hc-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sung-Gu" <jeri...@apache.org>
Subject URI parsing - strict and tolerant? Re: DO NOT REPLY [Bug 21689] - Exception in get method
Date Tue, 22 Jul 2003 15:50:10 GMT

----- Original Message ----- 

> ------- Additional Comments From ortwin.glueck@nose.ch  2003-07-17 15:12 -------
> Claus,
> This is a re-thrown exception from a URIException. The colon character (:) must
> be escaped (%3a) as it delimits the protocol identifier from the host part in a
> URL. HttpClient is very strict about URIs. IE beeing able to parse this URL only

I assume you mean 'strict'is an escaped ones only.

> means that IE is very tolerant. But HttpClient complies to standards.

I assume you mean that 'tolerant' is ok to be dealed with both escaped and unescaped ones.

1st step: an URI is regarded as an escaped one. (only the case currently..)

2nd step: (if URIException reason code is ESCAPING thingy,)
              it could be as an unescaped one.

Then, it's possbile HttpClient to be tolerant about various type of URIs...

(Probably, aleady  you know though... ) 
Just let you know,

View raw message