mirror of
https://github.com/espressif/esp-idf.git
synced 2024-10-05 20:47:46 -04:00
8d05d2511c
For Secure Boot v2 case, unsigned image is first padded to next 64K
aligned boundary and then a signature block of 4K gets appended. Thus
an app partition whose size is 4K aligned should be allowed here.
For Secure Boot v1 case, app partition size must be 64K aligned as the
signature block lies at the very end of 64K boundary.
Relevant:
57b601ab7f
615 lines
23 KiB
Python
Executable File
615 lines
23 KiB
Python
Executable File
#!/usr/bin/env python
|
|
#
|
|
# ESP32 partition table generation tool
|
|
#
|
|
# Converts partition tables to/from CSV and binary formats.
|
|
#
|
|
# See https://docs.espressif.com/projects/esp-idf/en/latest/api-guides/partition-tables.html
|
|
# for explanation of partition table structure and uses.
|
|
#
|
|
# SPDX-FileCopyrightText: 2016-2021 Espressif Systems (Shanghai) CO LTD
|
|
# SPDX-License-Identifier: Apache-2.0
|
|
|
|
from __future__ import division, print_function, unicode_literals
|
|
|
|
import argparse
|
|
import binascii
|
|
import errno
|
|
import hashlib
|
|
import os
|
|
import re
|
|
import struct
|
|
import sys
|
|
|
|
MAX_PARTITION_LENGTH = 0xC00 # 3K for partition data (96 entries) leaves 1K in a 4K sector for signature
|
|
MD5_PARTITION_BEGIN = b'\xEB\xEB' + b'\xFF' * 14 # The first 2 bytes are like magic numbers for MD5 sum
|
|
PARTITION_TABLE_SIZE = 0x1000 # Size of partition table
|
|
|
|
MIN_PARTITION_SUBTYPE_APP_OTA = 0x10
|
|
NUM_PARTITION_SUBTYPE_APP_OTA = 16
|
|
|
|
SECURE_NONE = None
|
|
SECURE_V1 = 'v1'
|
|
SECURE_V2 = 'v2'
|
|
|
|
__version__ = '1.2'
|
|
|
|
APP_TYPE = 0x00
|
|
DATA_TYPE = 0x01
|
|
|
|
TYPES = {
|
|
'app': APP_TYPE,
|
|
'data': DATA_TYPE,
|
|
}
|
|
|
|
|
|
def get_ptype_as_int(ptype):
|
|
""" Convert a string which might be numeric or the name of a partition type to an integer """
|
|
try:
|
|
return TYPES[ptype]
|
|
except KeyError:
|
|
try:
|
|
return int(ptype, 0)
|
|
except TypeError:
|
|
return ptype
|
|
|
|
|
|
# Keep this map in sync with esp_partition_subtype_t enum in esp_partition.h
|
|
SUBTYPES = {
|
|
APP_TYPE: {
|
|
'factory': 0x00,
|
|
'test': 0x20,
|
|
},
|
|
DATA_TYPE: {
|
|
'ota': 0x00,
|
|
'phy': 0x01,
|
|
'nvs': 0x02,
|
|
'coredump': 0x03,
|
|
'nvs_keys': 0x04,
|
|
'efuse': 0x05,
|
|
'undefined': 0x06,
|
|
'esphttpd': 0x80,
|
|
'fat': 0x81,
|
|
'spiffs': 0x82,
|
|
},
|
|
}
|
|
|
|
|
|
def get_subtype_as_int(ptype, subtype):
|
|
""" Convert a string which might be numeric or the name of a partition subtype to an integer """
|
|
try:
|
|
return SUBTYPES[get_ptype_as_int(ptype)][subtype]
|
|
except KeyError:
|
|
try:
|
|
return int(subtype, 0)
|
|
except TypeError:
|
|
return subtype
|
|
|
|
|
|
ALIGNMENT = {
|
|
APP_TYPE: 0x10000,
|
|
DATA_TYPE: 0x1000,
|
|
}
|
|
|
|
|
|
def get_alignment_offset_for_type(ptype):
|
|
return ALIGNMENT.get(ptype, ALIGNMENT[DATA_TYPE])
|
|
|
|
|
|
def get_alignment_size_for_type(ptype):
|
|
if ptype == APP_TYPE and secure == SECURE_V1:
|
|
# For secure boot v1 case, app partition must be 64K aligned
|
|
# signature block (68 bytes) lies at the very end of 64K block
|
|
return 0x10000
|
|
if ptype == APP_TYPE and secure == SECURE_V2:
|
|
# For secure boot v2 case, app partition must be 4K aligned
|
|
# signature block (4K) is kept after padding the unsigned image to 64K boundary
|
|
return 0x1000
|
|
# No specific size alignement requirement as such
|
|
return 0x1
|
|
|
|
|
|
def get_partition_type(ptype):
|
|
if ptype == 'app':
|
|
return APP_TYPE
|
|
if ptype == 'data':
|
|
return DATA_TYPE
|
|
raise InputError('Invalid partition type')
|
|
|
|
|
|
def add_extra_subtypes(csv):
|
|
for line_no in csv:
|
|
try:
|
|
fields = [line.strip() for line in line_no.split(',')]
|
|
for subtype, subtype_values in SUBTYPES.items():
|
|
if (int(fields[2], 16) in subtype_values.values() and subtype == get_partition_type(fields[0])):
|
|
raise ValueError('Found duplicate value in partition subtype')
|
|
SUBTYPES[TYPES[fields[0]]][fields[1]] = int(fields[2], 16)
|
|
except InputError as err:
|
|
raise InputError('Error parsing custom subtypes: %s' % err)
|
|
|
|
|
|
quiet = False
|
|
md5sum = True
|
|
secure = SECURE_NONE
|
|
offset_part_table = 0
|
|
|
|
|
|
def status(msg):
|
|
""" Print status message to stderr """
|
|
if not quiet:
|
|
critical(msg)
|
|
|
|
|
|
def critical(msg):
|
|
""" Print critical message to stderr """
|
|
sys.stderr.write(msg)
|
|
sys.stderr.write('\n')
|
|
|
|
|
|
class PartitionTable(list):
|
|
def __init__(self):
|
|
super(PartitionTable, self).__init__(self)
|
|
|
|
@classmethod
|
|
def from_file(cls, f):
|
|
data = f.read()
|
|
data_is_binary = data[0:2] == PartitionDefinition.MAGIC_BYTES
|
|
if data_is_binary:
|
|
status('Parsing binary partition input...')
|
|
return cls.from_binary(data), True
|
|
|
|
data = data.decode()
|
|
status('Parsing CSV input...')
|
|
return cls.from_csv(data), False
|
|
|
|
@classmethod
|
|
def from_csv(cls, csv_contents):
|
|
res = PartitionTable()
|
|
lines = csv_contents.splitlines()
|
|
|
|
def expand_vars(f):
|
|
f = os.path.expandvars(f)
|
|
m = re.match(r'(?<!\\)\$([A-Za-z_][A-Za-z0-9_]*)', f)
|
|
if m:
|
|
raise InputError("unknown variable '%s'" % m.group(1))
|
|
return f
|
|
|
|
for line_no in range(len(lines)):
|
|
line = expand_vars(lines[line_no]).strip()
|
|
if line.startswith('#') or len(line) == 0:
|
|
continue
|
|
try:
|
|
res.append(PartitionDefinition.from_csv(line, line_no + 1))
|
|
except InputError as err:
|
|
raise InputError('Error at line %d: %s\nPlease check extra_partition_subtypes.inc file in build/config directory' % (line_no + 1, err))
|
|
except Exception:
|
|
critical('Unexpected error parsing CSV line %d: %s' % (line_no + 1, line))
|
|
raise
|
|
|
|
# fix up missing offsets & negative sizes
|
|
last_end = offset_part_table + PARTITION_TABLE_SIZE # first offset after partition table
|
|
for e in res:
|
|
if e.offset is not None and e.offset < last_end:
|
|
if e == res[0]:
|
|
raise InputError('CSV Error at line %d: Partitions overlap. Partition sets offset 0x%x. '
|
|
'But partition table occupies the whole sector 0x%x. '
|
|
'Use a free offset 0x%x or higher.'
|
|
% (e.line_no, e.offset, offset_part_table, last_end))
|
|
else:
|
|
raise InputError('CSV Error at line %d: Partitions overlap. Partition sets offset 0x%x. Previous partition ends 0x%x'
|
|
% (e.line_no, e.offset, last_end))
|
|
if e.offset is None:
|
|
pad_to = get_alignment_offset_for_type(e.type)
|
|
if last_end % pad_to != 0:
|
|
last_end += pad_to - (last_end % pad_to)
|
|
e.offset = last_end
|
|
if e.size < 0:
|
|
e.size = -e.size - e.offset
|
|
last_end = e.offset + e.size
|
|
|
|
return res
|
|
|
|
def __getitem__(self, item):
|
|
""" Allow partition table access via name as well as by
|
|
numeric index. """
|
|
if isinstance(item, str):
|
|
for x in self:
|
|
if x.name == item:
|
|
return x
|
|
raise ValueError("No partition entry named '%s'" % item)
|
|
else:
|
|
return super(PartitionTable, self).__getitem__(item)
|
|
|
|
def find_by_type(self, ptype, subtype):
|
|
""" Return a partition by type & subtype, returns
|
|
None if not found """
|
|
# convert ptype & subtypes names (if supplied this way) to integer values
|
|
ptype = get_ptype_as_int(ptype)
|
|
subtype = get_subtype_as_int(ptype, subtype)
|
|
|
|
for p in self:
|
|
if p.type == ptype and p.subtype == subtype:
|
|
yield p
|
|
return
|
|
|
|
def find_by_name(self, name):
|
|
for p in self:
|
|
if p.name == name:
|
|
return p
|
|
return None
|
|
|
|
def verify(self):
|
|
# verify each partition individually
|
|
for p in self:
|
|
p.verify()
|
|
|
|
# check on duplicate name
|
|
names = [p.name for p in self]
|
|
duplicates = set(n for n in names if names.count(n) > 1)
|
|
|
|
# print sorted duplicate partitions by name
|
|
if len(duplicates) != 0:
|
|
critical('A list of partitions that have the same name:')
|
|
for p in sorted(self, key=lambda x:x.name):
|
|
if len(duplicates.intersection([p.name])) != 0:
|
|
critical('%s' % (p.to_csv()))
|
|
raise InputError('Partition names must be unique')
|
|
|
|
# check for overlaps
|
|
last = None
|
|
for p in sorted(self, key=lambda x:x.offset):
|
|
if p.offset < offset_part_table + PARTITION_TABLE_SIZE:
|
|
raise InputError('Partition offset 0x%x is below 0x%x' % (p.offset, offset_part_table + PARTITION_TABLE_SIZE))
|
|
if last is not None and p.offset < last.offset + last.size:
|
|
raise InputError('Partition at 0x%x overlaps 0x%x-0x%x' % (p.offset, last.offset, last.offset + last.size - 1))
|
|
last = p
|
|
|
|
# check that otadata should be unique
|
|
otadata_duplicates = [p for p in self if p.type == TYPES['data'] and p.subtype == SUBTYPES[DATA_TYPE]['ota']]
|
|
if len(otadata_duplicates) > 1:
|
|
for p in otadata_duplicates:
|
|
critical('%s' % (p.to_csv()))
|
|
raise InputError('Found multiple otadata partitions. Only one partition can be defined with type="data"(1) and subtype="ota"(0).')
|
|
|
|
if len(otadata_duplicates) == 1 and otadata_duplicates[0].size != 0x2000:
|
|
p = otadata_duplicates[0]
|
|
critical('%s' % (p.to_csv()))
|
|
raise InputError('otadata partition must have size = 0x2000')
|
|
|
|
def flash_size(self):
|
|
""" Return the size that partitions will occupy in flash
|
|
(ie the offset the last partition ends at)
|
|
"""
|
|
try:
|
|
last = sorted(self, reverse=True)[0]
|
|
except IndexError:
|
|
return 0 # empty table!
|
|
return last.offset + last.size
|
|
|
|
def verify_size_fits(self, flash_size_bytes: int) -> None:
|
|
""" Check that partition table fits into the given flash size.
|
|
Raises InputError otherwise.
|
|
"""
|
|
table_size = self.flash_size()
|
|
if flash_size_bytes < table_size:
|
|
mb = 1024 * 1024
|
|
raise InputError('Partitions tables occupies %.1fMB of flash (%d bytes) which does not fit in configured '
|
|
"flash size %dMB. Change the flash size in menuconfig under the 'Serial Flasher Config' menu." %
|
|
(table_size / mb, table_size, flash_size_bytes / mb))
|
|
|
|
@classmethod
|
|
def from_binary(cls, b):
|
|
md5 = hashlib.md5()
|
|
result = cls()
|
|
for o in range(0,len(b),32):
|
|
data = b[o:o + 32]
|
|
if len(data) != 32:
|
|
raise InputError('Partition table length must be a multiple of 32 bytes')
|
|
if data == b'\xFF' * 32:
|
|
return result # got end marker
|
|
if md5sum and data[:2] == MD5_PARTITION_BEGIN[:2]: # check only the magic number part
|
|
if data[16:] == md5.digest():
|
|
continue # the next iteration will check for the end marker
|
|
else:
|
|
raise InputError("MD5 checksums don't match! (computed: 0x%s, parsed: 0x%s)" % (md5.hexdigest(), binascii.hexlify(data[16:])))
|
|
else:
|
|
md5.update(data)
|
|
result.append(PartitionDefinition.from_binary(data))
|
|
raise InputError('Partition table is missing an end-of-table marker')
|
|
|
|
def to_binary(self):
|
|
result = b''.join(e.to_binary() for e in self)
|
|
if md5sum:
|
|
result += MD5_PARTITION_BEGIN + hashlib.md5(result).digest()
|
|
if len(result) >= MAX_PARTITION_LENGTH:
|
|
raise InputError('Binary partition table length (%d) longer than max' % len(result))
|
|
result += b'\xFF' * (MAX_PARTITION_LENGTH - len(result)) # pad the sector, for signing
|
|
return result
|
|
|
|
def to_csv(self, simple_formatting=False):
|
|
rows = ['# ESP-IDF Partition Table',
|
|
'# Name, Type, SubType, Offset, Size, Flags']
|
|
rows += [x.to_csv(simple_formatting) for x in self]
|
|
return '\n'.join(rows) + '\n'
|
|
|
|
|
|
class PartitionDefinition(object):
|
|
MAGIC_BYTES = b'\xAA\x50'
|
|
|
|
# dictionary maps flag name (as used in CSV flags list, property name)
|
|
# to bit set in flags words in binary format
|
|
FLAGS = {
|
|
'encrypted': 0
|
|
}
|
|
|
|
# add subtypes for the 16 OTA slot values ("ota_XX, etc.")
|
|
for ota_slot in range(NUM_PARTITION_SUBTYPE_APP_OTA):
|
|
SUBTYPES[TYPES['app']]['ota_%d' % ota_slot] = MIN_PARTITION_SUBTYPE_APP_OTA + ota_slot
|
|
|
|
def __init__(self):
|
|
self.name = ''
|
|
self.type = None
|
|
self.subtype = None
|
|
self.offset = None
|
|
self.size = None
|
|
self.encrypted = False
|
|
|
|
@classmethod
|
|
def from_csv(cls, line, line_no):
|
|
""" Parse a line from the CSV """
|
|
line_w_defaults = line + ',,,,' # lazy way to support default fields
|
|
fields = [f.strip() for f in line_w_defaults.split(',')]
|
|
|
|
res = PartitionDefinition()
|
|
res.line_no = line_no
|
|
res.name = fields[0]
|
|
res.type = res.parse_type(fields[1])
|
|
res.subtype = res.parse_subtype(fields[2])
|
|
res.offset = res.parse_address(fields[3])
|
|
res.size = res.parse_address(fields[4])
|
|
if res.size is None:
|
|
raise InputError("Size field can't be empty")
|
|
|
|
flags = fields[5].split(':')
|
|
for flag in flags:
|
|
if flag in cls.FLAGS:
|
|
setattr(res, flag, True)
|
|
elif len(flag) > 0:
|
|
raise InputError("CSV flag column contains unknown flag '%s'" % (flag))
|
|
|
|
return res
|
|
|
|
def __eq__(self, other):
|
|
return self.name == other.name and self.type == other.type \
|
|
and self.subtype == other.subtype and self.offset == other.offset \
|
|
and self.size == other.size
|
|
|
|
def __repr__(self):
|
|
def maybe_hex(x):
|
|
return '0x%x' % x if x is not None else 'None'
|
|
return "PartitionDefinition('%s', 0x%x, 0x%x, %s, %s)" % (self.name, self.type, self.subtype or 0,
|
|
maybe_hex(self.offset), maybe_hex(self.size))
|
|
|
|
def __str__(self):
|
|
return "Part '%s' %d/%d @ 0x%x size 0x%x" % (self.name, self.type, self.subtype, self.offset or -1, self.size or -1)
|
|
|
|
def __cmp__(self, other):
|
|
return self.offset - other.offset
|
|
|
|
def __lt__(self, other):
|
|
return self.offset < other.offset
|
|
|
|
def __gt__(self, other):
|
|
return self.offset > other.offset
|
|
|
|
def __le__(self, other):
|
|
return self.offset <= other.offset
|
|
|
|
def __ge__(self, other):
|
|
return self.offset >= other.offset
|
|
|
|
def parse_type(self, strval):
|
|
if strval == '':
|
|
raise InputError("Field 'type' can't be left empty.")
|
|
return parse_int(strval, TYPES)
|
|
|
|
def parse_subtype(self, strval):
|
|
if strval == '':
|
|
if self.type == TYPES['app']:
|
|
raise InputError('App partition cannot have an empty subtype')
|
|
return SUBTYPES[DATA_TYPE]['undefined']
|
|
return parse_int(strval, SUBTYPES.get(self.type, {}))
|
|
|
|
def parse_address(self, strval):
|
|
if strval == '':
|
|
return None # PartitionTable will fill in default
|
|
return parse_int(strval)
|
|
|
|
def verify(self):
|
|
if self.type is None:
|
|
raise ValidationError(self, 'Type field is not set')
|
|
if self.subtype is None:
|
|
raise ValidationError(self, 'Subtype field is not set')
|
|
if self.offset is None:
|
|
raise ValidationError(self, 'Offset field is not set')
|
|
if self.size is None:
|
|
raise ValidationError(self, 'Size field is not set')
|
|
offset_align = get_alignment_offset_for_type(self.type)
|
|
if self.offset % offset_align:
|
|
raise ValidationError(self, 'Offset 0x%x is not aligned to 0x%x' % (self.offset, offset_align))
|
|
if self.type == APP_TYPE and secure is not SECURE_NONE:
|
|
size_align = get_alignment_size_for_type(self.type)
|
|
if self.size % size_align:
|
|
raise ValidationError(self, 'Size 0x%x is not aligned to 0x%x' % (self.size, size_align))
|
|
|
|
if self.name in TYPES and TYPES.get(self.name, '') != self.type:
|
|
critical("WARNING: Partition has name '%s' which is a partition type, but does not match this partition's "
|
|
'type (0x%x). Mistake in partition table?' % (self.name, self.type))
|
|
all_subtype_names = []
|
|
for names in (t.keys() for t in SUBTYPES.values()):
|
|
all_subtype_names += names
|
|
if self.name in all_subtype_names and SUBTYPES.get(self.type, {}).get(self.name, '') != self.subtype:
|
|
critical("WARNING: Partition has name '%s' which is a partition subtype, but this partition has "
|
|
'non-matching type 0x%x and subtype 0x%x. Mistake in partition table?' % (self.name, self.type, self.subtype))
|
|
|
|
STRUCT_FORMAT = b'<2sBBLL16sL'
|
|
|
|
@classmethod
|
|
def from_binary(cls, b):
|
|
if len(b) != 32:
|
|
raise InputError('Partition definition length must be exactly 32 bytes. Got %d bytes.' % len(b))
|
|
res = cls()
|
|
(magic, res.type, res.subtype, res.offset,
|
|
res.size, res.name, flags) = struct.unpack(cls.STRUCT_FORMAT, b)
|
|
if b'\x00' in res.name: # strip null byte padding from name string
|
|
res.name = res.name[:res.name.index(b'\x00')]
|
|
res.name = res.name.decode()
|
|
if magic != cls.MAGIC_BYTES:
|
|
raise InputError('Invalid magic bytes (%r) for partition definition' % magic)
|
|
for flag,bit in cls.FLAGS.items():
|
|
if flags & (1 << bit):
|
|
setattr(res, flag, True)
|
|
flags &= ~(1 << bit)
|
|
if flags != 0:
|
|
critical('WARNING: Partition definition had unknown flag(s) 0x%08x. Newer binary format?' % flags)
|
|
return res
|
|
|
|
def get_flags_list(self):
|
|
return [flag for flag in self.FLAGS.keys() if getattr(self, flag)]
|
|
|
|
def to_binary(self):
|
|
flags = sum((1 << self.FLAGS[flag]) for flag in self.get_flags_list())
|
|
return struct.pack(self.STRUCT_FORMAT,
|
|
self.MAGIC_BYTES,
|
|
self.type, self.subtype,
|
|
self.offset, self.size,
|
|
self.name.encode(),
|
|
flags)
|
|
|
|
def to_csv(self, simple_formatting=False):
|
|
def addr_format(a, include_sizes):
|
|
if not simple_formatting and include_sizes:
|
|
for (val, suffix) in [(0x100000, 'M'), (0x400, 'K')]:
|
|
if a % val == 0:
|
|
return '%d%s' % (a // val, suffix)
|
|
return '0x%x' % a
|
|
|
|
def lookup_keyword(t, keywords):
|
|
for k,v in keywords.items():
|
|
if simple_formatting is False and t == v:
|
|
return k
|
|
return '%d' % t
|
|
|
|
def generate_text_flags():
|
|
""" colon-delimited list of flags """
|
|
return ':'.join(self.get_flags_list())
|
|
|
|
return ','.join([self.name,
|
|
lookup_keyword(self.type, TYPES),
|
|
lookup_keyword(self.subtype, SUBTYPES.get(self.type, {})),
|
|
addr_format(self.offset, False),
|
|
addr_format(self.size, True),
|
|
generate_text_flags()])
|
|
|
|
|
|
def parse_int(v, keywords={}):
|
|
"""Generic parser for integer fields - int(x,0) with provision for
|
|
k/m/K/M suffixes and 'keyword' value lookup.
|
|
"""
|
|
try:
|
|
for letter, multiplier in [('k', 1024), ('m', 1024 * 1024)]:
|
|
if v.lower().endswith(letter):
|
|
return parse_int(v[:-1], keywords) * multiplier
|
|
return int(v, 0)
|
|
except ValueError:
|
|
if len(keywords) == 0:
|
|
raise InputError('Invalid field value %s' % v)
|
|
try:
|
|
return keywords[v.lower()]
|
|
except KeyError:
|
|
raise InputError("Value '%s' is not valid. Known keywords: %s" % (v, ', '.join(keywords)))
|
|
|
|
|
|
def main():
|
|
global quiet
|
|
global md5sum
|
|
global offset_part_table
|
|
global secure
|
|
parser = argparse.ArgumentParser(description='ESP32 partition table utility')
|
|
|
|
parser.add_argument('--flash-size', help='Optional flash size limit, checks partition table fits in flash',
|
|
nargs='?', choices=['1MB', '2MB', '4MB', '8MB', '16MB', '32MB', '64MB', '128MB'])
|
|
parser.add_argument('--disable-md5sum', help='Disable md5 checksum for the partition table', default=False, action='store_true')
|
|
parser.add_argument('--no-verify', help="Don't verify partition table fields", action='store_true')
|
|
parser.add_argument('--verify', '-v', help='Verify partition table fields (deprecated, this behaviour is '
|
|
'enabled by default and this flag does nothing.', action='store_true')
|
|
parser.add_argument('--quiet', '-q', help="Don't print non-critical status messages to stderr", action='store_true')
|
|
parser.add_argument('--offset', '-o', help='Set offset partition table', default='0x8000')
|
|
parser.add_argument('--secure', help='Require app partitions to be suitable for secure boot', nargs='?', const=SECURE_V1, choices=[SECURE_V1, SECURE_V2])
|
|
parser.add_argument('--extra-partition-subtypes', help='Extra partition subtype entries', nargs='*')
|
|
parser.add_argument('input', help='Path to CSV or binary file to parse.', type=argparse.FileType('rb'))
|
|
parser.add_argument('output', help='Path to output converted binary or CSV file. Will use stdout if omitted.',
|
|
nargs='?', default='-')
|
|
|
|
args = parser.parse_args()
|
|
|
|
quiet = args.quiet
|
|
md5sum = not args.disable_md5sum
|
|
secure = args.secure
|
|
offset_part_table = int(args.offset, 0)
|
|
if args.extra_partition_subtypes:
|
|
add_extra_subtypes(args.extra_partition_subtypes)
|
|
|
|
table, input_is_binary = PartitionTable.from_file(args.input)
|
|
|
|
if not args.no_verify:
|
|
status('Verifying table...')
|
|
table.verify()
|
|
|
|
if args.flash_size:
|
|
size_mb = int(args.flash_size.replace('MB', ''))
|
|
table.verify_size_fits(size_mb * 1024 * 1024)
|
|
|
|
# Make sure that the output directory is created
|
|
output_dir = os.path.abspath(os.path.dirname(args.output))
|
|
|
|
if not os.path.exists(output_dir):
|
|
try:
|
|
os.makedirs(output_dir)
|
|
except OSError as exc:
|
|
if exc.errno != errno.EEXIST:
|
|
raise
|
|
|
|
if input_is_binary:
|
|
output = table.to_csv()
|
|
with sys.stdout if args.output == '-' else open(args.output, 'w') as f:
|
|
f.write(output)
|
|
else:
|
|
output = table.to_binary()
|
|
try:
|
|
stdout_binary = sys.stdout.buffer # Python 3
|
|
except AttributeError:
|
|
stdout_binary = sys.stdout
|
|
with stdout_binary if args.output == '-' else open(args.output, 'wb') as f:
|
|
f.write(output)
|
|
|
|
|
|
class InputError(RuntimeError):
|
|
def __init__(self, e):
|
|
super(InputError, self).__init__(e)
|
|
|
|
|
|
class ValidationError(InputError):
|
|
def __init__(self, partition, message):
|
|
super(ValidationError, self).__init__(
|
|
'Partition %s invalid: %s' % (partition.name, message))
|
|
|
|
|
|
if __name__ == '__main__':
|
|
try:
|
|
main()
|
|
except InputError as e:
|
|
print(e, file=sys.stderr)
|
|
sys.exit(2)
|