Cython 3.3.0a1__cp315-cp315-win_amd64.whl

Cython/Includes/posix/stdlib.pxd

@@ -0,0 +1,29 @@
+# POSIX additions to <stdlib.h>
+# https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/stdlib.h.html
+
+cdef extern from "<stdlib.h>" nogil:
+    void   _Exit(int)
+    double drand48()
+    double erand48(unsigned short *)
+    int    getsubopt(char **, char *const *, char **)
+    void   lcong48(unsigned short *)
+    long   lrand()
+    char  *mkdtemp(char *)
+    int    mkstemp(char *)
+    long   mrand()
+    long   nrand48(unsigned short *)
+    int    posix_memalign(void **, size_t, size_t)
+    int    posix_openpt(int)
+    char  *ptsname(int)
+    int    putenv(char *)
+    int    rand_r(unsigned *)
+    long   random()
+    char  *realpath(const char *, char *)
+    unsigned short *seed48(unsigned short *)
+    int    setenv(const char *, const char *, int)
+    void   setkey(const char *)
+    char  *setstate(char *)
+    void   srand48(long)
+    void   srandom(unsigned)
+    int    unlockpt(int)
+    int    unsetenv(const char *)

Cython/Includes/posix/strings.pxd

@@ -0,0 +1,9 @@
+cdef extern from "<strings.h>" nogil:
+    int    bcmp(const void *, const void *, size_t)
+    void   bcopy(const void *, void *, size_t)
+    void   bzero(void *, size_t)
+    int    ffs(int)
+    char   *index(const char *, int)
+    char   *rindex(const char *, int)
+    int    strcasecmp(const char *, const char *)
+    int    strncasecmp(const char *, const char *, size_t)

Cython/Includes/posix/time.pxd

@@ -0,0 +1,71 @@
+# https://pubs.opengroup.org/onlinepubs/009695399/basedefs/sys/time.h.html
+
+from posix.types cimport suseconds_t, time_t, clockid_t, timer_t
+from posix.signal cimport sigevent
+
+cdef extern from "<sys/time.h>" nogil:
+    enum: CLOCK_REALTIME
+    enum: TIMER_ABSTIME
+    enum: CLOCK_MONOTONIC
+
+    # FreeBSD-specific clocks
+    enum: CLOCK_UPTIME
+    enum: CLOCK_UPTIME_PRECISE
+    enum: CLOCK_UPTIME_FAST
+    enum: CLOCK_REALTIME_PRECISE
+    enum: CLOCK_REALTIME_FAST
+    enum: CLOCK_MONOTONIC_PRECISE
+    enum: CLOCK_MONOTONIC_FAST
+    enum: CLOCK_SECOND
+
+    # Linux-specific clocks
+    enum: CLOCK_PROCESS_CPUTIME_ID
+    enum: CLOCK_THREAD_CPUTIME_ID
+    enum: CLOCK_MONOTONIC_RAW
+    enum: CLOCK_REALTIME_COARSE
+    enum: CLOCK_MONOTONIC_COARSE
+    enum: CLOCK_BOOTTIME
+    enum: CLOCK_REALTIME_ALARM
+    enum: CLOCK_BOOTTIME_ALARM
+
+    enum: ITIMER_REAL
+    enum: ITIMER_VIRTUAL
+    enum: ITIMER_PROF
+
+    cdef struct timezone:
+        int tz_minuteswest
+        int dsttime
+
+    cdef struct timeval:
+        time_t      tv_sec
+        suseconds_t tv_usec
+
+    cdef struct timespec:
+        time_t tv_sec
+        long   tv_nsec
+
+    cdef struct itimerval:
+        timeval it_interval
+        timeval it_value
+
+    cdef struct itimerspec:
+        timespec it_interval
+        timespec it_value
+
+    int nanosleep(const timespec *, timespec *)
+
+    int getitimer(int, itimerval *)
+    int gettimeofday(timeval *tp, timezone *tzp)
+    int setitimer(int, const itimerval *, itimerval *)
+
+    int clock_getcpuclockid(pid_t, clockid_t *)
+    int clock_getres(clockid_t, timespec *)
+    int clock_gettime(clockid_t, timespec *)
+    int clock_nanosleep(clockid_t, int, const timespec *, timespec *)
+    int clock_settime(clockid_t, const timespec *)
+
+    int timer_create(clockid_t, sigevent *, timer_t *)
+    int timer_delete(timer_t)
+    int timer_gettime(timer_t, itimerspec *)
+    int timer_getoverrun(timer_t)
+    int timer_settime(timer_t, int, const itimerspec *, itimerspec *)

Cython/Includes/posix/types.pxd

@@ -0,0 +1,30 @@
+# Note that the actual size of these types is system-dependent, and
+# cannot be detected before C compile time.  However, the generated C code
+# will correctly use the actual size of these types *except* for
+# determining promotion in binary arithmetic expressions involving
+# mixed types.  In this case, operands are promoted to the declared
+# larger type, with a bias towards typedef types.  Thus, with the
+# declarations below, long + time_t will result in a time_t whereas
+# long long + time_t will result in a long long which should be
+# acceptable for either 32-bit or 64-bit signed time_t (though admittedly
+# the POSIX standard doesn't even specify that time_t must be an integral
+# type).
+
+cdef extern from "<sys/types.h>":
+    ctypedef long blkcnt_t
+    ctypedef long blksize_t
+    ctypedef long clockid_t
+    ctypedef long dev_t
+    ctypedef long gid_t
+    ctypedef long id_t
+    ctypedef unsigned long ino_t
+    ctypedef long mode_t
+    ctypedef long nlink_t
+    ctypedef long off_t
+    ctypedef long pid_t
+    ctypedef struct sigset_t:
+        pass
+    ctypedef long suseconds_t
+    ctypedef long time_t
+    ctypedef long timer_t
+    ctypedef long uid_t

Cython/Includes/posix/uio.pxd

@@ -0,0 +1,26 @@
+# https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/sys_uio.h.html
+
+from posix.types cimport off_t
+
+
+cdef extern from "<sys/uio.h>" nogil:
+
+    cdef struct iovec:
+        void  *iov_base
+        size_t iov_len
+
+    ssize_t readv (int fd, const iovec *iov, int iovcnt)
+    ssize_t writev(int fd, const iovec *iov, int iovcnt)
+
+    # Linux-specific, https://man7.org/linux/man-pages/man2/readv.2.html
+    ssize_t preadv (int fd, const iovec *iov, int iovcnt, off_t offset)
+    ssize_t pwritev(int fd, const iovec *iov, int iovcnt, off_t offset)
+
+    enum: RWF_DSYNC
+    enum: RWF_HIPRI
+    enum: RWF_SYNC
+    enum: RWF_NOWAIT
+    enum: RWF_APPEND
+
+    ssize_t preadv2 (int fd, const iovec *iov, int iovcnt, off_t offset, int flags)
+    ssize_t pwritev2(int fd, const iovec *iov, int iovcnt, off_t offset, int flags)

Cython/Includes/posix/unistd.pxd

@@ -0,0 +1,271 @@
+# http://www.opengroup.org/onlinepubs/009695399/basedefs/unistd.h.html
+
+from posix.types cimport gid_t, pid_t, off_t, uid_t
+
+cdef extern from "<unistd.h>" nogil:
+
+    #:NULL
+
+    enum: R_OK
+    enum: W_OK
+    enum: X_OK
+    enum: F_OK
+
+    enum: _CS_PATH
+    enum: _CS_POSIX_V6_ILP32_OFF32_CFLAGS
+    enum: _CS_POSIX_V6_ILP32_OFF32_LDFLAGS
+    enum: _CS_POSIX_V6_ILP32_OFF32_LIBS
+    enum: _CS_POSIX_V6_ILP32_OFFBIG_CFLAGS
+    enum: _CS_POSIX_V6_ILP32_OFFBIG_LDFLAGS
+    enum: _CS_POSIX_V6_ILP32_OFFBIG_LIBS
+    enum: _CS_POSIX_V6_LP64_OFF64_CFLAGS
+    enum: _CS_POSIX_V6_LP64_OFF64_LDFLAGS
+    enum: _CS_POSIX_V6_LP64_OFF64_LIBS
+    enum: _CS_POSIX_V6_LPBIG_OFFBIG_CFLAGS
+    enum: _CS_POSIX_V6_LPBIG_OFFBIG_LDFLAGS
+    enum: _CS_POSIX_V6_LPBIG_OFFBIG_LIBS
+    enum: _CS_POSIX_V6_WIDTH_RESTRICTED_ENVS
+
+    enum: SEEK_SET
+    enum: SEEK_CUR
+    enum: SEEK_END
+
+    enum: F_LOCK
+    enum: F_TEST
+    enum: F_TLOCK
+    enum: F_ULOCK
+
+    enum: _PC_2_SYMLINKS
+    enum: _PC_ALLOC_SIZE_MIN
+    enum: _PC_ASYNC_IO
+    enum: _PC_CHOWN_RESTRICTED
+    enum: _PC_FILESIZEBITS
+    enum: _PC_LINK_MAX
+    enum: _PC_MAX_CANON
+    enum: _PC_MAX_INPUT
+    enum: _PC_NAME_MAX
+    enum: _PC_NO_TRUNC
+    enum: _PC_PATH_MAX
+    enum: _PC_PIPE_BUF
+    enum: _PC_PRIO_IO
+    enum: _PC_REC_INCR_XFER_SIZE
+    enum: _PC_REC_MIN_XFER_SIZE
+    enum: _PC_REC_XFER_ALIGN
+    enum: _PC_SYMLINK_MAX
+    enum: _PC_SYNC_IO
+    enum: _PC_VDISABLE
+
+    enum: _SC_2_C_BIND
+    enum: _SC_2_C_DEV
+    enum: _SC_2_CHAR_TERM
+    enum: _SC_2_FORT_DEV
+    enum: _SC_2_FORT_RUN
+    enum: _SC_2_LOCALEDEF
+    enum: _SC_2_PBS
+    enum: _SC_2_PBS_ACCOUNTING
+    enum: _SC_2_PBS_CHECKPOINT
+    enum: _SC_2_PBS_LOCATE
+    enum: _SC_2_PBS_MESSAGE
+    enum: _SC_2_PBS_TRACK
+    enum: _SC_2_SW_DEV
+    enum: _SC_2_UPE
+    enum: _SC_2_VERSION
+    enum: _SC_ADVISORY_INFO
+    enum: _SC_AIO_LISTIO_MAX
+    enum: _SC_AIO_MAX
+    enum: _SC_AIO_PRIO_DELTA_MAX
+    enum: _SC_ARG_MAX
+    enum: _SC_ASYNCHRONOUS_IO
+    enum: _SC_ATEXIT_MAX
+    enum: _SC_BARRIERS
+    enum: _SC_BC_BASE_MAX
+    enum: _SC_BC_DIM_MAX
+    enum: _SC_BC_SCALE_MAX
+    enum: _SC_BC_STRING_MAX
+    enum: _SC_CHILD_MAX
+    enum: _SC_CLK_TCK
+    enum: _SC_CLOCK_SELECTION
+    enum: _SC_COLL_WEIGHTS_MAX
+    enum: _SC_CPUTIME
+    enum: _SC_DELAYTIMER_MAX
+    enum: _SC_EXPR_NEST_MAX
+    enum: _SC_FSYNC
+    enum: _SC_GETGR_R_SIZE_MAX
+    enum: _SC_GETPW_R_SIZE_MAX
+    enum: _SC_HOST_NAME_MAX
+    enum: _SC_IOV_MAX
+    enum: _SC_IPV6
+    enum: _SC_JOB_CONTROL
+    enum: _SC_LINE_MAX
+    enum: _SC_LOGIN_NAME_MAX
+    enum: _SC_MAPPED_FILES
+    enum: _SC_MEMLOCK
+    enum: _SC_MEMLOCK_RANGE
+    enum: _SC_MEMORY_PROTECTION
+    enum: _SC_MESSAGE_PASSING
+    enum: _SC_MONOTONIC_CLOCK
+    enum: _SC_MQ_OPEN_MAX
+    enum: _SC_MQ_PRIO_MAX
+    enum: _SC_NGROUPS_MAX
+    enum: _SC_OPEN_MAX
+    enum: _SC_PAGE_SIZE
+    enum: _SC_PAGESIZE
+    enum: _SC_PRIORITIZED_IO
+    enum: _SC_PRIORITY_SCHEDULING
+    enum: _SC_RAW_SOCKETS
+    enum: _SC_RE_DUP_MAX
+    enum: _SC_READER_WRITER_LOCKS
+    enum: _SC_REALTIME_SIGNALS
+    enum: _SC_REGEXP
+    enum: _SC_RTSIG_MAX
+    enum: _SC_SAVED_IDS
+    enum: _SC_SEM_NSEMS_MAX
+    enum: _SC_SEM_VALUE_MAX
+    enum: _SC_SEMAPHORES
+    enum: _SC_SHARED_MEMORY_OBJECTS
+    enum: _SC_SHELL
+    enum: _SC_SIGQUEUE_MAX
+    enum: _SC_SPAWN
+    enum: _SC_SPIN_LOCKS
+    enum: _SC_SPORADIC_SERVER
+    enum: _SC_SS_REPL_MAX
+    enum: _SC_STREAM_MAX
+    enum: _SC_SYMLOOP_MAX
+    enum: _SC_SYNCHRONIZED_IO
+    enum: _SC_THREAD_ATTR_STACKADDR
+    enum: _SC_THREAD_ATTR_STACKSIZE
+    enum: _SC_THREAD_CPUTIME
+    enum: _SC_THREAD_DESTRUCTOR_ITERATIONS
+    enum: _SC_THREAD_KEYS_MAX
+    enum: _SC_THREAD_PRIO_INHERIT
+    enum: _SC_THREAD_PRIO_PROTECT
+    enum: _SC_THREAD_PRIORITY_SCHEDULING
+    enum: _SC_THREAD_PROCESS_SHARED
+    enum: _SC_THREAD_SAFE_FUNCTIONS
+    enum: _SC_THREAD_SPORADIC_SERVER
+    enum: _SC_THREAD_STACK_MIN
+    enum: _SC_THREAD_THREADS_MAX
+    enum: _SC_THREADS
+    enum: _SC_TIMEOUTS
+    enum: _SC_TIMER_MAX
+    enum: _SC_TIMERS
+    enum: _SC_TRACE
+    enum: _SC_TRACE_EVENT_FILTER
+    enum: _SC_TRACE_EVENT_NAME_MAX
+    enum: _SC_TRACE_INHERIT
+    enum: _SC_TRACE_LOG
+    enum: _SC_TRACE_NAME_MAX
+    enum: _SC_TRACE_SYS_MAX
+    enum: _SC_TRACE_USER_EVENT_MAX
+    enum: _SC_TTY_NAME_MAX
+    enum: _SC_TYPED_MEMORY_OBJECTS
+    enum: _SC_TZNAME_MAX
+    enum: _SC_V6_ILP32_OFF32
+    enum: _SC_V6_ILP32_OFFBIG
+    enum: _SC_V6_LP64_OFF64
+    enum: _SC_V6_LPBIG_OFFBIG
+    enum: _SC_VERSION
+    enum: _SC_XBS5_ILP32_OFF32
+    enum: _SC_XBS5_ILP32_OFFBIG
+    enum: _SC_XBS5_LP64_OFF64
+    enum: _SC_XBS5_LPBIG_OFFBIG
+    enum: _SC_XOPEN_CRYPT
+    enum: _SC_XOPEN_ENH_I18N
+    enum: _SC_XOPEN_LEGACY
+    enum: _SC_XOPEN_REALTIME
+    enum: _SC_XOPEN_REALTIME_THREADS
+    enum: _SC_XOPEN_SHM
+    enum: _SC_XOPEN_STREAMS
+    enum: _SC_XOPEN_UNIX
+    enum: _SC_XOPEN_VERSION
+
+    enum: STDIN_FILENO	#0
+    enum: STDOUT_FILENO	#1
+    enum: STDERR_FILENO	#2
+
+    ctypedef unsigned useconds_t
+
+    int          access(const char *, int)
+    unsigned     alarm(unsigned)
+    int          chdir(const char *)
+    int          chown(const char *, uid_t, gid_t)
+    int          close(int)
+    size_t       confstr(int, char *, size_t)
+    char        *crypt(const char *, const char *)
+    char        *ctermid(char *)
+    int          dup(int)
+    int          dup2(int, int)
+    void         encrypt(char[64], int)
+    int          execl(const char *, const char *, ...)
+    int          execle(const char *, const char *, ...)
+    int          execlp(const char *, const char *, ...)
+    int          execv(const char *, char *[])
+    int          execve(const char *, char *[], char *[])
+    int          execvp(const char *, char *[])
+    void        _exit(int)
+    int          fchown(int, uid_t, gid_t)
+    int          fchdir(int)
+    int          fdatasync(int)
+    pid_t        fork()
+    long         fpathconf(int, int)
+    int          fsync(int)
+    int          ftruncate(int, off_t)
+    char        *getcwd(char *, size_t)
+    gid_t        getegid()
+    uid_t        geteuid()
+    gid_t        getgid()
+    int          getgroups(int, gid_t [])
+    long         gethostid()
+    int          gethostname(char *, size_t)
+    char        *getlogin()
+    int          getlogin_r(char *, size_t)
+    int          getopt(int, char * [], const char *)
+    pid_t        getpgid(pid_t)
+    pid_t        getpgrp()
+    pid_t        getpid()
+    pid_t        getppid()
+    pid_t        getsid(pid_t)
+    uid_t        getuid()
+    char        *getwd(char *)
+    int          isatty(int)
+    int          lchown(const char *, uid_t, gid_t)
+    int          link(const char *, const char *)
+    int          lockf(int, int, off_t)
+    off_t        lseek(int, off_t, int)
+    int          nice(int)
+    long         pathconf(char *, int)
+    int          pause()
+    int          pipe(int [2])
+    ssize_t      pread(int, void *, size_t, off_t)
+    ssize_t      pwrite(int, const void *, size_t, off_t)
+    ssize_t      read(int, void *, size_t)
+    ssize_t      readlink(const char *, char *, size_t)
+    int          rmdir(const char *)
+    int          setegid(gid_t)
+    int          seteuid(uid_t)
+    int          setgid(gid_t)
+    int          setpgid(pid_t, pid_t)
+    pid_t        setpgrp()
+    int          setregid(gid_t, gid_t)
+    int          setreuid(uid_t, uid_t)
+    pid_t        setsid()
+    int          setuid(uid_t)
+    unsigned     sleep(unsigned)
+    void         swab(const void *, void *, ssize_t)
+    int          symlink(const char *, const char *)
+    void         sync()
+    long         sysconf(int)
+    pid_t        tcgetpgrp(int)
+    int          tcsetpgrp(int, pid_t)
+    int          truncate(const char *, off_t)
+    char        *ttyname(int)
+    int          ttyname_r(int, char *, size_t)
+    useconds_t   ualarm(useconds_t, useconds_t)
+    int          unlink(const char *)
+    int          usleep(useconds_t)
+    pid_t        vfork()
+    ssize_t      write(int, const void *, size_t)
+    char         *optarg
+    int          optind
+    int          opterr
+    int          optopt

Cython/Includes/posix/wait.pxd

@@ -0,0 +1,38 @@
+# https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/sys_wait.h.html
+
+from posix.types cimport pid_t, id_t
+from posix.signal cimport siginfo_t
+from posix.resource cimport rusage
+
+cdef extern from "<sys/wait.h>" nogil:
+    enum: WNOHANG
+    enum: WUNTRACED
+    enum: WCONTINUED
+    enum: WEXITED
+    enum: WSTOPPED
+    enum: WNOWAIT
+
+    int WEXITSTATUS(int status)
+    int WIFCONTINUED(int status)
+    int WIFEXITED(int status)
+    int WIFSIGNALED(int status)
+    int WIFSTOPPED(int status)
+    int WSTOPSIG(int status)
+    int WTERMSIG(int status)
+
+    ctypedef int idtype_t
+    enum: P_ALL             # idtype_t values
+    enum: P_PID
+    enum: P_PGID
+
+    pid_t wait(int *stat_loc)
+    pid_t waitpid(pid_t pid, int *status, int options)
+    int waitid(idtype_t idtype, id_t id, siginfo_t *infop, int options)
+
+# wait3 was in POSIX until 2008 while wait4 was never standardized.
+# Even so, these calls are in almost every Unix, always in sys/wait.h.
+# Hence, posix.wait is the least surprising place to declare them for Cython.
+# libc may require _XXX_SOURCE to be defined at C-compile time to provide them.
+
+    pid_t wait3(int *status, int options, rusage *rusage)
+    pid_t wait4(pid_t pid, int *status, int options, rusage *rusage)

Cython/LZSS.py

@@ -0,0 +1,170 @@
+"""
+LZSS-like compressor.
+
+See decompression function in StringTools.c, "DecompressString_LZSS".
+"""
+
+from __future__ import annotations
+
+from collections import defaultdict
+
+import cython
+
+PRINT_STATS = False
+
+
+def lzss_compress(data: bytes) -> bytes:
+    """
+    LZSS-like compressor.
+
+    Uses a byte triplet dict for fast match finding.
+    Outputs a bit stream with small overhead.
+
+    Args:
+        data: Input bytes to compress
+
+    Returns:
+        Compressed bytes (no header/metadata, decompressor must know output size)
+    """
+    if not data:
+        return b''
+
+    input_size = len(data)
+
+    # Hash table: maps 3-byte sequences to a list of positions starting with that sequence.
+    hash_table = defaultdict(list)
+
+    def find_longest_match(pos: cython.Py_ssize_t) -> tuple[cython.Py_ssize_t, cython.Py_ssize_t]:
+        """Find longest match in sliding window. Returns (offset, length)."""
+        if pos + 3 > input_size:
+            return (0, 0)
+
+        # 14-bit offset (max 16KiB lookback)
+        WINDOW_SIZE: cython.long = 1 << 14
+        # 8-bit length + 3, allowing up to 258 bytes.
+        MAX_MATCH: cython.long = min(255 + 3, input_size - pos)
+
+        best_len: cython.Py_ssize_t = 0
+        best_offset: cython.Py_ssize_t = 0
+        window_start: cython.Py_ssize_t = max(0, pos - WINDOW_SIZE - MAX_MATCH)
+
+        prev_pos: cython.Py_ssize_t
+        match_len: cython.long
+        max_len: cython.long
+
+        # Check all positions starting with the same 3-byte sequence.
+        key = data[pos:pos+3]
+        if key not in hash_table:
+            return (0, 0)
+
+        for prev_pos in hash_table[key]:
+            if prev_pos < window_start or prev_pos >= pos:
+                continue
+
+            # Extend match as far as possible.
+            match_len = 3
+            max_len = min(MAX_MATCH, pos - prev_pos)
+            while match_len < max_len and data[prev_pos + match_len] == data[pos + match_len]:
+                match_len += 1
+
+            # Prefer longer matches, break ties by preferring closer offsets.
+            if match_len > best_len:
+                best_len = match_len
+                best_offset = pos - prev_pos
+
+        # Lazy matching: check if next position has a better match.
+        next_key = data[pos + 1:pos + 4]
+        if 0 < best_len < MAX_MATCH and next_key in hash_table and pos + best_len + 1 < input_size:
+            next_best_len: cython.Py_ssize_t = 0
+            window_start = max(0, pos + 1 - WINDOW_SIZE - MAX_MATCH)
+
+            for prev_pos in hash_table[next_key]:
+                if prev_pos < window_start:
+                    continue
+
+                match_len = 3
+                max_len = min(MAX_MATCH, pos - prev_pos, input_size - pos - 1)
+                while match_len < max_len and data[prev_pos + match_len] == data[pos + 1 + match_len]:
+                    match_len += 1
+
+                if match_len > next_best_len:
+                    next_best_len = match_len
+
+            # If the next position has a significantly better match, ignore this one.
+            if next_best_len > best_len + 1:
+                best_offset = best_len = 0
+
+        return (best_offset, best_len)
+
+    # Build output as bit stream.
+    output = bytearray(b'\0')
+    pos: cython.Py_ssize_t = 0
+    flags_pos: cython.Py_ssize_t = 0
+    flags: cython.int = 0xFF0000
+
+    stats = [0, 0, 0, 0]
+
+    while pos < input_size:
+        offset, length = find_longest_match(pos)
+
+        #if offset > 0:
+        #    assert data[pos:pos+length] == data[pos-offset:pos-offset+length], (
+        #        data[pos:pos+3], data[pos-offset:pos-offset+length])
+
+        # Update hash table for current position.
+        key = data[pos:pos+3]
+        hash_table[key].append(pos)
+
+        flag = 0
+        offset -= length  # offset >= length if interesting, so remove redundancy.
+
+        # See if the match is worth it and store a single literal byte otherwise.
+
+        if length > 2 and 0 <= offset <= 0x7F:
+            # Store 7 bit offset and length as two bytes.
+            output.append(offset)
+            #assert output[-1] & 0x80 == 0
+            output.append(length - 3)
+            stats[1] += 1
+        elif length > 2 and length - 3 <= 0x1F and 0 <= offset <= 0x1FF:
+            # Store a longer 7+2 bit offset at the cost of a shorter 5 bit length.
+            output.append((offset & 0x7F) | 0x80)
+            output.append(((offset & 0x180) >> 2) | (length - 3))
+            #assert output[-1] & 0x80 == 0
+            stats[2] += 1
+        elif length > 3 and 0 <= offset < (1 << 14):
+            # Store a 7+7 bit offset with a separate 8 bit length.
+            output.append(offset & 0x7F | 0x80)
+            output.append((offset >> 7) & 0x7F | 0x80)
+            output.append(length - 3)
+            stats[3] += 1
+        else:
+            # Encode 8 bit literal + 1 bit flag.
+            flag = 1
+            length = 1
+            output.append(data[pos])
+            stats[0] += 1
+
+        pos += length
+
+        flags = (flag << 7) | (flags >> 1)
+        if flags < 0x10000:
+            output[flags_pos] = flags & 0xFF
+            flags_pos = len(output)
+            output.append(0)  # next flags
+            flags = 0xFF0000
+
+    if flags_pos == len(output) - 1:
+        output.pop()
+    else:
+        # Pad flags with empty non-literals
+        # - decompressor must stop at the end of the output buffer !
+        while flags >= 0x10000:
+            flags >>= 1
+        output[flags_pos] = flags & 0xFF
+
+    if PRINT_STATS:
+        print("BYTES:", input_size, "->", len(output))
+        print("ENCODINGS:", stats)
+
+    return bytes(output)

Cython/Plex/Actions.pxd

@@ -0,0 +1,24 @@
+cdef class Action:
+    cdef perform(self, token_stream, text)
+
+cdef class Return(Action):
+    cdef object value
+    cdef perform(self, token_stream, text)
+
+cdef class Call(Action):
+    cdef object function
+    cdef perform(self, token_stream, text)
+
+cdef class Method(Action):
+    cdef str name
+    cdef dict kwargs
+
+cdef class Begin(Action):
+    cdef object state_name
+    cdef perform(self, token_stream, text)
+
+cdef class Ignore(Action):
+    cdef perform(self, token_stream, text)
+
+cdef class Text(Action):
+    cdef perform(self, token_stream, text)