-
Notifications
You must be signed in to change notification settings - Fork 37.7k
utils: Use _wfopen and _wfreopen on Windows #13866
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
@Empact They are the same if we ignore the errno. |
No more conflicts as of last run. |
src/utilstrencodings.cpp
Outdated
#ifdef WIN32 | ||
std::wstring Utf8ToWide(const std::string& utf8_string) | ||
{ | ||
size_t size = MultiByteToWideChar(CP_UTF8, 0, &*utf8_string.begin(), utf8_string.size(), nullptr, 0); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I find the &* unnecessarily complicated. Can you use utf8_string.data() or even the more usual c_str() instead that gives a direct pointer back.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This avoid the unnecessary copy, &*utf8_string.begin()
point to where the exact string data are. c_str()
and data()
is a copy of the string data.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Not sure what implementation you are referring to but c_str and data have always returned the pointer to the internal data. Not a copy. Anyway it was just a nitpick. Thanks.
src/utilstrencodings.cpp
Outdated
size_t size = MultiByteToWideChar(CP_UTF8, 0, &*utf8_string.begin(), utf8_string.size(), nullptr, 0); | ||
std::wstring wide_string(size, 0); | ||
size = MultiByteToWideChar(CP_UTF8, 0, &*utf8_string.begin(), utf8_string.size(), &*wide_string.begin(), wide_string.size()); | ||
assert(size == wide_string.size()); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is this assert necessary? MultiByteToWideChar won't change the size of the string.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
They won't change, so that's why I use assert instead of a bunch of error handling.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I didn't spot the size= part before the asset. Sorry :)
src/fs.cpp
Outdated
@@ -1,15 +1,24 @@ | |||
#include <fs.h> | |||
#include <utilstrencodings.h> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
nit: Make this include conditional on WIN32
?
Update: use std codecvt instead of custom converter. |
The fopen function does not support unicode filename on Windows, so use Windows specific function do deal with it
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
utACK b9babc8
utACK b9babc8 |
b9babc8 utils: Use _wfopen and _wreopen on Windows (Chun Kuan Lee) Pull request description: The fopen function does not support unicode filename on Windows, so use Windows specific function do deal with it. Tree-SHA512: 4dcf14dcf9ec6307b9fdf95404e5b6b6b3df640949fd4b0c4ac7fecf8ea03a64fa25285fc319c4ff8a28e586eee106f1861116c181694955497402b2bf575f22
Summary: The fopen function does not support unicode filename on Windows, so use Windows specific function do deal with it This is a backport of Core [[bitcoin/bitcoin#13866 | PR13866]] Test Plan: `ninja && ninja check` Use the build bot to run tests on windows. Reviewers: #bitcoin_abc, Fabien Reviewed By: #bitcoin_abc, Fabien Differential Revision: https://reviews.bitcoinabc.org/D8343
Merge bitcoin#19660, bitcoin#19373, bitcoin#19841, bitcoin#13862, bitcoin#13866, bitcoin#17280, bitcoin#17682 and partial bitcoin#19326, bitcoin#14978: Auxiliary Backports
63e0be6 [Remove] By-pass logprint-scanner restriction. (furszy) 280ced3 utils: Fix broken Windows filelock (Chun Kuan Lee) be89860 utils: Convert Windows args to utf-8 string (Chun Kuan Lee) e8cfa6e Call unicode API on Windows (Chun Kuan Lee) 1a02a8a tests: Add test case for std::ios_base::ate (Chun Kuan Lee) 2e57cd4 Move boost/std fstream to fsbridge (furszy) 9d8bcd4 utils: Add fsbridge fstream function wrapper (Chun Kuan Lee) d59d48d utils: Convert fs error messages from multibyte to utf-8 (ken2812221) 9ef58cc Logging: use "fmterr" variable name for errors instead of general "e" that can be used by any other function. (furszy) dd94241 utils: Use _wfopen and _wreopen on Windows (Chun Kuan Lee) 3993641 add unicode compatible file_lock for Windows (Chun Kuan Lee) 48349f8 Provide relevant error message if datadir is not writable. (murrayn) Pull request description: As the software is currently using the ANSI encoding on Windows, the user's language settings could affect the proper functioning of the node/wallet, to the point of not be able to open some non-ASCII name files and directories. This solves the Windows encoding issues, completing the entire bitcoin#13869 work path (and some other required backports). Enabling for example users that use non-english characters in directories and file names to be accepted. Backported PRs: * bitcoin#12422. * bitcoin#12630. * bitcoin#13862. * bitcoin#13866. * bitcoin#13877. * bitcoin#13878. * bitcoin#13883. * bitcoin#13884. * bitcoin#13886. * bitcoin#13888. * bitcoin#14192. * bitcoin#13734. * bitcoin#14426. This is built on top of other two PRs that i have open #2423 and #2369. Solves old issues #940 and #2163. TODO: * Backport `assert_start_raises_init_error` and `ErrorMatch` in TestNode` (bitcoin#12718) ACKs for top commit: Fuzzbawls: ACK 63e0be6 random-zebra: ACK 63e0be6 and merging... Tree-SHA512: cb1f7c23abb5b7b3af50bba18652cc2cad93fd7c2fca9c16ffd3fee34c4c152a3b666dfa87fe6b44c430064dcdee4367144dcb4a41203c91b0173b805bdb3d7d
The fopen function does not support unicode filename on Windows, so use Windows specific function do deal with it.