On Fri, 2018-04-20 at 15:29 +0200, Philippe Mouawad wrote: > Is this approximation acceptable ? > context.setAttribute(CONTEXT_ATTRIBUTE_RECEIVED_BYTES, > metrics.getReceivedBytesCount()+ > (entity != null ? entity.getContentLength(): > 0L)); >
It depends what consider acceptable but this will produce complete garbage for chunked messages as their entity content length is always -1. Oleg > On Fri, Apr 20, 2018 at 3:28 PM, Oleg Kalnichevski <ol...@apache.org> > wrote: > > > On Fri, 2018-04-20 at 15:01 +0200, Philippe Mouawad wrote: > > > I am looking for raw content length > > > > > > > This cannot be done reliably with HttpClient 4.x (or I cannot think > > of > > a way at the moment) as it requires a direct access to the > > underlying > > connection. It would require dropping to HttpCore level and a > > manual > > connection management. > > > > Oleg > > > > > > > On Fri, Apr 20, 2018 at 2:58 PM, Oleg Kalnichevski <olegk@apache. > > > org> > > > wrote: > > > > > > > On Fri, 2018-04-20 at 14:31 +0200, Philippe Mouawad wrote: > > > > > On Fri, Apr 20, 2018 at 1:01 PM, Oleg Kalnichevski <oleg@ok2c > > > > > onsu > > > > > ltin > > > > > g.com> > > > > > wrote: > > > > > > > > > > > On Fri, 2018-04-20 at 11:40 +0200, Philippe Mouawad wrote: > > > > > > > Hello Oleg, > > > > > > > Thanks for rapid answer. > > > > > > > The body size including entity. > > > > > > > > > > > > > > > > > > > > > > > > > Should this number represent raw bytes (including the size > > > > > > of > > > > > > transfer > > > > > > encoding elements) > > > > > > > > > > > > > > > Can you illustrate so that I am sure to understand what you > > > > > mean > > > > > by > > > > > size of > > > > > transfer > > > > > encoding elements ? > > > > > > > > > > > > > --- > > > > HTTP/1.1 200 OK\r\n > > > > Content-Length: 5\r\n > > > > \r\n > > > > stuff > > > > --- > > > > > > > > total message length: 43 > > > > raw content length: 5 > > > > content length: 5 > > > > > > > > --- > > > > HTTP/1.1 200 OK\r\n > > > > Transfer-Encoding: chunked\r\n > > > > \r\n > > > > 5\r\n > > > > stuff0\r\n > > > > \r\ > > > > n > > > > --- > > > > > > > > total message length: 60 > > > > raw content length: 16 > > > > content length: 5 > > > > > > > > What number are you after? > > > > > > > > Oleg > > > > > > > > > > > > > > > > > > > or content bytes only (same of Content-Length when > > > > > > present)? > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > It would be much easier if you took the size of the entire > > > > > > message. > > > > > > > > > > > > > > > > How should I proceed ? > > > > > > > > > > > > > > > > > Oleg > > > > > > > > > > > > > > > > > > > > > > > > > I'd like the header size to be aside, I already know how > > > > > > > to > > > > > > > compute > > > > > > > header > > > > > > > size. > > > > > > > > > > > > > > Regards > > > > > > > > > > > > > > > > > > > > > On Fri, Apr 20, 2018 at 11:21 AM, Oleg Kalnichevski <oleg > > > > > > > k@ap > > > > > > > ache > > > > > > > .org > > > > > > > > > > > > > > > > > > > > > > wrote: > > > > > > > > > > > > > > > On Fri, 2018-04-20 at 11:09 +0200, Philippe Mouawad > > > > > > > > wrote: > > > > > > > > > Hello , > > > > > > > > > > > > > > > > > > In current Live version of JMeter we use this way of > > > > > > > > > computing > > > > > > > > > size > > > > > > > > > of > > > > > > > > > responses for each request: > > > > > > > > > > > > > > > > > > - > > > > > > > > > https://github.com/apache/jmeter/blob/v4_0/src/pro > > > > > > > > > toco > > > > > > > > > l/ht > > > > > > > > > tp/o > > > > > > > > > rg/a > > > > > > > > > pache/jmeter/protocol/http/sampler/HTTPHC4Impl.java#L > > > > > > > > > 888 > > > > > > > > > > > > > > > > > > > > > > > > > > > As we migrated to last APIs we based implementation > > > > > > > > > on > > > > > > > > > this > > > > > > > > > SO > > > > > > > > > response > > > > > > > > > from Oleg: > > > > > > > > > > > > > > > > > > - > > > > > > > > > https://stackoverflow.com/questions/26166469/measu > > > > > > > > > re-b > > > > > > > > > andw > > > > > > > > > idth > > > > > > > > > -usa > > > > > > > > > ge-with-apache-httpcomponents-httpclient > > > > > > > > > > > > > > > > > > And ended up with this: > > > > > > > > > > > > > > > > > > - > > > > > > > > > https://github.com/apache/jmeter/blob/trunk/src/pr > > > > > > > > > otoc > > > > > > > > > ol/h > > > > > > > > > ttp/ > > > > > > > > > org/ > > > > > > > > > apache/jmeter/protocol/http/sampler/HTTPHC4Impl.java# > > > > > > > > > L382 > > > > > > > > > > > > > > > > > > But it appears that computation is wrong and we would > > > > > > > > > need to > > > > > > > > > do > > > > > > > > > this > > > > > > > > > to be > > > > > > > > > correct: > > > > > > > > > > > > > > > > > > protected HttpResponse doReceiveResponse( > > > > > > > > > final HttpRequest request, > > > > > > > > > final HttpClientConnection conn, > > > > > > > > > final HttpContext context) throws > > > > > > > > > HttpException, > > > > > > > > > IOException { > > > > > > > > > HttpResponse response = > > > > > > > > > super.doReceiveResponse(request, > > > > > > > > > conn, > > > > > > > > > context); > > > > > > > > > HttpConnectionMetrics metrics = > > > > > > > > > conn.getMetrics(); > > > > > > > > > HttpEntity entity = response.getEntity(); > > > > > > > > > context.setAttribute(CONTEXT_ATTRIBUTE_RE > > > > > > > > > CEIV > > > > > > > > > ED_B > > > > > > > > > YTES > > > > > > > > > , > > > > > > > > > metrics.getReceivedBytesCount()+ > > > > > > > > > (entity != null ? > > > > > > > > > entity.getContentLength(): > > > > > > > > > 0L)); > > > > > > > > > metrics.reset(); > > > > > > > > > return response; > > > > > > > > > } > > > > > > > > > > > > > > > > > > > > > > > > > Philippe > > > > > > > > > > > > > > > > Are you trying to calculate the size of an entire > > > > > > > > response > > > > > > > > (message > > > > > > > > head + message entity body) or a response entity only? > > > > > > > > > > > > > > > > Oleg > > > > > > > > > > > > > > > > > > > > > > > > > > Is my understanding correct ? or am I missing > > > > > > > > > something ? > > > > > > > > > > > > > > > > > > Thanks > > > > > > > > > > > > > > > > > > Regards > > > > > > > > > > > > > > > > ----------------------------------------------------- > > > > > > > > ---- > > > > > > > > ---- > > > > > > > > ---- > > > > > > > > ---- > > > > > > > > To unsubscribe, e-mail: httpclient-users-unsubscribe@hc > > > > > > > > .apa > > > > > > > > che. > > > > > > > > org > > > > > > > > For additional commands, e-mail: httpclient-users-help@ > > > > > > > > hc.a > > > > > > > > pach > > > > > > > > e.or > > > > > > > > g > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > --------------------------------------------------------- > > > > > > ---- > > > > > > ---- > > > > > > ---- > > > > > > To unsubscribe, e-mail: httpclient-users-unsubscr...@hc.apa > > > > > > che. > > > > > > org > > > > > > For additional commands, e-mail: httpclient-users-help@hc.a > > > > > > pach > > > > > > e.or > > > > > > g > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > ------------------------------------------------------------- > > > > ---- > > > > ---- > > > > To unsubscribe, e-mail: httpclient-users-unsubscribe@hc.apache. > > > > org > > > > For additional commands, e-mail: httpclient-users-help@hc.apach > > > > e.or > > > > g > > > > > > > > > > > > > > > > > > ----------------------------------------------------------------- > > ---- > > To unsubscribe, e-mail: httpclient-users-unsubscr...@hc.apache.org > > For additional commands, e-mail: httpclient-users-h...@hc.apache.or > > g > > > > > > --------------------------------------------------------------------- To unsubscribe, e-mail: httpclient-users-unsubscr...@hc.apache.org For additional commands, e-mail: httpclient-users-h...@hc.apache.org
