does somebody known the vcl : SuperObject is thread safe???
I use it in threads, and never noticed thread safety problems.
It uses the Interlocked* mechanism were needed, so as long you also use
it safely...
--
To unsubscribe or change your settings for TWSocket mailing list
please goto http
A quick fix that seems to correct the issue is to add a FTargetPort:=
FPort to the procedure THttpCli.StartRelocation in the
OverbyteIcsHttpProt unit.
procedure THttpCli.StartRelocation;
...
CleanupSendStream;
LoginDelayed;
end
else begin
+FTargetPort:= FPo
On 11-12-2012 17:30, Arno Garrels wrote:
- Original Message -
From: "RTT"
To: "ICS support mailing"
Sent: Tuesday, December 11, 2012 5:59 PM
Subject: Re: [twsocket] Did I find a bug in THttpCli?
The port after the redirect is the correct 8080. I
just don&
On 11-12-2012 16:38, Arno Garrels wrote:
- Original Message -
From: "Angus Robertson - Magenta Systems Ltd"
To:
Sent: Tuesday, December 11, 2012 5:25 PM
Subject: Re: [twsocket] Did I find a bug in THttpCli?
If you GET this URL:
http://www.htmlvalidator.com/test/cookies/test-redirect.
I tested with current ICSv8/v7 and it works perfectly (checked with Wireshark).
The only problem is that property Location doesn't include the non-default
port number. A workaround would be to get the Location header manually.
In a browser I get an html that says:
Server port is 8080.
With ICS
I have put some more log files on the website:
www.deephaven.co.uk/downloads/ftpuplog.zip
Two files "pass" (small file) and "fail" (large file) contain more logging
info.
Not sure if this tells us more about the end of the transfer though.
Maybe what you need to do is to keep the control ch
Is it possible to integrate CGI/PHP in any of the ICS components?
http://lists.elists.org/pipermail/twsocket/2011-February/043191.html
Check the idRunner project
http://sourceforge.net/projects/psvlib/files/
Not for ICS, but not so difficult to adapt to the ICS THttpWebserver
--
To unsubscrib
On 28-11-2012 08:51, Graham Powell wrote:
1: Can the FTP client be modified to generate the ftpPutAsync event on
something else as well as the Transfer Complete. Wireshark always shows some
message at the end that contains FIN ACK.
Sometimes, when I'm updating my web site using Filezilla, and s
What's the scenario where I need to change the mime type after I
have searched for it already?
So the application can easily check what actual MIME type has been used,
and dynamically change it if incorrect.
That's an unneeded double-check. If you are providing the MimeTypesList,
you will ma
On 10-11-2012 14:42, Stan wrote:
I think loading hundreds of types from Registry when 'js' is missing
in the
default list is overkill, because most of them are not Internet-involved.
It don't takes that long, but even so It's also my opinion..., If it's
not needed, don't do it.
Because of t
On 22-10-2012 19:04, LeNif wrote:
My problem is when I disconnect the socket and I immediately reconnects, I get
an error EStackOverflow
And if the reconnect is not "immediately" (you wait more time), the
error does not occur?
Have you checked the stack trace, in that stack overflow condition
GetPeerAddr
onHTTPRequestDone is also a good place, if you want to log bytes
delivered, etc..
Hi,
Version 7.39
I need to log the IP address of the client when it connects to the server.
OnClientConnect seems to be the best place but I do not see any way to get
the IP address of the conn
On 21-07-2012 21:22, zayin wrote:
I was wondering if there is a better solution. Some answer code that tells
the browser the data was received but not to change page or to reload the
button host page.
http://www.w3schools.com/ajax/default.asp
--
To unsubscribe or change your settings for TWSocke
Yep. Just change the line
Result := _FileDateToDateTime(SearchRec.Time);
to
Result :=
_FileDateToDateTime(SearchRec.FindData.ftLastWriteTime);
The correct fix should be
Result :=
_FileDateToDateTime(_FileTimeToDosDateTime(SearchRec.FindData.ftLastWriteTime));
.
Last-Modified: header should always be displayed time as GMT or UTC
Not an HTTP imposition. As long it's a valid http-date, it's fine.
There is only the need to use the same convention when sending, and
comparing these dates.
However, there is a bug in OverbyteIcsHttpSrv.pas in functio
On 28-03-2012 19:59, François Piette wrote:
I just
want to have an idea about how many of you are still using an old Delphi
version.
Still using the old good D7.
If you are asking this them I suppose you haven't coded yet any
significant part of the planned new version?
What changes you have in
On 04-03-2012 15:24, François Piette wrote:
I don't really understand the real world use case for such
"WebSockets" in the context of a IC/Delphi/WebServer application
Now that web browsers are implementing it
(http://en.wikipedia.org/wiki/WebSocket#Browser_support), it would be
nice to have
On 05-02-2012 17:38, Arno Garrels wrote:
Problem is that AnswerStreamAcceptRange can only be used
for OK-responses.
If I'm understanding this correctly, can't the status parameter be
added, and then response handled accordingly?
Right now I don't need this functionality. I'm just trying to under
On 05-02-2012 12:15, Arno Garrels wrote:
So there must be an option to allow
ranges on a per request basis. A new method would provide such an option.
I see. A new method then (and "AnswerStreamAcceptRange" seems more
appropriate).
Or, probably better, replace the AnswerStream code with the ne
On 04-02-2012 16:45, Arno Garrels wrote:
That's a question too, IMO accept/handle ranges should be optional and
silently changing the default behaviour of an existing method is
questionable, no?
I totally agree with that. The server should have an options property to
specify this kind of fun
On 04-02-2012 11:25, Arno Garrels wrote:
Tobias Rapp wrote:
Hi!
I have added a new function "AnswerStreamPart"
Shouldn't that method be named "AnswerStreamAcceptRange",
or at least "AnswerStreamRange"?
How about if the this ranges handling code is added to the AnswerStream
method? Why and
On 29-01-2012 16:19, François Piette wrote:
Please provide the link to the API you are trying to implement. Just that
part is not enough.
https://content.etrade.com/etrade/estation/pdf/API_Technical_Documentation.pdf
--
To unsubscribe or change your settings for TWSocket mailing list
please goto
On 29-01-2012 16:00, Iqbal Husain wrote:
3. Getting Account List. {REST resource HTTP Method GET}
4. Getting Account Balance. {REST resource HTTP Method GET}
In all of my above example, I put the OAuth parameters in
TSslHTTPCli.SendStream and it works fine.
How's that possible?!
Why this fix has not been added?
The first one is not really important, even if completes the logic
already taken by the ParseURL for URLs missing the protocol , but the
second change, used by the THttpCli when parsing redirection URLs, is
IMO mandatory, because URLs beginning with '//' are val
This fix applied to the ExtractURLEncodedValue, unit OverbyteIcsHttpSrv
Jul 06, 2002 V1.05 Thomas Smyth fixed
ExtractURLEncodedValue which had problem with names
beginning with
same sequence of chars.
Needs to be applied to the GetCookieValue function too
On 06-01-2012 17:31, Arno Garrels wrote:
What do you mean by "cURL"?
http://curl.haxx.se/
--
To unsubscribe or change your settings for TWSocket mailing list
please goto http://lists.elists.org/cgi-bin/mailman/listinfo/twsocket
Visit our website at http://www.overbyte.be
On 13-12-2011 19:21, Arno Garrels wrote:
We don't want to open and load a text file each
> time a new type is found.
True, that would be dog-slow.
For the generality of the HTTP server applications, other than the 10 or
15 most common MIME types (that should always be in the list) and the
on
On 13-12-2011 14:43, Arno Garrels wrote:
I would not link headers with MIME types (in the same list) but rather
keep it clear and simple. This new class should also be usable from
other components such as the SMTP client.
Totally agree with that. Associating header elements to the content-type
i
On 13-12-2011 08:31, Angus Robertson - Magenta Systems Ltd wrote:
I'm just about to change to HttpServer MIME handling again, to use a
look-up list read from the Windows classes registry or a text file) when
the server starts, to replace the current hard coded MIME list.
If filling from the reg
On 13-12-2011 00:16, Lars Gehre wrote:
Overriding said function would mean, I would have to duplicate the code and
with each update of ICS I would have to check all my projects for this
overriden function. Not very practical. It's easier to change the ICS code
once every update...
What code are
If you call the SendDocument yourself, you can use this overload version
procedure SendDocument(SendType: THttpSendType; const CustomHeaders:
string); overload; virtual; { V7.29 }
If not, you can override it and pass your custom header in the
CustomHeaders parameter, calling the inherited Se
RFC2616 references RFC2396 as a source of information about URI in the
description of a general syntax (Chapter 3.2.1). Then the next chapter
(3.2.2) specifies the http URL which /must/ start with the "http:"
scheme.
The RFC2396 also don't say noting about a default protocol for such
situat
On 27-10-2011 18:33, Arno Garrels wrote:
I'd be happy to see more contributions from the ICS users in general.
1st change
In unit OverbyteIcsUrl, procedure ParseURL(..., replace
if (url[1] = '/') then begin
On 27-10-2011 19:28, Francois PIETTE wrote:
If I'm wrong, please point me to the exact text in the /HTTP/ standard
(RFC2616).
From RFC2396, that merge RFC1808 with two others, and that is
referenced in the RFC2616.
The bellow provided example isn't producing anymore the relocation with
such "begin with //" address, but this don't invalidate the fact that
the THTTPCli cant resolve a url such as:
//www.google.com
nor handle a relocation with the "location" field set to equal URL
According to the rfc180
According to the rfc1808.txt, an URL can start with //
The ParseURL function will fail in such cases. Also, during a
relocation, the THTTPCli fail to parse a "location" field with such URLs.
Here is an example, that result in a relocation with these characteristics
http://twitpic.com/show/thumb/
On 21-10-2011 19:18, Angus Robertson - Magenta Systems Ltd wrote:
Oct 21, 2011 V7.41 Angus added OnHttpMimeContentType to allow custom
ContentTypes to be supported for unusual file extensions.
Seems to be working, but why the heck you call the internal
DocumentToContentType, even if it is assign
I also find myself with the need to edit the DocumentToContentType
function, each time I update my local copy of ICS.
This has not being addressed yet, so how about if, instead of the more
complex implementation proposed here, an assignable
OnDocumentToContentType property is added to the http
On 17-09-2011 09:56, Francois PIETTE wrote:
I see. The idea is the usage of the widestring just as a easy way to
call the, behind the scene, SysAllocString and SysFreeString, that we
need in this case. But we can always treat it as a buffer, setting
its length, and moving data, explicitly, with
On 16-09-2011 20:20, Arno Garrels wrote:
As far as I read this thread the OP uses a non-Unicode Delphi version.
So "string" maps to AnsiString. Any assignment of an AnsiString to a
WideString leads to an implicit string cast to Unicode (Win API
WideCharToMultiByte) internally, that's one of the
On 16-09-2011 18:58, Arno Garrels wrote:
That would work which plain text
Why your reference to "plaint text"? A WideString can carry any data, or
I'm missing something?
And WideStrings are now relatively fast, in Vista and 7, if compared to XP.
But I really don't know if two calls, one to ge
I think you could also use a WideString var parameter.
function ReadMessage(var msg:WideString):boolean; stdcall;
var
Note: AnsiString;
begin
EnterCriticalSection(CritSectn);
if (NotesList.Count> 0) then
begin
msg := NotesList.Strings[0];
NotesList.Delete(0);
Result := true;
end
I would loose control on the group
Loose control in what sense? My understanding is that the nabble service
will just archive the mailling list, and preset it in a forum alike
manner. Similar, but better, to the already available here
http://www.archivum.info/twsocket@elists.org/
And in syste
When time allows I'll have
to find a final solution with or without Live Mail.
Try Thunderbird. Simple, full featured, and with plenty of add-ons
(https://addons.mozilla.org/en-US/thunderbird/) to extend it further.
And, as developer, you can create your own extensions ;-)
--
To unsubscribe or
You just need to put your Server application accessible from the
Internet, opening the used TCP port in the firewall, and routing it
(port forwarding) in the router that that PC use to connect to the
Internet.
Your Clients should then connect to the public IP of that router.
It's exactly the sa
Anyway I never though names would be a problem in a non official unit...
Official or not, function names should reflect the propose, or readers
will fail to find it, of find wrong code for what they are looking for.
Anyway, I was just digging to see if something else was behind this
incorrec
If this is HTML related, why you use "XML" in the names?!
Why not use the faster, and memory efficient, HTTPApp.pas code methods?
Ok, I took your example and created a (I suppose) complete encode-decode
unit.
Daniele
--
To unsubscribe or change your settings for TWSocket mailing list
please go
Delphi HTTPApp unit, function HTMLDecode
Hi all,
I need to convert text that uses special HTML encoding such as "&" or
" " to their corresponding charachter.
Do any of you know if such function exists?
'Cause at the moment I'm thinking of writing it myself but it's gonna take
quite a while...
On 11-04-2011 16:49, Francois PIETTE wrote:
I don't know if there's a better way
That is fine.
So the HttpCli built-in timeout is not enough?!
--
To unsubscribe or change your settings for TWSocket mailing list
please goto http://lists.elists.org/cgi-bin/mailman/listinfo/twsocket
Visit our web
On 18-03-2011 17:03, zayin wrote:
It all shows up fine but when the user hits the accept button I get back the
'Tagname' with the added '+
' symbols " Stop+Pump+CRT". Why not just " Stop Pump CRT"?
http://www.cs.cf.ac.uk/Dave/PERL/node201.html
Use ICS ExtractURLEncodedValue function to decode t
On 17-03-2011 17:12, FrancoGG wrote:
the application's memory usage keeps increasing every time a user connects
If you are using a custom THttpConnection, check if you are missing to
free any of its owned objects in its destructor method.
--
To unsubscribe or change your settings for TWSocket m
On 07-03-2011 06:31, Arno Garrels wrote:
TWSocket property KeepAliveOnOff may be used to setup winsock to send
keep-alive packets in the background in order to detect such brocken
connections, however that's not reliable since both peers must support
it and routers have to route the keep-alive pa
On 07-03-2011 06:31, Arno Garrels wrote:
That's sounds like a bug. I would expect that THttpCli is reset to
default values after that error?
I don't see any reference to "10053" in any of the ICS code, so I
suppose this situation is not being handled internally by the THttpCli.
Because the THttp
Isn't the CtrlSocket.State reliable to know if a connection is still on?
I'm trying to reuse an THttpCli, but I'm getting 10053 errors if a
persistent connection is idle for some time, I suppose the server
keepalive timeout.
The ICS THttpCli code check this CtrlSocket.State property, to try to
On 01-03-2011 10:06, A Drent wrote:
I've looked into the usermade page, but which sample do you refer to?
HTTPCLIEXAMPLE.ZIP ?
ICS_WebInterface.zip
This one may help too
http://users.telenet.be/ws36637/
--
To unsubscribe or change your settings for TWSocket mailing list
please goto http://list
On 27-02-2011 23:31, brian - wrote:
I added the ShellExecute import to avoid adding another unit; I rather
not include entire units for things where I need only a couple procs and
such, it adds up to the exe unnecesarily.
The smart linking feature should work ok for the shellapi unit.
As for
On 27-02-2011 19:29, brian - wrote:
I don't really like the idea of embedding the TWebBrowser object, bulky and
buggy activex stuff.
The WebBrowser control works very well for this simple "navigate to a
page" task.
If you want to test, just replace your RequestPIN code with the next one.
On 27-02-2011 18:39, brian - wrote:
I like it tbh, even with the
annoyance of havign to login and retrieve the pass for the user, it's only
required once.
As I said, Twitter don't require the PIN, so you can ignore that step.
User only need to authorize the application, and forget about the PI
On 27-02-2011 17:18, brian - wrote:
use an alternate method to wont require the user to visit the twitter site
to get the PIN code, the app will simulate a browser login, catch the
cookies and proceed to auth on its own to retrieve the PIN.
This will work until twitter decide to change these htm
You can remove the b64ASM.pas dependency using the ICS equivalent
//b64asm.Base64Encode(HMAC_SHA1_EX(SignBase,signkey));
OverbyteIcsMimeUtils.Base64Encode(HMAC_SHA1_EX(SignBase,signkey));
Probably not so fast, but not really important in this case of small
length strings.
--
To unsubscr
On 27-02-2011 16:01, brian - wrote:
Simply added deflate to the result. Without that, I don't get any data back.
Juts tested your component here, without that modification, and works
just fine. What exactly happen in your case. What data you don't get?
--
To unsubscribe or change your settings
it's time for a rewrite from
scratch IMO.
And give it a different name, to bypass the "no brake early code" rule,
or some of the annoyances will have to remain. Or start a new branch of
renewed components.
And, IMHO, if technically possible, the THttpCli should mix HTTP and
HTTPS, and handle in
On 18-02-2011 13:52, Francois PIETTE wrote:
It starts to become interesting :-)
What other dependency is there ?
None
What is the license for that component ? Could it be added to ICS
distribution or ICS usermade web page ?
From what I can see, MIT License for all the code from the author
On 17-02-2011 18:50, Angus Robertson - Magenta Systems Ltd wrote:
Nothing mine, but a rapid Google search reveals this two
> http://www.lakraven.com/delphi-stuff/ttwitter/
which use the Indy HTTP component, not ICS.
It's a complex component, not trivial to convert to ICS.
I'm already tweeting f
On 17-02-2011 22:35, brian - wrote:
I got this working, httpcli is just working as expect, was nothing wrong
with it. I will keep that extra variable for the headers anyway, it's a lot
more convenient than the notify event.
Great, thanks for the followup.
Are you using the Indy components, for
On 17-02-2011 18:50, Angus Robertson - Magenta Systems Ltd wrote:
> http://sourceforge.net/projects/simplerss/
also use Indy HTTP, a complex XML project.
In this case the Indy stuff is used only to format datetime values, very
easy to replace.
There is also an TIdHTTP component, used only in t
On 17-02-2011 18:50, Angus Robertson - Magenta Systems Ltd wrote:
Nothing mine, but a rapid Google search reveals this two
http://www.lakraven.com/delphi-stuff/ttwitter/
which use the Indy HTTP component, not ICS.
It's a complex component, not trivial to convert to ICS.
http://sourceforge.net/
On 17-02-2011 17:38, Angus Robertson - Magenta Systems Ltd wrote:
Or are you offering to contribute them?
The Twitter component could be provided by brian - , that the other day
asked in this list about passing OAuth parameters in the request header.
--
To unsubscribe or change your settings fo
Because ICS does not have Twitter or RSS components!
Or are you offering to contribute them?
Nothing mine, but a rapid Google search reveals this two
http://www.lakraven.com/delphi-stuff/ttwitter/
http://sourceforge.net/projects/simplerss/
--
To unsubscribe or change your settings for TWSocke
On 17-02-2011 17:18, Arno Garrels wrote:
Since we already have an internal e-mail notification sent to TeamICS
members (based on TSmtpCli) it would be easy to send another copy to
some mailing list.
How about upgrade that program to add a new entry to a RSS feed, and
post a new Twitter message?
During a relocation an intermediary RequestDone events is fired. Is
this really needed, and for what propose?
Relocation is considered as a another automatic request. You may
cancel it on the fly or simply disable the feature with
FollowRelocation property.
Yes, but if I want to follow it,
During a relocation an intermediary RequestDone events is fired. Is this
really needed, and for what propose?
If yes, then I think it should be fired with the correct response status
code, not zero as now, so it can be clearly detected as not the final
request done even. Also, an "httpRelocating
On 17-02-2011 15:17, Francois PIETTE wrote:
Depending on the number of interested peoples, we will either send
individual emails or use another mailing list.
How about an RSS feed and a Twitter ICSSVN account? :-)
--
To unsubscribe or change your settings for TWSocket mailing list
please goto h
On 12-02-2011 22:31, brian - wrote:
Authorization: OAuth oauth_callback="oob", realm="",
oauth_nonce="5B2407849960FC2B4EC23007EA63E8ED",
oauth_timestamp="1297549627", oauth_consumer_key="5BZYtiAzGvQTL0ZEhAn45w",
oauth_signature_method="HMAC-SHA1", oauth_version="1.0",
oauth_signature="RjpaedDdvs6
On 12-02-2011 22:31, brian - wrote:
there's no option in httpcli to manually modify the headers
before a request
You can do it from the OnBeforeHeaderSend event.
--
To unsubscribe or change your settings for TWSocket mailing list
please goto http://lists.elists.org/cgi-bin/mailman/listinfo/twso
On 27-01-2011 18:27, Arno Garrels wrote:
Without the certificate(s) and private key(s) he may intercept transparently
as long as he likes. When he wants to decrypt the session on the fly he
has to go thru the handshake process on behave of the victim by presenting
the stolen certificate(s), actin
With a stolen key that's easy.
Sure, and this is exactly what SSL try to circumvent.
But not so easy if the encrypt key is not a fixed value, but a variable
one. The attacker will need to stole the client or server code and
reverse engineering it too.
This is also valid for SSL.
No, the
It is a very simple monitoring system.
SSL is okay for this as far as I am concerned.
As others have say to you already, just concentrate your efforts in the
development of the client and server code. Leave the data encryption to
the last stage of the project.
Starting with the SSL componen
It's easy to implement encryption but you'll have to add key exchange also and
exclude possibility of key sniffing what is quite harder.
Just start with a fixed strong password, and add to it a variable salt
that can be a hash of some of the client/server header fields. One of
this fields c
"Man in the Middle" attacks don't work if the "man in the middle"
don't know how to handle the encrypted data/protocol he is
intercepting.
True, and how do you manage that is not happening?
Can't be happening because the man in the middle can't generate valid
data, or alter intercepted data
On 26-01-2011 20:15, daniel cc wrote:
I Insist because I have no other options because I don't have
knowledge of doing it in the other way.
So, you just need to learn how to use a common web search engine and how
to make the right questions at the right places.
Reading old replies to your que
SSL/TLS aktually uses common, powerful and strong encryption algorithms.
However secure peer to peer communication is much more than just that.
SSL security includes, for instance, peer verification and protects
against "Man in the Middle" attacks.
"Man in the Middle" attacks don't work if the
(I am using SSL sockets in both sites)
Why you insist in use SSL if the client and server applications are
coded by you?!! Why use a standard that only exist because of the need
to connect many different implementations of clients to many different
implementations of servers?
You will be mu
On 18-01-2011 13:50, daniel cc wrote:
Clients are not having static IP and the place where clients are gonna
be doesn't allow top open any ports for incoming trafic etc.
Connecting from the server to the server placed in clients isn't going
to work in my case or perhaps I am mistaking?
If t
On 18-01-2011 11:25, daniel cc wrote:
So you build your combobox, wait 10 seconds, one client disconnects, all
the clients numbers reduce by one and your combobox is out of date. When
you try to send a command to one of them it may be the wrong client or
you'll get an exception because there ar
It gives error when clicking the connect button
What's the connect error?
however,
I will be able to fix the errors but,
this still doesn't help me much because,
I need to be able to send the commands from the server to the client
as well.
I can find nothing about how to send anything from
Now,
The project "OverbyteIcsSrvTcp" has absolutely nothing in inside to
help me. I can't find anything about sending or receiving in this
project.
The OverbyteIcsSrvTcp uses the OverbyteIcsTcpCmd.pas, where you will
find the code of the server.
The project "OverbyteIcsClient7" gives m
Are the headers as you guys would expect?
The "Vary: *" header prevents the response from being cached, so IE
"gets confused" when processing the save attachment request.
--
To unsubscribe or change your settings for TWSocket mailing list
please goto http://lists.elists.org/cgi-bin/mailman
Here is a sample that work.
Get the 'multipart/form-data' encode tstream class code from this unit:
http://read.pudn.com/downloads116/sourcecode/internet/495621/Gz235_200671623235689/cn700_del549790557/IndyMultipartFormData/MsMultiPartFormData.pas__.htm
Start a new Delphi application, and in th
Any kind of demos for the person who has never done such things before?
But what kind of data are you sending/receiving. Plain text, binary?
You just need to encrypt it before send, and decrypt at receive.
Here is a good cryptography library
http://www.cityinthesky.co.uk/cryptography.html
--
To
Is it possible to encrypt the connection between the client and server?
The connection, or your custom data?
--
To unsubscribe or change your settings for TWSocket mailing list
please goto http://lists.elists.org/cgi-bin/mailman/listinfo/twsocket
Visit our website at http://www.overbyte.be
On 11-09-2010 09:22, Angus Robertson - Magenta Systems Ltd wrote:
What is the purpose of the OPTIONS method?
From http://www.w3.org/Protocols/rfc2616/rfc2616-sec9.html
"The OPTIONS method represents a request for information about the
communication options available on the request/response ch
Done. I've now sent you a PM with the .diff file.
Feel free to apply your changes to the latest SVN revision and send me
the patch file by PM. Changes MUST not break existing code and a change
log in unit's comment section was helpful too.
--
To unsubscribe or change your settings for TWSocke
Thank you for the explanation Angus.
But, why don't keep the naming schema? TriggerAfterProcessRequest ( and
respective OnAfterProcessRequest) would be less confuse.
Rui
--
To unsubscribe or change your settings for TWSocket mailing list
please goto http://lists.elists.org/cgi-bin/mailman/list
On 10-09-2010 16:43, Arno Garrels wrote:
Another thing that would be useful, if added to the code base of the
> THttpConnection, is the possibility to send additional headers from
> the THttpConnection.SendDocument, to define cache control, etc..
> procedure THttpConnection.SendDocument(SendT
Hi everyone!
Is there any simple way to handle other than the GET, POST and HEADER
request methods,... as OPTIONS,...?
The way it is now, the THttpConnection.ProcessRequest procedure always
respond with a Answer501, not giving any chance to handle other methods.
I think a TriggerOnUnknownMeth
Is happening here too, but only if I continue with the step over
debugging (F8) after the 4 thread are created. Never happened while they
are created. No idea if that's what you want to say with "repeat
pressing F8 until the AV is raised."
The debugger will keep stepping the WmTest function, un
, default set to true, parameter, DetectUtf8, to define if
the auto-detection, and decoding, of UTF8 encoded strings must be made
automatically by the function after the URLDecode conversion
Hi RTT
Thanks for your tips. Finally I get it work but I should write code a
bit different
String usStr
Obviously what works fine is the sequence of decode transformations on
your encoded filename, not in the already decoded string as I posted. Sorry.
This is the correct example:
UTF8Decode(URLDEcode('Sin%20t%C3%ADtulo%201_html_m5b7e3440.jpg'));
This seems to work fine
UTF8Decode(URLDecode('Sin
This seems to work fine
UTF8Decode(URLDecode('Sin título1_html_m5b7e3440.jpg'));
Curiosity struck, so I googled it...
Apparently you aren't the only one with this issue.
If you have the string before it gets encoded in the 1st place you can
convert to UTF8 first, then encode it to URL so you ca
1 - 100 of 144 matches
Mail list logo
