for something so simple. I guess I could always do
> some recursion. Thanks.

Sebastian Hungerecker

1/20/2009 4:33:00 PM

0

Simon Tan wrote:
> I'm not sure why but I assumed gsub! would allow me to replace
> \n with line feeds

"\n" *is* a line feed.
>> string = "hello\nworld"
=> "hello\nworld"
>> puts string
hello
world
=> nil

HTH,
Sebastian
--
NP: Dire Straits - Brothers In Arms
Jabber: sepp2k@jabber.org
ICQ: 205544826

Albert Schlef

1/20/2009 9:49:00 PM

0

Simon Tan wrote:
> I have a string(I believe...I'm
> using Hpricot to retreive web pages and it likes to add slashes to
> everything). For example, I want to change "Hello, \nWorld" by changing
> \n into a newline.

Perhaps there's a misunderstanding here. I doubt Hpricot adds anything.
If you see this "Hello, \nWorld" in 'irb' then that's because 'irb'
shows you the string in the way you'd write it in your source-code. The
two characters "\n" are actually a newline.
--
Posted via http://www.ruby-....

Simon Tan

1/25/2009 12:25:00 AM

0

Ok, I just realized I've been using p to print everything out! "Puts"
works a whole lot better. Doh! Thanks for all the help. Sorry for
phrasing the question poorly :)

--
Posted via http://www.ruby-....

list. rb

1/26/2009 12:06:00 PM

0





On Jan 24, 2009, at 7:25 PM, Simon Tan <simon1tan@yahoo.com> wrote:

> Ok, I just realized I've been using p to print everything out! "Puts"
> works a whole lot better. Doh! Thanks for all the help. Sorry for
> phrasing the question poorly :)
>
> --
> Posted via http://www.ruby-....
>

LOL!