#!/usr/bin/env python3

# OpenEmbedded pkgdata utility
#
# Written by: Paul Eggleton <paul.eggleton@linux.intel.com>
#
# Copyright 2012-2015 Intel Corporation
#
# SPDX-License-Identifier: GPL-2.0-only
#

import sys
import os
import os.path
import fnmatch
import re
import argparse
import logging
from collections import defaultdict, OrderedDict

scripts_path = os.path.dirname(os.path.realpath(__file__))
lib_path = scripts_path + '/lib'
sys.path = sys.path + [lib_path]
import scriptutils
import argparse_oe
logger = scriptutils.logger_create('pkgdatautil')

def tinfoil_init():
    import bb.tinfoil
    import logging
    tinfoil = bb.tinfoil.Tinfoil()
    tinfoil.logger.setLevel(logging.WARNING)
    tinfoil.prepare(True)
    return tinfoil


def glob(args):
    # Handle both multiple arguments and multiple values within an arg (old syntax)
    globs = []
    for globitem in args.glob:
        globs.extend(globitem.split())

    if not os.path.exists(args.pkglistfile):
        logger.error('Unable to find package list file %s' % args.pkglistfile)
        sys.exit(1)

    skipval = "-locale-|^locale-base-|-dev$|-doc$|-dbg$|-staticdev$|^kernel-module-"
    if args.exclude:
        skipval += "|" + args.exclude
    skipregex = re.compile(skipval)

    skippedpkgs = set()
    mappedpkgs = set()
    with open(args.pkglistfile, 'r') as f:
        for line in f:
            fields = line.rstrip().split()
            if not fields:
                continue
            pkg = fields[0]
            # We don't care about other args (used to need the package architecture but the
            # new pkgdata structure avoids the need for that)

            # Skip packages for which there is no point applying globs
            if skipregex.search(pkg):
                logger.debug("%s -> !!" % pkg)
                skippedpkgs.add(pkg)
                continue

            # Skip packages that already match the globs, so if e.g. a dev package
            # is already installed and thus in the list, we don't process it any further
            # Most of these will be caught by skipregex already, but just in case...
            already = False
            for g in globs:
                if fnmatch.fnmatchcase(pkg, g):
                    already = True
                    break
            if already:
                skippedpkgs.add(pkg)
                logger.debug("%s -> !" % pkg)
                continue

            # Define some functions
            def revpkgdata(pkgn):
                return os.path.join(args.pkgdata_dir, "runtime-reverse", pkgn)
            def fwdpkgdata(pkgn):
                return os.path.join(args.pkgdata_dir, "runtime", pkgn)
            def readpn(pkgdata_file):
                pn = ""
                with open(pkgdata_file, 'r') as f:
                    for line in f:
                        if line.startswith("PN:"):
                            pn = line.split(': ')[1].rstrip()
                return pn
            def readrenamed(pkgdata_file):
                renamed = ""
                pn = os.path.basename(pkgdata_file)
                with open(pkgdata_file, 'r') as f:
                    for line in f:
                        if line.startswith("PKG:%s:" % pn):
                            renamed = line.split(': ')[1].rstrip()
                return renamed

            # Main processing loop
            for g in globs:
                mappedpkg = ""
                # First just try substitution (i.e. packagename -> packagename-dev)
                newpkg = g.replace("*", pkg)
                revlink = revpkgdata(newpkg)
                if os.path.exists(revlink):
                    mappedpkg = os.path.basename(os.readlink(revlink))
                    fwdfile = fwdpkgdata(mappedpkg)
                    if os.path.exists(fwdfile):
                        mappedpkg = readrenamed(fwdfile)
                    if not os.path.exists(fwdfile + ".packaged"):
                        mappedpkg = ""
                else:
                    revlink = revpkgdata(pkg)
                    if os.path.exists(revlink):
                        # Check if we can map after undoing the package renaming (by resolving the symlink)
                        origpkg = os.path.basename(os.readlink(revlink))
                        newpkg = g.replace("*", origpkg)
                        fwdfile = fwdpkgdata(newpkg)
                        if os.path.exists(fwdfile):
                            mappedpkg = readrenamed(fwdfile)
                        else:
                            # That didn't work, so now get the PN, substitute that, then map in the other direction
                            pn = readpn(revlink)
                            newpkg = g.replace("*", pn)
                            fwdfile = fwdpkgdata(newpkg)
                            if os.path.exists(fwdfile):
                                mappedpkg = readrenamed(fwdfile)
                        if not os.path.exists(fwdfile + ".packaged"):
                            mappedpkg = ""
                    else:
                        # Package doesn't even exist...
                        logger.debug("%s is not a valid package!" % (pkg))
                        break

                if mappedpkg:
                    logger.debug("%s (%s) -> %s" % (pkg, g, mappedpkg))
                    mappedpkgs.add(mappedpkg)
                else:
                    logger.debug("%s (%s) -> ?" % (pkg, g))

    logger.debug("------")

    print("\n".join(mappedpkgs - skippedpkgs))

def read_value(args):
    # Handle both multiple arguments and multiple values within an arg (old syntax)
    packages = []
    if args.file:
        with open(args.file, 'r') as f:
            for line in f:
                splitline = line.split()
                if splitline:
                    packages.append(splitline[0])
    else:
        for pkgitem in args.pkg:
            packages.extend(pkgitem.split())
        if not packages:
            logger.error("No packages specified")
            sys.exit(1)

    def readvar(pkgdata_file, valuename, mappedpkg):
        val = ""
        with open(pkgdata_file, 'r') as f:
            for line in f:
                if (line.startswith(valuename + ":") or
                    line.startswith(valuename + "_" + mappedpkg + ":")):
                    val = line.split(': ', 1)[1].rstrip()
        return val

    logger.debug("read-value('%s', '%s' '%s')" % (args.pkgdata_dir, args.valuenames, packages))
    for package in packages:
        pkg_split = package.split('_')
        pkg_name = pkg_split[0]
        logger.debug("package: '%s'" % pkg_name)
        revlink = os.path.join(args.pkgdata_dir, "runtime-reverse", pkg_name)
        logger.debug(revlink)
        if os.path.exists(revlink):
            mappedpkg = os.path.basename(os.readlink(revlink))
            qvars = args.valuenames
            val_names = qvars.split(',')
            values = []
            for qvar in val_names:
                if qvar == "PACKAGE":
                    value = mappedpkg
                else:
                    value = readvar(revlink, qvar, mappedpkg)
                if qvar == "PKGSIZE":
                    # PKGSIZE is now in bytes, but we we want it in KB
                    pkgsize = (int(value) + 1024 // 2) // 1024
                    value = "%d" % pkgsize
                if args.unescape:
                    import codecs
                    # escape_decode() unescapes backslash encodings in byte streams
                    value = codecs.escape_decode(bytes(value, "utf-8"))[0].decode("utf-8")
                values.append(value)

            values_str = ' '.join(values)
            if args.prefix_name:
                print('%s %s' % (pkg_name, values_str))
            else:
                print(values_str)
        else:
            logger.debug("revlink %s does not exist", revlink)

def lookup_pkglist(pkgs, pkgdata_dir, reverse):
    if reverse:
        mappings = OrderedDict()
        for pkg in pkgs:
            revlink = os.path.join(pkgdata_dir, "runtime-reverse", pkg)
            logger.debug(revlink)
            if os.path.exists(revlink):
                mappings[pkg] = os.path.basename(os.readlink(revlink))
    else:
        mappings = defaultdict(list)
        for pkg in pkgs:
            pkgfile = os.path.join(pkgdata_dir, 'runtime', pkg)
            if os.path.exists(pkgfile):
                with open(pkgfile, 'r') as f:
                    for line in f:
                        fields = line.rstrip().split(': ')
                        if fields[0] == 'PKG:%s' % pkg:
                            mappings[pkg].append(fields[1])
                            break
    return mappings

def lookup_pkg(args):
    # Handle both multiple arguments and multiple values within an arg (old syntax)
    pkgs = []
    for pkgitem in args.pkg:
        pkgs.extend(pkgitem.split())

    mappings = lookup_pkglist(pkgs, args.pkgdata_dir, args.reverse)

    if len(mappings) < len(pkgs):
        missing = list(set(pkgs) - set(mappings.keys()))
        logger.error("The following packages could not be found: %s" % ', '.join(missing))
        sys.exit(1)

    if args.reverse:
        items = list(mappings.values())
    else:
        items = []
        for pkg in pkgs:
            items.extend(mappings.get(pkg, []))

    print('\n'.join(items))

def lookup_recipe(args):
    def parse_pkgdatafile(pkgdatafile):
        with open(pkgdatafile, 'r') as f:
            found = False
            for line in f:
                if line.startswith('PN:'):
                    print("%s" % line.split(':', 1)[1].strip())
                    found = True
                    break
            if not found:
                logger.error("Unable to find PN entry in %s" % pkgdatafile)
                sys.exit(1)

    # Handle both multiple arguments and multiple values within an arg (old syntax)
    pkgs = []
    for pkgitem in args.pkg:
        pkgs.extend(pkgitem.split())

    for pkg in pkgs:
        providepkgpath = os.path.join(args.pkgdata_dir, "runtime-rprovides", pkg)
        if os.path.exists(providepkgpath):
            for f in os.listdir(providepkgpath):
                if f != pkg:
                    print("%s is in the RPROVIDES of %s:" % (pkg, f))
                pkgdatafile = os.path.join(args.pkgdata_dir, "runtime", f)
                parse_pkgdatafile(pkgdatafile)
            continue
        pkgdatafile = os.path.join(args.pkgdata_dir, 'runtime-reverse', pkg)
        if os.path.exists(pkgdatafile):
            parse_pkgdatafile(pkgdatafile)
        else:
            if args.carryon:
                print("The following packages could not be found: %s" % pkg)
            else:
                logger.error("The following packages could not be found: %s" % pkg)
                sys.exit(1)

def package_info(args):
    def parse_pkgdatafile(pkgdatafile):
        vars = ['PKGV', 'PKGE', 'PKGR', 'PN', 'PV', 'PE', 'PR', 'PKGSIZE']
        if args.extra:
            vars += args.extra
        with open(pkgdatafile, 'r') as f:
            vals = dict()
            extra = ''
            for line in f:
                for var in vars:
                    m = re.match(var + '(?:_\S+)?:\s*(.+?)\s*$', line)
                    if m:
                        vals[var] = m.group(1)
            pkg_version = vals['PKGV'] or ''
            recipe = vals['PN'] or ''
            recipe_version = vals['PV'] or ''
            pkg_size = vals['PKGSIZE'] or ''
            if 'PKGE' in vals:
                pkg_version = vals['PKGE'] + ":" + pkg_version
            if 'PKGR' in vals:
                pkg_version = pkg_version + "-" + vals['PKGR']
            if 'PE' in vals:
                recipe_version = vals['PE'] + ":" + recipe_version
            if 'PR' in vals:
                recipe_version = recipe_version + "-" + vals['PR']
            if args.extra:
                for var in args.extra:
                    if var in vals:
                        val = re.sub(r'\s+', ' ', vals[var])
                        extra += ' "%s"' % val
            print("%s %s %s %s %s%s" % (pkg, pkg_version, recipe, recipe_version, pkg_size, extra))

    # Handle both multiple arguments and multiple values within an arg (old syntax)
    packages = []
    if args.file:
        with open(args.file, 'r') as f:
            for line in f:
                splitline = line.split()
                if splitline:
                    packages.append(splitline[0])
    else:
        for pkgitem in args.pkg:
            packages.extend(pkgitem.split())
        if not packages:
            logger.error("No packages specified")
            sys.exit(1)

    for pkg in packages:
        providepkgpath = os.path.join(args.pkgdata_dir, "runtime-rprovides", pkg)
        if os.path.exists(providepkgpath):
            for f in os.listdir(providepkgpath):
                if f != pkg:
                    print("%s is in the RPROVIDES of %s:" % (pkg, f))
                pkgdatafile = os.path.join(args.pkgdata_dir, "runtime", f)
                parse_pkgdatafile(pkgdatafile)
            continue
        pkgdatafile = os.path.join(args.pkgdata_dir, "runtime-reverse", pkg)
        if not os.path.exists(pkgdatafile):
            logger.error("Unable to find any built runtime package named %s" % pkg)
            sys.exit(1)
        parse_pkgdatafile(pkgdatafile)

def get_recipe_pkgs(pkgdata_dir, recipe, unpackaged):
    recipedatafile = os.path.join(pkgdata_dir, recipe)
    if not os.path.exists(recipedatafile):
        logger.error("Unable to find packaged recipe with name %s" % recipe)
        sys.exit(1)
    packages = []
    with open(recipedatafile, 'r') as f:
        for line in f:
            fields = line.rstrip().split(': ')
            if fields[0] == 'PACKAGES':
                packages = fields[1].split()
                break

    if not unpackaged:
        pkglist = []
        for pkg in packages:
            if os.path.exists(os.path.join(pkgdata_dir, 'runtime', '%s.packaged' % pkg)):
                pkglist.append(pkg)
        return pkglist
    else:
        return packages

def list_pkgs(args):
    found = False

    def matchpkg(pkg):
        if args.pkgspec:
            matched = False
            for pkgspec in args.pkgspec:
                if fnmatch.fnmatchcase(pkg, pkgspec):
                    matched = True
                    break
            if not matched:
                return False
        if not args.unpackaged:
            if args.runtime:
                revlink = os.path.join(args.pkgdata_dir, "runtime-reverse", pkg)
                if os.path.exists(revlink):
                    # We're unlikely to get here if the package was not packaged, but just in case
                    # we add the symlinks for unpackaged files in the future
                    mappedpkg = os.path.basename(os.readlink(revlink))
                    if not os.path.exists(os.path.join(args.pkgdata_dir, 'runtime', '%s.packaged' % mappedpkg)):
                        return False
                else:
                    return False
            else:
                if not os.path.exists(os.path.join(args.pkgdata_dir, 'runtime', '%s.packaged' % pkg)):
                    return False
        return True

    pkglist = []
    if args.recipe:
        packages = get_recipe_pkgs(args.pkgdata_dir, args.recipe, args.unpackaged)

        if args.runtime:
            runtime_pkgs = lookup_pkglist(packages, args.pkgdata_dir, False)
            for rtpkgs in runtime_pkgs.values():
                pkglist.extend(rtpkgs)
        else:
            pkglist = packages
    else:
        if args.runtime:
            searchdir = 'runtime-reverse'
        else:
            searchdir = 'runtime'

        for root, dirs, files in os.walk(os.path.join(args.pkgdata_dir, searchdir)):
            for fn in files:
                if fn.endswith('.packaged'):
                    continue
                pkglist.append(fn)

    for pkg in sorted(pkglist):
        if matchpkg(pkg):
            found = True
            print("%s" % pkg)

    if not found:
        if args.pkgspec:
            logger.error("Unable to find any package matching %s" % args.pkgspec)
        else:
            logger.error("No packages found")
        sys.exit(1)

def list_pkg_files(args):
    import json
    def parse_pkgdatafile(pkgdatafile, long=False):
        with open(pkgdatafile, 'r') as f:
            found = False
            for line in f:
                if line.startswith('FILES_INFO:'):
                    found = True
                    val = line.split(': ', 1)[1].strip()
                    dictval = json.loads(val)
                    if long:
                        width = max(map(len, dictval), default=0)
                        for fullpth in sorted(dictval):
                            print("\t{:{width}}\t{}".format(fullpth, dictval[fullpth], width=width))
                    else:
                        for fullpth in sorted(dictval):
                            print("\t%s" % fullpth)
                    break
            if not found:
                logger.error("Unable to find FILES_INFO entry in %s" % pkgdatafile)
                sys.exit(1)


    if args.recipe:
        if args.pkg:
            logger.error("list-pkg-files: If -p/--recipe is specified then a package name cannot be specified")
            sys.exit(1)
        recipepkglist = get_recipe_pkgs(args.pkgdata_dir, args.recipe, args.unpackaged)
        if args.runtime:
            pkglist = []
            runtime_pkgs = lookup_pkglist(recipepkglist, args.pkgdata_dir, False)
            for rtpkgs in runtime_pkgs.values():
                pkglist.extend(rtpkgs)
        else:
            pkglist = recipepkglist
    else:
        if not args.pkg:
            logger.error("list-pkg-files: If -p/--recipe is not specified then at least one package name must be specified")
            sys.exit(1)
        pkglist = args.pkg

    for pkg in sorted(pkglist):
        print("%s:" % pkg)
        if args.runtime:
            pkgdatafile = os.path.join(args.pkgdata_dir, "runtime-reverse", pkg)
            if not os.path.exists(pkgdatafile):
                if args.recipe:
                    # This package was empty and thus never packaged, ignore
                    continue
                logger.error("Unable to find any built runtime package named %s" % pkg)
                sys.exit(1)
            parse_pkgdatafile(pkgdatafile, args.long)

        else:
            providepkgpath = os.path.join(args.pkgdata_dir, "runtime-rprovides", pkg)
            if os.path.exists(providepkgpath):
                for f in os.listdir(providepkgpath):
                    if f != pkg:
                        print("%s is in the RPROVIDES of %s:" % (pkg, f))
                    pkgdatafile = os.path.join(args.pkgdata_dir, "runtime", f)
                    parse_pkgdatafile(pkgdatafile, args.long)
                continue
            pkgdatafile = os.path.join(args.pkgdata_dir, "runtime", pkg)
            if not os.path.exists(pkgdatafile):
                logger.error("Unable to find any built recipe-space package named %s" % pkg)
                sys.exit(1)
            parse_pkgdatafile(pkgdatafile, args.long)

def find_path(args):
    import json

    found = False
    for root, dirs, files in os.walk(os.path.join(args.pkgdata_dir, 'runtime')):
        for fn in files:
            with open(os.path.join(root,fn)) as f:
                for line in f:
                    if line.startswith('FILES_INFO:'):
                        val = line.split(': ', 1)[1].strip()
                        dictval = json.loads(val)
                        for fullpth in dictval.keys():
                            if fnmatch.fnmatchcase(fullpth, args.targetpath):
                                found = True
                                print("%s: %s" % (fn, fullpth))
                        break
    if not found:
        logger.error("Unable to find any package producing path %s" % args.targetpath)
        sys.exit(1)


def main():
    parser = argparse_oe.ArgumentParser(description="OpenEmbedded pkgdata tool - queries the pkgdata files written out during do_package",
                                        epilog="Use %(prog)s <subcommand> --help to get help on a specific command")
    parser.add_argument('-d', '--debug', help='Enable debug output', action='store_true')
    parser.add_argument('-p', '--pkgdata-dir', help='Path to pkgdata directory (determined automatically if not specified)')
    subparsers = parser.add_subparsers(title='subcommands', metavar='<subcommand>')
    subparsers.required = True

    parser_lookup_pkg = subparsers.add_parser('lookup-pkg',
                                          help='Translate between recipe-space package names and runtime package names',
                                          description='Looks up the specified recipe-space package name(s) to see what the final runtime package name is (e.g. glibc becomes libc6), or with -r/--reverse looks up the other way.')
    parser_lookup_pkg.add_argument('pkg', nargs='+', help='Package name to look up')
    parser_lookup_pkg.add_argument('-r', '--reverse', help='Switch to looking up recipe-space package names from runtime package names', action='store_true')
    parser_lookup_pkg.set_defaults(func=lookup_pkg)

    parser_list_pkgs = subparsers.add_parser('list-pkgs',
                                          help='List packages',
                                          description='Lists packages that have been built')
    parser_list_pkgs.add_argument('pkgspec', nargs='*', help='Package name to search for (wildcards * ? allowed, use quotes to avoid shell expansion)')
    parser_list_pkgs.add_argument('-r', '--runtime', help='Show runtime package names instead of recipe-space package names', action='store_true')
    parser_list_pkgs.add_argument('-p', '--recipe', help='Limit to packages produced by the specified recipe')
    parser_list_pkgs.add_argument('-u', '--unpackaged', help='Include unpackaged (i.e. empty) packages', action='store_true')
    parser_list_pkgs.set_defaults(func=list_pkgs)

    parser_list_pkg_files = subparsers.add_parser('list-pkg-files',
                                          help='List files within a package',
                                          description='Lists files included in one or more packages')
    parser_list_pkg_files.add_argument('pkg', nargs='*', help='Package name to report on (if -p/--recipe is not specified)')
    parser_list_pkg_files.add_argument('-r', '--runtime', help='Specified package(s) are runtime package names instead of recipe-space package names', action='store_true')
    parser_list_pkg_files.add_argument('-p', '--recipe', help='Report on all packages produced by the specified recipe')
    parser_list_pkg_files.add_argument('-u', '--unpackaged', help='Include unpackaged (i.e. empty) packages (only useful with -p/--recipe)', action='store_true')
    parser_list_pkg_files.add_argument('-l', '--long', help='Show more information per file', action='store_true')
    parser_list_pkg_files.set_defaults(func=list_pkg_files)

    parser_lookup_recipe = subparsers.add_parser('lookup-recipe',
                                          help='Find recipe producing one or more packages',
                                          description='Looks up the specified runtime package(s) to see which recipe they were produced by')
    parser_lookup_recipe.add_argument('pkg', nargs='+', help='Runtime package name to look up')
    parser_lookup_recipe.add_argument('-c', '--continue', dest="carryon", help='Continue looking up recipes even if we can not find one', action='store_true')
    parser_lookup_recipe.set_defaults(func=lookup_recipe)

    parser_package_info = subparsers.add_parser('package-info',
                                          help='Show version, recipe and size information for one or more packages',
                                          description='Looks up the specified runtime package(s) and display information')
    parser_package_info.add_argument('pkg', nargs='*', help='Runtime package name to look up')
    parser_package_info.add_argument('-f', '--file', help='Read package names from the specified file (one per line, first field only)')
    parser_package_info.add_argument('-e', '--extra', help='Extra variables to display, e.g., LICENSE (can be specified multiple times)', action='append')
    parser_package_info.set_defaults(func=package_info)

    parser_find_path = subparsers.add_parser('find-path',
                                          help='Find package providing a target path',
                                          description='Finds the recipe-space package providing the specified target path')
    parser_find_path.add_argument('targetpath', help='Path to find (wildcards * ? allowed, use quotes to avoid shell expansion)')
    parser_find_path.set_defaults(func=find_path)

    parser_read_value = subparsers.add_parser('read-value',
                                          help='Read any pkgdata value for one or more packages',
                                          description='Reads the named value from the pkgdata files for the specified packages')
    parser_read_value.add_argument('valuenames', help='Name of the value/s to look up (separated by commas, no spaces)')
    parser_read_value.add_argument('pkg', nargs='*', help='Runtime package name to look up')
    parser_read_value.add_argument('-f', '--file', help='Read package names from the specified file (one per line, first field only)')
    parser_read_value.add_argument('-n', '--prefix-name', help='Prefix output with package name', action='store_true')
    parser_read_value.add_argument('-u', '--unescape', help='Expand escapes such as \\n', action='store_true')
    parser_read_value.set_defaults(func=read_value)

    parser_glob = subparsers.add_parser('glob',
                                          help='Expand package name glob expression',
                                          description='Expands one or more glob expressions over the packages listed in pkglistfile')
    parser_glob.add_argument('pkglistfile', help='File listing packages (one package name per line)')
    parser_glob.add_argument('glob', nargs="+", help='Glob expression for package names, e.g. *-dev')
    parser_glob.add_argument('-x', '--exclude', help='Exclude packages matching specified regex from the glob operation')
    parser_glob.set_defaults(func=glob)


    args = parser.parse_args()

    if args.debug:
        logger.setLevel(logging.DEBUG)

    if not args.pkgdata_dir:
        import scriptpath
        bitbakepath = scriptpath.add_bitbake_lib_path()
        if not bitbakepath:
            logger.error("Unable to find bitbake by searching parent directory of this script or PATH")
            sys.exit(1)
        logger.debug('Found bitbake path: %s' % bitbakepath)
        if not os.environ.get('BUILDDIR', ''):
            logger.error("This script can only be run after initialising the build environment (e.g. by using oe-init-build-env)")
            sys.exit(1)
        tinfoil = tinfoil_init()
        try:
            args.pkgdata_dir = tinfoil.config_data.getVar('PKGDATA_DIR')
        finally:
            tinfoil.shutdown()
        logger.debug('Value of PKGDATA_DIR is "%s"' % args.pkgdata_dir)
        if not args.pkgdata_dir:
            logger.error('Unable to determine pkgdata directory from PKGDATA_DIR')
            sys.exit(1)

    if not os.path.exists(args.pkgdata_dir):
        logger.error('Unable to find pkgdata directory %s' % args.pkgdata_dir)
        sys.exit(1)

    ret = args.func(args)

    return ret


if __name__ == "__main__":
    main()