tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sushil Prusty <sushil.pru...@gmail.com>
Subject Re: Problem with Transfer-Encoding
Date Tue, 08 Jul 2014 04:12:02 GMT
Hello Users

Thanks for all your various input.
Terence :- Yes you found the root cause of the issue ,I did same changes in
my jsp page and it's worked fine.

I was confused with FIREBUG tool but after using HTTPFOX tool it showed me
what value going from my browser.


Thanks user for your time.



On Mon, Jul 7, 2014 at 10:37 PM, Terence M. Bandoian <terence@tmbsw.com>
wrote:

> On 7/5/2014 6:36 PM, André Warnier wrote:
>
>> Sushil Prusty wrote:
>>
>>> Dear User
>>>
>>> Thanks for you input.
>>>
>>
>> You're welcome.
>> First, a foreword : I will try my best to help you, but doing this is
>> very difficult, and doing it via email is even more difficult.
>> I was not kidding when I wrote earlier that even looking at the data may
>> make it change.
>> Of course, that is not really true, but the fact of cutting and pasting
>> this data, from your saved HTTPFox trace into an email that you send to the
>> Tomcat list, and then the Tomcat list server forwarding this to other
>> people in a new email, may again decode and re-encode this data several
>> times, and confuse the situation totally.
>> So we need to be very, very systematic, and make sure that what we see is
>> really what we get, ok ?
>>
>> What you should really do, is to save the original HttpFox data to a
>> file, then save that file, then zip that file, then post it somewhere where
>> we can get this zip-file.
>> So that we can download it, unzip it, and then be sure that we are really
>> seeing the same data as you do.
>>
>> In the meantime, a question :
>>
>>  I just debugged using HttpFox here is below you find header
>>>
>>> (Request-Line)    POST /test/testUserEditAction.do?dojoIframeSend=true
>>>
>>
>> The above request line is triggered by something.
>> By what ?
>> Is that a link or button on a HTML page which is currently loaded in your
>> browser ?
>>
>> If yes, then before you actually click this link, can you in your browser
>> use the "View..Character set" function, and tells us what the browser
>> thinks about the current page loaded in the browser, before you even send
>> this request to the server ?
>>
>> The reason why I am asking, is that this is the character set which the
>> browser will most probably use to encode the text data that it sends to the
>> server (when you click the link).
>>
>>
>> Then see the note below, in the text.
>>
>
>
> I agree with André about the difficulties of debugging character
> encodings.  A couple of things you might check are the character encodings
> of the page and the form.  The character encoding of the page may be set
> with the Content-type meta tag:
>
> <meta http-equiv="Content-type" content="text/html;charset=UTF-8"/>
>
> For the form, I believe the character encoding defaults to the character
> encoding of the page but may be explicitly set with the accept-charset
> attribute:
>
> <form method="post" action="action.do" accept-charset="utf-8"></form>
>
> Hope that helps.
>
> -Terence Bandoian
>
>
>
>
>>  HTTP/1.1
>>> Host    **********
>>> User-Agent    Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:30.0)
>>> Gecko/20100101 Firefox/30.0
>>> Accept text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
>>> Accept-Language    en-US,en;q=0.5
>>> Accept-Encoding    gzip, deflate
>>> Referer    https://s
>>> ************Action.do?action=login&flashVersion=9.0.47.0&
>>> hasCorrectFlashVersion=false&product=xyz
>>> Cookie    JSESSIONID=******************; doNotShowStartupOnLoad=true
>>> Connection    keep-alive
>>> Content-Type    multipart/form-data;
>>> boundary=---------------------------*******************
>>> Content-Length    4039
>>>
>>>
>>> In Post body
>>>
>>>
>>> -----------------------------1550434539176507601876254213
>>> Content-Disposition: form-data; name="disclaimerText"
>>>
>>> Zażółć gęślÄ jaźń! ta funkcjonalność nie
jest wspierana
>>>
>>
>> The line above may or may not have been further corrupted (compared to
>> the original that you see), by the simple fact of copying this text into
>> your email.
>> But assuming for a moment that it was not, and that it really is what it
>> looks like above, there is some kind of a problem :
>>
>> (You'll have to follow carefully here)
>> If I take the original text line which you posted in your first message :
>>
>> Zażółć gęślą jaźń! ta funkcjonalność nie jest wspierana*
>>
>> and I imagine that internally, this is encoded as UTF-8;
>> Then if I look at that same series of UTF-8 characters, but now examine
>> the *bytes* that compose these characters and view them in ASCII, I should
>> see this :
>>
>> Zażółć gęślą jaźń! ta funkcjonalność nie
jest wspierana*
>>
>> But if you compare this carefully, with the string as it appears in your
>> HttpFox trace, you will see that it does not match exactly. For example,
>> look at the last 2 letters of the word "funkcjonalność", in both versions.
>>
>> So there appears to be some discrepancy between the character set which
>> your browser is really using (to send data to the server), and the UTF-8
>> that your server seems to expect.
>>
>> Furthermore (and put this on account of my suspicious nature if you want)
>> :
>> The second part of that message, in Polish, means : "This functionality
>> is not supported".
>> Which triggers the question : what kind of HTML page would be sending
>> this phrase, as part of the data, in a POST to a server ? Can you give us
>> some context as to what you are trying to do here ?
>>
>>
>>
>>  -----------------------------1550434539176507601876254213
>>>
>>> I believe psot data got changed before   reaching   to server .
>>>
>>>   Do you have any ideas what's wrong here, where the error might be ?
>>>
>>>
>>>
>>>
>>> On Sat, Jul 5, 2014 at 9:08 PM, André Warnier <aw@ice-sa.com> wrote:
>>>
>>>  Konstantin Kolinko wrote:
>>>>
>>>>  2014-07-05 9:24 GMT+04:00 Sushil Prusty <sushil.prusty@gmail.com>:
>>>>>
>>>>>  Hello,
>>>>>>
>>>>>> summary of my Problem:
>>>>>>
>>>>>> When a client POSTs Tranfer-Encoding data   using browser ,
>>>>>> my server is not processing the request character encoding properly
.
>>>>>>
>>>>>>
>>>>>> I send the following request:
>>>>>> Content-Type text/html;charset=UTF-8
>>>>>> Date Sat, 05 Jul 2014 05:10:09 GMT
>>>>>> Server Apache-Coyote/1.1
>>>>>> Transfer-Encoding chunked
>>>>>> *disclaimerTextZażółć gęślą jaźń! ta funkcjonalność nie
jest
>>>>>> wspierana*
>>>>>>
>>>>>>   Full details:
>>>>>>
>>>>>> My application running on  apache-tomcat-7.0.40
>>>>>>  and Java
>>>>>> 1.6.x)  in linux box.
>>>>>>
>>>>>> Below response is changed once it's reach to my controller
>>>>>> *ZażóÅÄ gÄÅlÄ jaźÅ! ta funkcjonalnoÅÄ nie jest wspierana*
>>>>>>
>>>>>> I have below configuration  in server.xml
>>>>>>
>>>>>>  <Connector port="80" protocol="HTTP/1.1" connectionTimeout="20000"
>>>>>> maxPostSize="5242880" maxParameterCount="25000"/>
>>>>>>     <Connector
>>>>>>             port="443"
>>>>>>             protocol="HTTP/1.1"
>>>>>>             scheme="https"
>>>>>>             noCompressionUserAgents="gozilla, traviata"
>>>>>>
>>>>>> compressableMimeType="text/html,text/xml,text/javascript,
>>>>>> text/css,application/javascript,application/json"
>>>>>>             URIEncoding="UTF-8"
>>>>>>     />
>>>>>> and in my catalina.sh
>>>>>> set JAVA_OPTS=-Djavax.servlet.request.encoding=UTF-8
>>>>>> -Dfile.encoding=UTF-8
>>>>>> (...)
>>>>>>
>>>>>>  As a sanity check:
>>>>> 1) That "I send the following request" listing looks more like a
>>>>> response, not a request. (E.g. the "Server Apache-Coyote/1.1" header
>>>>> makes no sense in a request).
>>>>>
>>>>> So you are lying somewhere.
>>>>>
>>>>> There is no point for me to try guessing what you are doing. You may
>>>>> have confused "reading" with "writing" somewhere, and without source
>>>>> code one cannot verify your words.
>>>>>
>>>>> You have to provide a step-by-step instruction and enough source code
>>>>> so that a person who is not familiar with your system were able to
>>>>> reproduce your problem.
>>>>>
>>>>>
>>>>> 2) Content-Type says "text/html", but that line of text is not a valid
>>>>> HTML document.
>>>>>
>>>>>
>>>>>  +1
>>>> Character encoding/decoding issues are hell to debug as it is, because
>>>> they are like quantum physics : even looking at them can change them.(*)
>>>> So you need to provide *accurate* and "raw" information, otherwise it is
>>>> just a loss of time for everyone.
>>>>
>>>> Use a browser plugin like HttpFox, LiveHttpHeaders, HttpFox or similar
>>>> to
>>>> monitor the requests being sent and responses being received, at the
>>>> browser level.  All these plugins allow you to selectively dump
>>>> requests/responses to a file.  Do that.
>>>> Also, check in your browser that when you receive a response page back
>>>> from the server, your browser is really seeing this response in the
>>>> proper
>>>> character set (use "View.. Character encoding..").
>>>>
>>>> "Transfer Encoding" has nothing to do with the *character encoding* of
>>>> either the request or the response.  The little imprecise data that the
>>>> OP
>>>> provided above /suggests/ that there is some double encoding taking
>>>> place
>>>> /somewhere/, but so far it could as well be in the email client that he
>>>> used to post to the list, as anywhere else.
>>>>
>>>> (*) with the wrong editor, or the wrong locale e.g.
>>>>
>>>>
>>>>
>>>> ---------------------------------------------------------------------
>>>> To unsubscribe, e-mail: users-unsubscribe@tomcat.apache.org
>>>> For additional commands, e-mail: users-help@tomcat.apache.org
>>>>
>>>>
>>>>
>>>
>>
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@tomcat.apache.org
> For additional commands, e-mail: users-help@tomcat.apache.org
>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message