I am testing REST API using rest-assured. When I send POST request it looks like rest-assured is making extra call. Here is the output from /var/log/httpd/access_log:
11.31.41.111 - - [26/Nov/2019:19:39:14 +0000] "POST /rest/v1/contact HTTP/1.1" 401 340 "-" "Apache-HttpClient/4.5.3 (Java/1.8.0_221)"
11.31.41.111 - - [26/Nov/2019:19:39:14 +0000] "POST /rest/v1/contact HTTP/1.1" 200 515 "-" "Apache-HttpClient/4.5.3 (Java/1.8.0_221)"
When I send exactly same request using Postman, access log shows only one request comes to the server:
11.31.41.111 - - [26/Nov/2019:19:40:44 +0000] "POST /rest/v1/contact/ HTTP/1.1" 200 529 "-" "PostmanRuntime/7.19.0"
Any idea why this is happening?
You should use Preemptive Authentication when building RestAssured request specification. Here's an example:
RestAssured.given().auth().preemptive().basic("username", "password")
.when().get("http://example.com")
.then().statusCode(200);
Using Eclipse's internal representation of Tomcat8 (do not know if that is relevant), the following is a series of page requests. This comes from the file:
C:\Users\BS\Eclipse\workspace.metadata.plugins\org.eclipse.wst.server.core\tmp0\logs\localhost_access_log.2016-04-25.txt:
"GET /ProjK/ HTTP/1.1" 200 16354
"GET /ProjK/WEFiles/Css/v02/openElement.css?v=50491098000 HTTP/1.1" 200 16354
"GET /ProjK/Templates/BaseLayer.css?v=50491098000 HTTP/1.1" 200 16354
"GET /ProjK/index.css?v=50491098000 HTTP/1.1" 200 16354
"GET /ProjK/WEFiles/Client/jQuery/migrate.js?v=50491098000 HTTP/1.1" 200 16354
"GET /ProjK/WEFiles/Client/Common/oe.min.js?v=50491098000 HTTP/1.1" 200 16354
"GET /ProjK/Files/Image/Template/Top.png HTTP/1.1" 200 16354
"GET /ProjK/Files/Image/Template/Logo.png HTTP/1.1" 200 16354
"GET /ProjK/Files/Image/BioTransparent.png HTTP/1.1" 200 16354
"GET /ProjK/WEFiles/Image/WEImage/PuceImg-3f204a74.png HTTP/1.1" 200 16354
"GET /ProjK/WEFiles/Image/empty.png HTTP/1.1" 200 16354
Observe that each response is the same byte count. The browser receives the content of the same file regardless of the request.
The content is from "index.html" that has been renamed to "index.ftl" and renamed back to "index.html". I have been making simple edits to this file hoping to see any change in what the browser shows.
I see that the file "index.html" has been updated with the new content in both the workspace and the /tmp0/webapps/ folder.
Yet, there are no changes in what is delivered to the browser.
A number of conversations suggests that Tomcat caching be dealt with. I would have to wonder why, during app development, this is an actual consideration.
But caching does not explain why every page and page resource request is served the same unchanging content.
What common mistake would a new java web app developer be making with respect to the setup/configuration of Tomcat and/or Eclipse and or the webapp that would explain this behavior?
I'm using 2 agents to consume a REST webservice : Apache HttpClient and CURL.
Both the clients receive different size of data for the same REST webservice. Moreover the CURL is receiving inconsistent amount of bytes all the time while the Apache HttpClient is consistent.
Below is the access log snippet with number to bytes received.
"POST <<url>> HTTP/1.1" 200 10083053 "-" "Apache-HttpClient/4.3.1 (java 1.5)"
"POST <<url>> HTTP/1.1" 200 10083053 "-" "Apache-HttpClient/4.3.1 (java 1.5)"
"POST <<url>> HTTP/1.1" 200 10083053 "-" "Apache-HttpClient/4.3.1 (java 1.5)"
"POST <<url>> HTTP/1.1" 200 10128377 "-" "curl/7.19.7 (x86_64-redhat-linux-gnu)
"POST <<url>> HTTP/1.1" 200 10128674 "-" "curl/7.19.7 (x86_64-redhat-linux-gnu)
"POST <<url>> HTTP/1.1" 200 8192 "-" "curl/7.19.7 (x86_64-redhat-linux-gnu)
What could be the problem with CURL receiving inconsistent amount of bytes?
Hi, all,
i think this is a bug about the constant defined in sinatra, let's look my code.
route.rb
require 'sinatra'
get '/' do
C = "this is a test for constant"
"Hello World!"
end
Gemfile
source 'http://rubygems.org'
gem 'rack'
gem 'sinatra'
config.ru
require './route'
run Sinatra::Application
Starting the web server, we will see the below
$ rackup
[2011-10-08 19:54:36] INFO WEBrick 1.3.1
[2011-10-08 19:54:36] INFO ruby 1.9.2 (2011-07-09) [i686-linux]
[2011-10-08 19:54:36] INFO WEBrick::HTTPServer#start: pid=3268 port=9292
127.0.0.1 - - [08/Oct/2011 19:54:42] "GET / HTTP/1.1" 200 25 0.0059
127.0.0.1 - - [08/Oct/2011 19:54:42] "GET / HTTP/1.1" 200 25 0.0142
/home/zcdny/repo/test/route.rb:4: warning: already initialized constant C
127.0.0.1 - - [08/Oct/2011 19:54:43] "GET / HTTP/1.1" 200 25 0.0094
127.0.0.1 - - [08/Oct/2011 19:54:43] "GET / HTTP/1.1" 200 25 0.0098
/home/zcdny/repo/test/route.rb:4: warning: already initialized constant C
127.0.0.1 - - [08/Oct/2011 19:54:55] "GET / HTTP/1.1" 200 25 0.0003
127.0.0.1 - - [08/Oct/2011 19:54:55] "GET / HTTP/1.1" 200 25 0.0006
/home/zcdny/repo/test/route.rb:4: warning: already initialized constant C
127.0.0.1 - - [08/Oct/2011 19:54:56] "GET / HTTP/1.1" 200 25 0.0003
127.0.0.1 - - [08/Oct/2011 19:54:56] "GET / HTTP/1.1" 200 25 0.0005
Eidt
Fixing the file route.rb
require 'sinatra'
configure do
C = "this is a test for constant"
end
get '/' do
"Hello World!"
end
Now, the server no longer warning the constant be initialized.
But the log of server still appears double 'GET' require ,
i just want it requires for one to every client required, that is my question, How to solve it.
Thanks in advance.
What's wrong about that? If you define the constant twice (which happens if you have two GET requests or a GET and a HEAD request) then that warning will be displayed. You global variables instead. But if you don't have to, try to avoid global state at all cost, otherwise you might run in architectural issues (what if you want to serve more endpoints and globals clash) and make it hard to scale: if you rely on the internal state of a process, will you be able to serve the website from two processes? what about two machines?
I'm desperatly trying to find a solution for a web application that has to run on an iOS-Safari (e.g. on iPad, iPad2 and iPhone 4):
It's a web application I wrote some time ago which lets the user search for and listen to short music samples (MP3s, all from ~100 kB to ~1.5 MB). The audio player is Flash-based, so it doesn't work on iOS-devices at the moment and I'll have to implement an alternative either in HTML 5 or with a "direct" QuickTime-object.
Both my HTML 5- and QuickTime-alternatives for iOS-devices work fine so far, but there's one major problem I can't find a solution to:
Unlike Flash and most HTML 5-capable browsers on Windows Safari on my iPad 2 won't store the audiofiles in the browsercache after loading and playing them - neither with HTML 5 audio-tags nor with a QuickTime-Object. Every time I load an audio file for playback from the server (with JavaScript-Commands, so without changing or reloading the whole page) it's downloaded again completely.
If a user listens to sample A and then to sample B, Safari forgot about having played sample A and downloads the whole MP3 again if I like to listen to it again. On a mobile device with a potentially narrow bandwith this behaviour is out of the question.
Is there a way of storing downloaded audiofiles opened by HTML 5 or QuickTime in Safari's cache so it remembers already having downloaded them - like it caches usual "web-files" like HTML, CSS or JPEG-images, or like Flash stores such objects in its local cache?
My first attempt was trying to use the Application Cache with a manifest file - although this is not really the intended purpose for my application... I don't have a static set of files that I want to have cached or "available offline" - I just want to cache MP3s which the user has played yet.
It should be possible to use a "dynamic manifest": One that is parsed by the Apache PHP module and listing the files played so far from a PHP session - something like this:
session_start();
header("Content-Type: text/cache-manifest, charset=UTF-8");
echo "CACHE MANIFEST\n";
foreach($_SESSION['playedSongs'] as $song)
{
echo $song."\n";
}
So whenever a song is loaded / played, I could access the PHP session with AJAX, insert the filename of the played file and manually update the manifest by calling window.applicationCache.update() or .swapCache().
There are two problems with this:
First of all: It doesn't work. And I didn't even get to the point of trying to use a dynamic manifest:
<!DOCTYPE html>
<html manifest="cache.manifest">
<head>
<title>Test</title>
<script type="text/javascript">
function playStuff(id)
{
if(id == 1)
{
window.document.getElementById("audio").innerHTML = '<audio controls preload="automatic" autobuffer><source src="song01.mp3" type="audio/mp3" /></audio>';
}
else if(id == 2)
{
window.document.getElementById("audio").innerHTML = '<audio controls preload="automatic" autobuffer><source src="song02.mp3" type="audio/mp3" /></audio>';
}
}
</script>
</head>
<body>
<div id="audio"></div><br />
<br />
<input type="button" value="playStuff(1)" onclick="playStuff(1)" />
<input type="button" value="playStuff(2)" onclick="playStuff(2)" />
</body>
</html>
The cache.manifest looks like this:
CACHE MANIFEST
song01.mp3
song02.mp3
and is properly returned from Apache as "text/cache-manifest" by adding
AddType text/cache-manifest manifest
to the .htaccess of this directory.
The Apache-logs clearly show that the Safari (respectively "AppleCoreMedia") doesn't care about the application cache when it comes to audio-files:
Safari itself seems to acknowledge the manifest and indeed preload the files:
192.168.0.40 - - [23/Jul/2011:12:45:46 +0200] "GET /websql/index2.html HTTP/1.1" 200 2619 "-" "Mozilla/5.0 (iPad; U; CPU OS 4_3_3 like Mac OS X; de-de) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8J2 Safari/6533.18.5"
192.168.0.40 - - [23/Jul/2011:12:45:46 +0200] "GET /websql/cache.manifest HTTP/1.1" 200 79 "-" "Mozilla/5.0 (iPad; U; CPU OS 4_3_3 like Mac OS X; de-de) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8J2 Safari/6533.18.5"
192.168.0.40 - - [23/Jul/2011:12:45:46 +0200] "GET /websql/cache.manifest?%3E HTTP/1.1" 200 79 "-" "Mozilla/5.0 (iPad; U; CPU OS 4_3_3 like Mac OS X; de-de) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8J2 Safari/6533.18.5"
192.168.0.40 - - [23/Jul/2011:12:45:46 +0200] "GET /websql/song02.mp3 HTTP/1.1" 200 120525 "-" "Mozilla/5.0 (iPad; U; CPU OS 4_3_3 like Mac OS X; de-de) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8J2 Safari/6533.18.5"
192.168.0.40 - - [23/Jul/2011:12:45:46 +0200] "GET /websql/song01.mp3 HTTP/1.1" 200 120525 "-" "Mozilla/5.0 (iPad; U; CPU OS 4_3_3 like Mac OS X; de-de) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8J2 Safari/6533.18.5"
Up to this point I did nothing but opening my test-application in Safari.
Playing song01.mp3:
192.168.0.40 - - [23/Jul/2011:12:47:25 +0200] "GET /websql/song01.mp3 HTTP/1.1" 206 2 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:47:25 +0200] "GET /websql/song01.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:47:25 +0200] "GET /websql/song01.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:47:25 +0200] "GET /websql/song01.mp3 HTTP/1.1" 304 - "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:47:25 +0200] "GET /websql/song01.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:47:29 +0200] "GET /websql/song01.mp3 HTTP/1.1" 304 - "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:47:29 +0200] "GET /websql/song01.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
Playing song2.mp3:
192.168.0.40 - - [23/Jul/2011:12:48:04 +0200] "GET /websql/song02.mp3 HTTP/1.1" 206 2 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:48:04 +0200] "GET /websql/song02.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:48:04 +0200] "GET /websql/song02.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:48:04 +0200] "GET /websql/song02.mp3 HTTP/1.1" 304 - "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:48:04 +0200] "GET /websql/song02.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:48:05 +0200] "GET /websql/song02.mp3 HTTP/1.1" 304 - "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:48:05 +0200] "GET /websql/song02.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
Playing song1.mp3 again:
192.168.0.40 - - [23/Jul/2011:12:48:38 +0200] "GET /websql/song01.mp3 HTTP/1.1" 304 - "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:48:38 +0200] "GET /websql/song01.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:48:38 +0200] "GET /websql/song01.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:48:38 +0200] "GET /websql/song01.mp3 HTTP/1.1" 304 - "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:48:38 +0200] "GET /websql/song01.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:48:40 +0200] "GET /websql/song01.mp3 HTTP/1.1" 304 - "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:48:40 +0200] "GET /websql/song01.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
Playing song2.mp3 again:
192.168.0.40 - - [23/Jul/2011:12:49:12 +0200] "GET /websql/song02.mp3 HTTP/1.1" 304 - "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:49:12 +0200] "GET /websql/song02.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:49:12 +0200] "GET /websql/song02.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:49:12 +0200] "GET /websql/song02.mp3 HTTP/1.1" 304 - "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:49:12 +0200] "GET /websql/song02.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:49:13 +0200] "GET /websql/song02.mp3 HTTP/1.1" 304 - "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
192.168.0.40 - - [23/Jul/2011:12:49:13 +0200] "GET /websql/song02.mp3 HTTP/1.1" 206 120525 "-" "AppleCoreMedia/1.0.0.8J2 (iPad; U; CPU OS 4_3_3 like Mac OS X; de_de)"
Every file is downloaded again completely when playing it. So "AppleCoreMedia" (whatever this may be exactly, the QuickTime-plugin that is triggered by the HTML 5 audio-element I suppose?) either doesn't have access to the Application Cache or simply just doesn't realize the files in it. So if I switch my iPad to "Airplane Mode" now, Safari is unable to access / load / play the files.
I also tried using a QuickTime-object instead of an HTML 5 audio-tag (as far as I know HTML 5 audio and video in Safari always use QuickTime?) and controlling it with something like:
document.movie1.SetURL('song02.mp3');
Nothing changes, it's just like using HTML 5 audio and everything gets downloaded again when loading/playing it.
And even if this would work there'd still be a problem:
To properly implement that I would have to load the MP3-file in the Application Cache before playing it. When doing this it seems impossible to show a "real" progress: The ProgressEvent that is fired from the Application Cache after updating it doesn't seem to provide any information about data loaded so far and the complete size of a file. It's just "File 1 from 2" and so on, and not a "real" progress where I could determine something like: "100 kB of 1.2 MB loaded" as I can do with the audio-element.
All the other storage-approaches like Web SQL / Web Database or Local Storage are no help either:
I don't see any way of getting the MP3 data into Local Storage or Web Database and/or getting it out again to play it. The HTML 5 Canvas-element has a toDataURL()-function to produce a Base64-encoded representation and use it for storage - the Audio-element doesn't seem to have anything like this.
My last really "dirty" approach was trying to load "manually" Base64-encoded-MP3s with a combination of AJAX and PHP: A PHP-script outputs a Base64-representation of a MP3-file and is loaded by AJAX, so I could store the Base64-representation e.g. as Local Storage or in Web Database:
$infile = 'song01.mp3';
$contents = file_get_contents($infile);
$base64 = base64_encode($contents);
$audio = 'data:audio/mp3;base64,'.$base64;
echo $audio;
I tried using the resulting AJAX responseText as the source-argument in an audio-source-tag. Suprise: It doesn't work in Safari on my iPad 2, the player just fails to load the "file", although this works fine in Chrome on Windows. Possibly a size limitation for Base64-URIs on Safari / iOS?
And again: Even if this was working in iOS/Safari I don't know of a way to determine a real progress from an AJAX-query...
The last thing I was thinking of was not replacing the audio- or source-tags when loading a song but leaving them in the DOM-structure, check if it's already there when a song is loaded and just add a new audio-tag if a song hasn't been loaded yet. Doesn't work...
If you add multiple player-instances dynamically (again no matter if HTML 5 -tags or QuickTime-objects) instead of "overwriting" them, Safari forgets about ever having loaded the first MP3 as soon as you even insert a new audio- or QuickTime-Element into the DOM tree - you don't even have to load / play something in the new instance! Repeated playback without complete reloading of the file just works as long as you don't playback or insert anything else audio / media related. BTW: Just using Audio-objects in JavaScript and "saving" them into an array doesn't work either / doesn't make Safari cache anything.
This produces lots of unnecessary traffic and takes unnecessary long time if you're in a cellular network with a low bandwith!
I'm working at this problem for three days now without even coming close to a solution...
Any ideas?
I'm almost certain that this is by design and can't be overridden; the CoreMedia stuff purposefully streams the file as needed and throws it away without caching it when the player is dismissed. This is due to limited storage, battery life, etc since the vast majority of the time, you load one media file, play it however many times, then get rid of it. Whenever the content-type is a media type this will happen.
Another post referenced the idea of encoding the data in a PNG to get the browser to cache it, but I haven't attempted that.
You might try merging the various audio samples into one file then transmitting the start/stop times for each sample (an index basically); then you can load the file in a single audio player and jump to the necessary location and play for just the specified time. If that location hasn't been downloaded yet, I believe Safari will use range headers to jump forward in the file (but that can depend on exactly what type of container and whether the container has an index).
Another alternative would be using a streaming media server that can dynamically play audio. Just have the stream going when the player is engaged but streaming silence (the right protocol should use minimal bandwidth for this case), then requests for a sample trigger the streaming server to play that sample. Not ideal or terribly efficient unfortunately.
In case anyone's still pulling their hair out over this stuff (I lost a weekend to it recently), Safari on iOS 6 has a Web Audio API.
https://developer.apple.com/technologies/ios6/
Try this trick I have seen on my twitter some weeks ago but never actually got the time to try: Append an iframe, and set the source to the media file URL. It sounds weird, but it was from a very popular JS guy tweet...
Oh and I see it here: How can I autoplay media in iOS >= 4.2.1 Mobile Safari?
Def worth a try