Various hacks to make this build on clang on OS X 10.11 and Boost 1.60. #2
Conversation
Don't just blindly pull it. I have not tested this.
|
HFS for reals!? On Apr 30, 2016, at 17:20, Jay Carlson [email protected] wrote:
|
|
Does this also mean you’re getting a 64-bit build to work?? -dvd On Apr 30, 2016, at 17:20, Jay Carlson [email protected] wrote:
|
|
It builds, application launches, shows splash screen, generates keys, show a $0.00 window. That’s all I know. The most tedious problem was the |
|
Try firing up the console and see if you can setgenerate true I can also send you 1000 coins for SNG (low cost in case they go missing) if you send me your Give* address -dvd On Apr 30, 2016, at 17:32, Jay Carlson [email protected] wrote:
|
|
I did Testing with small transactions makes sense. Try |
|
YES? http://xtc.inter.com:2751/block/0000071943046c9bba73990b92ee91196584f399c772ccc1322d7494490a2547 On Apr 30, 2016, at 17:47, Jay Carlson [email protected] wrote:
|
|
|
|
BOOM BABY BOOM! So freaking cool — 99% of the ALTS out there have to be built 32bit and with boost 1.55 — this would be a coup! Qt 4 or 5 ?? This is worth 1M 2GIVE to me — it’s what I sent to the dev who got my cross compile environment up How are you with IOS dev ;-) https://github.com/voisine/breadwallet I’ve got a friends kid who has XP with swift and ios develoment looking into forking BreadWallet to make us an IOS 2GIVE wallet On Apr 30, 2016, at 17:51, Jay Carlson [email protected] wrote:
|
|
It’s still Qt 4. Qt 5 was initially unhappy with me, but it’s possible that cleaning up this stuff makes it easier. I’ll take a look after dinner. I haven’t done iOS development (shell scripts on jailbroken phones don’t count), but “how hard could it be". Swift does sound like the way to go, but I haven’t written more than hello world in it. |
I would think that if Breadwallet can actually compile and deploy as it, its a straight skin job to re-brand and make sure it has the Scrypt support and whatever server side indexing thing needs to be in place. I’m just as happy to outsource all of that and the Android wallet as well — but who knows? There is a ton of stuff to work on with the whole Gift* paper wallet option and the charitable giving directory — plus build the website and message board and eventually go talk to my buddy who owns coinoutlet about adding support for the paperwallets to his BTM network… |
|
Okay I’ve reviewed the changes — all that NULL stuff was my change to the use of “throw runtime_err()” as a means to return RPC command help strings and shift that into the call table vRPCCommands[] directly as a next step towards externalizing the strings for future language translation options. The main place where this comes together is in bitcoinrcp.cpp: If you type ‘help' in the console plus test a few commands like 'help getblockcoun’t et al we can confirm that it didn’t break that. I’ll do a test on my local machine make sure that the return Value(); works too and if so I’ll pull your changes in on those files. As for the +#ifndef Q_MOC_RUN changes I’ll check those as well to make sure they do not break anything on Win32 and Linux In src/scrypt-x86_64.S I see the .align shift from 32 to 16 — any idea why this was necessary? Do we know if it makes any performance difference? And finally in src/serialize.h The +#if 0 Is that just a short cut to completely disable / comment off the function...Is there some native function that this was superceding? Is this one implemented with some alternate behavior? Assuming it doesn’t break Boost 1.55 it should be a no brainer to fold it all in — otherwise we might need a few #ifdef BOOST_VERSION tests inserted. Otherwise — MONSTER THANKS for taking a swing at this!! As I relayed to you, the only reliable way we have had to date to build Mac wallets was to acquire a 32-bit mac running 10.6 so having a modern up to date tool chain to be able to build true 64-bit wallets would be a coup for 2GIVE and a real contribution to the ALT community! Sync up later? -dvd On Apr 30, 2016, at 17:20, Jay Carlson [email protected] wrote:
|
|
I’ve folded in most of these changes by hand since I’m still getting back up to speed on GIT and PULL requests. Have not do the align 16 yet — wanted to make sure it all still worked as is on cross build for windows -dvd On Apr 30, 2016, at 18:04, Jay Carlson [email protected] wrote:
|
|
I’m not much of a git pro either, as I found out when I tried to do some repeated merges. I just did |
|
On some combination of compiler version, this counted as a duplicate declaration: If you set it back to |
|
I’ll try it and see if it breaks mining… On May 1, 2016, at 20:49, Jay Carlson [email protected] wrote:
|
Don't just blindly pull it. I have not tested this.