|How to implement schema.org on https pages|
tl;dr: Use http URIs.
In this answer on Webmasters SE I explained why
you should favor http over https Schema.org URIs:
The http URIs seem to be canonical, as the actual
definition of the Schema.org vocabulary only
defines http, not https. In addition: all examples
(even on HTTPS) use the HTTP variant, the authors
mentioned that they prefer to see the use of the
HTTP variant, and RDFa’s Initial
|gwan dynamic buffer, how often to use and when should you not use it?|
As an application server, G-WAN is expected to
generate dynamic contents.
In this case, the server is building a reply
served to clients.
Part of these dynamic contents are binary (like
pictures) and this is why G-WAN offers a native
ultra-fast in-memory GIF, charts, and frame-buffer
API. More complex images can be generated with
general-purpose libraries like Cairo (used by
|HTTP restriction: oversized cookie|
I found the solution by just looking at the
response headers locally using Chrome devtools. I
realized that Flask sessions are built on top of
cookies which caused the oversizing issue (see
here). I just got rid of it and now everything
|Does a HTTP resource that accepts range requests always specify content-length?|
As per any other response with a 200 status code,
such a response SHOULD (in the RFC 2119 sense of
"SHOULD" which can be summarised as "better have a
damn good reason why not, and be able to say what
that damn good reason is") include a length (RFC
2616 §14.13) bar a few scenarios in which it is
prohibited (§4.4) (in particular, the value must
be both the entity and transfer length and as s
|Do we need to follow http accept header on an error?|
the server SHOULD send a representation
containing an explanation of the error situation
What this means is that the body of the response
should contain the error, and the Content-Type
identifies its type. For example, by default
Tomcat returns an HTML page, so the Content-Type
is set appropriately (to text/html, I believe).
If your erro
|Go http.Request Header on Redirect and Proxy|
I looked briefly at the code, at it looks like
goproxy doesn't handle 3XX responses in any
special way – they're simply routed back to your
client, so it can react accordingly. In your case,
you'll issue another request with all the headers
If it wasn't your concern and you were just
wondering if simply existence of proxy requires
any hack on the client side, then no, http.Client
|request.getParameter() fails on correctly encoded URL|
request.setCharacterEncoding( "UTF-8" ); has no
effect on doGet() . In doGet() the queryString
gets parsed by the container before it reaches
You should use doPost and request.getInputStream()
and parse the queryString yourself. And do not
use request.getParameter() before
request.getInputStream() or it won't work.
Java by default encodes String in utf-16 .. So you
will have to
|Digest Authentication not working in JMeter|
Realm should not be required for digest
authentication, domain is quite enough
Make sure that you're using either HttpClient3.1
or HttpClient4 implementation
See Windows Authentication with Apache JMeter
guide for details on how to configure HTTP
Authorization Manager for digest and kerberos
|What is the difference between Etag and Expires header?|
For what you have described
It is clear that ETag works as it expected to be
by responding with 304 not modified for the
request with If-None-Match field and ETag value.
so now the browser will load the image from cache
instead of getting a new image from server costing
bandwidth and time.
It seems that caching is disabled in your
browser.That's why a new request has been sent
before the cache
|Are all spiders supposed to use +http in their user agent string?|
It's just a convention that some spiders follow.
There is no constraint on what people can put in a
user agent header.
Take a look at this list of user agents that
contain "GoogleBot". You'll notice that many of
these don't contain "+http".
|Abort HTTP chunk encoded response with Error Page|
Unfortunately, there is no chunk you can send that
says to the browser "Hey, forget everything I sent
so far - something went wrong". Sending a TCP
reset packet will cause most browsers to display a
"Server reset connection" page, but that requires
access to one level down the network stack.
The only other I can think of is to try sending a
chunk with a negative length. I don't know how
|Browser image caching confusion|
The Cache-Control value is the one that will be
used - the image will expire from the cache after
604800 seconds. See this question for more detail.
See also here:
When both Cache-Control and Expires are present,
Cache-Control takes precedence
|How to prevent Denial of Service (DoS) using IIS?|
DOS attacks cannot be prevented at web server
level. It requires a commercial firewall, or cloud
based services such as CloudFlare.
Like you discovered, manual configuration on IIS,
or any simple scripting won't save you from that.
|Golang negroni and http.NewServeMux() issue|
I am not sure what sure what is the problem with
this code, but it appears to be not the negorni
way. negroni might not behave as expected if we
wrap its handler with another http.Handlerfunc.
So, I modified my code and got the work done by
My current code looks something like as below:
mux := http.NewServeMux()
|Unable to get http header|
By default, the Authorization header is removed by
the Authentication handler. That's why you cant'
seem to access it or log it. Handlers act before
the mediation inSequence.
The Authentication header can be preserved by
editing the api-manager.xml config file in
/repository/conf. Simply uncomment the following
node and change the value to false: