cosmopolitan/libc/str/str.h

395 lines
21 KiB
C
Raw Normal View History

2020-06-15 14:18:57 +00:00
#ifndef COSMOPOLITAN_LIBC_STR_STR_H_
#define COSMOPOLITAN_LIBC_STR_STR_H_
#if !(__ASSEMBLER__ + __LINKER__ + 0)
COSMOPOLITAN_C_START_
/*───────────────────────────────────────────────────────────────────────────│─╗
Add /statusz page to redbean plus other enhancements redbean improvements: - Explicitly disable corking - Simulate Python regex API for Lua - Send warmup requests in main process on startup - Add Class-A granular IPv4 network classification - Add /statusz page so you can monitor your redbean's health - Fix regressions on OpenBSD/NetBSD caused by recent changes - Plug Authorization header into Lua GetUser and GetPass APIs - Recognize X-Forwarded-{For,Host} from local reverse proxies - Add many additional functions to redbean Lua server page API - Report resource usage of child processes on `/` listing page - Introduce `-a` flag for logging child process resource usage - Introduce `-t MILLIS` flag and `ProgramTimeout(ms)` init API - Introduce `-H "Header: value"` flag and `ProgramHeader(k,v)` API Cosmopolitan Libc improvements: - Make strerror() simpler - Make inet_pton() not depend on sscanf() - Fix OpenExecutable() which broke .data section earlier - Fix stdio in cases where it overflows kernel tty buffer - Fix bugs in crash reporting w/o .com.dbg binary present - Add polyfills for SO_LINGER, SO_RCVTIMEO, and SO_SNDTIMEO - Polyfill TCP_CORK on BSD and XNU using TCP_NOPUSH magnums New netcat clone in examples/nc.c: While testing some of the failure conditions for redbean, I noticed that BusyBox's `nc` command is pretty busted, if you use it as an interactive tool, rather than having it be part of a pipeline. Unfortunately this'll only work on UNIX since Windows doesn't let us poll on stdio and sockets at the same time because I don't think they want tools like this running on their platform. So if you want forbidden fruit, it's here so enjoy it
2021-04-23 17:45:19 +00:00
cosmopolitan § characters » usas x3.4-1967
2020-06-15 14:18:57 +00:00
fourth age telecommunications */
extern const uint8_t gperf_downcase[256];
extern const uint8_t kToLower[256];
Undiamond Python headers This change gets the Python codebase into a state where it conforms to the conventions of this codebase. It's now possible to include headers from Python, without worrying about ordering. Python has traditionally solved that problem by "diamonding" everything in Python.h, but that's problematic since it means any change to any Python header invalidates all the build artifacts. Lastly it makes tooling not work. Since it is hard to explain to Emacs when I press C-c C-h to add an import line it shouldn't add the header that actually defines the symbol, and instead do follow the nonstandard Python convention. Progress has been made on letting Python load source code from the zip executable structure via the standard C library APIs. System calss now recognizes zip!FILENAME alternative URIs as equivalent to zip:FILENAME since Python uses colon as its delimiter. Some progress has been made on embedding the notice license terms into the Python object code. This is easier said than done since Python has an extremely complicated ownership story. - Some termios APIs have been added - Implement rewinddir() dirstream API - GetCpuCount() API added to Cosmopolitan Libc - More bugs in Cosmopolitan Libc have been fixed - zipobj.com now has flags for mangling the path - Fixed bug a priori with sendfile() on certain BSDs - Polyfill F_DUPFD and F_DUPFD_CLOEXEC across platforms - FIOCLEX / FIONCLEX now polyfilled for fast O_CLOEXEC changes - APE now supports a hybrid solution to no-self-modify for builds - Many BSD-only magnums added, e.g. O_SEARCH, O_SHLOCK, SF_NODISKIO
2021-08-12 07:42:14 +00:00
extern const uint8_t kToUpper[256];
2020-06-15 14:18:57 +00:00
extern const uint8_t kBase36[256];
extern const char16_t kCp437[256];
2020-06-15 14:18:57 +00:00
int isascii(int);
int isspace(int);
int isalpha(int);
int isdigit(int);
int isalnum(int);
int isxdigit(int);
int isprint(int);
int islower(int);
int isupper(int);
int isblank(int);
int iscntrl(int);
int isgraph(int);
int tolower(int);
int ispunct(int);
int toupper(int);
int hextoint(int);
2020-10-27 10:39:46 +00:00
int cescapec(int);
2020-06-15 14:18:57 +00:00
/*───────────────────────────────────────────────────────────────────────────│─╗
cosmopolitan § characters » thompson-pike encoding
fifth age telecommunications
0123456789abcdef
0 Control
1§
2 !"#$%&'()*+,-./┐
3 0123456789:;<=>?
4 @ABCDEFGHIJKLMNOASA x3.4-1967
5 PQRSTUVWXYZ[\]^_
6 `abcdefghijklmno
7pqrstuvwxyz{|}~
8ÇüéâäàåçêëèïîìÄÅ
9 ÉæÆôöòûùÿÖÜ¢£¥ƒThompson-Pike Continuation
a á¡óúñѪº¿¬½¼¡«» (not really characters)
b
c1 Continuation will follow
d
eαßΓπΣσμτΦΘΩδφε2 Continuations will follow
f±÷°·²λ
5 Continuations follow (and is negative)
5 Continuations follow (note: -1=λ)
4 Continuations follow
3 Continuations follow */
#define INVALID_CODEPOINT 0xfffd
2020-06-15 14:18:57 +00:00
int iswalnum(wint_t);
int iswalpha(wint_t);
int iswblank(wint_t);
int iswcntrl(wint_t);
int iswdigit(wint_t);
int iswgraph(wint_t);
int iswlower(wint_t);
int iswspace(wint_t);
int iswupper(wint_t);
int iswxdigit(wint_t);
int iswpunct(wint_t);
int iswprint(wint_t);
wint_t towlower(wint_t);
wint_t towupper(wint_t);
2020-06-15 14:18:57 +00:00
/*───────────────────────────────────────────────────────────────────────────│─╗
cosmopolitan § strings
*/
Undiamond Python headers This change gets the Python codebase into a state where it conforms to the conventions of this codebase. It's now possible to include headers from Python, without worrying about ordering. Python has traditionally solved that problem by "diamonding" everything in Python.h, but that's problematic since it means any change to any Python header invalidates all the build artifacts. Lastly it makes tooling not work. Since it is hard to explain to Emacs when I press C-c C-h to add an import line it shouldn't add the header that actually defines the symbol, and instead do follow the nonstandard Python convention. Progress has been made on letting Python load source code from the zip executable structure via the standard C library APIs. System calss now recognizes zip!FILENAME alternative URIs as equivalent to zip:FILENAME since Python uses colon as its delimiter. Some progress has been made on embedding the notice license terms into the Python object code. This is easier said than done since Python has an extremely complicated ownership story. - Some termios APIs have been added - Implement rewinddir() dirstream API - GetCpuCount() API added to Cosmopolitan Libc - More bugs in Cosmopolitan Libc have been fixed - zipobj.com now has flags for mangling the path - Fixed bug a priori with sendfile() on certain BSDs - Polyfill F_DUPFD and F_DUPFD_CLOEXEC across platforms - FIOCLEX / FIONCLEX now polyfilled for fast O_CLOEXEC changes - APE now supports a hybrid solution to no-self-modify for builds - Many BSD-only magnums added, e.g. O_SEARCH, O_SHLOCK, SF_NODISKIO
2021-08-12 07:42:14 +00:00
void bzero(void *, size_t) libcesque;
2020-06-15 14:18:57 +00:00
void *memset(void *, int, size_t) memcpyesque;
void *memcpy(void *restrict, const void *restrict, size_t) memcpyesque;
void *mempcpy(void *restrict, const void *restrict, size_t) memcpyesque;
void *memccpy(void *restrict, const void *restrict, int, size_t) memcpyesque;
void *memmove(void *, const void *, size_t) memcpyesque;
void *memeqmask(void *, const void *, const void *, size_t) memcpyesque;
Undiamond Python headers This change gets the Python codebase into a state where it conforms to the conventions of this codebase. It's now possible to include headers from Python, without worrying about ordering. Python has traditionally solved that problem by "diamonding" everything in Python.h, but that's problematic since it means any change to any Python header invalidates all the build artifacts. Lastly it makes tooling not work. Since it is hard to explain to Emacs when I press C-c C-h to add an import line it shouldn't add the header that actually defines the symbol, and instead do follow the nonstandard Python convention. Progress has been made on letting Python load source code from the zip executable structure via the standard C library APIs. System calss now recognizes zip!FILENAME alternative URIs as equivalent to zip:FILENAME since Python uses colon as its delimiter. Some progress has been made on embedding the notice license terms into the Python object code. This is easier said than done since Python has an extremely complicated ownership story. - Some termios APIs have been added - Implement rewinddir() dirstream API - GetCpuCount() API added to Cosmopolitan Libc - More bugs in Cosmopolitan Libc have been fixed - zipobj.com now has flags for mangling the path - Fixed bug a priori with sendfile() on certain BSDs - Polyfill F_DUPFD and F_DUPFD_CLOEXEC across platforms - FIOCLEX / FIONCLEX now polyfilled for fast O_CLOEXEC changes - APE now supports a hybrid solution to no-self-modify for builds - Many BSD-only magnums added, e.g. O_SEARCH, O_SHLOCK, SF_NODISKIO
2021-08-12 07:42:14 +00:00
void explicit_bzero(void *, size_t);
2020-06-15 14:18:57 +00:00
size_t strlen(const char *) strlenesque;
size_t strnlen(const char *, size_t) strlenesque;
size_t strnlen_s(const char *, size_t);
char *strchr(const char *, int) strlenesque;
char *index(const char *, int) strlenesque;
void *memchr(const void *, int, size_t) strlenesque;
char *strchrnul(const char *, int) strlenesque returnsnonnull;
void *rawmemchr(const void *, int) strlenesque returnsnonnull;
size_t strlen16(const char16_t *) strlenesque;
size_t strnlen16(const char16_t *, size_t) strlenesque;
size_t strnlen16_s(const char16_t *, size_t);
char16_t *strchr16(const char16_t *, int) strlenesque;
void *memchr16(const void *, int, size_t) strlenesque;
char16_t *strchrnul16(const char16_t *, int) strlenesque returnsnonnull;
void *rawmemchr16(const void *, int) strlenesque returnsnonnull;
size_t wcslen(const wchar_t *) strlenesque;
size_t wcsnlen(const wchar_t *, size_t) strlenesque;
size_t wcsnlen_s(const wchar_t *, size_t);
wchar_t *wcschr(const wchar_t *, wchar_t) strlenesque;
wchar_t *wmemchr(const wchar_t *, wchar_t, size_t) strlenesque;
wchar_t *wcschrnul(const wchar_t *, wchar_t) strlenesque returnsnonnull;
char *strstr(const char *, const char *) strlenesque;
char16_t *strstr16(const char16_t *, const char16_t *) strlenesque;
wchar_t *wcsstr(const wchar_t *, const wchar_t *) strlenesque;
void *rawwmemchr(const void *, wchar_t) strlenesque returnsnonnull;
int memcmp(const void *, const void *, size_t) strlenesque;
int strcmp(const char *, const char *) strlenesque;
int strncmp(const char *, const char *, size_t) strlenesque;
int strcmp16(const char16_t *, const char16_t *) strlenesque;
int strncmp16(const char16_t *, const char16_t *, size_t) strlenesque;
int wcscmp(const wchar_t *, const wchar_t *) strlenesque;
int wcsncmp(const wchar_t *, const wchar_t *, size_t) strlenesque;
int wmemcmp(const wchar_t *, const wchar_t *, size_t) strlenesque;
int strcasecmp(const char *, const char *) strlenesque;
int memcasecmp(const void *, const void *, size_t) strlenesque;
2020-06-15 14:18:57 +00:00
int strcasecmp16(const char16_t *, const char16_t *) strlenesque;
int wcscasecmp(const wchar_t *, const wchar_t *) strlenesque;
int strncasecmp(const char *, const char *, size_t) strlenesque;
int strncasecmp16(const char16_t *, const char16_t *, size_t) strlenesque;
int wcsncasecmp(const wchar_t *, const wchar_t *, size_t) strlenesque;
char *strrchr(const char *, int) strlenesque;
void *memrchr(const void *, int, size_t) strlenesque;
char16_t *strrchr16(const char16_t *, int) strlenesque;
void *memrchr16(const void *, int, size_t) strlenesque;
wchar_t *wcsrchr(const wchar_t *, int) strlenesque;
void *wmemrchr(const void *, wchar_t, size_t) strlenesque;
char *strpbrk(const char *, const char *) strlenesque;
char16_t *strpbrk16(const char16_t *, const char16_t *) strlenesque;
wchar_t *wcspbrk(const wchar_t *, const wchar_t *) strlenesque;
size_t strspn(const char *, const char *) strlenesque;
size_t strspn16(const char16_t *, const char16_t *) strlenesque;
size_t wcsspn(const wchar_t *, const wchar_t *) strlenesque;
size_t strcspn(const char *, const char *) strlenesque;
size_t strcspn16(const char16_t *, const char16_t *) strlenesque;
size_t wcscspn(const wchar_t *, const wchar_t *) strlenesque;
void *memfrob(void *, size_t) memcpyesque;
int strcoll(const char *, const char *) strlenesque;
char *strsep(char **, const char *) paramsnonnull();
int strcmpzbw(const uint16_t *, const char *) strlenesque;
int strcasecmpzbw(const uint16_t *, const char *) strlenesque;
char *stpcpy(char *, const char *) memcpyesque;
char *stpncpy(char *, const char *, size_t) memcpyesque;
char *strcat(char *, const char *) memcpyesque;
2021-01-30 16:54:12 +00:00
char16_t *strcat16(char16_t *, const char16_t *) memcpyesque;
wchar_t *wcscat(wchar_t *, const wchar_t *) memcpyesque;
2020-06-15 14:18:57 +00:00
size_t strlcpy(char *, const char *, size_t);
size_t strlcat(char *, const char *, size_t);
2021-05-01 11:53:23 +00:00
size_t strxfrm(char *, const char *, size_t);
2020-06-15 14:18:57 +00:00
char *strcpy(char *, const char *) memcpyesque;
char16_t *strcpy16(char16_t *, const char16_t *) memcpyesque;
2021-01-30 16:54:12 +00:00
wchar_t *wcscpy(wchar_t *, const wchar_t *) memcpyesque;
2020-06-15 14:18:57 +00:00
char *strncat(char *, const char *, size_t) memcpyesque;
Undiamond Python headers This change gets the Python codebase into a state where it conforms to the conventions of this codebase. It's now possible to include headers from Python, without worrying about ordering. Python has traditionally solved that problem by "diamonding" everything in Python.h, but that's problematic since it means any change to any Python header invalidates all the build artifacts. Lastly it makes tooling not work. Since it is hard to explain to Emacs when I press C-c C-h to add an import line it shouldn't add the header that actually defines the symbol, and instead do follow the nonstandard Python convention. Progress has been made on letting Python load source code from the zip executable structure via the standard C library APIs. System calss now recognizes zip!FILENAME alternative URIs as equivalent to zip:FILENAME since Python uses colon as its delimiter. Some progress has been made on embedding the notice license terms into the Python object code. This is easier said than done since Python has an extremely complicated ownership story. - Some termios APIs have been added - Implement rewinddir() dirstream API - GetCpuCount() API added to Cosmopolitan Libc - More bugs in Cosmopolitan Libc have been fixed - zipobj.com now has flags for mangling the path - Fixed bug a priori with sendfile() on certain BSDs - Polyfill F_DUPFD and F_DUPFD_CLOEXEC across platforms - FIOCLEX / FIONCLEX now polyfilled for fast O_CLOEXEC changes - APE now supports a hybrid solution to no-self-modify for builds - Many BSD-only magnums added, e.g. O_SEARCH, O_SHLOCK, SF_NODISKIO
2021-08-12 07:42:14 +00:00
char16_t *strncat16(char16_t *, const char16_t *, size_t) memcpyesque;
wchar_t *wcsncat(wchar_t *, const wchar_t *, size_t) memcpyesque;
2020-06-15 14:18:57 +00:00
char *strncpy(char *, const char *, size_t) memcpyesque;
char *strtok(char *, const char *) paramsnonnull((2)) libcesque;
char *strtok_r(char *, const char *, char **) paramsnonnull((2, 3));
uint16_t *strcpyzbw(uint16_t *, const char *) memcpyesque;
2021-06-10 02:34:27 +00:00
wchar_t *wcstok(wchar_t *, const wchar_t *, wchar_t **) paramsnonnull((2, 3));
2020-06-15 14:18:57 +00:00
char *wstrtrunc(uint16_t *) memcpyesque;
char *wstrntrunc(uint16_t *, size_t) memcpyesque;
bool startswith(const char *, const char *) strlenesque;
2021-08-14 13:17:56 +00:00
bool startswithi(const char *, const char *) strlenesque;
2020-06-15 14:18:57 +00:00
bool startswith16(const char16_t *, const char16_t *) strlenesque;
bool wcsstartswith(const wchar_t *, const wchar_t *) strlenesque;
bool endswith(const char *, const char *) strlenesque;
bool endswith16(const char16_t *, const char16_t *) strlenesque;
bool wcsendswith(const wchar_t *, const wchar_t *) strlenesque;
2020-10-27 10:39:46 +00:00
const char *IndexDoubleNulString(const char *, unsigned) strlenesque;
int strverscmp(const char *, const char *);
2020-06-15 14:18:57 +00:00
wchar_t *wmemset(wchar_t *, wchar_t, size_t) memcpyesque;
char16_t *memset16(char16_t *, char16_t, size_t) memcpyesque;
compatfn wchar_t *wmemcpy(wchar_t *, const wchar_t *, size_t) memcpyesque;
compatfn wchar_t *wmempcpy(wchar_t *, const wchar_t *, size_t) memcpyesque;
compatfn wchar_t *wmemmove(wchar_t *, const wchar_t *, size_t) memcpyesque;
2020-12-29 04:01:15 +00:00
int timingsafe_memcmp(const void *, const void *, size_t);
2021-02-27 18:33:32 +00:00
void *tinymemccpy(void *, const void *, int, size_t) memcpyesque;
Undiamond Python headers This change gets the Python codebase into a state where it conforms to the conventions of this codebase. It's now possible to include headers from Python, without worrying about ordering. Python has traditionally solved that problem by "diamonding" everything in Python.h, but that's problematic since it means any change to any Python header invalidates all the build artifacts. Lastly it makes tooling not work. Since it is hard to explain to Emacs when I press C-c C-h to add an import line it shouldn't add the header that actually defines the symbol, and instead do follow the nonstandard Python convention. Progress has been made on letting Python load source code from the zip executable structure via the standard C library APIs. System calss now recognizes zip!FILENAME alternative URIs as equivalent to zip:FILENAME since Python uses colon as its delimiter. Some progress has been made on embedding the notice license terms into the Python object code. This is easier said than done since Python has an extremely complicated ownership story. - Some termios APIs have been added - Implement rewinddir() dirstream API - GetCpuCount() API added to Cosmopolitan Libc - More bugs in Cosmopolitan Libc have been fixed - zipobj.com now has flags for mangling the path - Fixed bug a priori with sendfile() on certain BSDs - Polyfill F_DUPFD and F_DUPFD_CLOEXEC across platforms - FIOCLEX / FIONCLEX now polyfilled for fast O_CLOEXEC changes - APE now supports a hybrid solution to no-self-modify for builds - Many BSD-only magnums added, e.g. O_SEARCH, O_SHLOCK, SF_NODISKIO
2021-08-12 07:42:14 +00:00
void *memmem(const void *, size_t, const void *, size_t) libcesque nosideeffect;
char *strerror(int) returnsnonnull nothrow nocallback;
2021-02-21 19:06:18 +00:00
long a64l(const char *);
char *l64a(long);
2020-06-15 14:18:57 +00:00
char *strntolower(char *, size_t);
char *strtolower(char *) paramsnonnull();
char *strntoupper(char *, size_t);
char *strtoupper(char *) paramsnonnull();
char *chomp(char *);
char16_t *chomp16(char16_t *);
wchar_t *wchomp(wchar_t *);
2021-08-14 13:17:56 +00:00
bool IsText(const void *, size_t);
bool IsUtf8(const void *, size_t);
Improve ZIP filesystem and change its prefix The ZIP filesystem has a breaking change. You now need to use /zip/ to open() / opendir() / etc. assets within the ZIP structure of your APE binary, instead of the previous convention of using zip: or zip! URIs. This is needed because Python likes to use absolute paths, and having ZIP paths encoded like URIs simply broke too many things. Many more system calls have been updated to be able to operate on ZIP files and file descriptors. In particular fcntl() and ioctl() since Python would do things like ask if a ZIP file is a terminal and get confused when the old implementation mistakenly said yes, because the fastest way to guarantee native file descriptors is to dup(2). This change also improves the async signal safety of zipos and ensures it doesn't maintain any open file descriptors beyond that which the user has opened. This change makes a lot of progress towards adding magic numbers that are specific to platforms other than Linux. The philosophy here is that, if you use an operating system like FreeBSD, then you should be able to take advantage of FreeBSD exclusive features, even if we don't polyfill them on other platforms. For example, you can now open() a file with the O_VERIFY flag. If your program runs on other platforms, then Cosmo will automatically set O_VERIFY to zero. This lets you safely use it without the need for #ifdef or ifstatements which detract from readability. One of the blindspots of the ASAN memory hardening we use to offer Rust like assurances has always been that memory passed to the kernel via system calls (e.g. writev) can't be checked automatically since the kernel wasn't built with MODE=asan. This change makes more progress ensuring that each system call will verify the soundness of memory before it's passed to the kernel. The code for doing these checks is fast, particularly for buffers, where it can verify 64 bytes a cycle. - Correct O_LOOP definition on NT - Introduce program_executable_name - Add ASAN guards to more system calls - Improve termios compatibility with BSDs - Fix bug in Windows auxiliary value encoding - Add BSD and XNU specific errnos and open flags - Add check to ensure build doesn't talk to internet
2021-08-22 08:04:18 +00:00
bool _isabspath(const char *) strlenesque;
2020-10-27 10:39:46 +00:00
bool escapedos(char16_t *, unsigned, const char16_t *, unsigned);
void *memset_pure(void *, int, size_t) memcpyesque;
void *memmove_pure(void *, const void *, size_t) memcpyesque;
void *mempcpy_pure(void *, const void *, size_t) memcpyesque;
size_t strlen_pure(const char *) strlenesque;
size_t strcspn_pure(const char *, const char *) strlenesque;
2020-06-15 14:18:57 +00:00
/*───────────────────────────────────────────────────────────────────────────│─╗
cosmopolitan § strings » multibyte
*/
typedef unsigned mbstate_t;
2020-12-01 11:43:40 +00:00
axdx_t tprecode8to16(char16_t *, size_t, const char *);
axdx_t tprecode16to8(char *, size_t, const char16_t *);
2020-06-15 14:18:57 +00:00
int strcmp8to16(const char *, const char16_t *) strlenesque;
int strncmp8to16(const char *, const char16_t *, size_t) strlenesque;
int strcasecmp8to16(const char *, const char16_t *) strlenesque;
int strncasecmp8to16(const char *, const char16_t *, size_t) strlenesque;
int strcmp16to8(const char16_t *, const char *) strlenesque;
int strncmp16to8(const char16_t *, const char *, size_t) strlenesque;
int strcasecmp16to8(const char16_t *, const char *) strlenesque;
int strncasecmp16to8(const char16_t *, const char *, size_t) strlenesque;
wchar_t *wcsncpy(wchar_t *, const wchar_t *, size_t);
int mbtowc(wchar_t *, const char *, size_t);
size_t mbrtowc(wchar_t *, const char *, size_t, mbstate_t *);
size_t mbsrtowcs(wchar_t *, const char **, size_t, mbstate_t *);
size_t mbstowcs(wchar_t *, const char *, size_t);
size_t wcrtomb(char *, wchar_t, mbstate_t *);
size_t c32rtomb(char *, char32_t, mbstate_t *);
size_t mbrtoc32(char32_t *, const char *, size_t, mbstate_t *);
size_t c16rtomb(char *, char16_t, mbstate_t *);
size_t mbrtoc16(char16_t *, const char *, size_t, mbstate_t *);
size_t mbrlen(const char *, size_t, mbstate_t *);
size_t mbsnrtowcs(wchar_t *, const char **, size_t, size_t, mbstate_t *);
size_t wcsnrtombs(char *, const wchar_t **, size_t, size_t, mbstate_t *);
size_t wcsrtombs(char *, const wchar_t **, size_t, mbstate_t *);
size_t wcstombs(char *, const wchar_t *, size_t);
int mbsinit(const mbstate_t *);
int mblen(const char *, size_t);
2020-06-15 14:18:57 +00:00
int wctomb(char *, wchar_t);
int wctob(wint_t);
wint_t btowc(int);
2020-06-15 14:18:57 +00:00
size_t strclen(const char *) nosideeffect;
size_t strnclen(const char *, size_t) nosideeffect;
size_t strclen16(const char16_t *) nosideeffect;
size_t strnclen16(const char16_t *, size_t) nosideeffect;
typedef unsigned wctype_t;
wctype_t wctype(const char *) strlenesque;
int iswctype(wint_t, wctype_t) pureconst;
2020-06-15 14:18:57 +00:00
/*───────────────────────────────────────────────────────────────────────────│─╗
cosmopolitan § strings » system
*/
char *strsignal(int) returnsnonnull libcesque;
2020-12-24 07:42:56 +00:00
#if defined(__GNUC__) && !defined(__STRICT_ANSI__)
2020-06-15 14:18:57 +00:00
/*───────────────────────────────────────────────────────────────────────────│─╗
cosmopolitan § strings » optimizations
*/
#define __memcpy_isgoodsize(SIZE) \
(__builtin_constant_p(SIZE) && ((SIZE) <= __BIGGEST_ALIGNMENT__ && \
2020-12-24 07:42:56 +00:00
__builtin_popcountl((unsigned)(SIZE)) == 1))
2020-06-15 14:18:57 +00:00
2020-12-24 07:42:56 +00:00
#define __memset_isgoodsize(SIZE) \
(__builtin_constant_p(SIZE) && \
(((SIZE) <= __BIGGEST_ALIGNMENT__ && \
__builtin_popcountl((unsigned)(SIZE)) == 1) || \
((SIZE) % __BIGGEST_ALIGNMENT__ == 0 && \
(SIZE) / __BIGGEST_ALIGNMENT__ <= 3)))
2020-06-15 14:18:57 +00:00
#define memcpy(DEST, SRC, SIZE) \
(__memcpy_isgoodsize(SIZE) ? __builtin_memcpy(DEST, SRC, SIZE) \
: __memcpy("MemCpy", DEST, SRC, SIZE))
2020-06-15 14:18:57 +00:00
#define memset(DEST, BYTE, SIZE) \
(__memset_isgoodsize(SIZE) ? __builtin_memset(DEST, BYTE, SIZE) \
: __memset(DEST, BYTE, SIZE))
2020-12-24 07:42:56 +00:00
#if defined(__STDC_HOSTED__) && defined(__SSE2__)
#define strlen(STR) \
(__builtin_constant_p(STR) ? __builtin_strlen(STR) : ({ \
size_t LeN; \
const char *StR = (STR); \
asm("call\tstrlen" \
: "=a"(LeN) \
: "D"(StR), "m"(*(char(*)[0x7fffffff])StR) \
: "rcx", "rdx", "xmm3", "xmm4", "cc"); \
LeN; \
}))
2020-06-15 14:18:57 +00:00
#define memmove(DEST, SRC, SIZE) __memcpy("MemMove", (DEST), (SRC), (SIZE))
2020-09-03 12:44:37 +00:00
#define mempcpy(DEST, SRC, SIZE) \
({ \
size_t SIze = (SIZE); \
(void *)((char *)memcpy((DEST), (SRC), SIze) + SIze); \
})
2020-06-15 14:18:57 +00:00
#define __memcpy(FN, DEST, SRC, SIZE) \
({ \
void *DeSt = (DEST); \
const void *SrC = (SRC); \
size_t SiZe = (SIZE); \
asm("call\t" FN \
: "=m"(*(char(*)[SiZe])(DeSt)) \
: "D"(DeSt), "S"(SrC), "d"(SiZe), "m"(*(const char(*)[SiZe])(SrC)) \
2020-12-24 07:42:56 +00:00
: "xmm3", "xmm4", "rcx", "cc"); \
2020-06-15 14:18:57 +00:00
DeSt; \
})
2020-06-15 14:18:57 +00:00
#define __memset(DEST, BYTE, SIZE) \
({ \
void *DeSt = (DEST); \
size_t SiZe = (SIZE); \
asm("call\tMemSet" \
2020-06-15 14:18:57 +00:00
: "=m"(*(char(*)[SiZe])(DeSt)) \
: "D"(DeSt), "S"(BYTE), "d"(SiZe) \
2020-12-24 07:42:56 +00:00
: "xmm3", "xmm4", "rcx", "cc"); \
2020-06-15 14:18:57 +00:00
DeSt; \
})
2020-12-24 07:42:56 +00:00
#else /* hosted+sse2 */
2020-06-15 14:18:57 +00:00
#define mempcpy(DEST, SRC, SIZE) \
({ \
void *Rdi, *Dest = (DEST); \
const void *Rsi, *Src = (SRC); \
size_t SiZe = (SIZE); \
size_t Rcx; \
asm("rep movsb" \
: "=D"(Rdi), "=S"(Rsi), "=c"(Rcx), "=m"(*(char(*)[SiZe])(Dest)) \
2020-06-15 14:18:57 +00:00
: "0"(Dest), "1"(Src), "2"(SiZe), "m"(*(const char(*)[SiZe])(Src)) \
: "cc"); \
Rdi; \
})
2020-06-15 14:18:57 +00:00
#define __memcpy(FN, DEST, SRC, SIZE) \
({ \
void *Rdi, *Dest = (DEST); \
const void *Rsi, *Src = (SRC); \
size_t SiZe = (SIZE); \
size_t Rcx; \
asm("rep movsb" \
: "=D"(Rdi), "=S"(Rsi), "=c"(Rcx), "=m"(*(char(*)[SiZe])(Dest)) \
: "0"(Dest), "1"(Src), "2"(SiZe), "m"(*(const char(*)[SiZe])(Src)) \
: "cc"); \
Dest; \
})
2020-06-15 14:18:57 +00:00
#define __memset(DEST, BYTE, SIZE) \
({ \
void *Rdi, *Dest = (DEST); \
size_t SiZe = (SIZE); \
size_t Rcx; \
asm("rep stosb" \
: "=D"(Rdi), "=c"(Rcx), "=m"(*(char(*)[SiZe])(Dest)) \
2020-09-03 12:44:37 +00:00
: "0"(Dest), "1"(SiZe), "a"(BYTE) \
2020-06-15 14:18:57 +00:00
: "cc"); \
Dest; \
})
#endif /* hosted/sse2/unbloat */
/*───────────────────────────────────────────────────────────────────────────│─╗
cosmopolitan § strings » address sanitizer
*/
#if defined(__FSANITIZE_ADDRESS__)
#undef memcpy
#undef memmove
#undef mempcpy
#undef memset
#undef strlen
#define memcpy memmove_pure
#define memmove memmove_pure
#define mempcpy mempcpy_pure
#define memset memset_pure
#define strcspn strcspn_pure
#define strlen strlen_pure
#endif /* __FSANITIZE_ADDRESS__ */
2020-06-15 14:18:57 +00:00
#endif /* __GNUC__ && !__STRICT_ANSI__ */
COSMOPOLITAN_C_END_
#endif /* !(__ASSEMBLER__ + __LINKER__ + 0) */
#endif /* COSMOPOLITAN_LIBC_STR_STR_H_ */