| Age | Commit message (Collapse) | Author |
|---|
|
c53d48a BIP70: Chainparams: DRY: Make qt/guiutil.cpp fit BIP70 chain name strings (Jorge Timón)
|
|
6dd3a44 translations: Don't translate markdown or force English grammar (MarcoFalke)
|
|
|
|
Fix a null pointer dereference in validateaddress with -disablewallet. Also add a regression testcase.
|
|
e167af2 [doc] Remove excessive white space (MarcoFalke)
e0eeb67 [trivial] clang-format: Set AlignAfterOpenBracket: false (MarcoFalke)
0af8fe4 devtools: Update README.md (MarcoFalke)
|
|
|
|
22e7807 Always flush block and undo when switching to new file (Pieter Wuille)
|
|
d1c3762 Revert "Revert "Enable policy enforcing GetMedianTimePast as the end point of lock-time constraints"" (Gregory Maxwell)
e4e5334 Restore MedianTimePast for locktime. (Gregory Maxwell)
|
|
53238ff Clarify what minrelaytxfee does (MarcoFalke)
abd8b76 [qt] Properly display required fee instead of minTxFee (MarcoFalke)
|
|
Previously, the undo weren't being flushed during a reindex because
fKnown was set to true in FindBlockPos. That is the correct behaviour
for block files as they aren't being touched, but undo files are
touched.
This changes the behaviour to always flush when switching to a new file
(even for block files, though that isn't really necessary).
|
|
|
|
As a side effect, the qt user will see "test" instead of "testnet"
|
|
268b79e [qt] rpcconsole: Scale monospace font to 95% (MarcoFalke)
28313b8 [qt] Use fixed pitch font for the rpc console (MarcoFalke)
|
|
6342a48 Init: Use DEFAULT_TRANSACTION_MINFEE in help message (MarcoFalke)
a9c73a1 [wallet] Add comments for doxygen (MarcoFalke)
6b0e622 [wallet] Refactor to use new MIN_CHANGE (MarcoFalke)
|
|
de0499d Fix ZMQ Notification initialization and shutdown (João Barbosa)
|
|
a6efc01 Bugfix: Omit wallet-related options from -help when wallet is disabled (Luke Dashjr)
5f9260f Bugfix: If genproclimit is omitted to RPC setgenerate, don't change it; also show correct default in getmininginfo (Luke Dashjr)
420a82f Bugfix: Describe dblogsize option correctly (it refers to the wallet database, not memory pool) (Luke Dashjr)
caa3d42 Bugfix: RPC: blockchain: Display correct defaults in help for verifychain method (Luke Dashjr)
|
|
Moves the call Initialize() from init.cpp to CreateWithArguments() and handles the
return value. Moves the call Shutdown() from init.cpp to destructor.
Changes Initialize() and Shutdown() to protected members.
|
|
of lock-time constraints""
This reverts commit 8537ecdfc40181249ec37556015a99cfae4b21fd.
|
|
Revert "Revert "Add rules--presently disabled--for using GetMedianTimePast as endpoint for lock-time calculations""
This reverts commit 40cd32e835092c3158175511da5193193ec54939.
After careful analysis it was determined that the change was, in fact, safe and several people were suffering
momentary confusion about locktime semantics.
|
|
|
|
|
|
|
|
40cd32e Revert "Add rules--presently disabled--for using GetMedianTimePast as endpoint for lock-time calculations" (Gregory Maxwell)
8537ecd Revert "Enable policy enforcing GetMedianTimePast as the end point of lock-time constraints" (Gregory Maxwell)
|
|
7497e80 tests: Initialize networking on windows (Wladimir J. van der Laan)
|
|
endpoint for lock-time calculations"
This reverts commit 9d55050773d57c0e12005e524f2e54d9e622c6e2.
As noted by Luke-Jr, under some conditions this will accept transactions which are invalid by the network
rules. This happens when the current block time is head of the median time past and a transaction's
locktime is in the middle.
This could be addressed by changing the rule to MAX(this_block_time, MTP+offset) but this solution and
the particular offset used deserve some consideration.
|
|
lock-time constraints"
This reverts commit dea8d21fc63e9f442299c97010e4740558f4f037.
|
|
Also update transifex slug for new version.
|
|
|
|
30d9662 Reject invalid pubkeys when reading ckey items from the wallet. (Gregory Maxwell)
|
|
e9e6163 Make -checkmempool=1 not fail through int32 overflow (Pieter Wuille)
|
|
|
|
This makes the behavior more consistent with key objects and will
reject some corrupted pubkeys (e.g. zero length).
|
|
Also:
* Preserve white space
* Make fixed font as large as default font
|
|
|
|
0be387a unittest: fix test for null tx input (Daniel Kraft)
|
|
a83f3c2 Add explicit shared_ptr constructor due to C++11 error (Bob McElrath)
|
|
040c0ea Init: Cleanup error and warning strings (MarcoFalke)
6782f58 [trivial] Latest config.guess (MarcoFalke)
bf68191 [trivial] rpcnet: fix typo (MarcoFalke)
95f4291 [trivial] Rewrite help text for feature enabled by default (MarcoFalke)
|
|
|
|
Also update doc/translation_strings_policy.md
|
|
|
|
* Introduce new constant MIN_CHANGE and use it instead of the
hardcoded "CENT"
* Add test case for MIN_CHANGE
* Introduce new constant for -mintxfee default:
DEFAULT_TRANSACTION_MINFEE = 1000
|
|
|
|
ad5aae1 constify missing catch cases (Philip Kaufmann)
|
|
214de7e [Trivial] ensure minimal header conventions (Philip Kaufmann)
|
|
ab1f560 Support -checkmempool=N, which runs checks on average once every N transactions (Pieter Wuille)
|
|
|
|
Update the unittest that is meant to catch a transaction that is invalid
because it has a null input. The old test failed not because of that
but because it was considered a coinbase with too large script. This is
already checked with a different test, though.
The new test is *not* a coinbase since it has two inputs, but one of
them is null. This really checks the corresponding code path in
CheckTransaction.
|
|
- ensure header namespaces and end comments are correct
- add missing header end comments
- ensure minimal formatting (add newlines etc.)
|
|
- ensure all missing catch cases are constant where possible
|
|
2d8c49d Clean up tx prioritization when conflict mined (Casey Rodarmor)
|
