All most of the ring sample I can see have a response map as
{"Content-Type" "text/html"} without adding a kind of charset.
I will spend some time to test Rack to see if there is the same issue.
Concretely, to see Rack will add a charset if there is no one in
response.
or It will action like Ring to see those response pass through
directly.
I know Ring looks like Rack.
On 8月13日, 上午10时36分, limux <liumengji...@gmail.com> wrote:
> Perhaps Jetty add a charset of iso-8859-1 if there isn't one in
> response.
> At the same time, in Compojure, it add none of the charset when a
> string is rendered. The headers has only a few info exactly as
> {"Content-Type" "text/html"}.
> So perhaps Jetty will add a old charset iso-8859-1. That's all! Maybe?
>
> Regards.
>
> On 8月13日, 上午2时25分, Rasmus Svensson <r...@lysator.liu.se> wrote:
>
>
>
>
>
>
>
> > 2010/8/12 James Reeves <jree...@weavejester.com>:
>
> > > On 12 August 2010 14:33, limux <liumengji...@gmail.com> wrote:
> > >> The solution inhttp://tiny.cc/3cmrxisuseful, thanks.
> > >> That what cause the issue should be compojure. That thread's time is
> > >> 6, June.
> > >> and compjure haven't fixed it.
>
> > > The solution you mention is some middleware that sets the content-type
> > > charset header to a specific value.
>
> > > Has this fixed the issue? I was under the impression from Rasmus's
> > > post that raw strings worked fine, and it was just an issue with
> > > Hiccup.
>
> > > However, that in itself is odd, as Hiccup only uses raw strings and
> > > the str function to join them together. I believe this should maintain
> > > the correct string encoding. So assuming both str and literal strings
> > > work, Hiccup should work.
>
> > > I guess we need to determine whether the string itself has the wrong
> > > encoding, or whether an incorrect encoding has been specified in the
> > > content type.
>
> > > - James
>
> > > --
> > > You received this message because you are subscribed to the Google
> > > Groups "Clojure" group.
> > > To post to this group, send email to clojure@googlegroups.com
> > > Note that posts from new members are moderated - please be patient with
> > > your first post.
> > > To unsubscribe from this group, send email to
> > > clojure+unsubscr...@googlegroups.com
> > > For more options, visit this group at
> > >http://groups.google.com/group/clojure?hl=en
>
> > From what I can tell, the problem he had was caused by compojure's
> > default content type "text/html" being replaced by "text/html;
> > charset=iso8859-1". If he added the charset attribute (with the
> > middleware proposed in the link) the problem went away.
>
> > I asked him to check the page info in firefox to see what content-type
> > the web server served. Without the middleware, it was "text/html;
> > charset=iso8859-1" and with it, it was "text/html; charset=utf-8", as
> > expected. The only value existing in the compojure code is
> > "text/html", iirc.
>
> > It appears that Jetty rewrites any text/html content type it serves
> > and adds a charset attribute (maybe with the dreaded "OS default
> > charset" as its value) if there isn't one.
>
> > Time for some tests, maybe?
>
> > // Rasmus
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
clojure+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
