mirror of
https://github.com/espressif/esp-idf.git
synced 2024-10-05 20:47:46 -04:00
cd9249970f
(cherry picked from commit 8b96668c9123fbbf8c78b347479215f471b102ed)
550 lines
22 KiB
Python
Executable File
550 lines
22 KiB
Python
Executable File
#!/usr/bin/env python
|
|
#
|
|
# Copyright 2018 Espressif Systems (Shanghai) PTE LTD
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
#
|
|
|
|
from __future__ import print_function
|
|
from builtins import range
|
|
from future.moves.itertools import zip_longest
|
|
import sys
|
|
import os
|
|
import csv
|
|
import argparse
|
|
import distutils.dir_util
|
|
|
|
try:
|
|
sys.path.insert(0, os.getenv('IDF_PATH') + "/components/nvs_flash/nvs_partition_generator/")
|
|
import nvs_partition_gen
|
|
except Exception as e:
|
|
print(e)
|
|
sys.exit("Please check IDF_PATH")
|
|
|
|
|
|
def verify_values_exist(input_values_file, keys_in_values_file):
|
|
""" Verify all keys have corresponding values in values file
|
|
"""
|
|
line_no = 1
|
|
key_count_in_values_file = len(keys_in_values_file)
|
|
|
|
values_file = open(input_values_file, 'r')
|
|
values_file_reader = csv.reader(values_file, delimiter=',')
|
|
next(values_file_reader)
|
|
|
|
for values_data in values_file_reader:
|
|
line_no += 1
|
|
if len(values_data) != key_count_in_values_file:
|
|
raise SystemExit("\nOops...Number of values is not equal to number of keys in file: %s at line No:%s\n"
|
|
% (str(input_values_file), str(line_no)))
|
|
|
|
|
|
def verify_keys_exist(values_file_keys, input_config_file):
|
|
""" Verify all keys from config file are present in values file
|
|
"""
|
|
keys_missing = []
|
|
|
|
config_file = open(input_config_file,'r')
|
|
config_file_reader = csv.reader(filter(lambda row: row[0] != '#',config_file), delimiter=',')
|
|
for line_no, config_data in enumerate(config_file_reader,1):
|
|
if 'namespace' not in config_data:
|
|
if values_file_keys:
|
|
if config_data[0] == values_file_keys[0]:
|
|
del values_file_keys[0]
|
|
else:
|
|
keys_missing.append([config_data[0], line_no])
|
|
else:
|
|
keys_missing.append([config_data[0], line_no])
|
|
|
|
if keys_missing:
|
|
for key, line_no in keys_missing:
|
|
print("Key:`", str(key), "` at line no:", str(line_no),
|
|
" in config file is not found in values file.")
|
|
config_file.close()
|
|
raise SystemExit(1)
|
|
|
|
config_file.close()
|
|
|
|
|
|
def verify_datatype_encoding(input_config_file):
|
|
""" Verify datatype and encodings from config file is valid
|
|
"""
|
|
valid_encodings = ["string", "binary", "hex2bin","u8", "i8", "u16", "u32", "i32","base64"]
|
|
valid_datatypes = ["file","data","namespace"]
|
|
line_no = 0
|
|
|
|
config_file = open(input_config_file,'r')
|
|
config_file_reader = csv.reader(filter(lambda row: row[0] != '#',config_file), delimiter=',')
|
|
for config_data in config_file_reader:
|
|
line_no += 1
|
|
if config_data[1] not in valid_datatypes:
|
|
raise SystemExit("Oops...config file: %s has invalid datatype at line no:%s\n`"
|
|
% (str(input_config_file), str(line_no)))
|
|
if 'namespace' not in config_data:
|
|
if config_data[2] not in valid_encodings:
|
|
raise SystemExit("Oops...config file: %s has invalid encoding at line no:%s\n`"
|
|
% (str(input_config_file), str(line_no)))
|
|
|
|
|
|
def verify_file_data_count(input_config_file, keys_repeat):
|
|
""" Verify count of data on each line in config file is equal to 3
|
|
(as format must be: <key,type and encoding>)
|
|
"""
|
|
line_no = 0
|
|
config_file = open(input_config_file, 'r')
|
|
config_file_reader = csv.reader(filter(lambda row: row[0] != '#',config_file), delimiter=',')
|
|
for line in config_file_reader:
|
|
line_no += 1
|
|
if len(line) != 3 and line[0] not in keys_repeat:
|
|
raise SystemExit("Oops...data missing in config file at line no:%s <format needed:key,type,encoding>\n"
|
|
% str(line_no))
|
|
config_file.close()
|
|
|
|
|
|
def verify_data_in_file(input_config_file, input_values_file, config_file_keys, keys_in_values_file, keys_repeat):
|
|
""" Verify count of data on each line in config file is equal to 3 \
|
|
(as format must be: <key,type and encoding>)
|
|
Verify datatype and encodings from config file is valid
|
|
Verify all keys from config file are present in values file and \
|
|
Verify each key has corresponding value in values file
|
|
"""
|
|
try:
|
|
values_file_keys = []
|
|
|
|
verify_file_data_count(input_config_file, keys_repeat)
|
|
|
|
verify_datatype_encoding(input_config_file)
|
|
|
|
# Get keys from values file present in config files
|
|
values_file_keys = get_keys(keys_in_values_file, config_file_keys)
|
|
|
|
verify_keys_exist(values_file_keys, input_config_file)
|
|
|
|
verify_values_exist(input_values_file, keys_in_values_file)
|
|
|
|
except Exception as err:
|
|
print(err)
|
|
raise
|
|
|
|
|
|
def get_keys(keys_in_values_file, config_file_keys):
|
|
""" Get keys from values file present in config file
|
|
"""
|
|
values_file_keys = []
|
|
for key in range(len(keys_in_values_file)):
|
|
if keys_in_values_file[key] in config_file_keys:
|
|
values_file_keys.append(keys_in_values_file[key])
|
|
|
|
return values_file_keys
|
|
|
|
|
|
def add_config_data_per_namespace(input_config_file):
|
|
""" Add config data per namespace to `config_data_to_write` list
|
|
"""
|
|
config_data_to_write = []
|
|
config_data_per_namespace = []
|
|
|
|
csv_config_file = open(input_config_file,'r')
|
|
config_file_reader = csv.reader(filter(lambda row: row[0] != '#',csv_config_file), delimiter=',')
|
|
|
|
# `config_data_per_namespace` is added to `config_data_to_write` list after reading next namespace
|
|
for config_data in config_file_reader:
|
|
if 'REPEAT' in config_data:
|
|
config_data.remove('REPEAT')
|
|
if 'namespace' in config_data:
|
|
if config_data_per_namespace:
|
|
config_data_to_write.append(config_data_per_namespace)
|
|
config_data_per_namespace = []
|
|
config_data_per_namespace.append(config_data)
|
|
else:
|
|
config_data_per_namespace.append(config_data)
|
|
else:
|
|
config_data_per_namespace.append(config_data)
|
|
|
|
# `config_data_per_namespace` is added to `config_data_to_write` list as EOF is reached
|
|
if (not config_data_to_write) or (config_data_to_write and config_data_per_namespace):
|
|
config_data_to_write.append(config_data_per_namespace)
|
|
|
|
csv_config_file.close()
|
|
|
|
return config_data_to_write
|
|
|
|
|
|
def get_fileid_val(file_identifier, keys_in_config_file, keys_in_values_file,
|
|
values_data_line, key_value_data, fileid_value):
|
|
""" Get file identifier value
|
|
"""
|
|
file_id_found = False
|
|
|
|
for key in key_value_data:
|
|
if file_identifier and not file_id_found and file_identifier in key:
|
|
fileid_value = key[1]
|
|
file_id_found = True
|
|
|
|
if not file_id_found:
|
|
fileid_value = str(int(fileid_value) + 1)
|
|
|
|
return fileid_value
|
|
|
|
|
|
def add_data_to_file(config_data_to_write, key_value_pair, output_csv_file):
|
|
""" Add data to csv target file
|
|
"""
|
|
header = ['key', 'type', 'encoding', 'value']
|
|
data_to_write = []
|
|
|
|
target_csv_file = open(output_csv_file, 'w')
|
|
|
|
output_file_writer = csv.writer(target_csv_file, delimiter=',')
|
|
output_file_writer.writerow(header)
|
|
|
|
for namespace_config_data in config_data_to_write:
|
|
for data in namespace_config_data:
|
|
data_to_write = data[:]
|
|
if 'namespace' in data:
|
|
data_to_write.append('')
|
|
output_file_writer.writerow(data_to_write)
|
|
else:
|
|
key = data[0]
|
|
while key not in key_value_pair[0]:
|
|
del key_value_pair[0]
|
|
if key in key_value_pair[0]:
|
|
value = key_value_pair[0][1]
|
|
data_to_write.append(value)
|
|
del key_value_pair[0]
|
|
output_file_writer.writerow(data_to_write)
|
|
|
|
# Set index to start of file
|
|
target_csv_file.seek(0)
|
|
|
|
target_csv_file.close()
|
|
|
|
|
|
def create_dir(filetype, output_dir_path):
|
|
""" Create new directory(if doesn't exist) to store file generated
|
|
"""
|
|
output_target_dir = output_dir_path + filetype
|
|
if not os.path.isdir(output_target_dir):
|
|
distutils.dir_util.mkpath(output_target_dir)
|
|
|
|
return output_target_dir
|
|
|
|
|
|
def set_repeat_value(total_keys_repeat, keys, csv_file, target_filename):
|
|
key_val_pair = []
|
|
key_repeated = []
|
|
with open(csv_file, 'r') as read_from, open(target_filename,'w') as write_to:
|
|
csv_file_reader = csv.reader(read_from, delimiter=',')
|
|
headers = next(csv_file_reader)
|
|
values = next(csv_file_reader)
|
|
csv_file_writer = csv.writer(write_to, delimiter=',')
|
|
csv_file_writer.writerow(headers)
|
|
csv_file_writer.writerow(values)
|
|
total_keys_values = list(zip_longest(keys, values))
|
|
|
|
# read new data, add value if key has repeat tag, write to new file
|
|
for row in csv_file_reader:
|
|
index = -1
|
|
key_val_new = list(zip_longest(keys, row))
|
|
key_val_pair = total_keys_values[:]
|
|
key_repeated = total_keys_repeat[:]
|
|
while key_val_new and key_repeated:
|
|
index = index + 1
|
|
# if key has repeat tag, get its corresponding value, write to file
|
|
if key_val_new[0][0] == key_repeated[0]:
|
|
val = key_val_pair[0][1]
|
|
row[index] = val
|
|
del key_repeated[0]
|
|
del key_val_new[0]
|
|
del key_val_pair[0]
|
|
csv_file_writer.writerow(row)
|
|
|
|
return target_filename
|
|
|
|
|
|
def main(input_config_file=None,input_values_file=None,target_file_name_prefix=None,
|
|
file_identifier=None,output_dir_path=None,part_size=None,input_version=None,
|
|
input_is_keygen=None,input_is_encrypt=None,input_is_keyfile=None):
|
|
try:
|
|
if all(arg is None for arg in [input_config_file,input_values_file,target_file_name_prefix,
|
|
file_identifier,output_dir_path]):
|
|
parser = argparse.ArgumentParser(prog='./mfg_gen.py',
|
|
description="Create binary files from input config and values file",
|
|
formatter_class=argparse.RawDescriptionHelpFormatter)
|
|
|
|
parser.add_argument('--conf',
|
|
dest='config_file',
|
|
help='the input configuration csv file',
|
|
default=None)
|
|
|
|
parser.add_argument('--values',
|
|
dest='values_file',
|
|
help='the input values csv file',
|
|
default=None)
|
|
|
|
parser.add_argument('--prefix',
|
|
dest='prefix',
|
|
help='the unique name as each filename prefix')
|
|
|
|
parser.add_argument('--fileid',
|
|
dest='fileid',
|
|
help='the unique file identifier(any key in values file) \
|
|
as each filename suffix (Default: numeric value(1,2,3...)')
|
|
|
|
parser.add_argument('--outdir',
|
|
dest='outdir',
|
|
default=os.getcwd(),
|
|
help='the output directory to store the files created\
|
|
(Default: current directory)')
|
|
|
|
parser.add_argument("--size",
|
|
dest='part_size',
|
|
help='Size of NVS Partition in bytes (must be multiple of 4096)')
|
|
|
|
parser.add_argument("--version",
|
|
dest="version",
|
|
help='Set version. Default: v2',
|
|
choices=['v1','v2'],
|
|
default='v2',
|
|
type=str.lower)
|
|
|
|
parser.add_argument("--keygen",
|
|
dest="keygen",
|
|
help='Generate keys for encryption. Default: false',
|
|
choices=['true','false'],
|
|
default='false',
|
|
type=str.lower)
|
|
|
|
parser.add_argument("--encrypt",
|
|
dest="encrypt",
|
|
help='Set encryption mode. Default: false',
|
|
choices=['true','false'],
|
|
default='false',
|
|
type=str.lower)
|
|
|
|
parser.add_argument("--keyfile",
|
|
dest="keyfile",
|
|
help='File having key for encryption (Applicable only if encryption mode is true)',
|
|
default=None)
|
|
|
|
args = parser.parse_args()
|
|
|
|
args.outdir = os.path.join(args.outdir, '')
|
|
|
|
input_config_file = args.config_file
|
|
input_values_file = args.values_file
|
|
target_file_name_prefix = args.prefix
|
|
output_dir_path = args.outdir
|
|
part_size = args.part_size
|
|
input_version = args.version
|
|
input_is_keygen = args.keygen
|
|
input_is_encrypt = args.encrypt
|
|
input_is_keyfile = args.keyfile
|
|
file_identifier = ''
|
|
print_arg_str = "Invalid.\nTo generate binary --conf, --values, --prefix and --size arguments are mandatory.\
|
|
\nTo generate encryption keys --keygen argument is mandatory."
|
|
print_encrypt_arg_str = "Missing parameter. Enter --keygen or --keyfile."
|
|
|
|
if args.fileid:
|
|
file_identifier = args.fileid
|
|
|
|
if input_config_file and input_is_encrypt.lower() == 'true' and input_is_keygen.lower() == 'true' and input_is_keyfile:
|
|
sys.exit('Invalid. Cannot provide both --keygen and --keyfile argument together.')
|
|
|
|
nvs_partition_gen.check_input_args(input_config_file, input_values_file, part_size, input_is_keygen,
|
|
input_is_encrypt, input_is_keyfile, input_version, print_arg_str,
|
|
print_encrypt_arg_str, output_dir_path)
|
|
|
|
if not input_config_file and input_is_keygen:
|
|
if input_is_encrypt == 'true':
|
|
sys.exit("Invalid.\nOnly --keyfile and --outdir arguments allowed.\n")
|
|
# Generate Key Only
|
|
nvs_partition_gen.nvs_part_gen(input_filename=input_config_file, output_filename=input_values_file,
|
|
input_part_size=part_size, is_key_gen=input_is_keygen,
|
|
encrypt_mode=input_is_encrypt, key_file=input_is_keyfile,
|
|
version_no=input_version, output_dir=output_dir_path)
|
|
exit(0)
|
|
|
|
if not (input_config_file and input_values_file and target_file_name_prefix and part_size):
|
|
sys.exit(print_arg_str)
|
|
|
|
keys_in_values_file = []
|
|
keys_in_config_file = []
|
|
config_data_to_write = []
|
|
key_value_data = []
|
|
csv_file_list = []
|
|
keys_repeat = []
|
|
is_empty_line = False
|
|
files_created = False
|
|
file_identifier_value = '0'
|
|
output_target_dir = ''
|
|
target_values_file = None
|
|
output_file_prefix = None
|
|
|
|
# Verify config file is not empty
|
|
if os.stat(input_config_file).st_size == 0:
|
|
raise SystemExit("Oops...config file: %s is empty." % input_config_file)
|
|
|
|
# Verify values file is not empty
|
|
if os.stat(input_values_file).st_size == 0:
|
|
raise SystemExit("Oops...values file: %s is empty." % input_values_file)
|
|
|
|
# Verify config file does not have empty lines
|
|
csv_config_file = open(input_config_file,'r')
|
|
try:
|
|
config_file_reader = csv.reader(csv_config_file, delimiter=',')
|
|
for config_data in config_file_reader:
|
|
for data in config_data:
|
|
empty_line = data.strip()
|
|
if empty_line is '':
|
|
is_empty_line = True
|
|
else:
|
|
is_empty_line = False
|
|
break
|
|
if is_empty_line:
|
|
raise SystemExit("Oops...config file: %s cannot have empty lines. " % input_config_file)
|
|
if not config_data:
|
|
raise SystemExit("Oops...config file: %s cannot have empty lines." % input_config_file)
|
|
|
|
csv_config_file.seek(0)
|
|
|
|
# Extract keys from config file
|
|
for config_data in config_file_reader:
|
|
if 'namespace' not in config_data:
|
|
keys_in_config_file.append(config_data[0])
|
|
if 'REPEAT' in config_data:
|
|
keys_repeat.append(config_data[0])
|
|
|
|
csv_config_file.close()
|
|
except Exception as e:
|
|
print(e)
|
|
finally:
|
|
csv_config_file.close()
|
|
|
|
is_empty_line = False
|
|
# Verify values file does not have empty lines
|
|
csv_values_file = open(input_values_file, 'r')
|
|
try:
|
|
values_file_reader = csv.reader(csv_values_file, delimiter=',')
|
|
for values_data in values_file_reader:
|
|
for data in values_data:
|
|
empty_line = data.strip()
|
|
if empty_line is '':
|
|
is_empty_line = True
|
|
else:
|
|
is_empty_line = False
|
|
break
|
|
if is_empty_line:
|
|
raise SystemExit("Oops...values file: %s cannot have empty lines." % input_values_file)
|
|
if not values_data:
|
|
raise SystemExit("Oops...values file: %s cannot have empty lines." % input_values_file)
|
|
|
|
csv_values_file.seek(0)
|
|
|
|
# Extract keys from values file
|
|
keys_in_values_file = next(values_file_reader)
|
|
|
|
csv_values_file.close()
|
|
except Exception as e:
|
|
print(e)
|
|
exit(1)
|
|
finally:
|
|
csv_values_file.close()
|
|
|
|
# Verify file identifier exists in values file
|
|
if file_identifier:
|
|
if file_identifier not in keys_in_values_file:
|
|
raise SystemExit('Oops...target_file_identifier: %s does not exist in values file.\n' % file_identifier)
|
|
|
|
# Verify data in the input_config_file and input_values_file
|
|
verify_data_in_file(input_config_file, input_values_file, keys_in_config_file,
|
|
keys_in_values_file, keys_repeat)
|
|
|
|
# Add config data per namespace to `config_data_to_write` list
|
|
config_data_to_write = add_config_data_per_namespace(input_config_file)
|
|
|
|
try:
|
|
with open(input_values_file, 'r') as csv_values_file:
|
|
values_file_reader = csv.reader(csv_values_file, delimiter=',')
|
|
keys = next(values_file_reader)
|
|
|
|
filename, file_ext = os.path.splitext(input_values_file)
|
|
target_filename = filename + "_created" + file_ext
|
|
if keys_repeat:
|
|
target_values_file = set_repeat_value(keys_repeat, keys, input_values_file, target_filename)
|
|
else:
|
|
target_values_file = input_values_file
|
|
|
|
csv_values_file = open(target_values_file, 'r')
|
|
|
|
values_file_reader = csv.reader(csv_values_file, delimiter=',')
|
|
next(values_file_reader)
|
|
for values_data_line in values_file_reader:
|
|
key_value_data = list(zip_longest(keys_in_values_file,values_data_line))
|
|
|
|
# Get file identifier value from values file
|
|
file_identifier_value = get_fileid_val(file_identifier, keys_in_config_file,
|
|
keys_in_values_file, values_data_line, key_value_data, file_identifier_value)
|
|
|
|
key_value_pair = key_value_data[:]
|
|
|
|
# Create new directory(if doesn't exist) to store csv file generated
|
|
output_target_dir = create_dir("csv/", output_dir_path)
|
|
|
|
# Verify if output csv file does not exist
|
|
csv_filename = target_file_name_prefix + "-" + file_identifier_value + ".csv"
|
|
csv_file_list.append(csv_filename)
|
|
output_csv_file = output_target_dir + csv_filename
|
|
if os.path.isfile(output_csv_file):
|
|
raise SystemExit("Target csv file: %s already exists.`" % output_csv_file)
|
|
|
|
# Add values corresponding to each key to csv target file
|
|
add_data_to_file(config_data_to_write, key_value_pair, output_csv_file)
|
|
|
|
# Create new directory(if doesn't exist) to store bin file generated
|
|
output_target_dir = create_dir("bin/", output_dir_path)
|
|
|
|
# Verify if output bin file does not exist
|
|
output_file_prefix = target_file_name_prefix + "-" + file_identifier_value
|
|
output_bin_file = output_target_dir + output_file_prefix + ".bin"
|
|
if os.path.isfile(output_bin_file):
|
|
raise SystemExit("Target csv file: %s already exists.`" % output_bin_file)
|
|
|
|
# Create output csv and bin file
|
|
if input_is_keygen.lower() == 'true' and input_is_keyfile:
|
|
input_is_keyfile = os.path.basename(input_is_keyfile)
|
|
nvs_partition_gen.nvs_part_gen(input_filename=output_csv_file, output_filename=output_bin_file,
|
|
input_part_size=part_size, is_key_gen=input_is_keygen,
|
|
encrypt_mode=input_is_encrypt, key_file=input_is_keyfile,
|
|
version_no=input_version, encr_key_prefix=output_file_prefix, output_dir=output_dir_path)
|
|
print("CSV Generated: ", str(output_csv_file))
|
|
|
|
files_created = True
|
|
|
|
csv_values_file.close()
|
|
except Exception as e:
|
|
print(e)
|
|
exit(1)
|
|
finally:
|
|
csv_values_file.close()
|
|
return csv_file_list, files_created, target_values_file
|
|
|
|
except ValueError as err:
|
|
print(err)
|
|
except Exception:
|
|
raise
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|