Re: wget | error 403 (#10)

[@rockdaboot]

Tim Rühsen commented:


The point is, if you are not using a utf-8 charset and --iri is enabled (e.g. 
by default), each URL is translated into UTF-8 by `locale_to_utf8()`. This 
function calls `url_unescape_except_reserved()` which unescapes all characters 
that doesn't need to be escaped. In your example it is %21 and %7E (exclamation 
mark and tilde). When the GET request is generated, these characters are not 
escaped again. So the URL is a different string (same URL but slightly 
differently encoded) send to the cloudfront proxy, which does not do a 
normalization/decoding and thus can't find the new string in his lookup table.

IMO, this is pretty much a cloudfront stupidity in combination with wget trying 
to do the best for the user.

-- 
Reply to this email directly or view it on GitLab: 
https://gitlab.com/gnuwget/wget/-/issues/10#note_955371872
You're receiving this email because of your account on gitlab.com.