url: use foreach-style C++ loop #23138
Conversation
nodejs-github-bot
added
c++
src/node_url.h
Outdated
| @@ -159,9 +159,9 @@ class URL { | |||
|
|
|||
| std::string path() const { | |||
| std::string ret; | |||
| for (auto i = context_.path.begin(); i != context_.path.end(); i++) { | |||
| for (const auto& i : context_.path) { | |||
There was a problem hiding this comment.
Can we replace auto with std::string?
There was a problem hiding this comment.
I would ask not to.
As per the discussion in #23028
src/node_url.h
Outdated
| ret += '/'; | ||
| ret += *i; | ||
| ret += i; |
There was a problem hiding this comment.
There are two possible minor improvements.
One is to use stringstream
The other is to hint to the compiler the the / always goes with the next part:
ret += '/' + i;
YMMV
There was a problem hiding this comment.
Theoretically this should be the best solution, but I'm not sure I like it:
return std::accumulate(cbegin(_path), cend(_path), std::string(),
[](auto const& a, auto const& b) { return a + '/' + b; });
There was a problem hiding this comment.
I’d prefer std::accumulate over stringstream but otherwise the suggestions in these two comments seem fine (as does the current state of the PR)
There was a problem hiding this comment.
The idea behind using a range-based for loop was to improve readability, and using accumulate for such a trivial thing seems like doing the opposite.
src/node_url.h
Outdated
| ret += '/'; | ||
| ret += *i; | ||
| ret += i; |
There was a problem hiding this comment.
I’d prefer std::accumulate over stringstream but otherwise the suggestions in these two comments seem fine (as does the current state of the PR)
tniessen
force-pushed
the
url-use-foreach-loop
branch
from
bcad6b4
to
b3f40ee
Compare
tniessen
force-pushed
the
url-use-foreach-loop
branch
from
b3f40ee
to
2440d9c
Compare
src/node_url.h
Outdated
| for (auto i = context_.path.begin(); i != context_.path.end(); i++) { | ||
| ret += '/'; | ||
| ret += *i; | ||
| for (const std::string& i : context_.path) { |
There was a problem hiding this comment.
Maybe rename it to something like path_element?
To me, i means "index" and that's how it's always used in JS.
|
Landed in 1d9ec04 |
PR-URL: #23138 Reviewed-By: Refael Ackermann <[email protected]> Reviewed-By: Tiancheng "Timothy" Gu <[email protected]> Reviewed-By: Anna Henningsen <[email protected]> Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: Joyee Cheung <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Sakthipriyan Vairamani <[email protected]>
PR-URL: #23138 Reviewed-By: Refael Ackermann <[email protected]> Reviewed-By: Tiancheng "Timothy" Gu <[email protected]> Reviewed-By: Anna Henningsen <[email protected]> Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: Joyee Cheung <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Sakthipriyan Vairamani <[email protected]>
PR-URL: #23138 Reviewed-By: Refael Ackermann <[email protected]> Reviewed-By: Tiancheng "Timothy" Gu <[email protected]> Reviewed-By: Anna Henningsen <[email protected]> Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: Joyee Cheung <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Sakthipriyan Vairamani <[email protected]>
Change
URL::parseto use a range-based for loop for better readability.CI on commit before opening PR: https://ci.nodejs.org/job/node-test-commit/21840/
Checklist
make -j4 test(UNIX), orvcbuild test(Windows) passes