Age | Commit message (Collapse) | Author | |
---|---|---|---|
2014-04-23 | (svn r26488) -Codechange: perform the appropriate length checks when getting ↵ | rubidium | |
a client name | |||
2014-04-23 | (svn r26486) -Codechange: replace a number of snprintfs with seprintf | rubidium | |
2014-04-23 | (svn r26482) -Codechange: add an include that allows us to undefine/redefine ↵ | rubidium | |
"unsafe" functions to prevent them from being used, and thus having to care about certain aspects of their return values | |||
2013-01-08 | (svn r24900) -Fix [FS#5389]: Comments with typos (most fixes supplied by ↵ | planetmaker | |
Eagle_rainbow) | |||
2012-01-17 | (svn r23817) -Fix [FS#4962]: desync due to different NewGRF version. So ↵ | rubidium | |
reduce the chance that it happens significantly with betas/RCs/nightlies by doing the same as is done for stable releases: check the NewGRF version of server vs client. Previously this check was not done for nightlies/betas/RCs due to missing versioning information in the source tarballs, but they have that for a while now. So just force the NewGRF version check for all versions, and remove the broken --revision configure option | |||
2012-01-15 | (svn r23802) -Fix [FS#4968] (r23601): game lobby gui not updated when new ↵ | yexo | |
company information becomes available | |||
2012-01-09 | (svn r23780) -Fix [FS#4963] (r23764): also name the two new errors ↵ | truebrain | |
server-side. As extra, split up one of the errors in 3 errors, to be more specific what goes wrong. As cherry on top, make sure on all sides we can never again forget to add such entries on both sides, by introducing an assert_compile() | |||
2012-01-06 | (svn r23764) -Fix [FS#4955]: make default timeouts for certain states lower ↵ | rubidium | |
and configurable | |||
2012-01-03 | (svn r23740) -Codechange: remove some 300 unneeded includes from the .cpp files | rubidium | |
2011-12-19 | (svn r23601) -Fix: fix the conflict in window number | truebrain | |
2011-12-10 | (svn r23476) -Codechange: use the error queue to replace switch mode error ↵ | rubidium | |
strings, again making it possible to return multiple errors | |||
2011-12-10 | (svn r23475) -Codechange: queue critical error messages, so when multiple ↵ | rubidium | |
happen you won't miss any | |||
2011-12-10 | (svn r23474) -Codechange: move the declaration error related functions to ↵ | rubidium | |
error.h | |||
2011-08-21 | (svn r22805) -Codechange: move use of magic number for version checking to ↵ | rubidium | |
more logical location | |||
2011-05-04 | (svn r22423) -Document: some network stuff | rubidium | |
2011-05-01 | (svn r22400) -Codechange: replace some defines in the tcp/game code so ↵ | rubidium | |
doxygen can create better documentation | |||
2011-04-30 | (svn r22387) -Fix-ish [FS#4601]: Windows' recv seems to return "graceful ↵ | rubidium | |
closed" before having passed the remaining buffer which causes OpenTTD to think all connections are "incorrectly" terminated, i.e. without the "I'm leaving" packet from the client. So let the client wait a tiny bit after sending the "I'm leaving" packet and before gracefully closing the connection | |||
2011-04-22 | (svn r22364) -Codechange: don't show the (unknown) client address in the ↵ | rubidium | |
client list at clients | |||
2011-04-22 | (svn r22362) -Codechange: NetworkFindClientInfoFromClientID -> ↵ | rubidium | |
NetworkClientInfo::GetByClientID | |||
2011-04-09 | (svn r22308) -Fix [FS#4574]: waiting on a server could kick the client, or ↵ | rubidium | |
rather the client would kick itself due to an unexpected packet | |||
2011-02-27 | (svn r22154) -Fix [FS#4529]: _current_company was modified when moving a ↵ | terkhen | |
client to spectators, causing issues in bankruptcy (Rubidium) | |||
2011-02-11 | (svn r22063) -Change: make the "has network" check also check whether the ↵ | rubidium | |
client is actually fully connected | |||
2011-02-08 | (svn r22031) -Fix: bad servers could crash a client | rubidium | |
2011-01-22 | (svn r21890) -Cleanup: remove some unneeded includes | rubidium | |
2011-01-21 | (svn r21875) -Codechange: indentation of some comments was wrong | smatz | |
2011-01-19 | (svn r21854) -Codechange: refactor the password setting methods to make it ↵ | rubidium | |
possible to change the password of other companies (on the server) | |||
2011-01-19 | (svn r21852) -Codechange: generalise GenerateCompanyPasswordHash (dihedral) | rubidium | |
2011-01-19 | (svn r21851) -Codechange: rename NetworkClientSetPassword to ↵ | rubidium | |
NetworkClientSetCompanyPassword (dihedral) | |||
2011-01-19 | (svn r21850) -Codechange: move password hashing to a more general location ↵ | rubidium | |
(dihedral) | |||
2011-01-03 | (svn r21701) -Codechange: ConsoleColour = TextColour, so make it that way ↵ | rubidium | |
and remove some unneeded casts | |||
2011-01-01 | (svn r21693) -Codechange: no need to cast char[] to const uint8* | smatz | |
2011-01-01 | (svn r21692) -Cleanup: remove unused extern declaration of ↵ | smatz | |
HashCurrentCompanyPassword(), make it static | |||
2011-01-01 | (svn r21691) -Codechange: make sure GenerateCompanyPasswordHash() behaves ↵ | smatz | |
deterministically when/if the value of NETWORK_SERVER_ID_LENGTH is changed | |||
2011-01-01 | (svn r21687) -Fix: verify the colour code we received from the server is valid | smatz | |
2010-12-05 | (svn r21401) -Fix: ofcourse MSVC x64 has something to complain about... | rubidium | |
2010-12-05 | (svn r21400) -Fix [FS#4284]: when loading a savegame takes more than 2 ↵ | rubidium | |
seconds locally, don't see that as lag of the server. Also increase the client side's lag thresholds a bit | |||
2010-12-05 | (svn r21397) -Change/Feature-ish: when getting a savegame from the server, ↵ | rubidium | |
just keep it in memory instead of writing it to a file | |||
2010-12-05 | (svn r21396) -Codechange: rename SafeSaveOrLoad to SaveLoad (it only did ↵ | rubidium | |
load), and add support for (safe) loading with a LoadFilter | |||
2010-12-05 | (svn r21392) -Change: prepare the network protocol for getting the file size ↵ | rubidium | |
later in the download process | |||
2010-11-30 | (svn r21361) -Change: make sure the client is listening, or rather ↵ | rubidium | |
receiving, our frames | |||
2010-11-30 | (svn r21358) -Codechange: make some network function names conform to coding ↵ | rubidium | |
style | |||
2010-11-26 | (svn r21334) -Fix [FS#4271]: make (more) sure that the savegame and ↵ | rubidium | |
transferred file are the same file and not different ones | |||
2010-11-19 | (svn r21256) -Fix [FS#4189]: Clients weren't always notified when the ↵ | rubidium | |
connection was lost | |||
2010-10-24 | (svn r21033) -Codechange: be more strict with the packets a client accepts ↵ | rubidium | |
at a given moment | |||
2010-10-24 | (svn r21029) -Codechange: split the map downloading packet + 3-state enum ↵ | rubidium | |
into 3 separate packets | |||
2010-10-24 | (svn r21027) -Change/Fix: under some circumstances the file handle of the ↵ | rubidium | |
downloaded savegame wouldn't be closed, and validity of the handled wasn't checked in all cases | |||
2010-10-22 | (svn r21008) -Change: don't make client reconnect waiting time depend on the ↵ | rubidium | |
company; in coop games that does not spread clients at all, and most companies have a low number causing it not to be spread out either. Use the ClientID instead. Based on a patch by dihedral. | |||
2010-10-18 | (svn r20993) -Codechange: some shuffling of game protocol packet description ↵ | rubidium | |
so they're documented in the "same" place as UDP, content and admin packets (dihedral) | |||
2010-10-17 | (svn r20957) -Codechange: Add another parameter to FindGRFConfig() to define ↵ | frosch | |
search restrictions. | |||
2010-10-15 | (svn r20933) -Codechange: move some more client related methods and such to ↵ | rubidium | |
network_client.cpp |