| --- Log opened Mon Dec 29 00:00:26 2014 |
| 00:54 | | io\PACKERS is now known as iospace |
| 01:57 | | HotShot [HotShot@Nightstar-39c4pa.nap.wideopenwest.com] has joined #code |
| 02:11 | | Turaiel[Offline] is now known as Turaiel |
| 02:11 | | HotShot [HotShot@Nightstar-39c4pa.nap.wideopenwest.com] has quit [Ping timeout: 121 seconds] |
| 02:29 | | Checkmate [Z@Nightstar-484uip.cust.comxnet.dk] has quit [Ping timeout: 121 seconds] |
| 03:46 | <@celticminstrel> | Getting this code to compile in VS2013 is harder than I expected. |
| 03:48 | <@celticminstrel> | Now it's complaining I can't access a protected member... even though it's in a public base class. Maybe confused because it's a pointer-to-member? |
| 03:48 | <@celticminstrel> | Hm. |
| 03:49 | | * celticminstrel suddenly wonders whether I need to tell it to use c++11. |
| 03:50 | <@celticminstrel> | ...no, doesn't look like there's an option for that. |
| 03:56 | <@celticminstrel> | Well, it looks like I can reference the member using the derived class name instead of the base class name. |
| 03:57 | <@celticminstrel> | The real issue I need to fix involves Boost.Spirit, and I have absolutely no idea what on earth is wrong. |
| 04:09 | <@celticminstrel> | It's kinda weird how it explodes whenever a file doesn't end in a newline. |
| 04:10 | | Turaiel is now known as Turaiel[Offline] |
| 05:14 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: Leaving] |
| 07:18 | | Kindamoody[zZz] is now known as Kindamoody |
| 08:44 | | Checkmate [Z@Nightstar-484uip.cust.comxnet.dk] has joined #code |
| 08:44 | | mode/#code [+o Checkmate] by ChanServ |
| 10:16 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has quit [Ping timeout: 121 seconds] |
| 10:18 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has joined #code |
| 10:18 | | mode/#code [+ao McMartin McMartin] by ChanServ |
| 10:33 | | Orthia [orthianz@Nightstar-4j4.kvk.224.119.IP] has quit [Ping timeout: 121 seconds] |
| 10:38 | | Orthia [orthianz@Nightstar-d1u2qm.callplus.net.nz] has joined #code |
| 10:38 | | mode/#code [+o Orthia] by ChanServ |
| 10:57 | | Orth [orthianz@Nightstar-sgfn3g.callplus.net.nz] has joined #code |
| 10:58 | | Orthia [orthianz@Nightstar-d1u2qm.callplus.net.nz] has quit [Ping timeout: 121 seconds] |
| 11:43 | | celticminstrel [celticminst@Nightstar-jeiu0g.dsl.bell.ca] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] |
| 12:30 | | Kindamoody [Kindamoody@Nightstar-180u8i.tbcn.telia.com] has quit [Ping timeout: 121 seconds] |
| 12:33 | | Kindamoody [Kindamoody@Nightstar-180u8i.tbcn.telia.com] has joined #code |
| 12:33 | | mode/#code [+o Kindamoody] by ChanServ |
| 12:55 | | Kindamoody [Kindamoody@Nightstar-180u8i.tbcn.telia.com] has quit [Ping timeout: 121 seconds] |
| 13:21 | | gnolam [lenin@Nightstar-dl1h4n.cust.bredbandsbolaget.se] has joined #code |
| 13:21 | | mode/#code [+o gnolam] by ChanServ |
| 13:27 | | Orth [orthianz@Nightstar-sgfn3g.callplus.net.nz] has quit [Ping timeout: 121 seconds] |
| 13:32 | | Orthia [orthianz@Nightstar-4j4.kvk.224.119.IP] has joined #code |
| 13:32 | | mode/#code [+o Orthia] by ChanServ |
| 13:33 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 13:41 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has joined #code |
| 13:41 | | mode/#code [+o himi] by ChanServ |
| 13:56 | <@Tarinaky> | Argh, I think Windows is being too clevere. |
| 14:27 | < RobinStamer> | Impossiburu |
| 15:35 | | gnolaptop [lenin@Nightstar-oru2ae.priv.bahnhof.se] has joined #code |
| 15:35 | | gnolaptop is now known as gnoLAN |
| 15:39 | < RobinStamer> | ... that was an invite to explain more there Tarinaky |
| 15:41 | <@Tarinaky> | Ah. |
| 15:42 | <@Tarinaky> | So. I think Windows 8.1 does some echo cancellation stuff under the hood to stop sound from the speakers overwhelming the micrphone. |
| 15:42 | <@Tarinaky> | However, for reasons of testing I actually /want/ to capture the speakers with the microphone. |
| 15:43 | <@Tarinaky> | So I cranked the Microphone gain up in Windows. |
| 15:43 | | Orthia [orthianz@Nightstar-4j4.kvk.224.119.IP] has quit [Ping timeout: 121 seconds] |
| 15:43 | <@Tarinaky> | However, part of the testing involves muting all but one channel of the playback/capture device |
| 15:43 | <@Tarinaky> | To test that individual sensors/speaker works. |
| 15:44 | <@Tarinaky> | When I mute the channels of the microphone, this has the side-effect of resetting the Microphone level in Windows to 0 |
| 15:44 | <@Tarinaky> | Which then means my test sound isn't recorded. |
| 15:45 | <@Tarinaky> | And I don't see an easy work around this issue without seriously rethinking my initial assumptions about the system. |
| 15:48 | | Orthia [orthianz@Nightstar-n8uab4.callplus.net.nz] has joined #code |
| 15:48 | | mode/#code [+o Orthia] by ChanServ |
| 15:51 | < RobinStamer> | Ah,yeah. There's probably a barely documented way to turn that feature off. |
| 15:52 | < RobinStamer> | But that means you'd have to look through MS docs, so... |
| 15:52 | < RobinStamer> | Good luck. |
| 15:52 | <@Tarinaky> | I'd ask one of the resident Windows team but it's the holidays so everyone's escaped. |
| 15:56 | <@Tarinaky> | And now forsomething completely different: http://www.lofibucket.com/articles/oscilloscope_quake.html |
| 15:57 | <@ErikMesoy> | Tarinaky: rightclick your sound control, open "Recording devices", then right-click in the list there and select "Show disabled devices". |
| 15:57 | <@ErikMesoy> | I'm not sure if this actually answers the problem you want, but it seems like a barely documented thing that might help. >_> |
| 15:58 | <@ErikMesoy> | Right-clicking in the lists in the sound control panel will get you special access to a bunch of playback and recording settings. |
| 15:58 | <@Tarinaky> | I'm not seeing any disabled devices. |
| 16:52 | | gnoLAN [lenin@Nightstar-oru2ae.priv.bahnhof.se] has quit [[NS] Quit: Restart] |
| 16:52 | | gnoLAN [lenin@Nightstar-oru2ae.priv.bahnhof.se] has joined #code |
| 17:00 | | celticminstrel [celticminst@Nightstar-jeiu0g.dsl.bell.ca] has joined #code |
| 17:00 | | mode/#code [+o celticminstrel] by ChanServ |
| 17:00 | | celticminstrel [celticminst@Nightstar-jeiu0g.dsl.bell.ca] has quit [[NS] Quit: KABOOM! It seems that I have exploded. Please wait while I reinstall the universe.] |
| 17:00 | | celticminstrel [celticminst@Nightstar-jeiu0g.dsl.bell.ca] has joined #code |
| 17:00 | | mode/#code [+o celticminstrel] by ChanServ |
| 17:05 | | macdjord|slep is now known as macdjord|wurk |
| 19:14 | <@celticminstrel> | Apparently cl.exe doesn't understand the syntax for "function returning reference to array" but doesn't have a problem if I first typedef the array type. |
| 19:15 | <&McMartin> | Could be |
| 19:15 | <&McMartin> | Welcome to Having To Support Compilers That Aren't gcc |
| 19:15 | | Kindamoody|autojoin [Kindamoody@Nightstar-180u8i.tbcn.telia.com] has joined #code |
| 19:15 | | mode/#code [+o Kindamoody|autojoin] by ChanServ |
| 19:15 | | Kindamoody|autojoin is now known as Kindamoody |
| 19:15 | <&McMartin> | (Also to programming in a language where typechecking is Turing-complete and I think parsing might be too) |
| 19:18 | <@celticminstrel> | Heh. |
| 19:18 | <@celticminstrel> | I've actually been using clang, not gcc. |
| 19:19 | <@celticminstrel> | Somehow I had a double static. |
| 19:19 | <@Tamber> | Extra staticky variables, in a nylon jumper? |
| 19:19 | <@celticminstrel> | I'm not sure if clang just accepted it, or if I had simply not built that target yet. |
| 19:19 | <@celticminstrel> | No, like "static static void blah()". |
| 19:20 | <@celticminstrel> | And for some reason, VS can't convert boost::path::c_str() to const char*. |
| 19:21 | <@celticminstrel> | So I had to change everything from .c_str() to .string().c_str() |
| 19:21 | <&McMartin> | You should make sure that one or both of those is not actually std::basic_string<WCHAR>. |
| 19:21 | <@celticminstrel> | I suppose it's possible. |
| 19:21 | <&McMartin> | Since I think boost::path because boost::wpath on Windows in a semi-recent version of boost::filesystem |
| 19:22 | <&McMartin> | s/because/became/ |
| 19:22 | <@celticminstrel> | Though, if it were the case, I'd've expected both versions to fail... |
| 19:22 | <&McMartin> | Also IIRC one of the things you can do with paths no longer roundtrips |
| 19:22 | <&McMartin> | There's a .native() in addition to a .string |
| 19:23 | <@celticminstrel> | Some of my issues were simply missing includes. I guess they were implicitly included when compiling with clang, but not when compiling with cl.exe. |
| 19:24 | <@celticminstrel> | Also, it seems the Mac's linker doesn't type-check but Windows does. Specifically, I had some things defined as different types in two files (not intentionally). |
| 19:25 | <@celticminstrel> | I didn't get link errors on Mac, but I did in VS. |
| 19:25 | <&McMartin> | Huh. I wonder if that's an ELF vs. COFF trait. |
| 19:25 | <@celticminstrel> | Well, in this case I'd say it was helpful. |
| 19:26 | <&McMartin> | Yeah, but it's also odd that it would happen at the link phase. |
| 19:26 | <@celticminstrel> | Two different files? |
| 19:26 | <&McMartin> | Especially since I thought C++ stuff *had* to be... oh wait |
| 19:26 | <@celticminstrel> | Defined in one file, declared extern in the other with a different type. |
| 19:26 | <&McMartin> | Was this C code with this? Because yeah, Windows compilers have had type mangling on C symbols by default for ages. |
| 19:26 | <@celticminstrel> | It wasn't C code. |
| 19:27 | <@celticminstrel> | It was global variables. |
| 19:27 | <&McMartin> | Very odd that it linked on Mac then, because type mangling in C++ is kinda mandatory for overloads to work |
| 19:27 | <&McMartin> | Oh, I see |
| 19:27 | <&McMartin> | Less sure then |
| 19:27 | <&McMartin> | Could be CL is type-mangling fields too just to make sure |
| 19:28 | | Orthia is now known as Reivles |
| 19:29 | <@celticminstrel> | ..okay, now I want to pull from my Windows computer to my Mac. I can do the reverse easily, but not sure if I can figure this one out. |
| 19:29 | <@celticminstrel> | I could instead push from Windows to Mac, but that means fiddling with branch names and stuff. |
| 19:36 | <&ToxicFrog> | Pull as in git pull? |
| 19:36 | <&ToxicFrog> | I do this by running sshd on the windows machine. |
| 19:37 | <&ToxicFrog> | You could also go the smb route. |
| 19:50 | <@celticminstrel> | How would that work? |
| 19:50 | <@celticminstrel> | The smb route. |
| 19:51 | <@celticminstrel> | (Since that doesn't sound like it requires installing software. >_> ) |
| 19:51 | <&ToxicFrog> | On windows: export the drive containing your git repo as an smb share. |
| 19:52 | <&ToxicFrog> | On OSX: mount said drive, then add the git repo as a remote |
| 19:52 | <@celticminstrel> | ...I should probably add my Mac repository as a remote on the Windows side so I don't have to type out the full URL every time. |
| 19:52 | <&ToxicFrog> | Yes. |
| 20:03 | <@celticminstrel> | Yay, it works. |
| 20:08 | <@celticminstrel> | Now I need to remember the URL for pulling from Mac to Windows over ssh... |
| 20:10 | <@celticminstrel> | Ah, got it. |
| 20:11 | <@celticminstrel> | ...it'd be easier if I could set up a keypair, but I have no idea how that would be done in Windows. |
| 20:18 | <&ToxicFrog> | You're doing the push/pull over ssh? |
| 20:20 | <@celticminstrel> | When on Windows, yes. |
| 20:20 | <@celticminstrel> | I have the sshd running on my Mac. |
| 20:21 | <@celticminstrel> | (ie, Remote Login enabled in sharing preferences) |
| 20:22 | <&ToxicFrog> | What git implementation on windows? |
| 20:23 | <&ToxicFrog> | If it's the msys-based one, with gitbash, you should be able to open a gitbash shell and "ssh-keygen", then "ssh-copy-id" as normal |
| 20:23 | <@celticminstrel> | Okay... |
| 20:24 | <@celticminstrel> | I'll try that. |
| 21:46 | <@celticminstrel> | It looks like McM was correct about boost::path containing a wstring instead of a string. |
| 21:48 | <@celticminstrel> | But boost::path::string() always returns std::string. |
| 21:49 | <&McMartin> | Aha |
| 21:49 | <&McMartin> | OK then |
| 21:49 | <@celticminstrel> | This is probably not the cause of the error I'm currently getting, though... |
| 21:49 | <&McMartin> | Right, I think boost::path::c_str() is now equivalent to boost::path::native().c_str(), which would be UTF-16 on Windows. |
| 21:54 | <@celticminstrel> | It seems to be an error in SFML. I guess I can try updating that. |
| 21:57 | <@celticminstrel> | ...hm, I have the same version on Windows and Mac though... |
| 21:59 | <@celticminstrel> | Using the 32-bit version shouldn't cause problems, right? |
| 22:00 | <&McMartin> | Well, it restricts you to 2GB of mallocable space |
| 22:00 | <&McMartin> | And you can't link a 32-bit DLL into a 64-bit executable |
| 22:01 | <@celticminstrel> | I'm not sure if this is being built as a 64-bit executable... |
| 22:01 | <@celticminstrel> | It probably should be. |
| 22:01 | | Kindamoody [Kindamoody@Nightstar-180u8i.tbcn.telia.com] has quit [[NS] Quit: Switching networks, brb] |
| 22:02 | | Kindamoody|autojoin [Kindamoody@Nightstar-180u8i.tbcn.telia.com] has joined #code |
| 22:02 | | mode/#code [+o Kindamoody|autojoin] by ChanServ |
| 22:02 | | Kindamoody|autojoin is now known as Kindamoody |
| 22:03 | <@celticminstrel> | I'd like this to be runnable on XP. |
| 22:08 | <&McMartin> | Then it probably shouldn't be. |
| 22:08 | <@celticminstrel> | ...that's what I meant to say. Whoops. |
| 22:09 | <@celticminstrel> | The crash seems to be in msvcr110.dll, through sfml-graphics.dll. |
| 22:10 | <@celticminstrel> | Though I'm not sure if it's completely consistent... at first I didn't get msvcr110 anywhere. |
| 22:10 | <&McMartin> | Do you perchance have a double-free somewhere corrupting your heap |
| 22:10 | <&McMartin> | msvcr110.dll is libc |
| 22:12 | <@celticminstrel> | Not that I'm aware of, at least. >_> |
| 22:13 | <@celticminstrel> | Usually a double-free causes an access violation when running on the Mac, I think. |
| 22:15 | <@celticminstrel> | For some reason the boost::path starts with "D:/" and then uses backslashes as the path separator in the rest of the path. |
| 22:15 | <@celticminstrel> | I have no idea if that's remotely relevant. |
| 22:16 | <&McMartin> | That's legal |
| 22:16 | <&McMartin> | boost::filesystem canonicalizes at the last minute, more or less. |
| 22:16 | <&McMartin> | And Win32 accepts slashes as path separators anyway for the most part. |
| 22:16 | <@celticminstrel> | The context of the crash is loading an image from a file using SFML's Image::loadFromFile routine; I have an Image* and am passing path.string() to the aforementioned routine. |
| 22:16 | <&McMartin> | What happens if you assign it to a std::string local first? Could you somehow be getting destructed too soon? |
| 22:17 | <@celticminstrel> | And the Image* was initialized the line before. |
| 22:17 | <@celticminstrel> | I tried that, it didn't help. |
| 22:17 | <&McMartin> | Is it *expecting* a UTF-16 string? |
| 22:17 | <@celticminstrel> | Hmm, that's a good question... |
| 22:18 | <@celticminstrel> | Ah, no, loadFromFile takes std::string as its parameter. |
| 22:18 | <@celticminstrel> | (Or const std::string& if you want to be precise.) |
| 22:32 | | Kindamoody is now known as Kindamoody[zZz] |
| 22:33 | | Turaiel[Offline] is now known as Turaiel |
| 22:39 | | * celticminstrel has gotten temporarily distracted from that issue by trying to figure out if there's a way to get VS to display printf/cout output in the VS output window. |
| 22:43 | <@celticminstrel> | All the answers I'm finding tell me to use OutputDebugString instead. :/ |
| 22:48 | <&McMartin> | ODS is a good function~ |
| 22:48 | <@celticminstrel> | Eh, it's kinda plain. Doesn't even take a format string. |
| 22:57 | <@celticminstrel> | ... "can't convert from string interator to char*" |
| 22:57 | <&McMartin> | I don't believe that's automatic |
| 22:58 | <&McMartin> | Only std::vector<> makes that guarantee, IIRC and even there you have to do like &vec[0] |
| 22:58 | <@celticminstrel> | It's begin() and end() in this case. |
| 22:58 | <&McMartin> | Yeah |
| 22:59 | <&McMartin> | Those aren't char*s, they're std::basic_string<char>::iterators. |
| 22:59 | <&McMartin> | In particular I don't believe that strings guarantee that they are stored in contiguous RAM. |
| 22:59 | <&McMartin> | I think .data() does though |
| 23:00 | | Turaiel is now known as Turaiel[Offline] |
| 23:14 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 23:14 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 23:14 | <@celticminstrel> | ...okay, calling OutputDebugString doesn't actually seem to work. |
| 23:14 | <@celticminstrel> | Or maybe I'm doing something wrong. |
| 23:15 | <&McMartin> | OutputDebugString dumps to the debugging ringbuffer that you watch in DbgView, not to a stream |
| 23:16 | <@celticminstrel> | I'd like the output to appear in the Output pane at the bottom right. |
| 23:17 | <@celticminstrel> | ...ah, it's that setting in Options that one of the other answers suggested setting. |
| 23:17 | <@celticminstrel> | It works now. |
| 23:18 | | * celticminstrel made a custom streambuf and injected it into cout and cerr. Now I just need to find a way to catch printf as well... >_> |
| 23:20 | <@celticminstrel> | (And fprintf(stderr,...)) |
| 23:27 | | Turaiel[Offline] is now known as Turaiel |
| 23:28 | | gnoLAN [lenin@Nightstar-oru2ae.priv.bahnhof.se] has quit [A TLS packet with unexpected length was received.] |
| 23:33 | <@celticminstrel> | The closest thing I found so far is "redirect stdout to a pipe and have another thread copy from the pipe to OutputDebugString". |
| 23:46 | <@celticminstrel> | (Which probably renders the other solution obsolete.) |
| 23:46 | <@celticminstrel> | (I mean the custom streambuf.) |
| 23:59 | | Turaiel is now known as Turaiel[Offline] |
| --- Log closed Tue Dec 30 00:00:42 2014 |