fa61365a13 wallet: Translate all initErrors in CreateWalletFromFile (MarcoFalke)
fa70d199d0 util: Make util/error bilingual_str (refactor) (MarcoFalke)
Pull request description:
The translations are going to close in three days (#15940), so I am submitting this as a standalone pull request.
Those changes are part of a bugfix #16661, which includes a test. The first change (the refactor) is required, the second commit is not. I am happy to drop it, if needed.
ACKs for top commit:
laanwj:
utACK fa61365a13
hebasto:
ACK fa61365a13, I have tested the code on Linux Mint 19.2.
Tree-SHA512: a7616cc38b9ffd301c6b915ea808a65815c3d97e9f57ec091772eb260e5cf0d75a13a6e4dfa3913e236833677c7929b9a748cb7d7a0e406d51749944b614e11b
057fac3453 gui: Update English source translation (Wladimir J. van der Laan)
072f37a7ed gui: Update transifex slug for 0.19 (Wladimir J. van der Laan)
Pull request description:
- Update transifex slug for 0.19
- Update English source translation
ACKs for top commit:
fanquake:
ACK 057fac3453 - I see the same diff when running `make translate`.
Tree-SHA512: e6800011c2cc642b656d2a8fa652e94c665183424321eb4f1f3506e90a65f730794e5e9d49243538c0710c66fec1ed6f30e099dbc8a847e840817e894782866f
9965940e35 doc: Add release note for the new gettransaction argument (darosior)
b8b3f0435a tests: Add a new functional test for gettransaction (darosior)
7f3bb247a8 gettransaction: add an argument to decode the transaction (darosior)
Pull request description:
This PR adds a new parameter to the `gettransaction` call : `decode`. If set to `true`, it will add a new `decoded` field to the response. This mimics the behavior of `getrawtransaction`'s `verbose` argument to avoid using 2 calls if we want to decode a wallet transaction (`gettransaction` then `decoderawtransaction`).
Fix#16181 .
ACKs for top commit:
meshcollider:
re-utACK 9965940e35
Tree-SHA512: bcb6b4bd252b3488d6afc77659c499c2ad99fd58661eb24b6a2e17014c74f22e47fde70e00fedb4f4754915786622ad02483b2cf2c4dea0ab0eb4ac8276dbeee
9b92538ade Remove unused fScriptChecks parameter from CheckInputs (Matt Corallo)
Pull request description:
fScriptChecks = false just short-circuits the entire function, so
passing it in is entirely useless.
This is extracted from #13233 /cc TheBlueMatt.
Recommend reviewing with `git show --ignore-all-space`, i.e.:
https://github.com/bitcoin/bitcoin/pull/13868/files?w=1
ACKs for top commit:
TheBlueMatt:
utACK 9b92538ade. Checked diff had no functional change and new comment copy looks correct.
kallewoof:
ACK 9b92538ade
ajtowns:
ACK 9b92538ade ; code review, checked tests work. Looks right to me, and fanquake's notes make sense. Could change the coinbase early exit to `assert(!tx.IsCoinBase());`.
fanquake:
ACK 9b92538ade - Notes / testing below.
Tree-SHA512: add253a3e8cf4b33eddbc49efcec333c14b5ea61c7d34e43230351d40cff6adc919a75b91c72c4de8647a395284db74a61639f4c67848d4b2fec3a705b557790
0bb33b5348 qt: Replace objc_msgSend with native syntax (Hennadii Stepanov)
Pull request description:
Changes in Xcode 11 Objective-C Runtime cause an error (#16387) during building on MacOS 10.15 Catalina.
This PR fixes this issue by replacing `objc_msgSend()` function calls with the native Objective-C syntax.
Refs:
- [changes in `objc_msgSend` function](https://developer.apple.com/documentation/objectivec/1456712-objc_msgsend?changes=latest_minor&language=objc)
- [`OBJC_OLD_DISPATCH_PROTOTYPES` macro](https://developer.apple.com/documentation/objectivec/objc_old_dispatch_prototypes?language=objc)
ACKs for top commit:
l2a5b1:
ACK 0bb33b5 - Diff looks good. Sending messages via native Objective-C code feels more robust and is more readable than casting all the `objc_msgSend` function calls to the appropriate function signature (which would also have fixed the issue).
jonasschnelli:
utACK 0bb33b5348 - Confirmed that the called macOS framework function is available on our build targets.
fanquake:
ACK 0bb33b5348 - Still works as expected.
Tree-SHA512: c09cb684d06bd1da053a17c182b7bb1642e45bb347d26c76e1c5d835c320567caee366d85e34bb7f2be38e63ed041e0d06a56c2a9d89f7e5bece9b19cc5c6772
8340763dc3 Check for codespell in lint-spelling.sh (Kristaps Kaupe)
Pull request description:
Similar check for `spellcheck` already exists in `lint-shell.sh`
ACKs for top commit:
practicalswift:
ACK 8340763dc3 -- diff looks correct
fanquake:
ACK 8340763dc3 - Test is the same as in `lint-shell.sh`.
Tree-SHA512: f590fc4eae391c5246e86cb70cf79431bd3763ac872f82d3b2222c3a095235ca046a7072c5e61bbaeb3ed6ecd5ff85979769ee8b585de4fb97ac8cfbbf73b956
d9d8984270 wallet: Use wallet name instead of pointer on unload/release (João Barbosa)
Pull request description:
Fixes#16668. Wallet name is unique so it can be used instead of pointer.
ACKs for top commit:
meshcollider:
utACK d9d8984270
instagibbs:
utACK d9d8984270
ryanofsky:
utACK d9d8984270. Alternately I think it might be possible to use an intptr_t set instead of a string set to get around the undefined behavior described in the issue.
Tree-SHA512: eccd4d260cd4c02b52c30deeb32dbfd190a1151a5340eb3aa4ece0dc6ae3b3ed746ce5617336461f6f27c437c435629cd07d20beb1c5450f23b75edde6728598
This adds a new boolean parameter 'decode' to the gettransaction call, which, if set to true, add a 'decoded' field to the result containing the decoded transaction
8b6f5aabb9 qt: Replace QFontMetrics::width() with TextWidth() (Hennadii Stepanov)
Pull request description:
Compiling master (d8fc997913) on macOS Catalina (with a patch from #16720) reveals one more instance of `QFontMetrics::width()` which is supposed to be replaced with `TextWidth()` in the merged #16701.
Sorry for incomplete solution provided in #16701. It’s especially sad that the line I missed lies in only 7 lines from the code touched in #16701.
ACKs for top commit:
fanquake:
ACK 8b6f5aabb9
Tree-SHA512: 65cd8bea550150e5ee47c1e906d8c2393547cf4feba3701a933a4f24fad5ecdb552ac2de4e1200ed14efaa0df0480150dd58fccbddc3b902f6c2141603874902
798a589aff wallet: extract PubKey from P2PK script with Solver (Sebastian Falbesoner)
Pull request description:
The function `ExtractPubKey()` checks if a given script matches the P2PK pattern
(`<PubKey> OP_CHECKSIG`), extracts the PubKey and additionally checks if it is
cryptographically valid (full validation with ECC library via `CPubKey::IsFullyValid()`).
Currently this is done manually in the following order:
1. check if first script OP is data push with valid PubKey length (first part of pattern match), extract PubKey
2. create `CPubKey` object with extracted PubKey
3. fully validate public key
4. check if last script OP is `OP_CHECKSIG` (second part of pattern match)
Using Solver, the pattern matching and PubKey extraction can be done via a
single step, leading to the following simplified order with shorter code:
1. check if given script matches P2PK pattern with Solver (also contains valid PubKey length check), extracts Pubkey
2. create `CPubKey` object with extracted Pubkey
3. fully validate public key
ACKs for top commit:
instagibbs:
utACK 798a589aff
theStack:
> utACK [798a589](798a589aff)
sipa:
ACK 798a589aff
achow101:
Code Review ACK 798a589aff
Tree-SHA512: 350358a89afed8c2a7967c50e9714a2d4a909259b50e694ce68dde3e7d0fa0bf3238d33642e73f2bdb53860f6d3f7327ca3eb6426b74eaffacfbca0a384d68cd
fa0b910486 [doc] chain: Declare BLOCK_VALID_HEADER reserved (MarcoFalke)
Pull request description:
`BLOCK_VALID_HEADER` was never used and the comment is confusing to me in several ways:
* It claims "version ok". However, without the previous header, it is not possible to check the validity of the version since the height needs to be known (c.f. BIP 90)
* It claims "hash satisfies claimed PoW". While it is possible to check against the claimed PoW, it is not possible without the previous header to check that the claimed PoW is itself valid.
* It claims "1 <= vtx count <= max". However, with the header alone and current consensus rules, the number of transactions is unknown.
ACKs for top commit:
sipa:
ACK fa0b910486
ryanofsky:
ACK fa0b910486
Tree-SHA512: 3972995a0a2f83aa55767bf8982af1fcb9493483f62aee6df27e58be9181a48d5968ae718b390cecc8be3ed4f26495683b1cffde8ef272dea0bd610ec169ef8b
c53667764e qt: Remove obsolete QModelIndex::child() (Hennadii Stepanov)
Pull request description:
The `QModelIndex::child()` member function is [obsolete](https://doc.qt.io/qt-5.12/qmodelindex-obsolete.html) since Qt 5.12.
This PR removes it, does not change behavior and keeps compatibility with [Qt 5.5.1](https://github.com/bitcoin/bitcoin/pull/15393).
Here is an excerpt from the master build log:
```
qt/receivecoinsdialog.cpp: In member function ‘void ReceiveCoinsDialog::copyColumnToClipboard(int)’:
qt/receivecoinsdialog.cpp:264:111: warning: ‘QModelIndex QModelIndex::child(int, int) const’ is deprecated: Use QAbstractItemModel::index [-Wdeprecated-declarations]
GUIUtil::setClipboard(model->getRecentRequestsTableModel()->data(firstIndex.child(firstIndex.row(), column), Qt::EditRole).toString());
^
In file included from /home/hebasto/Qt/5.13.0/gcc_64/include/QtWidgets/qabstractitemview.h:45:0,
from /home/hebasto/Qt/5.13.0/gcc_64/include/QtWidgets/qheaderview.h:44,
from /home/hebasto/Qt/5.13.0/gcc_64/include/QtWidgets/QHeaderView:1,
from ./qt/guiutil.h:12,
from ./qt/receivecoinsdialog.h:8,
from qt/receivecoinsdialog.cpp:7:
/home/hebasto/Qt/5.13.0/gcc_64/include/QtCore/qabstractitemmodel.h:457:20: note: declared here
inline QModelIndex QModelIndex::child(int arow, int acolumn) const
^~~~~~~~~~~
```
ACKs for top commit:
laanwj:
Code review ACK c53667764e
promag:
ACK c53667764e, just read the change.
jonasschnelli:
utACK c53667764e
Tree-SHA512: 99fcb6ff60a6d47b925bda9f14006269eaad09ba4f7a41ac4975c6cf04bd906b58aed721cbfa0be7da8e6613d92e30d4be18b7e4d3960f026c7226558a4c3196
Add a LockChain method to CWallet to know if we can lock or query
chain state safely.
At tx loading, we rely on chain to know if hashBlock of tx is still
in main chain. If not, we set its status to unconfirmed and reset
its hashBlock/nIndex.
If wallet loaded is the wallet-tool one, all wallet txn will
show up with a height of zero. It doesn't matter as status is not
used by wallet-tool.
We take lock prematurely in CWallet::LoadWallet and CWallet::Verify
to ensure that lock order is respected between cs_main an cs_wallet.
d48c1e837a Add window final block height to getchaintxstats (Jonathan "Duke" Leto)
Pull request description:
This patch is motivated by the desire to make the output of `getchaintxstats` more useful and optimized for applications to consume and render the data.
Firstly, this data is already available to the RPC, no additional work is done. Currently additional RPC calls will be needed to look up the height of the final block in the window or the block height that began the window.
By adding the block height of the final block in the window, the JSON is "self-contained" and applications can calculate the exact block height range of the window with no additional RPC requests.
For example, a web application which wants to render historical information for `getchaintxstats` RPC on various window sizes might call the RPC with various window lengths, once per day, and store the JSON results somewhere. Because the final block height of each dataset is included, it's no extra work to determine the exact block window range of each JSON response.
ACKs for top commit:
promag:
ACK d48c1e837a.
Tree-SHA512: fd4952c125f81a4ad18f7c78498c6b3e265b93cb574832166ac25596321ce84957f971f3f78f37d7e42638dc65f2a5d4d760f289873c9c2f2a82eb00a0f87c3f
0912134039 qt: Remove QSignalMapper from TransactionView (Hennadii Stepanov)
9e0c1d676c qt: Remove QSignalMapper from RPCConsole (Hennadii Stepanov)
Pull request description:
The [`QSignalMapper`](https://doc.qt.io/qt-5/qsignalmapper.html) class has been [deprecated](https://doc-snapshots.qt.io/qt5-5.10/obsoleteclasses.html) since Qt 5.10.
This PR replaces it by lambdas and does not change behavior.
ACKs for top commit:
jonasschnelli:
utACK 0912134039
Tree-SHA512: 0c102d5cab4adc8b6252f72e07123ac87c65434c88cada3e72816ecea8fc4803f15b9c050fb5e1c7e8a96f709265521fd6813ab1890dbf5634032f7ee0d50675
d75e704ac0 Add log output during initial header sync (Jonas Schnelli)
Pull request description:
The non debug log output is completely quiet during the header sync. I see two main reasons to add infos about the state of the initial header sync...
* users may think the node did fail to start sync
* it's a little complicate to check if your getting throttled during header sync (repeatedly calling `getchaintips` or similar)
ACKs for top commit:
fanquake:
Concept ACK d75e704ac0
practicalswift:
utACK d75e704ac0
laanwj:
Tested ACK d75e704ac0
Tree-SHA512: 2e738571b703d7251290864603c3a829729645962c2fa3187250bab0585e66a5f01fce892e9b5b98da451fab2b40a2e4784f9b2e5a9cad75ff62c535affe7430
c6dd32da69 qt: Replace obsolete functions of QDesktopWidget (Hennadii Stepanov)
1260ecd812 qt: Add TextWidth() wrapper (Hennadii Stepanov)
Pull request description:
The following functions are obsolete in Qt 5.13:
- [`QFontMetrics::width()`](https://doc.qt.io/qt-5/qfontmetrics-obsolete.html#width)
- [`QDesktopWidget::availableGeometry()`](https://doc.qt.io/qt-5/qdesktopwidget-obsolete.html#availableGeometry)
- [`QDesktopWidget::screenGeometry()`](https://doc.qt.io/qt-5/qdesktopwidget-obsolete.html#screenGeometry)
This PR replaces them and does not change behavior.
Here are some excerpts from the master build log:
```
qt/bitcoingui.cpp: In constructor ‘BitcoinGUI::BitcoinGUI(interfaces::Node&, const PlatformStyle*, const NetworkStyle*, QWidget*)’:
qt/bitcoingui.cpp:84:57: warning: ‘const QRect QDesktopWidget::availableGeometry(int) const’ is deprecated: Use QGuiApplication::screens() [-Wdeprecated-declarations]
move(QApplication::desktop()->availableGeometry().center() - frameGeometry().center());
^
In file included from /home/hebasto/Qt/5.13.0/gcc_64/include/QtWidgets/QDesktopWidget:1:0,
from qt/bitcoingui.cpp:43:
/home/hebasto/Qt/5.13.0/gcc_64/include/QtWidgets/qdesktopwidget.h:88:67: note: declared here
QT_DEPRECATED_X("Use QGuiApplication::screens()") const QRect availableGeometry(int screen = -1) const;
^~~~~~~~~~~~~~~~~
```
```
qt/bitcoingui.cpp:1410:74: warning: ‘int QFontMetrics::width(const QString&, int) const’ is deprecated: Use QFontMetrics::horizontalAdvance [-Wdeprecated-declarations]
max_width = qMax(max_width, fm.width(BitcoinUnits::longName(unit)));
^
In file included from /home/hebasto/Qt/5.13.0/gcc_64/include/QtWidgets/qwidget.h:50:0,
from /home/hebasto/Qt/5.13.0/gcc_64/include/QtWidgets/qdialog.h:44,
from /home/hebasto/Qt/5.13.0/gcc_64/include/QtWidgets/QDialog:1,
from ./qt/optionsdialog.h:8,
from ./qt/bitcoingui.h:12,
from qt/bitcoingui.cpp:5:
/home/hebasto/Qt/5.13.0/gcc_64/include/QtGui/qfontmetrics.h:108:9: note: declared here
int width(const QString &, int len = -1) const;
^~~~~
```
```
qt/splashscreen.cpp: In constructor ‘SplashScreen::SplashScreen(interfaces::Node&, Qt::WindowFlags, const NetworkStyle*)’:
qt/splashscreen.cpp:127:50: warning: ‘const QRect QDesktopWidget::screenGeometry(int) const’ is deprecated: Use QGuiApplication::screens() [-Wdeprecated-declarations]
move(QApplication::desktop()->screenGeometry().center() - r.center());
^
In file included from /home/hebasto/Qt/5.13.0/gcc_64/include/QtWidgets/QDesktopWidget:1:0,
from qt/splashscreen.cpp:24:
/home/hebasto/Qt/5.13.0/gcc_64/include/QtWidgets/qdesktopwidget.h:79:67: note: declared here
QT_DEPRECATED_X("Use QGuiApplication::screens()") const QRect screenGeometry(int screen = -1) const;
^~~~~~~~~~~~~~
```
ACKs for top commit:
jonasschnelli:
utACK c6dd32da69
Tree-SHA512: deb7bcbf86e1dcc6508bd91288772c2fe8811db79fa2011de37d0469cdd094fbf7fd8c4512c607bed0bd08dc2968e893c0bbc190732c43c69ed1085259df766c
The function ExtractPubKey() checks if a given script matches the P2PK pattern
(<PubKey> OP_CHECKSIG), extracts the PubKey and additionally checks if it is
cryptographically valid (full validation with ECC library via .IsFullyValid()).
Currently this is done manually in the following order:
1) check if first script OP is data push with valid PubKey length
(first part of pattern match), extract PubKey
2) create CPubKey object with extracted PubKey
3) fully validate public key
4) check if last script OP is OP_CHECKSIG
(second part of pattern match)
Using Solver, the pattern matching and PubKey extraction can be done via a
single step, leading to the following simplified order with shorter code:
1) check if given script matches P2PK pattern with Solver
(also contains valid PubKey length check), extracts Pubkey
2) create CPubKey object with extracted Pubkey
3) fully validate public key
d53d591230 Added the bench_bitcoin project to the list automatically produced by the msvc-autogen python script. (Aaron Clauson)
Pull request description:
As discussed in #16747.
The msbuild bench_bitcoin project is currently missing a number of classes. Rather than add them manually this PR adds bench_bitcoin to the list of project files that are auto generated from the main make files.
ACKs for top commit:
practicalswift:
ACK d53d591230
fanquake:
ACK d53d591230 - Thanks for following up quickly.
Tree-SHA512: dd767f3d8b1641a736a868b0ab5173369ab68dc42b5ac60f26e1d435a1813caad494aee175fc54b457a82e2faf36b9676ea35414c7d93c240d2e069434714f03
41d484d5c8 doc: Delete stale URL in test README (Michael Folkson)
Pull request description:
The resource on the Boost unit test framework previously linked to in src/test/README.md was a stale URL.
Instead of deleting it, I've replaced it with an alternative resource on the framework on [boost.org](https://www.boost.org/doc/libs/1_45_0/libs/test/doc/html/utf/tutorials.html).
ACKs for top commit:
promag:
ACK 41d484d5c8.
hebasto:
ACK 41d484d5c8, the removed link is really obsolete.
fanquake:
ACK 41d484d5c8 - Thanks.
Tree-SHA512: 764f12548441bde615f77b7a2ca7c5188b4ab936972d16b84960fbd8604d4cbd224415bc59ce839e7e63293aa84fd97f31a69e38734e531231cdb0e148d2e1bd
fa27372e65 ci: Move CCACHE_DIR and test_runner tmp dir into ./ci/scratch/ (MarcoFalke)
fa60583d23 ci: Pass down $MAKEJOBS to test_runner.py (MarcoFalke)
Pull request description:
Some changes to the ci system:
* Pass down MAKEJOBS to the test_runner, instead of falling back to the default of 4. Passing it down avoids OOM on weak machines and allows better use of resources on beefy machines.
* Move CCACHE_DIR to ./ci/scratch/ subfolder: `ccache` is executed with root permissions inside the docker, so the cache files are created with root as owner. So it might be wise to not put them in the $HOME of the host
* Use the scratch dir as prefix for the test runner, as opposed to `/tmp/`, which is often a ramdisk and thus leads to OOM on the host when either a lot of tests are run in parallel or when a lot of tests fail and the datadirs are not cleaned.
ACKs for top commit:
fanquake:
ACK fa27372e65
laanwj:
ACK fa27372e65
Tree-SHA512: 67834fbab282051ec81c319d460528b32870507e53df2b8a1ce9a1f3f6a685aaf8eb8ba03f5406918ca4a33adf736e6a4adad7134c54cf3a9e47a26c64a13442
f3b57f4a1c Unrecommend making config file owned by bitcoin (setpill)
870d4152df Set ProtectHome in systemd service file (setpill)
639a416e37 Chgrp config dir to bitcoin in systemd service (setpill)
aded0528f0 Improve clarity of systemd service file comments (setpill)
Pull request description:
Rationale: ran into a bug with the systemd service file, fixed it locally and figured I might as well contribute my fix.
Also fixed some unrelated confusing phrasing in the comments of the same file, after discussion in IRC.
ACKs for top commit:
sipsorcery:
tACK f3b57f4a1c (nothing changed since previous tACK).
ryanofsky:
utACK f3b57f4a1c. Only change since last review is removing ConfigurationDirectoryMode churn in early commits
Tree-SHA512: 2188345878925b9e8a5c2c3df8dfba443720e2252a164db54a8e1d8007846721497b2d98c56f1d9b60a9a9ed4fdb1156c7b02c699616b220a9b614671617d32a
1373fa7e3d doc: add default bitcoin.conf locations (Chuf)
Pull request description:
Added default bitcoin.conf data directories and paths
ACKs for top commit:
practicalswift:
ACK 1373fa7e3d
ryanofsky:
ACK 1373fa7e3d
fanquake:
ACK 1373fa7e3d - Already three ACKs and lots of discussion here, so I'm going to merge, and the other comment
Tree-SHA512: 8bb1ed9868c5d171b6791bd6dc9598eddfdf64977d327ff4f333323cef8e3e76b1a67da21e4199f008a12f5610ac6dc6f34f4a13235e8846754eb6d6e5075da4
9452802480 doc: Tidy up shadowing section (João Barbosa)
Pull request description:
Removes the example because it violates the code format.
ACKs for top commit:
MarcoFalke:
unsigned ACK 9452802480
ryanofsky:
ACK 9452802480
fanquake:
ACK 9452802480 - Thanks for following up.
Tree-SHA512: 1fc31355d368225713298da7803e39e99014fbfcd229f2d3b56c082de95ab2965e51c80b172a5abce4646c53f845fa62a6d94d5df714e7835cac07a8ec7d5da7
aff1c9c884 Adds an option to msbuild common configuration to ignore linker warning 4221. This warning is for object files that do not include any symbols. The warning is harmless and occurs due to some classes that are *nix only having no source to compile for an msvc build. (Aaron Clauson)
Pull request description:
Adds an option to msbuild common configuration to ignore linker warning 4221. This warning is for object files that do not include any symbols. The warning is harmless and occurs due to some classes that are *nix only having no source to compile for an msvc build.
ACKs for top commit:
MarcoFalke:
ACK aff1c9c884
hebasto:
ACK aff1c9c884, I have reviewed the code, it looks OK and I agree it can be merged.
fanquake:
ACK aff1c9c884
Tree-SHA512: fda3721462d927d8734e16fc0250c7c76b197a085219a9e5267cdd14e19f23f50807d6b5516dee8402b60f3fdf188677e3d618fe6acbdbb5ac3bb635aeb7b297
Translated strings should not end up in the debug log, stderr, or
returned by an RPC.
Changing the util methods in util/error to return a bilingual_str paves
the way to achieve this goal in the long term.
e4f4ea47eb lint: Catch use of [] or {} as default parameter values in Python functions (practicalswift)
25dd867150 Avoid using mutable default parameter values (practicalswift)
Pull request description:
Avoid common Python default parameter gotcha when mutable `dict`/`list`:s are used as default parameter values.
Examples of this gotcha caught during review:
* https://github.com/bitcoin/bitcoin/pull/16673#discussion_r317415261
* https://github.com/bitcoin/bitcoin/pull/14565#discussion_r241942304
Perhaps surprisingly this is how mutable list and dictionary default parameter values behave in Python:
```
>>> def f(i, j=[], k={}):
... j.append(i)
... k[i] = True
... return j, k
...
>>> f(1)
([1], {1: True})
>>> f(1)
([1, 1], {1: True})
>>> f(2)
([1, 1, 2], {1: True, 2: True})
```
In contrast to:
```
>>> def f(i, j=None, k=None):
... if j is None:
... j = []
... if k is None:
... k = {}
... j.append(i)
... k[i] = True
... return j, k
...
>>> f(1)
([1], {1: True})
>>> f(1)
([1], {1: True})
>>> f(2)
([2], {2: True})
```
The latter is typically the intended behaviour.
This PR fixes two instances of this and adds a check guarding against this gotcha going forward :-)
ACKs for top commit:
Sjors:
Oh Python... ACK e4f4ea47eb. Testing tip: swap the two commits.
Tree-SHA512: 56e14d24fc866211a20185c9fdb274ed046c3aed2dc0e07699e58b6f9fa3b79f6d0c880fb02d72b7fe5cc5eb7c0ff6da0ead33123344e1a872209370c2e49e3f
