summaryrefslogtreecommitdiff
path: root/compat/winansi.c
AgeCommit message (Collapse)AuthorFilesLines
2014-06-16Win32: reliably detect console pipe handlesLibravatar Karsten Blees1-18/+7
As of "Win32: Thread-safe windows console output", child processes may print to the console even if stdout has been redirected to a file. E.g.: git config tar.cat.command "cat" git archive -o test.cat HEAD Detecting whether stdout / stderr point to our console pipe is currently based on the assumption that OS HANDLE values are never reused. This is apparently not true if stdout / stderr is replaced via dup2() (as in builtin/archive.c:17). Instead of comparing handle values, check if the file descriptor isatty() backed by a pipe OS handle. This is only possible by swapping the handles in MSVCRT's internal data structures, as we do in winansi_init(). Reported-by: Johannes Sixt <j6t@kdbg.org> Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2014-06-10Win32: fix broken pipe detectionLibravatar Karsten Blees1-44/+70
As of "Win32: Thread-safe windows console output", git-log no longer terminates when the pager process dies. This is due to disabling buffering for the replaced stdout / stderr streams. Git-log will periodically fflush stdout (see write_or_die.c/mayble_flush_or_die()), but with no buffering, this is a NOP that always succeeds (so we never detect the EPIPE error). Exchange the original console handles with our console thread pipe handles by accessing the internal MSVCRT data structures directly (which are exposed via __pioinfo for some reason). Implement this with minimal assumptions about the actual data structure to make it work with different (hopefully even future) MSVCRT versions. While messing with internal data structures is ugly, this patch solves the problem at the source instead of adding more workarounds. We no longer need the special winansi_isatty override, and the limitations documented in "Win32: Thread-safe windows console output" are gone (i.e. fdopen(1/2) returns unbuffered streams now, and isatty() for duped console file descriptors works as expected). Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Stepan Kasal <kasal@ucw.cz> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2014-06-10Win32: Thread-safe windows console outputLibravatar Karsten Blees1-138/+263
Winansi.c has many static variables that are accessed and modified from the [v][f]printf / fputs functions overridden in the file. This may cause multi threaded git commands that print to the console to produce corrupted output or even crash. Additionally, winansi.c doesn't override all functions that can be used to print to the console (e.g. fwrite, write, fputc are missing), so that ANSI escapes don't work properly for some git commands (e.g. git-grep). Instead of doing ANSI emulation in just a few wrapped functions on top of the IO API, let's plug into the IO system and take advantage of the thread safety inherent to the IO system. Redirect stdout and stderr to a pipe if they point to the console. A background thread reads from the pipe, handles ANSI escape sequences and UTF-8 to UTF-16 conversion, then writes to the console. The pipe-based stdout and stderr replacements must be set to unbuffered, as MSVCRT doesn't support line buffering and fully buffered streams are inappropriate for console output. Due to the byte-oriented pipe, ANSI escape sequences and multi-byte UTF-8 sequences can no longer be expected to arrive in one piece. Replace the string-based ansi_emulate() with a simple stateful parser (this also fixes colored diff hunk headers, which were broken as of commit 2efcc977). Override isatty to return true for the pipes redirecting to the console. Exec/spawn obtain the original console handle to pass to the next process via winansi_get_osfhandle(). All other overrides are gone, the default stdio implementations work as expected with the piped stdout/stderr descriptors. Global variables are either initialized on startup (single threaded) or exclusively modified by the background thread. Threads communicate through the pipe, no further synchronization is necessary. The background thread is terminated by disonnecting the pipe after flushing the stdio and pipe buffers. This doesn't work for anonymous pipes (created via CreatePipe), as DisconnectNamedPipe only works on the read end, which discards remaining data. Thus we have to setup the pipe manually, with the write end beeing the server (opened with CreateNamedPipe) and the read end the client (opened with CreateFile). Limitations: doesn't track reopened or duped file descriptors, i.e.: - fdopen(1/2) returns fully buffered streams - dup(1/2), dup2(1/2) returns normal pipe descriptors (i.e. isatty() = false, winansi_get_osfhandle won't return the original console handle) Currently, only the git-format-patch command uses xfdopen(xdup(1)) (see "realstdout" in builtin/log.c), but works well with these limitations. Many thanks to Atsushi Nakagawa <atnak@chejz.com> for suggesting and reviewing the thread-exit-mechanism. Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Stepan Kasal <kasal@ucw.cz> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2014-06-10Win32: warn if the console font doesn't support UnicodeLibravatar Karsten Blees1-0/+66
Unicode console output won't display correctly with default settings because the default console font ("Terminal") only supports the system's OEM charset. Unfortunately, this is a user specific setting, so it cannot be easily fixed by e.g. some registry tricks in the setup program. This change prints a warning on exit if console output contained non-ascii characters and the console font is supposedly not a TrueType font (which usually have decent Unicode support). Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Stepan Kasal <kasal@ucw.cz> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2014-06-10Win32: detect console streams more reliablyLibravatar Karsten Blees1-24/+26
GetStdHandle(STD_OUTPUT_HANDLE) doesn't work for stderr if stdout is redirected. Use _get_osfhandle of the FILE* instead. _isatty() is true for all character devices (including parallel and serial ports). Check return value of GetConsoleScreenBufferInfo instead to reliably detect console handles (also don't initialize internal state from an uninitialized CONSOLE_SCREEN_BUFFER_INFO structure if the function fails). Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Stepan Kasal <kasal@ucw.cz> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2014-06-10Win32: support Unicode console outputLibravatar Karsten Blees1-6/+20
WriteConsoleW seems to be the only way to reliably print unicode to the console (without weird code page conversions). Also redirects vfprintf to the winansi.c version. Signed-off-by: Karsten Blees <blees@dcon.de> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Stepan Kasal <kasal@ucw.cz> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2009-09-18Make usage of windows.h lean and meanLibravatar Marius Storm-Olsen1-1/+0
Centralize the include of windows.h in git-compat-util.h, turn on WIN32_LEAN_AND_MEAN to avoid including plenty of other header files which is not needed in Git. Also ensure we load winsock2.h first, so we don't load the older winsock definitions at a later stage, since they contain duplicate definitions. When moving windows.h into git-compat-util.h, we need to protect the definition of struct pollfd in mingw.h, since this file is used by both MinGW and MSVC, and the latter defines this struct in winsock2.h. We need to keep the windows.h include in compat/win32.h, since its shared by both MinGW and Cygwin, and we're not touching Cygwin in this commit. The include in git-compat-util.h is protected with an ifdef WIN32, which is not the case when compiling for Cygwin. Signed-off-by: Marius Storm-Olsen <mstormo@gmail.com> Acked-by: Johannes Sixt <j6t@kdbg.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2009-06-01Work around a regression in Windows 7, causing erase_in_line() to crash ↵Libravatar Johannes Schindelin1-1/+2
sometimes The function FillConsoleOutputCharacterA() was pretty content in XP to take a NULL pointer if we did not want to store the number of written columns. In Windows 7, it crashes, but only when called from within Git Bash, not from within cmd.exe. Go figure. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Steffen Prohaska <prohaska@zib.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2009-03-10winansi: support ESC [ K (erase in line)Libravatar Johannes Schindelin1-3/+15
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Johannes Sixt <j6t@kdbg.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-07-19Add ANSI control code emulation for the Windows consoleLibravatar Peter Harris1-0/+345
This adds only the minimum necessary to keep git pull/merge's diffstat from wrapping. Notably absent is support for the K (erase) operation, and support for POSIX write. Signed-off-by: Peter Harris <git@peter.is-a-geek.org> Signed-off-by: Johannes Sixt <johannes.sixt@telecom.at> Signed-off-by: Junio C Hamano <gitster@pobox.com>