tools/userland-mangler
author Mike Sullivan <Mike.Sullivan@Oracle.COM>
Wed, 18 Dec 2013 10:04:53 -0800
branchs11-update
changeset 2855 576d5a7857b7
parent 2408 54e8d1ce1654
child 3532 b2e7186fdcce
permissions -rwxr-xr-x
16404201 serialize pkglinting of components 16666374 userland misses opensolaris.org very much, but must move on to java.net's love 16703472 more userland components could benefit from resolve.deps files 16921386 move to parfait 1.2.0.1 16991973 some things ain't parfait'n 17361780 LD_Z_TEXT should, ya know, contain -z text not -z direct 17389915 default userland gcc should be gcc4 17621943 move to build 32 breaks userland-incorporation 17785723 gdb shouldn't override CC/CXX 17852021 userland-incorporator should detect duplicate packages 17899000 libtool doesn't build properly in non-C locale
#!/usr/bin/python2.6
#
# CDDL HEADER START
#
# The contents of this file are subject to the terms of the
# Common Development and Distribution License (the "License").
# You may not use this file except in compliance with the License.
#
# You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
# or http://www.opensolaris.org/os/licensing.
# See the License for the specific language governing permissions
# and limitations under the License.
#
# When distributing Covered Code, include this CDDL HEADER in each
# file and include the License file at usr/src/OPENSOLARIS.LICENSE.
# If applicable, add the following below this CDDL HEADER, with the
# fields enclosed by brackets "[]" replaced with your own identifying
# information: Portions Copyright [yyyy] [name of copyright owner]
#
# CDDL HEADER END
#
# Copyright (c) 2011, 2013, Oracle and/or its affiliates. All rights reserved.
#
#
# userland-mangler - a file mangling utility
#
#  A simple program to mangle files to conform to Solaris WOS or Consoldation
#  requirements.
#

import os
import sys
import re

import pkg.fmri
import pkg.manifest
import pkg.actions
import pkg.elf as elf

attribute_oracle_table_header = """
.\\\" Oracle has added the ARC stability level to this manual page"""

attribute_table_header = """
.SH ATTRIBUTES
See
.BR attributes (5)
for descriptions of the following attributes:
.sp
.TS
box;
cbp-1 | cbp-1
l | l .
ATTRIBUTE TYPE	ATTRIBUTE VALUE """

attribute_table_availability = """
=
Availability	%s"""

attribute_table_stability = """
=
Stability	%s"""

attribute_table_footer = """
.TE 
.PP
"""
def attributes_section_text(availability, stability, modified_date):
	result = ''

	# is there anything to do?
	if availability is not None or stability is not None:
		result = attribute_oracle_table_header
		if modified_date is not None:
			result += ("\n.\\\" on %s" % modified_date)
		result += attribute_table_header

		if availability is not None:
			result += (attribute_table_availability % availability)
		if stability is not None:
			result += (attribute_table_stability % stability.capitalize())
		result += attribute_table_footer

	return result

notes_oracle_comment = """
.\\\" Oracle has added source availability information to this manual page"""

notes_header = """
.SH NOTES
"""

notes_community = """
Further information about this software can be found on the open source community website at %s.
"""
notes_source = """
This software was built from source available at https://java.net/projects/solaris-userland.  The original community source was downloaded from  %s
"""

def notes_section_text(header_seen, community, source, modified_date):
	result = ''

	# is there anything to do?
	if community is not None or source is not None:
		if header_seen == False:
			result += notes_header
		result += notes_oracle_comment
		if modified_date is not None:
			result += ("\n.\\\" on %s" % modified_date)
		if source is not None:
			result += (notes_source % source)
		if community is not None:
			result += (notes_community % community)

	return result

so_re = re.compile('^\.so.+$', re.MULTILINE)
section_re = re.compile('\.SH "?([^"]+).*$', re.IGNORECASE)
#
# mangler.man.stability = (mangler.man.stability)
# mangler.man.modified_date = (mangler.man.modified-date)
# mangler.man.availability = (pkg.fmri)
# mangler.man.source-url = (pkg.source-url)
# mangler.man.upstream-url = (pkg.upstream-url)
#
def mangle_manpage(manifest, action, text):
	# manpages must have a taxonomy defined
	stability = action.attrs.pop('mangler.man.stability', None)
	if stability is None:
		sys.stderr.write("ERROR: manpage action missing mangler.man.stability: %s" % action)
		sys.exit(1)

	# manpages may have a 'modified date'
	modified_date = action.attrs.pop('mangler.man.modified-date', None)

	attributes_written = False
	notes_seen = False

	if 'pkg.fmri' in manifest.attributes:
		fmri = pkg.fmri.PkgFmri(manifest.attributes['pkg.fmri'])
		availability = fmri.pkg_name

	community = None
	if 'info.upstream-url' in manifest.attributes:
		community = manifest.attributes['info.upstream-url']

	source = None
	if 'info.source-url' in manifest.attributes:
		source = manifest.attributes['info.source-url']
	elif 'info.repository-url' in manifest.attributes:
		source = manifest.attributes['info.repository-url']

	# skip reference only pages
	if so_re.match(text) is not None:
		return text

	# tell man that we want tables (and eqn)
	result = "'\\\" te\n"

	# write the orginal data
	for line in text.split('\n'):
		match = section_re.match(line)
		if match is not None:
			section = match.group(1)
			if section in ['SEE ALSO', 'NOTES']:
				if attributes_written == False:
					result += attributes_section_text(
								 availability,
								 stability,
								 modified_date)
					attributes_written = True
				if section == 'NOTES':
					notes_seen = True
		result += ("%s\n" % line)

	if attributes_written == False:
		result += attributes_section_text(availability, stability,
		    modified_date)

	result += notes_section_text(notes_seen, community, source,
	    modified_date)

	return result


#
# mangler.elf.strip = (true|false)
#
def mangle_elf(manifest, action, src, dest):
	pass

#
# mangler.script.file-magic =
#
def mangle_script(manifest, action, text):
	return text

#
# mangler.strip_cddl = false
#
def mangle_cddl(manifest, action, text):
	strip_cddl = action.attrs.pop('mangler.strip_cddl', 'true')
	if strip_cddl is 'false':
		return text
	cddl_re = re.compile('^[^\n]*CDDL HEADER START.+CDDL HEADER END[^\n]*$',
			     re.MULTILINE|re.DOTALL)
	return cddl_re.sub('', text)

def mangle_path(manifest, action, src, dest):
	if elf.is_elf_object(src):
		mangle_elf(manifest, action, src, dest)
	else:
		# a 'text' document (script, man page, config file, ...
		ifp = open(src, 'r')
		text = ifp.read()
		ifp.close()

		# remove the CDDL from files
		result = mangle_cddl(manifest, action, text)

		if 'facet.doc.man' in action.attrs:
		 	result = mangle_manpage(manifest, action, result)
		elif 'mode' in action.attrs and int(action.attrs['mode'], 8) & 0111 != 0:
			result = mangle_script(manifest, action, result)

		if text != result:
			destdir = os.path.dirname(dest)
			if not os.path.exists(destdir):
				os.makedirs(destdir)
			with open(dest, 'w') as ofp:
			    ofp.write(result)

#
# mangler.bypass = (true|false)
#
def mangle_paths(manifest, search_paths, destination):
	for action in manifest.gen_actions_by_type("file"):
		bypass = action.attrs.pop('mangler.bypass', 'false').lower()
		if bypass == 'true':
			continue

		path = None
		if 'path' in action.attrs:
			path = action.attrs['path']
		if action.hash and action.hash != 'NOHASH':
			path = action.hash
		if not path:
			continue

		if not os.path.exists(destination):
			os.makedirs(destination)

		dest = os.path.join(destination, path)
		for directory in search_paths:
			if directory != destination:
				src = os.path.join(directory, path)
				if os.path.isfile(src):
					mangle_path(manifest, action, src, dest)
					break

def load_manifest(manifest_file):
	manifest = pkg.manifest.Manifest()
	manifest.set_content(pathname=manifest_file)

	return manifest

def usage():
	print "Usage: %s [-m|--manifest (file)] [-d|--search-directory (dir)] [-D|--destination (dir)] " % (sys.argv[0].split('/')[-1])
	sys.exit(1)

def main():
	import getopt

	# FLUSH STDOUT 
	sys.stdout = os.fdopen(sys.stdout.fileno(), 'w', 0)

	search_paths = []
	destination = None
	manifests = []

	try:
		opts, args = getopt.getopt(sys.argv[1:], "D:d:m:",
			["destination=", "search-directory=", "manifest="])
	except getopt.GetoptError, err:
		print str(err)
		usage()

	for opt, arg in opts:
		if opt in [ "-D", "--destination" ]:
			destination = arg
		elif opt in [ "-d", "--search-directory" ]:
			search_paths.append(arg)
		elif opt in [ "-m", "--manifest" ]:
			try:
				manifest = load_manifest(arg)
			except IOError, err:
				print "oops, %s: %s" % (arg, str(err))
				usage()
			else:
				manifests.append(manifest)
		else:
			usage()

	if destination == None:
		usage()

	for manifest in manifests:
		mangle_paths(manifest, search_paths, destination)
		print manifest

	sys.exit(0)

if __name__ == "__main__":
	main()