Use a new module compileall2 for Python byte-compilation

compileall2.py

@@ -0,0 +1,464 @@
+"""Module/script to byte-compile all .py files to .pyc files.
+
+When called as a script with arguments, this compiles the directories
+given as arguments recursively; the -l option prevents it from
+recursing into directories.
+
+Without arguments, if compiles all modules on sys.path, without
+recursing into subdirectories.  (Even though it should do so for
+packages -- for now, you'll have to deal with packages separately.)
+
+See module py_compile for details of the actual byte-compilation.
+"""
+import os
+import sys
+import importlib.util
+import py_compile
+import struct
+
+from functools import partial
+from pathlib import Path
+
+# Python 3.7 and higher
+PY37 = sys.version_info[0:2] >= (3, 7)
+# Python 3.6 and higher
+PY36 = sys.version_info[0:2] >= (3, 6)
+# Python 3.5 and higher
+PY35 = sys.version_info[0:2] >= (3, 5)
+
+# Python 3.7 and above has a different structure and length
+# of pyc files header. Also, multiple ways how to invalidate pyc file was
+# introduced in Python 3.7. These cases are covered by variables here or by PY37
+# variable itself.
+if PY37:
+    pyc_struct_format = '<4sll'
+    pyc_header_lenght = 12
+    pyc_header_format = (pyc_struct_format, importlib.util.MAGIC_NUMBER, 0)
+else:
+    pyc_struct_format = '<4sl'
+    pyc_header_lenght = 8
+    pyc_header_format = (pyc_struct_format, importlib.util.MAGIC_NUMBER)
+
+RECURSION_LIMIT = sys.getrecursionlimit()
+
+__all__ = ["compile_dir","compile_file","compile_path"]
+
+def optimization_kwarg(opt):
+    """Returns opt as a dictionary {optimization: opt} for use as **kwarg
+    for Python >= 3.5 and empty dictionary for Python 3.4"""
+    if PY35:
+        return dict(optimization=opt)
+    else:
+        # `debug_override` is a way how to enable optimized byte-compiled files
+        # (.pyo) in Python <= 3.4
+        if opt:
+            return dict(debug_override=False)
+        else:
+            return dict()
+
+def _walk_dir(dir, maxlevels=RECURSION_LIMIT, quiet=0):
+    if PY36 and quiet < 2 and isinstance(dir, os.PathLike):
+        dir = os.fspath(dir)
+    else:
+        dir = str(dir)
+    if not quiet:
+        print('Listing {!r}...'.format(dir))
+    try:
+        names = os.listdir(dir)
+    except OSError:
+        if quiet < 2:
+            print("Can't list {!r}".format(dir))
+        names = []
+    names.sort()
+    for name in names:
+        if name == '__pycache__':
+            continue
+        fullname = os.path.join(dir, name)
+        if not os.path.isdir(fullname):
+            yield fullname
+        elif (maxlevels > 0 and name != os.curdir and name != os.pardir and
+              os.path.isdir(fullname) and not os.path.islink(fullname)):
+            yield from _walk_dir(fullname, maxlevels=maxlevels - 1,
+                                 quiet=quiet)
+
+def compile_dir(dir, maxlevels=RECURSION_LIMIT, ddir=None, force=False,
+                rx=None, quiet=0, legacy=False, optimize=-1, workers=1,
+                invalidation_mode=None, stripdir=None,
+                prependdir=None, limit_sl_dest=None):
+    """Byte-compile all modules in the given directory tree.
+
+    Arguments (only dir is required):
+
+    dir:       the directory to byte-compile
+    maxlevels: maximum recursion level (default `sys.getrecursionlimit()`)
+    ddir:      the directory that will be prepended to the path to the
+               file as it is compiled into each byte-code file.
+    force:     if True, force compilation, even if timestamps are up-to-date
+    quiet:     full output with False or 0, errors only with 1,
+               no output with 2
+    legacy:    if True, produce legacy pyc paths instead of PEP 3147 paths
+    optimize:  int or list of optimization levels or -1 for level of
+               the interpreter. Multiple levels leads to multiple compiled
+               files each with one optimization level.
+    workers:   maximum number of parallel workers
+    invalidation_mode: how the up-to-dateness of the pyc will be checked
+    stripdir:  part of path to left-strip from source file path
+    prependdir: path to prepend to beggining of original file path, applied
+               after stripdir
+    limit_sl_dest: ignore symlinks if they are pointing outside of
+                   the defined path
+    """
+    ProcessPoolExecutor = None
+    if workers is not None:
+        if workers < 0:
+            raise ValueError('workers must be greater or equal to 0')
+        elif workers != 1:
+            try:
+                # Only import when needed, as low resource platforms may
+                # fail to import it
+                from concurrent.futures import ProcessPoolExecutor
+            except ImportError:
+                workers = 1
+    files = _walk_dir(dir, quiet=quiet, maxlevels=maxlevels)
+    success = True
+    if workers is not None and workers != 1 and ProcessPoolExecutor is not None:
+        workers = workers or None
+        with ProcessPoolExecutor(max_workers=workers) as executor:
+            results = executor.map(partial(compile_file,
+                                           ddir=ddir, force=force,
+                                           rx=rx, quiet=quiet,
+                                           legacy=legacy,
+                                           optimize=optimize,
+                                           invalidation_mode=invalidation_mode,
+                                           stripdir=stripdir,
+                                           prependdir=prependdir,
+                                           limit_sl_dest=limit_sl_dest),
+                                   files)
+            success = min(results, default=True)
+    else:
+        for file in files:
+            if not compile_file(file, ddir, force, rx, quiet,
+                                legacy, optimize, invalidation_mode,
+                                stripdir=stripdir, prependdir=prependdir,
+                                limit_sl_dest=limit_sl_dest):
+                success = False
+    return success
+
+def compile_file(fullname, ddir=None, force=False, rx=None, quiet=0,
+                 legacy=False, optimize=-1,
+                 invalidation_mode=None, stripdir=None, prependdir=None,
+                 limit_sl_dest=None):
+    """Byte-compile one file.
+
+    Arguments (only fullname is required):
+
+    fullname:  the file to byte-compile
+    ddir:      if given, the directory name compiled in to the
+               byte-code file.
+    force:     if True, force compilation, even if timestamps are up-to-date
+    quiet:     full output with False or 0, errors only with 1,
+               no output with 2
+    legacy:    if True, produce legacy pyc paths instead of PEP 3147 paths
+    optimize:  int or list of optimization levels or -1 for level of
+               the interpreter. Multiple levels leads to multiple compiled
+               files each with one optimization level.
+    invalidation_mode: how the up-to-dateness of the pyc will be checked
+    stripdir:  part of path to left-strip from source file path
+    prependdir: path to prepend to beggining of original file path, applied
+               after stripdir
+    limit_sl_dest: ignore symlinks if they are pointing outside of
+                   the defined path.
+    """
+    success = True
+    if PY36 and quiet < 2 and isinstance(fullname, os.PathLike):
+        fullname = os.fspath(fullname)
+    else:
+        fullname = str(fullname)
+    name = os.path.basename(fullname)
+
+    dfile = None
+
+    if ddir is not None:
+        if not PY36:
+            ddir = str(ddir)
+        dfile = os.path.join(ddir, name)
+
+    if stripdir is not None:
+        fullname_parts = fullname.split(os.path.sep)
+        stripdir_parts = stripdir.split(os.path.sep)
+        ddir_parts = list(fullname_parts)
+
+        for spart, opart in zip(stripdir_parts, fullname_parts):
+            if spart == opart:
+                ddir_parts.remove(spart)
+
+        dfile = os.path.join(*ddir_parts)
+
+    if prependdir is not None:
+        if dfile is None:
+            dfile = os.path.join(prependdir, fullname)
+        else:
+            dfile = os.path.join(prependdir, dfile)
+
+    if isinstance(optimize, int):
+        optimize = [optimize]
+
+    if rx is not None:
+        mo = rx.search(fullname)
+        if mo:
+            return success
+
+    if limit_sl_dest is not None and os.path.islink(fullname):
+        if Path(limit_sl_dest).resolve() not in Path(fullname).resolve().parents:
+            return success
+
+    opt_cfiles = {}
+
+    if os.path.isfile(fullname):
+        for opt_level in optimize:
+            if legacy:
+                opt_cfiles[opt_level] = fullname + 'c'
+            else:
+                if opt_level >= 0:
+                    opt = opt_level if opt_level >= 1 else ''
+                    opt_kwarg = optimization_kwarg(opt)
+                    cfile = (importlib.util.cache_from_source(
+                             fullname, **opt_kwarg))
+                    opt_cfiles[opt_level] = cfile
+                else:
+                    cfile = importlib.util.cache_from_source(fullname)
+                    opt_cfiles[opt_level] = cfile
+
+        head, tail = name[:-3], name[-3:]
+        if tail == '.py':
+            if not force:
+                try:
+                    mtime = int(os.stat(fullname).st_mtime)
+                    expect = struct.pack(*(pyc_header_format + (mtime,)))
+                    for cfile in opt_cfiles.values():
+                        with open(cfile, 'rb') as chandle:
+                            actual = chandle.read(pyc_header_lenght)
+                        if expect != actual:
+                            break
+                    else:
+                        return success
+                except OSError:
+                    pass
+            if not quiet:
+                print('Compiling {!r}...'.format(fullname))
+            try:
+                for opt_level, cfile in opt_cfiles.items():
+                    if PY37:
+                        ok = py_compile.compile(fullname, cfile, dfile, True,
+                                                optimize=opt_level,
+                                                invalidation_mode=invalidation_mode)
+                    else:
+                        ok = py_compile.compile(fullname, cfile, dfile, True,
+                                                optimize=opt_level)
+            except py_compile.PyCompileError as err:
+                success = False
+                if quiet >= 2:
+                    return success
+                elif quiet:
+                    print('*** Error compiling {!r}...'.format(fullname))
+                else:
+                    print('*** ', end='')
+                # escape non-printable characters in msg
+                msg = err.msg.encode(sys.stdout.encoding,
+                                     errors='backslashreplace')
+                msg = msg.decode(sys.stdout.encoding)
+                print(msg)
+            except (SyntaxError, UnicodeError, OSError) as e:
+                success = False
+                if quiet >= 2:
+                    return success
+                elif quiet:
+                    print('*** Error compiling {!r}...'.format(fullname))
+                else:
+                    print('*** ', end='')
+                print(e.__class__.__name__ + ':', e)
+            else:
+                if ok == 0:
+                    success = False
+    return success
+
+def compile_path(skip_curdir=1, maxlevels=0, force=False, quiet=0,
+                 legacy=False, optimize=-1,
+                 invalidation_mode=None):
+    """Byte-compile all module on sys.path.
+
+    Arguments (all optional):
+
+    skip_curdir: if true, skip current directory (default True)
+    maxlevels:   max recursion level (default 0)
+    force: as for compile_dir() (default False)
+    quiet: as for compile_dir() (default 0)
+    legacy: as for compile_dir() (default False)
+    optimize: as for compile_dir() (default -1)
+    invalidation_mode: as for compiler_dir()
+    """
+    success = True
+    for dir in sys.path:
+        if (not dir or dir == os.curdir) and skip_curdir:
+            if quiet < 2:
+                print('Skipping current directory')
+        else:
+            success = success and compile_dir(
+                dir,
+                maxlevels,
+                None,
+                force,
+                quiet=quiet,
+                legacy=legacy,
+                optimize=optimize,
+                invalidation_mode=invalidation_mode,
+            )
+    return success
+
+
+def main():
+    """Script main program."""
+    import argparse
+
+    parser = argparse.ArgumentParser(
+        description='Utilities to support installing Python libraries.')
+    parser.add_argument('-l', action='store_const', const=0,
+                        default=RECURSION_LIMIT, dest='maxlevels',
+                        help="don't recurse into subdirectories")
+    parser.add_argument('-r', type=int, dest='recursion',
+                        help=('control the maximum recursion level. '
+                              'if `-l` and `-r` options are specified, '
+                              'then `-r` takes precedence.'))
+    parser.add_argument('-f', action='store_true', dest='force',
+                        help='force rebuild even if timestamps are up to date')
+    parser.add_argument('-q', action='count', dest='quiet', default=0,
+                        help='output only error messages; -qq will suppress '
+                             'the error messages as well.')
+    parser.add_argument('-b', action='store_true', dest='legacy',
+                        help='use legacy (pre-PEP3147) compiled file locations')
+    parser.add_argument('-d', metavar='DESTDIR',  dest='ddir', default=None,
+                        help=('directory to prepend to file paths for use in '
+                              'compile-time tracebacks and in runtime '
+                              'tracebacks in cases where the source file is '
+                              'unavailable'))
+    parser.add_argument('-s', metavar='STRIPDIR',  dest='stripdir',
+                        default=None,
+                        help=('part of path to left-strip from path '
+                              'to source file - for example buildroot. '
+                              'if `-d` and `-s` options are specified, '
+                              'then `-d` takes precedence.'))
+    parser.add_argument('-p', metavar='PREPENDDIR',  dest='prependdir',
+                        default=None,
+                        help=('path to add as prefix to path '
+                              'to source file - for example / to make '
+                              'it absolute when some part is removed '
+                              'by `-s` option'
+                              'if `-d` and `-a` options are specified, '
+                              'then `-d` takes precedence.'))
+    parser.add_argument('-x', metavar='REGEXP', dest='rx', default=None,
+                        help=('skip files matching the regular expression; '
+                              'the regexp is searched for in the full path '
+                              'of each file considered for compilation'))
+    parser.add_argument('-i', metavar='FILE', dest='flist',
+                        help=('add all the files and directories listed in '
+                              'FILE to the list considered for compilation; '
+                              'if "-", names are read from stdin'))
+    parser.add_argument('compile_dest', metavar='FILE|DIR', nargs='*',
+                        help=('zero or more file and directory names '
+                              'to compile; if no arguments given, defaults '
+                              'to the equivalent of -l sys.path'))
+    parser.add_argument('-j', '--workers', default=1,
+                        type=int, help='Run compileall concurrently')
+    parser.add_argument('-o', action='append', type=int, dest='opt_levels',
+                        help=('Optimization levels to run compilation with.'
+                              'Default is -1 which uses optimization level of'
+                              'Python interpreter itself (specified by -O).'))
+    parser.add_argument('-e', metavar='DIR', dest='limit_sl_dest',
+                        help='Ignore symlinks pointing outsite of the DIR')
+
+    if PY37:
+        invalidation_modes = [mode.name.lower().replace('_', '-')
+                              for mode in py_compile.PycInvalidationMode]
+        parser.add_argument('--invalidation-mode',
+                            choices=sorted(invalidation_modes),
+                            help=('set .pyc invalidation mode; defaults to '
+                                  '"checked-hash" if the SOURCE_DATE_EPOCH '
+                                  'environment variable is set, and '
+                                  '"timestamp" otherwise.'))
+
+    args = parser.parse_args()
+    compile_dests = args.compile_dest
+
+    if args.rx:
+        import re
+        args.rx = re.compile(args.rx)
+
+    if args.limit_sl_dest == "":
+        args.limit_sl_dest = None
+
+    if args.recursion is not None:
+        maxlevels = args.recursion
+    else:
+        maxlevels = args.maxlevels
+
+    if args.opt_levels is None:
+        args.opt_levels = [-1]
+
+    # if flist is provided then load it
+    if args.flist:
+        try:
+            with (sys.stdin if args.flist=='-' else open(args.flist)) as f:
+                for line in f:
+                    compile_dests.append(line.strip())
+        except OSError:
+            if args.quiet < 2:
+                print("Error reading file list {}".format(args.flist))
+            return False
+
+    if args.workers is not None:
+        args.workers = args.workers or None
+
+    if PY37 and args.invalidation_mode:
+        ivl_mode = args.invalidation_mode.replace('-', '_').upper()
+        invalidation_mode = py_compile.PycInvalidationMode[ivl_mode]
+    else:
+        invalidation_mode = None
+
+    success = True
+    try:
+        if compile_dests:
+            for dest in compile_dests:
+                if os.path.isfile(dest):
+                    if not compile_file(dest, args.ddir, args.force, args.rx,
+                                        args.quiet, args.legacy,
+                                        invalidation_mode=invalidation_mode,
+                                        stripdir=args.stripdir,
+                                        prependdir=args.prependdir,
+                                        optimize=args.opt_levels,
+                                        limit_sl_dest=args.limit_sl_dest):
+                        success = False
+                else:
+                    if not compile_dir(dest, maxlevels, args.ddir,
+                                       args.force, args.rx, args.quiet,
+                                       args.legacy, workers=args.workers,
+                                       invalidation_mode=invalidation_mode,
+                                       stripdir=args.stripdir,
+                                       prependdir=args.prependdir,
+                                       optimize=args.opt_levels,
+                                       limit_sl_dest=args.limit_sl_dest):
+                        success = False
+            return success
+        else:
+            return compile_path(legacy=args.legacy, force=args.force,
+                                quiet=args.quiet,
+                                invalidation_mode=invalidation_mode)
+    except KeyboardInterrupt:
+        if args.quiet < 2:
+            print("\n[interrupted]")
+        return False
+    return True
+
+
+if __name__ == '__main__':
+    exit_status = int(not main())
+    sys.exit(exit_status)

macros.pybytecompile

@@ -1,7 +1,7 @@
 # Note that the path could itself be a python file, or a directory
 
-# Python's compile_all module only works on directories, and requires a max
-# recursion depth
+# Note that the py_byte_compile macro should work for all Python versions
+# Which unfortunately makes the definition more complicated than it should be
 
 # Usage:
 #    %py_byte_compile <interpereter> <path>
@@ -13,12 +13,22 @@
 #    (%{py_byte_compile <interpereter> <path>}) || :
 
 %py_byte_compile()\
-python_binary="%1"\
-buildroot_path="%2"\
-bytecode_compilation_path=".${buildroot_path/#$RPM_BUILD_ROOT}"\
-failure=0\
-pushd $RPM_BUILD_ROOT\
-find $bytecode_compilation_path -type f -a -name "*.py" -print0 | xargs -0 $python_binary -O -m py_compile || failure=1\
-find $bytecode_compilation_path -type f -a -name "*.py" -print0 | xargs -0 $python_binary -m py_compile || failure=1\
-popd\
-test $failure -eq 0
+py2_byte_compile () {\
+    python_binary="%1"\
+    bytecode_compilation_path="%2"\
+    failure=0\
+    find $bytecode_compilation_path -type f -a -name "*.py" -print0 | xargs -0 $python_binary -c 'import py_compile, sys; [py_compile.compile(f, dfile=f.partition("'"$RPM_BUILD_ROOT"'")[2], doraise=True) for f in sys.argv[1:]]' || failure=1\
+    find $bytecode_compilation_path -type f -a -name "*.py" -print0 | xargs -0 $python_binary -O -c 'import py_compile, sys; [py_compile.compile(f, dfile=f.partition("'"$RPM_BUILD_ROOT"'")[2], doraise=True) for f in sys.argv[1:]]' || failure=1\
+    test $failure -eq 0\
+}\
+\
+py3_byte_compile () {\
+    python_binary="%1"\
+    bytecode_compilation_path="%2"\
+    PYTHONPATH="%{_rpmconfigdir}/redhat" $python_binary -B -m compileall2 -o 0 -o 1 -s $RPM_BUILD_ROOT -p / $bytecode_compilation_path \
+}\
+\
+# Get version without a dot (36 instead of 3.6), bash doesn't compare floats well \
+python_version=$(%1 -c "import sys; sys.stdout.write('{0.major}{0.minor}'.format(sys.version_info))") \
+# compileall2 Python module is not compatible with Python < 3.4 \
+if [ "$python_version" -ge 34 ]; then py3_byte_compile "%1" "%2"; else py2_byte_compile "%1" "%2"; fi

python-rpm-macros.spec

@@ -3,16 +3,18 @@ Version:        3
 Release:        46%{?dist}
 Summary:        The unversioned Python RPM macros
 
-License:        MIT
+# macros: MIT, compileall2.py: PSFv2
+License:        MIT and Python
 Source0:        macros.python
 Source1:        macros.python-srpm
 Source2:        macros.python2
 Source3:        macros.python3
 Source4:        macros.pybytecompile
+Source5:        https://github.com/fedora-python/compileall2/raw/v0.5.0/compileall2.py
 
 BuildArch:      noarch
-# For %%python3_pkgversion used in %%python_provide
-Requires:       python-srpm-macros
+# For %%python3_pkgversion used in %%python_provide and compileall2.py
+Requires:       python-srpm-macros >= 3-46
 Obsoletes:      python-macros < 3
 Provides:       python-macros = %{version}-%{release}
 
@@ -25,6 +27,7 @@ python?-devel packages require it. So install a python-devel package instead.
 
 %package -n python-srpm-macros
 Summary:        RPM macros for building Python source packages
+Requires:       redhat-rpm-config
 
 %description -n python-srpm-macros
 RPM macros for building Python source packages.
@@ -41,8 +44,6 @@ RPM macros for building Python 2 packages.
 %package -n python3-rpm-macros
 Summary:        RPM macros for building Python 3 packages
 Requires:       python-srpm-macros >= 3-38
-# Would need to be different for each release - worth it?
-#Conflicts:      python3-devel < 3.5.1-3
 
 %description -n python3-rpm-macros
 RPM macros for building Python 3 packages.
@@ -53,10 +54,13 @@ RPM macros for building Python 3 packages.
 %build
 
 %install
-mkdir -p %{buildroot}/%{rpmmacrodir}
+mkdir -p %{buildroot}%{rpmmacrodir}
 install -m 644 %{SOURCE0} %{SOURCE1} %{SOURCE2} %{SOURCE3} %{SOURCE4} \
-  %{buildroot}/%{rpmmacrodir}/
+  %{buildroot}%{rpmmacrodir}/
 
+mkdir -p %{buildroot}%{_rpmconfigdir}/redhat
+install -m 644 %{SOURCE5} \
+  %{buildroot}%{_rpmconfigdir}/redhat/
 
 %files
 %{rpmmacrodir}/macros.python
@@ -64,6 +68,7 @@ install -m 644 %{SOURCE0} %{SOURCE1} %{SOURCE2} %{SOURCE3} %{SOURCE4} \
 
 %files -n python-srpm-macros
 %{rpmmacrodir}/macros.python-srpm
+%{_rpmconfigdir}/redhat/compileall2.py
 
 %files -n python2-rpm-macros
 %{rpmmacrodir}/macros.python2
@@ -76,6 +81,9 @@ install -m 644 %{SOURCE0} %{SOURCE1} %{SOURCE2} %{SOURCE3} %{SOURCE4} \
 * Fri Jul 12 2019 Miro Hrončok <mhroncok@redhat.com> - 3-46
 - %%python_provide: Switch python2 and python3 behavior
 - https://fedoraproject.org/wiki/Changes/Python_means_Python3
+- Use compileall2 module for byte-compilation with Python >= 3.4
+- Do not allow passing arguments to Python during byte-compilation
+- Use `-s` argument for Python during byte-compilation
 
 * Tue Jul 09 2019 Miro Hrončok <mhroncok@redhat.com> - 3-45
 - %%python_provide: Don't try to obsolete %%_isa provides