Depends bumps #4767

pull theuni wants to merge 3 commits into bitcoin:master from theuni:depends-bumps changing 4 files +7 −6

theuni commented at 7:29 PM on August 26, 2014: member

Bumps our miniupnp/openssl versions to match what was done for 0.9.3.

build: add -DMINIUPNP_STATICLIB for new version

libminiupnpc changed their required static define to the much more sane
"MINIUPNP_STATICLIB". Sadly, they don't respect the old "STATICLIB" for
back-compat. Define them both since the old one didn't seem to be conflicting
anywhere.

Also go ahead and split out the cppflags so that they can be applied only where
they're needed. This will help us to build dll's from our libs without having
their import/export declspecs poisoned.

9f7f504efc

depends: bump openssl to 1.0.1i

This is the forward-port of bba01750226745d6666d587cabe57c321fde0875.

f628127887

TheBlueMatt commented at 1:42 AM on August 27, 2014: member

ACK, verified miniupnpc's new hash of 26f3985bad7768b8483b793448ae49414cdc4451d0ec83e7c1944367e15f9f07 and openssl's new hash of 3c179f46ca77069a6a0bac70212a9b3b838b2f66129cb52d568837fc79d8fcc7
laanwj commented at 6:52 AM on August 27, 2014: member

Untested ack, hashes match those in 0.9.3.
theuni force-pushed on Aug 27, 2014
theuni commented at 6:54 AM on August 27, 2014: member

Killed the whitespace so that @TheBlueMatt can sleep at night :)

Admittedly that would've bothered me too, I just didn't see it.

depends: bump miniupnpc to 1.9.20140701.

This is the forward-port of a9c6eef915ee264ccf4169d5e8769a6be1fa101a

b144a74a08

BitcoinPullTester commented at 7:06 AM on August 27, 2014: none

Automatic sanity-testing: PASSED, see http://jenkins.bluematt.me/pull-tester/p4767_b144a74a08a856997f48a3c83f0bd080cf9432f9/ for binaries and test log. This test script verifies pulls every time they are updated. It, however, dies sometimes and fails to test properly. If you are waiting on a test, please check timestamps to verify that the test.log is moving at http://jenkins.bluematt.me/pull-tester/current/ Contact BlueMatt on freenode if something looks broken.
theuni commented at 7:11 AM on August 27, 2014: member

Great.. http://miniupnp.free.fr/files is down so the pull-tester couldn't grab the new miniupnpc version to test.

Looks like #4752 will come in handy for cases like this one.
laanwj commented at 8:23 AM on August 27, 2014: member

I've merged #4752.

There's a concern that's brewing in the back of my head: is Travis hammering the upstream download URIs with requests? Or does the caching avoid this?
theuni commented at 9:49 AM on August 27, 2014: member

No, the caching avoids it. It'll only fetch the sources on PRs where the deps have changed (and even then, only those that have changed and those that depend on them). Once the deps have settled, it'll be very rare.
TheBlueMatt commented at 4:31 AM on August 28, 2014: member

@theuni In case you didnt notice, miniupnp is dual-hosted at free.fr and tuxfamily.org, so we could put in a second url for that.
laanwj added the label Build system on Aug 28, 2014
theuni commented at 7:01 PM on August 29, 2014: member

@TheBlueMatt I gave it some thought. I was getting ready to add the logic to try the first url, then an optional 2nd url, then the fallback url (#4752) if those failed. Realistically, the upstream url and our self-hosted fallback should be enough I'd think. If both of those fail, a backup link isn't likely to work either, and we'd just end up with a set of backup urls that are rarely tested, and are probably long-stale.
laanwj merged this on Sep 1, 2014
laanwj closed this on Sep 1, 2014
laanwj referenced this in commit 107922663b on Sep 1, 2014
MarcoFalke locked this on Sep 8, 2021

Contributors

Labels