Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.java.programmer > #28767 > unrolled thread

URLEncoder and #

Back to article view | Back to comp.lang.java.programmer

Contents

  URLEncoder and # Roedy Green <see_website@mindprod.com.invalid> - 2014-06-18 02:37 -0700
    Re: URLEncoder and # Nigel Wade <nmw@ion.le.ac.uk> - 2014-06-18 11:30 +0100
      Re: URLEncoder and # Roedy Green <see_website@mindprod.com.invalid> - 2014-06-20 08:06 -0700
        Re: URLEncoder and # markspace <markspace@nospam.nospam> - 2014-06-20 08:22 -0700
          Re: URLEncoder and # Arne Vajhøj <arne@vajhoej.dk> - 2014-06-21 23:10 -0400
    Re: URLEncoder and # Joerg Meier <joergmmeier@arcor.de> - 2014-06-18 16:42 +0200
    Re: URLEncoder and # Kevin McMurtrie <mcmurtrie@pixelmemory.us> - 2014-06-18 18:50 -0700

#28767 — URLEncoder and #

It seems URLEncoder converts # to %23

Why would there be any need to encode it?

The AppVisor people are complaining about URLs containing # saying
they need to be URLEncoded.  Surely that is not true.

-- 
Roedy Green Canadian Mind Products http://mindprod.com
Software gets slower faster than hardware gets faster. 
~ Niklaus Wirth (born: 1934-02-15 age: 80) Wirth's Law

[toc] | [next] | [standalone]

#28769

On 18/06/14 10:37, Roedy Green wrote:
> It seems URLEncoder converts # to %23
>
> Why would there be any need to encode it?

Because # is a special character in a URI. It separates the main URI from a fragment suffix. Have you never used "NAME" 
tag in your web pages to identify sub-sections? The browser shouldn't send anything after a # in a request, it's used by 
the browser to identify a fragment within the returned page.

The URI specification should satisfy your curiosity.

>
> The AppVisor people are complaining about URLs containing # saying
> they need to be URLEncoded.  Surely that is not true.
>

I think it probably is true.

[toc] | [prev] | [next] | [standalone]

#28792

On Wed, 18 Jun 2014 11:30:23 +0100, Nigel Wade <nmw@ion.le.ac.uk>
wrote, quoted or indirectly quoted someone who said :

>>
>> The AppVisor people are complaining about URLs containing # saying
>> they need to be URLEncoded.  Surely that is not true.
>>
>
>I think it probably is true.
  
If wanted to include # in a file name you would encode it, but if you
meant it to introduce an anchor on the page you would not, right?

-- 
Roedy Green Canadian Mind Products http://mindprod.com
Software gets slower faster than hardware gets faster. 
~ Niklaus Wirth (born: 1934-02-15 age: 80) Wirth's Law

[toc] | [prev] | [next] | [standalone]

#28797

On 6/20/2014 8:06 AM, Roedy Green wrote:
> On Wed, 18 Jun 2014 11:30:23 +0100, Nigel Wade <nmw@ion.le.ac.uk>
> wrote, quoted or indirectly quoted someone who said :
>
>>>
>>> The AppVisor people are complaining about URLs containing # saying
>>> they need to be URLEncoded.  Surely that is not true.
>>>
>>
>> I think it probably is true.
>
> If wanted to include # in a file name you would encode it, but if you
> meant it to introduce an anchor on the page you would not, right?
>

If by "file name" you mean the path part of the URI, then yes.  However 
watch out for including other special characters in the anchor:

http://foo.com/path#my#1anchor

Will confuse most parsers and is probably a syntax error anyway.  Also 
watch out for the query part of the URI, which needs to be encoded 
differently.

http://foo.com/path?a=#1,b=#2#anchor

Is also incorrectly encoded.

[toc] | [prev] | [next] | [standalone]

#28810

On 6/20/2014 11:22 AM, markspace wrote:
> On 6/20/2014 8:06 AM, Roedy Green wrote:
>> On Wed, 18 Jun 2014 11:30:23 +0100, Nigel Wade <nmw@ion.le.ac.uk>
>> wrote, quoted or indirectly quoted someone who said :
>>
>>>>
>>>> The AppVisor people are complaining about URLs containing # saying
>>>> they need to be URLEncoded.  Surely that is not true.
>>>>
>>>
>>> I think it probably is true.
>>
>> If wanted to include # in a file name you would encode it, but if you
>> meant it to introduce an anchor on the page you would not, right?
>>
>
> If by "file name" you mean the path part of the URI, then yes.  However
> watch out for including other special characters in the anchor:
>
> http://foo.com/path#my#1anchor
>
> Will confuse most parsers and is probably a syntax error anyway.  Also
> watch out for the query part of the URI, which needs to be encoded
> differently.
>
> http://foo.com/path?a=#1,b=#2#anchor
>
> Is also incorrectly encoded.

I think a lot of the confusion about this class comes from its name.

Contrary to its name it is not used to encode URL's.

It is documented. Java docs says:

<quote>
Utility class for HTML form encoding. This class contains static methods 
for converting a String to the application/x-www-form-urlencoded MIME 
format. For more information about HTML form encoding, consult the HTML 
specification.
</quote>

But just the name gives people the wrong associations.

Arne

[toc] | [prev] | [next] | [standalone]

#28771

On Wed, 18 Jun 2014 02:37:09 -0700, Roedy Green wrote:

> It seems URLEncoder converts # to %23

> Why would there be any need to encode it?

> The AppVisor people are complaining about URLs containing # saying
> they need to be URLEncoded.  Surely that is not true.

This link explains it, and in a weird twist of fate, also actually contains
the #:

<http://en.wikipedia.org/wiki/HTML_element#Anchor>

Liebe Gruesse,
		Joerg

-- 
Ich lese meine Emails nicht, replies to Email bleiben also leider
ungelesen.

[toc] | [prev] | [next] | [standalone]

#28779

In article <t8n2q9phsbha7e25j4tj0e53sqa3qqt0g0@4ax.com>,
 Roedy Green <see_website@mindprod.com.invalid> wrote:

> It seems URLEncoder converts # to %23
> 
> Why would there be any need to encode it?
> 
> The AppVisor people are complaining about URLs containing # saying
> they need to be URLEncoded.  Surely that is not true.

Like '/', it has different meanings in different parts of the URL.

There is no way to encode a simple string into a full URL.  That's why 
there are tools to build URLs.

Try finding a server that handles an encoded '/' in a path correctly.  
That incorrect handling breaks many kinds of REST APIs.

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.java.programmer

csiph-web