Make objects in range declarations immutable by default. Avoid unnecessary copying of objects in range declarations.
AbandonedPublic
Actions

Authored by fpelliccioni on Oct 1 2019, 16:32.

Details

Reviewers

Fabien
deadalnix

Group Reviewers

Restricted Project

Summary

Make objects in range declarations immutable by default.

Backport of Bitcoin Core PR13249
https://github.com/bitcoin/bitcoin/pull/13249

Test Plan

make check-all

Diff Detail

Repository

rABC Bitcoin ABC

Branch

feature-backport-5c24d3b98

Lint

Lint OK

Unit

No Unit Test Coverage

Build Status

Buildable 7702
Build 13443: Bitcoin ABC Buildbot (legacy)
Build 13442: arc lint + arc unit

Event Timeline

fpelliccioni created this revision.Oct 1 2019, 16:32

Herald added a reviewer: Restricted Project. · View Herald TranscriptOct 1 2019, 16:32

Herald added a reviewer: deadalnix. · View Herald Transcript

Build Bitcoin-ABC / Diffs / Diff Testing started.

Harbormaster completed remote builds in B7649: Diff 13293.Oct 1 2019, 16:50

Build Bitcoin-ABC / Diffs / Diff Testing passed.

Part of the PR are missing.

This revision now requires changes to proceed.Oct 1 2019, 18:46

In D4191#100495, @deadalnix wrote:

Part of the PR are missing.

It is a partial application of the PR, I forgot to clarify it, sorry. (summary edited).
I didn't apply the rule for built-in types, because it makes no sense:
for (int x : seq)
is preferred over
for (int const& x : seq).
Have I missed anything else?

fpelliccioni edited the summary of this revision. (Show Details)Oct 3 2019, 13:19

update the summary to clarify that it is a partial application of the PR.

Build Bitcoin-ABC / Diffs / Diff Testing started.

Build Bitcoin-ABC / Diffs / Diff Testing passed.

Harbormaster completed remote builds in B7678: Diff 13335.Oct 3 2019, 13:38

In D4191#100826, @fpelliccioni wrote:

In D4191#100495, @deadalnix wrote:

Part of the PR are missing.

It is a partial application of the PR, I forgot to clarify it, sorry. (summary edited).
I didn't apply the rule for built-in types, because it makes no sense:
for (int x : seq)
is preferred over
for (int const& x : seq).
Have I missed anything else?

for (const int x : seq) is even better. There are still some missing, for instance in qt/bitcoingui.cpp .

This revision now requires changes to proceed.Oct 3 2019, 15:41

In D4191#100850, @deadalnix wrote:

In D4191#100826, @fpelliccioni wrote:

In D4191#100495, @deadalnix wrote:

Part of the PR are missing.

It is a partial application of the PR, I forgot to clarify it, sorry. (summary edited).
I didn't apply the rule for built-in types, because it makes no sense:
for (int x : seq)
is preferred over
for (int const& x : seq).
Have I missed anything else?

for (const int x : seq) is even better. There are still some missing, for instance in qt/bitcoingui.cpp .