[HTTP client] - Could fail in retry_pipeline/2 due to an incorrect assumption of an internal returnvalue.

Own Id: OTP-6791

[HTTP client] - The check of the value of the transfer encoding has been updated so that it is case insensitive.

Own Id: OTP-6807

[HTTP client] - When receiving a 304 "Not Modified" reply, the httpc_handler will no longer expect to receive a HTTP body.

Own Id: OTP-6821

[HTTP client] - Parsing of the HTTP response failed when there was no headers fields in the response.

Own Id: OTP-6830

[HTTP server] - Started clean up of code so that it uses stdlib functions instead of reinventing them.

Own Id: OTP-6808

Minor Makefile changes.

Own Id: OTP-6689

http_base_64:encode/decode and httpd_util:to_upper/to_lower are now deprecated, as these functions has been moved to stdlib, using them will now cause an compiler warning and the documentation of them has been removed.

Own Id: OTP-6716

When making an asynchronous HTTP request and the underlying gen_tcp:connect failed with timeout (not a very common case) the return of the asynchronous HTTP request was delayed for "timeout" seconds. This happened due to the fact that when spawning a gen_server process the spawn will wait for the init function to complete. This is now avoided using proc_lib and gen_server:enter_loop/3, hence the asynchronous HTTP request return will not be delayed. Also the request timeout value is now propagated to gen_tcp rather that relying on the system default.

Own Id: OTP-6735

[tftp] The TFTP client/server implements now a more relaxed interpretation of the RFC 1350 regarding re-receive of acknowledgments. If multiple copies of the same acknowledgments is received the spurious ones are silently ignored. This fix was intended for inets-4.7.14 but accidentally it was not included in that release.

Own Id: OTP-6706 Aux Id: OTP-6691

Minor Makefile changes.

Own Id: OTP-6689

[tftp] The TFTP client/server implements now a more relaxed interpretation of the RFC 1350 regarding re-receive of acknowledgments. If multiple copies of the same acknowledgments is received the spurious ones are silently ignored.

Own Id: OTP-6691

[tftp] The TFTP client/server implements now a more relaxed interpretation of the RFC 1350 regarding re-receive of data packets. If multiple copies of the same data packet is received the spurious ones are silently ignored.

Own Id: OTP-6642

[httpd] - When calling httpd:restart there was a file descriptor and ets table leek due to the fact that the semantics of the function restart is not restart but reload.

Own Id: OTP-6573 Aux Id: seq10607

[tftp] Crash in tftp_engine:terminate/3 caused big crash report. The file was however transferred as normal, but the ugly printout is now gone.

Own Id: OTP-6596 Aux Id: seq10618

[http] - Chunked encoding updated to handle the empty chunk.

Own Id: OTP-6511

[http] - Removed minor bugs and dead code found by dialyzer.

Own Id: OTP-6522

[httpd] - The server no longer produces error messages when the client resets the the connection. This is not an error as far as the server is concerned.

Own Id: OTP-6484 Aux Id: seq10568

[tftp] The server is now silent by default. Error messages can however be displayed by setting the debug level to 'error' (new) or higher.

Own Id: OTP-6498

Bug in in handling of request timers has been corrected.

Own Id: OTP-6476

[tftp] - Daemon was only able to process one request when fd option was given.

Own Id: OTP-6480

[ftp, httpc] - Updated the internal ensure_started functions to handle that inets was started as an included application.

Own Id: OTP-6409 Aux Id: seq10546

[httpd] - Guard added to API function httpd_util:integer_to_hexlist/1.

Own Id: OTP-6397

[tftp] - Introduced ability to use prebound ports (see the option {port, Port} for more info). Added peer info (host and port) as new optional argument to prepare and open functions in tftp callback modules.

Own Id: OTP-6413

[httpd] - The http server could throw away pipelined requests leaving the client hanging.

Own Id: OTP-6310

[ftp] - Code for handling fallback to ipv4 was not exhaustive.

Own Id: OTP-6312

[httpd] - Incorrect handling of ipv6 address would crash the loading of the BindAddress parameter, which had the consequence that you could not start more than one HTTP server on an erlang node.

Own Id: OTP-6323

[httpc] - Some 30X codes, as for instance 302, should not always be automatically redirected but in inets-4.7.6 some restrictions where made too hard, never allowing an automatic redirection. (Automatic redirect should be allowed for get and head.)

Own Id: OTP-6332

[ftp] - The mode flag that could be used to force ftp active mode was ignored. Only the deprecated function force_active/1 works for prior releases to this one.

Own Id: OTP-6342

[httpc] - Added debug feature to turn on/off some basic erlang tracing on client processes.

Own Id: OTP-6326

[httpc] - The parsing of uri's was rewritten so that it should handle all types of uri's including ipv6 uri's.

Own Id: OTP-5677

[httpc, httpd] - Extensions and trailers where not properly handled by the chunk decoding implementation.

Own Id: OTP-6005 Aux Id: OTP-6264

[httpc] - A request resulting in an empty body is now returned without any delays.

Own Id: OTP-6243

[httpc] - When http:request/4 was used with a configured proxy, and the webserver returns 3XX code, http:request/4 entered an endless loop. Two problems was solved in this area, the absolute uri is now updated when a redirect is issued, so that the problem in this case will not arise, and the redirection endless loop detection was fixed so that will actually detect potential endless loops.

Own Id: OTP-6244

[httpc, httpd] - In some cases if a body contained the sequence "\r\n0" and was chunked encoded this sequence was incorrectly interpreted as the last chunk.

Own Id: OTP-6264 Aux Id: OTP-6005

[httpc, httpd] - http_request.erl didn't handle https URIs, which meant that redirects from ESI did not work.

Own Id: OTP-6274

[httpc, httpd] - The base 64 decoder was missing a guard so that invalid input lead to an emulator crash instead of a function clause as expected. Also the http server has been improved to handle the function clause error returning a bad credentials reply to the client.

Own Id: OTP-6279

[httpc] - Changed internal default value as it sometimes would be interpreted incorrectly causing the client to return an incomplete body.

Own Id: OTP-6283

[httpc] - Handling of 30X codes was changed so that it works according to the documentation. For instance 301 and 302 codes will not be automatically redirected.

Own Id: OTP-6297

[httpc] - A bug in the pipeline-handling code could cause a response to be sent to the client with an incorrect request id.

Own Id: OTP-6303

[httpc] - Added feature to send header values as they where typed by the user of the client. Note that the http standard requires them to be case insensitive. This feature should only be used if there is no other way to communicate with the server or for testing purpose.

Own Id: OTP-5527

[httpc] - When using asynchronous HTTP-request it is now possible to receive "200-responses" as streams instead of having to wait until the whole response has been delivered. It also possible to stream "200-response bodys" to a file both for synchronous and asynchronous requests.

Own Id: OTP-6263

[httpc] - Added option to generate Proxy-Authorization header from provided proxy username and password.

Own Id: OTP-6280

[FTP] Change documentation so that it agrees with the default behaviour of the code regarding the use of transfer type, which is right according to rfc959.

Own Id: OTP-6018

[ftp] The application handles the case if the owning process terminates with the reason 'shutdown'.

Own Id: OTP-6035

[ftp] Handle file errors from the FTP server.

Own Id: OTP-6036

[httpd] Header parsing of reply from cgi script incorrect.

Own Id: OTP-6145

[ftp] The timeout given in the ftp:open call was not properly used, which could leave the caller hanging forever.

Own Id: OTP-6184

Aux Id: seq10388

[httpd] HTTPD request handler does not handle unexpected info properly, which causes an unnecessarily obscure error message.

Own Id: OTP-6189

Aux Id: seq10395

[http] Misc fixes in the URI parsing module.

Igor Goryachev

Own Id: OTP-6191

Added application interface module: inets.

Own Id: OTP-6135

Removed code generating compiler warnings

Own Id: OTP-6069

[tftp] Added documentation (manual page) for TFTP.

Own Id: OTP-6082

[http,server] started dbg even though no debugging was desired.

Own Id: OTP-5984 Aux Id: seq10290

[http,server] request handler process died ugly due to a parse error when validating a bad request.

Own Id: OTP-6003 Aux Id: seq10260

[http,server] Server falsely sets ipv6 enabled when it gets an ipv4-mapped ipv6 address.

Own Id: OTP-5941 Aux Id: seq10221

[http,server] In case http version is unknown of client use HTTP1.0 to send status.

Own Id: OTP-5943 Aux Id: seq10198

[http,server] The process handling a request now ignores garbage messages.

Own Id: OTP-5961 Aux Id: seq10198

[http,server] Changed some actions taken if config data in httpd services was faulty.

Own Id: OTP-5962 Aux Id: seq10198

[http,server] It was possible to read arbitrary files on server by prepending ././ and ../../ in front of the file name.

Own Id: OTP-5938

[http,server] Handling of undefined file times (e.g. modification time: the mtime field on the file_info record).

Own Id: OTP-5865 Aux Id: OTP-5844

[http,server]It is now possible to set the wanted timeout for the server to setup a request connection. A new syntax for inets.config is provided in the users guide documentation. This syntax also allows to set tracing of the server for debug purposes.

Own Id: OTP-5913 Aux Id: seq10198

[http,server] Had earlier forgot to convert a value in entity_body to a binary.

Own Id: OTP-5796

[http,server] Now application checks whether the necessary directives under directive Directory exist.

Own Id: OTP-5821

[http, client] If an Ipv4-mapped ipv6 address is used the client now falls back on ipv4.

Own Id: OTP-5773 Aux Id: OTP-5765, OTP-5764

[http,server] Content-length may got a too low value, causing loss of data in clients.

Own Id: OTP-5775 Aux Id: seq10110

[http, client] The verbose mode prints what was sent and received during a request. Added a verbose option that can be used by http:set_options/1

Own Id: OTP-5766

[ftp, client] - Improvement of error handling if something goes wrong while handling the option list in ftp:open/[1,2,3].

Own Id: OTP-5711

[ftp, client] - Error when parsing a multiple FTP response line. The last line in a multiple response must be the response code followed by a space. A server may have intermediate lines that start with the response code even if this is not recommended. The parsing missed to make sure that that space was present in what it considered to be the last line.

Own Id: OTP-5712

[http, client] - The HTTP client will now retry a pipelined request that was unsuccessful due to the fact that the server unexpectedly closed the pipeline connection.

Own Id: OTP-5728

[http, server, esi] - Under some circumstances mod_esi would send a corrupted content-length header.

Own Id: OTP-5735

[http, server, get] - Removed debug printout which caused a confusing "Socket closed"-printout at high load.

Own Id: OTP-5762 Aux Id: seq10101

FTP: a data connection setup to the ftp server that failed caused a crash of the client. Now it is handled smoothly.

Own Id: OTP-5738

[ftp] If host name is a ipv4 tuple ftp erroneous tries to connect as a ipv6 address with the ipv4 address.

Own Id: OTP-5764

[ftp] Handles connect to ipv6/ipv4 address differently according to a change in inet:getaddr.

Own Id: OTP-5765

[http, server] - The HTTP request handling was remoduled to have a more straight forward error-handling. And the internal debug strategy was changed to use tracing instead of debug macros, which means we do not have to write special debug code.

Own Id: OTP-5729

ftp:ls towards different ftp servers resulted in different return results. E.g. the solaris 9 default server caused {ok,[]} while older servers caused {error,epath} as the result. For backwards compatibility the behaviour has been changed to the old result.

Own Id: OTP-5731

[http, server] - The documentation for the HTTP server has been partly rewritten and very restructured too provide a better overall picture. Lots of information provided by "semi manual pages" has been moved to the Users Guide.

Own Id: OTP-5752

FTP: verbose mode now also prints what the client sends on the control channel.

Own Id: OTP-5753

[ftp, client] - Timing related issues could sometimes cause the ftp response to be delayed.

Own Id: OTP-5705 Aux Id: seq10055

[http, server, esi] - The dispatching of the post body to the esi callback function was broken.

Own Id: OTP-5706

[ftp, client] - The FTP error code 550 was handled in an unexpected way. Some earlier versions of inets had a workaround for this in ftp:recv_bin/2 that was eliminated during restructuring of the ftp module while implementing ipv6 capabilities. The problem is now fixed.

Own Id: OTP-5682 Aux Id: seq10048

[http, client] Post request with a body in binary format failed as length was used instead of size.

Own Id: OTP-5686

[ftp, client] - For some FTP commands the FTP server will send more than one reply on the FTP control channel. In the case of a fast FTP server the client would sometimes wrongly disregard the second answer as trailing garbage attached to the first reply.

Own Id: OTP-5690 Aux Id: seq10055

[ftp, client] - A new option {progress, {CBmodule, CBFunction, InitProgressTerm}} has been added to allow users to create things such as progress bars in there GUI's. The option affects ftp:send/[3,4] and ftp:recv/[3,4].

Own Id: OTP-5680

[http, client] - Added new API function http:request/1

Own Id: OTP-5691

[httpd, server] - mod_cgi is implemented according to CGI-1.1 RFC 3875, an early implementation was based on some draft that is not totally compliant to the RFC. Documentation was updated. Also some code was restructured to facilitate testing and maintenance of the server.

Own Id: OTP-5694

[ftp, client] Calling ftp:recv/2 twice on the same connection failed due to that the last message on the ctrl channel was not appropriately taken care of. This could potentially cause a problem for any operation performed on the same connection where there had previously been an ftp:recv/2 call. Also, in some cases, when the process tries to close the data connection, it does not take into account that the data connection may actually not have been established.

Own Id: OTP-5662 Aux Id: seq10004, seq9988

[ftp, client] Enhanced error handling, mainly so the ftp client behaves gracefully when the user does strange things such as violate the user API.

Own Id: OTP-5665

[ftp, client] Added open option: mode.Deprecates function force_active/1.

Own Id: OTP-5663

[http, server] The server did not handle the config directive BindAddress value "*" properly. When creating the option list for the listen call, everything beside the atom undefined (if BindAddress was never given) and an 4-tuple (e.g. BindAddress value is 192.168.0.30) was incorrectly assumed to be an ipv6 address. For undefined (no BindAddress), Inets attempts to figure out if it is running on a ipv6-machine, and if so, add the inet6 option when calling listen. The same approach should be used when BindAddress is assigned the value "*".

Own Id: OTP-5642

Some data doesn't pass through http_base_64:decode/1 correctly. The decoding routine fails whenever a 4-character group of the encoding ends with "9" or "99". If it ends with 99, two bytes will be lost in the decode routine. If it ends with a single 9, one byte will be lost in the decode.

Own Id: OTP-5635 Aux Id: seq9971

[http, server,esi] Web server does not handle status-code returned by an esi function. I.e. the esi-function can no longer control the status code.

Own Id: OTP-5648 Aux Id: seq9982

[http, server,esi] Corrected header format. First character was lower case, and there where no space after the ":" character, example: content-length:0. Now, first character was upper case, and a space after the ":" character, example: Content-Length: 0 (To preserve backward compatibility with the de-facto standard as the new way does not break the HTTP standard!)

Own Id: OTP-5649 Aux Id: seq9982

[http, server, cgi] Parsing of the status header field could cause a crash.

Own Id: OTP-5650 Aux Id: seq9982

The internal design of using blocking gen_tcp:recv with a timeout and retries resulted in code that was hard to get a good overview of, and ultimate led to situations where the client got the wrong answer or no answer at all. The errors where many times very timing dependent and mainly effected the chunk-related functions, so if you where lucky you probably would not have noticed. The internal design was changed to use gen_tcp active once semantics. The API is not effected except for the function ftp:quote/2 which now returns a list of strings (ftp result lines) where the line endings "\r\n" has been removed. This was the original intention for the return value of ftp:quote/2 but it was non trivial to make a good such solution with the old design and a compromise was made.

*** POTENTIAL INCOMPATIBILITY ***

Own Id: OTP-5623

The new implementation of pipelining in inets-4.1 alas was slightly broken and unfortunately not caught by the test suite that apparently needs some additions. The result was that requests that ought to have been pipelined where not, this has now been fixed.

Own Id: OTP-5624

When using the latest esi interface with the callback interface of arity 3, HTTP Content-Type headers where ignored, this due to a subtle difference between this interface and the old one in how they viewed HTTP delimiters.

Own Id: OTP-5627

The HTTP server now supports ipv6 in the case that the underlying mechanisms also do so. (ssl does not yet support ipv6.)

Own Id: OTP-5141

The FTP client now supports ipv6 in the case that the underlying mechanisms also do so.

Own Id: OTP-5142

An option was added to disable the ipv6 support in the HTTP client. This to provide a workaround possibility for buggy ipv6-stacks.

Own Id: OTP-5625 Aux Id: seq9872

When generating dynamic HTTP response bodies the the default content-type is now set to "text/html" instead of "text/plain" which is more intuitive.

*** POTENTIAL INCOMPATIBILITY ***

Own Id: OTP-5626

Wrapper function http_transport:accept/3 did error-handling that it should not do because only the caller of the wrapper can determine what action to take. So timeouts where handled twice, once in http_transport and once in httpd_acceptor. Clean up of the wrapper module http_transport changed the action of the wrapper module and made the unwanted behavior noticeable in in OTP error logs. And now the unwanted error handling has been removed. The cleanup helped us find bad code but alas it also generates a lot of log printouts that are quite disturbing to the user of the HTTP-server.

Own Id: OTP-5549 Aux Id: seq9851

In the rewrite for 4.4 some mod_esi-environment values where mistaken for ordinary header values and where incorrectly transformed to strings. They are now atoms again.

Own Id: OTP-5551 Aux Id: seq9854

The HTTP server now handles "GET /\r\n\r\n" as well as "GET / \r\n\r\n". According to the RFC the whitespace is not needed.

Own Id: OTP-5552 Aux Id: seq8426

The ftp client now supports passive mode. Actually the ftp client will always try to use passive mode and if it fails it will use active mode instead. It is also possible to force the ftp-client to use active mode, if that is desired, by calling ftp:force_active/1 this way you can get the old behavior.

*** POTENTIAL INCOMPATIBILITY ***

Own Id: OTP-5148

The server did not handle HTTP-0.9 messages with an implicit version.

Own Id: OTP-5513

An internal server timeout killed the request handling process without sending a message back to the client. As this timeout only affects a single request it has been set to infinity (if the main server process dies the request handling process will also die and the client will receive an error). This might make a client that does not use a timeout hang for a longer period of time, but that is an expected behavior!

Own Id: OTP-5514 Aux Id: seq9806

That a third party closes the http servers accept socket is recoverable for inets, hence inets will only produce an info report as there was no error in Inets but measures where taken to avoid failure due to errors elsewhere.

Own Id: OTP-5516 Aux Id: seq9806

The HTTP client proxy settings where ignored. Bug introduced in inets-4.3.

Own Id: OTP-5517

Inets only sent the "WWW-Authenticate" header at the first attempt to get a page, if the user supplied the wrong user/password combination the header was not sent again. This forces the user to kill the browser entirely after a failed login attempt, before the user may try to login again. Inets now always send the authentication header.

Own Id: OTP-5521

A major rewrite of big parts of the HTTP server code was performed. There where many things that did not work satisfactory. Cgi script handling can never have worked properly and the cases when it did sort of work, a big unnecessary delay was enforced. Headers where not always treated as expected and HTTP version handling did not work, all responses where sent as version HTTP/1.1 no matter what.

Own Id: OTP-5537

When further testing the functionality of https requests that goes through a proxy. We realised that alas this can not currently be supported as it requires features from the ssl implementation that is not currently available. So for now an error message will be returned when trying to use this functionality.

Own Id: OTP-5453

When trying to get a url from a server that does not exist the client hanged instead of returning an error message. Bug introduced in inets-4.3.

Own Id: OTP-5454

Tunneling of SSL through a proxy has now been implemented. However due to lack of test sites this has only partially been verified, it is likely that there will have to be future improvements in this area.

Own Id: OTP-5443

The pipeline timeout was changed to be zero by default to avoid that people by accident would create connection processes that never dies and eats up the socket resources.

*** POTENTIAL INCOMPATIBILITY ***

Own Id: OTP-5442

Altered the way spawn_link is used in mod_esi to avoid getting, in this scenario unwanted error reports, from spawn_link. (The behavior of spawn_link was altered in a not backwards compatible way.)

Own Id: OTP-5444

Sometimes EWS modules where called with an Info record where the peername field was {-1, "unknown"}. This could happen when a client was making a lot of requests which it discards before they where answered by the server. The server now ignores such requests and does not call the EWS modules in this case.

Own Id: OTP-5380 Aux Id: seq9739

The HTTP-server now returns the 408 status code upon a request timeout as expected instead of the previous faulty behavior of sending a 500 status code.

Own Id: OTP-5409

The content length was put in to the HTTP-headers as an integer instead of as a string.

Own Id: OTP-5410

It was wrongly presumed that code:priv_dir would always be writable due to how the test-server works. The directory is now a configuration parameter in the inets-application configuration file. Failing to configure it will result in that all cookies are treated as session cookies.

Own Id: OTP-5411

An undocumented beta version of tftp is included.

Own Id: OTP-5419

When sending a request through a proxy the absolute URI must be used.

Own Id: OTP-5368

Basic support for cookies was implemented. Later some more functions to inspect cookies may be added.

Own Id: OTP-5331

A top tftp supervisor was added in preparation for adding a tftp service in a future Inets release.

Own Id: OTP-5379

The URI check that disables relative links that goes outside the server-root still missed a few cases, in spite of the improvement in OTP-5140.

Own Id: OTP-5249

The http client pipelining implementation has been rewritten as the old implementation was too optimistic about when to pipeline. In the process of doing this also the error handling was improved, better clean up is performed when the request handling process terminates and better handling of the case that the httpc_manager process dies and is restarted.

Own Id: OTP-5303

Improved handling of status codes 30X and 50X.

Own Id: OTP-5309

The Inets supervision tree has been reorganized to create a better balance between the Inets services. Preferably they should not effect each other. The ftp service has also been included in the Inets supervision tree, it was for reasons unknown, not included before.

Own Id: OTP-5188

The service concept in Inets is now better documented.

Own Id: OTP-5189

The Inets shutdown times have proven to be too short under some circumstances, as a heavy load, therefore they have been prolonged.

Own Id: OTP-5261 Aux Id: seq9624

Options for automatic redirection and pipelining is now available in the http client API.

Own Id: OTP-5304

A programming error could cause a badmatch in the http-client when the http response was chunk decoded.

Own Id: OTP-5101

The parsing of HTTP messages was missing a base case. This caused unexpected behavior when the separator CR and LF where received in different tcp packets.

Own Id: OTP-5239

When receiving a status 100 code, the client should only respond by sending the message body, if the client sent an expect header in the first place. Failing to do so may result in that the server receives the body twice.

Own Id: OTP-4848

mod_get now also handles http version HTTP/0.9

Own Id: OTP-4935 Aux Id: seq8426

"Last-modified" field was incorrectly set to local time with the tag GMT, it is now corrected so that the time reflected is in fact GMT.

Own Id: OTP-4936

The client will only add a host-field to the request if there is not one already present.

Own Id: OTP-4984

The Inets application tries to be compatible with Apache. To be more compatible the option 'MaxKeepAliveRequest' is renamed 'MaxKeepAliveRequests'. The old name is kept for backward compatibility.

Own Id: OTP-5024

Changing the base 64 decoding to not accept invalid input, uncovered a logical error in mod_security.erl An already decoded string was sent as input to decode. In this case, as it so happened, the two errors worked together creating the illusion that everything was right. This has now been corrected.

Own Id: OTP-5083

URLs where not properly scrutinised for relative paths. A malicious user could exploit this to read files outside the document root. This is no longer the case.

Own Id: OTP-5140

A HTTP 1.1 client is officially included in Inets. It is loosely based on the previously unsupported code contributed by Johan Blom. In this first version only the most basic HTTP functionality is supported. The user API has been changed.

*** POTENTIAL INCOMPATIBILITY ***

Own Id: OTP-5047

Fixed erroneous link in documentation.

Own Id: OTP-5089 Aux Id: seq8887

Added the function quote/2 that lets you send an arbitrary FTP command to the FTP client.

Own Id: OTP-5099 Aux Id: seq8961

Started integration of the HTTP client and server code too facilitate maintenance and further development.

Own Id: OTP-5110

Due to several possibilities to interpret the ftp standard some newer ftp-servers have interpreted the standard in such a way that the documented return value of ftp:nlist/2 does not always match the actual return value. Some extra checks have now been added to ensure the documented return value. This will also result in that ftp:nlist is not bug compatible in the case that nlist is given a filename instead of a directory it will now return an error instead of {ok, FileName}.

*** POTENTIAL INCOMPATIBILITY ***

Own Id: OTP-5165

Created a Users Guide for Inets. Earlier there where some fake manual pages and information was scattered everywhere and hard to find.

Own Id: OTP-5180

[httpd] - When calling httpd:restart there was a file descriptor and ets table leek due to the fact that the semantics of the function restart is not restart but reload. This is solved in inets-4.7.12 and this special inets-3.0.10 release is intended for old systems only.

Own Id: OTP-6579 Aux Id: OTP-6573, seq10607

[http,server] Illegal hexadecimal code in URL was not handled. The validation of URI:s are therefore updated.

Own Id: OTP-6078 Aux Id: seq10306

The URI check that disables relative links that goes outside the server-root missed a few cases.

Own Id: OTP-5249

The inets shutdown times have proven to be too short under some circumstances, as a heavy load, therefore they have been prolonged.

Own Id: OTP-5261 Aux Id: seq9624

[httpd] Fixed a number of transfer-encoding problems.

First part of the data received from a CGI-script was sent as chunked even if the client was HTTP/1.0.

Introduced new directive (DisableChunkedTransferEncodingSend) to turn off usage of chunked transfer-encoding (when sending) since it appear's some browser's have problems handling this. This applies if the client is HTTP/1.1.

(Own Id: OTP-4806)

Aux Id: Seq 8150

[httpc] HTTP client reformats some URLs (e.g. containing %20, space).

Updated client from sowap.sf.net as of 2003-09-08.

Johan Blom of Mobile Arts AB

(Own Id: OTP-4807)

[httpd] In module mod_browser, malformed search for parsed header, user-agent.

Also added new os and browser

(Own Id: OTP-4808)

[ftp] FTP client doesn't notice when disk is full.

(Own Id: OTP-4822)

Aux Id: Seq 8175

On Windows the ftp:ls function sometimes exits.

Workaround for a problem that seems to happen only on Windows when calling the ls function. Closing an already closed socket will result in enotsock returned which will result in an exit.

(Own Id: OTP-4770)

Web server does not handle econnaborted accept result.

This results in an unnecessary acceptor process restart.

(Own Id: OTP-4732)

ESI callback generates broken HTTP.

This was a problem for (at least) Netscape 4.75. It worked for Mozilla 1.4a (on Solaris 8) and rumor has it that it also worked for IE.

Sean Hinde

(Own Id: OTP-4696)

Log-size values ignored for security- and error disk-logs (SecurityDiskLogSize and ErrorDiskLogSize in mod_disk_log). Instead the TransferDiskLogSize was used.

Thomas Lange

(Own Id: OTP-4659)

HTTP client called undefined function in HTTP server (httpd_response:send_status/3).

(Own Id: OTP-4628)

HTTP client now updated to jnets-0.1.

Auther: Johan Blom of Mobile Arts AB.

(Own Id: OTP-4588)

FTP client now supports chunked receive. Proposal of Luke Gorrie provided inspiration but not algorithm.

(Own Id: OTP-4549)

HTTP client updated to jnets-0.1, now supporting proxy.

Auther: Johan Blom of Mobile Arts AB.

(Own Id: OTP-4552)

Body length calculation incorrect.

(Own Id: OTP-4548)

(Aux Id: OTP-4207, Seq 7209)

HTTP-request with a BODY length longer than 0 does not work.

(Own Id: OTP-4550)

(Aux Id: Seq 7653)

Calculation of remaining chunk size incorrect.

(Own Id: OTP-4551)

Wrong module name used when attempting to stop the security server (mod_sec_server instead of mod_security_server).

(Own Id: OTP-4556)

Updated to handle HTTP/1.1.

Change in the inets API.

(Own Id: OTP-4408)

(Aux Id: Seq 7485)

*** POTENTIAL INCOMPATIBILITY ***