diff options
Diffstat (limited to 'test/lint')
| -rwxr-xr-x | test/lint/lint-include-guards.py | 100 | ||||
| -rwxr-xr-x | test/lint/lint-include-guards.sh | 30 | ||||
| -rwxr-xr-x | test/lint/lint-locale-dependence.py | 259 | ||||
| -rwxr-xr-x | test/lint/lint-locale-dependence.sh | 241 | ||||
| -rwxr-xr-x | test/lint/lint-python-utf8-encoding.py | 73 | ||||
| -rwxr-xr-x | test/lint/lint-python-utf8-encoding.sh | 28 | ||||
| -rwxr-xr-x | test/lint/lint-submodule.py | 23 | ||||
| -rwxr-xr-x | test/lint/lint-submodule.sh | 20 |
8 files changed, 455 insertions, 319 deletions
diff --git a/test/lint/lint-include-guards.py b/test/lint/lint-include-guards.py new file mode 100755 index 0000000000..86284517d5 --- /dev/null +++ b/test/lint/lint-include-guards.py @@ -0,0 +1,100 @@ +#!/usr/bin/env python3 +# +# Copyright (c) 2018-2022 The Bitcoin Core developers +# Distributed under the MIT software license, see the accompanying +# file COPYING or http://www.opensource.org/licenses/mit-license.php. + +""" +Check include guards. +""" + +import re +import sys +from subprocess import check_output +from typing import List + + +HEADER_ID_PREFIX = 'BITCOIN_' +HEADER_ID_SUFFIX = '_H' + +EXCLUDE_FILES_WITH_PREFIX = ['src/crypto/ctaes', + 'src/leveldb', + 'src/crc32c', + 'src/secp256k1', + 'src/minisketch', + 'src/univalue', + 'src/tinyformat.h', + 'src/bench/nanobench.h', + 'src/test/fuzz/FuzzedDataProvider.h'] + + +def _get_header_file_lst() -> List[str]: + """ Helper function to get a list of header filepaths to be + checked for include guards. + """ + git_cmd_lst = ['git', 'ls-files', '--', '*.h'] + header_file_lst = check_output( + git_cmd_lst).decode('utf-8').splitlines() + + header_file_lst = [hf for hf in header_file_lst + if not any(ef in hf for ef + in EXCLUDE_FILES_WITH_PREFIX)] + + return header_file_lst + + +def _get_header_id(header_file: str) -> str: + """ Helper function to get the header id from a header file + string. + + eg: 'src/wallet/walletdb.h' -> 'BITCOIN_WALLET_WALLETDB_H' + + Args: + header_file: Filepath to header file. + + Returns: + The header id. + """ + header_id_base = header_file.split('/')[1:] + header_id_base = '_'.join(header_id_base) + header_id_base = header_id_base.replace('.h', '').replace('-', '_') + header_id_base = header_id_base.upper() + + header_id = f'{HEADER_ID_PREFIX}{header_id_base}{HEADER_ID_SUFFIX}' + + return header_id + + +def main(): + exit_code = 0 + + header_file_lst = _get_header_file_lst() + for header_file in header_file_lst: + header_id = _get_header_id(header_file) + + regex_pattern = f'^#(ifndef|define|endif //) {header_id}' + + with open(header_file, 'r', encoding='utf-8') as f: + header_file_contents = f.readlines() + + count = 0 + for header_file_contents_string in header_file_contents: + include_guard_lst = re.findall( + regex_pattern, header_file_contents_string) + + count += len(include_guard_lst) + + if count != 3: + print(f'{header_file} seems to be missing the expected ' + 'include guard:') + print(f' #ifndef {header_id}') + print(f' #define {header_id}') + print(' ...') + print(f' #endif // {header_id}\n') + exit_code = 1 + + sys.exit(exit_code) + + +if __name__ == '__main__': + main() diff --git a/test/lint/lint-include-guards.sh b/test/lint/lint-include-guards.sh deleted file mode 100755 index f14218aa74..0000000000 --- a/test/lint/lint-include-guards.sh +++ /dev/null @@ -1,30 +0,0 @@ -#!/usr/bin/env bash -# -# Copyright (c) 2018-2021 The Bitcoin Core developers -# Distributed under the MIT software license, see the accompanying -# file COPYING or http://www.opensource.org/licenses/mit-license.php. -# -# Check include guards. - -export LC_ALL=C -HEADER_ID_PREFIX="BITCOIN_" -HEADER_ID_SUFFIX="_H" - -REGEXP_EXCLUDE_FILES_WITH_PREFIX="src/(crypto/ctaes/|leveldb/|crc32c/|secp256k1/|minisketch/|test/fuzz/FuzzedDataProvider.h|tinyformat.h|bench/nanobench.h|univalue/)" - -EXIT_CODE=0 -for HEADER_FILE in $(git ls-files -- "*.h" | grep -vE "^${REGEXP_EXCLUDE_FILES_WITH_PREFIX}") -do - HEADER_ID_BASE=$(cut -f2- -d/ <<< "${HEADER_FILE}" | sed "s/\.h$//g" | tr / _ | tr - _ | tr "[:lower:]" "[:upper:]") - HEADER_ID="${HEADER_ID_PREFIX}${HEADER_ID_BASE}${HEADER_ID_SUFFIX}" - if [[ $(grep --count --extended-regexp "^#(ifndef|define|endif //) ${HEADER_ID}" "${HEADER_FILE}") != 3 ]]; then - echo "${HEADER_FILE} seems to be missing the expected include guard:" - echo " #ifndef ${HEADER_ID}" - echo " #define ${HEADER_ID}" - echo " ..." - echo " #endif // ${HEADER_ID}" - echo - EXIT_CODE=1 - fi -done -exit ${EXIT_CODE} diff --git a/test/lint/lint-locale-dependence.py b/test/lint/lint-locale-dependence.py new file mode 100755 index 0000000000..2abf1be6b3 --- /dev/null +++ b/test/lint/lint-locale-dependence.py @@ -0,0 +1,259 @@ +#!/usr/bin/env python3 +# Copyright (c) 2018-2022 The Bitcoin Core developers +# Distributed under the MIT software license, see the accompanying +# file COPYING or http://www.opensource.org/licenses/mit-license.php. +# +# Be aware that bitcoind and bitcoin-qt differ in terms of localization: Qt +# opts in to POSIX localization by running setlocale(LC_ALL, "") on startup, +# whereas no such call is made in bitcoind. +# +# Qt runs setlocale(LC_ALL, "") on initialization. This installs the locale +# specified by the user's LC_ALL (or LC_*) environment variable as the new +# C locale. +# +# In contrast, bitcoind does not opt in to localization -- no call to +# setlocale(LC_ALL, "") is made and the environment variables LC_* are +# thus ignored. +# +# This results in situations where bitcoind is guaranteed to be running +# with the classic locale ("C") whereas the locale of bitcoin-qt will vary +# depending on the user's environment variables. +# +# An example: Assuming the environment variable LC_ALL=de_DE then the +# call std::to_string(1.23) will return "1.230000" in bitcoind but +# "1,230000" in bitcoin-qt. +# +# From the Qt documentation: +# "On Unix/Linux Qt is configured to use the system locale settings by default. +# This can cause a conflict when using POSIX functions, for instance, when +# converting between data types such as floats and strings, since the notation +# may differ between locales. To get around this problem, call the POSIX function +# setlocale(LC_NUMERIC,"C") right after initializing QApplication, QGuiApplication +# or QCoreApplication to reset the locale that is used for number formatting to +# "C"-locale." +# +# See https://doc.qt.io/qt-5/qcoreapplication.html#locale-settings and +# https://stackoverflow.com/a/34878283 for more details. +# +# TODO: Reduce KNOWN_VIOLATIONS by replacing uses of locale dependent snprintf with strprintf. + +import re +import sys + +from subprocess import check_output, CalledProcessError + + +KNOWN_VIOLATIONS = [ + "src/dbwrapper.cpp:.*vsnprintf", + "src/test/dbwrapper_tests.cpp:.*snprintf", + "src/test/fuzz/locale.cpp:.*setlocale", + "src/test/fuzz/string.cpp:.*strtol", + "src/test/fuzz/string.cpp:.*strtoul", + "src/test/util_tests.cpp:.*strtoll" +] + +REGEXP_EXTERNAL_DEPENDENCIES_EXCLUSIONS = [ + "src/crypto/ctaes/", + "src/leveldb/", + "src/secp256k1/", + "src/minisketch/", + "src/tinyformat.h", + "src/univalue/" +] + +LOCALE_DEPENDENT_FUNCTIONS = [ + "alphasort", # LC_COLLATE (via strcoll) + "asctime", # LC_TIME (directly) + "asprintf", # (via vasprintf) + "atof", # LC_NUMERIC (via strtod) + "atoi", # LC_NUMERIC (via strtol) + "atol", # LC_NUMERIC (via strtol) + "atoll", # (via strtoll) + "atoq", + "btowc", # LC_CTYPE (directly) + "ctime", # (via asctime or localtime) + "dprintf", # (via vdprintf) + "fgetwc", + "fgetws", + "fold_case", # boost::locale::fold_case + "fprintf", # (via vfprintf) + "fputwc", + "fputws", + "fscanf", # (via __vfscanf) + "fwprintf", # (via __vfwprintf) + "getdate", # via __getdate_r => isspace // __localtime_r + "getwc", + "getwchar", + "is_digit", # boost::algorithm::is_digit + "is_space", # boost::algorithm::is_space + "isalnum", # LC_CTYPE + "isalpha", # LC_CTYPE + "isblank", # LC_CTYPE + "iscntrl", # LC_CTYPE + "isctype", # LC_CTYPE + "isdigit", # LC_CTYPE + "isgraph", # LC_CTYPE + "islower", # LC_CTYPE + "isprint", # LC_CTYPE + "ispunct", # LC_CTYPE + "isspace", # LC_CTYPE + "isupper", # LC_CTYPE + "iswalnum", # LC_CTYPE + "iswalpha", # LC_CTYPE + "iswblank", # LC_CTYPE + "iswcntrl", # LC_CTYPE + "iswctype", # LC_CTYPE + "iswdigit", # LC_CTYPE + "iswgraph", # LC_CTYPE + "iswlower", # LC_CTYPE + "iswprint", # LC_CTYPE + "iswpunct", # LC_CTYPE + "iswspace", # LC_CTYPE + "iswupper", # LC_CTYPE + "iswxdigit", # LC_CTYPE + "isxdigit", # LC_CTYPE + "localeconv", # LC_NUMERIC + LC_MONETARY + "mblen", # LC_CTYPE + "mbrlen", + "mbrtowc", + "mbsinit", + "mbsnrtowcs", + "mbsrtowcs", + "mbstowcs", # LC_CTYPE + "mbtowc", # LC_CTYPE + "mktime", + "normalize", # boost::locale::normalize + "printf", # LC_NUMERIC + "putwc", + "putwchar", + "scanf", # LC_NUMERIC + "setlocale", + "snprintf", + "sprintf", + "sscanf", + "std::locale::global", + "std::to_string", + "stod", + "stof", + "stoi", + "stol", + "stold", + "stoll", + "stoul", + "stoull", + "strcasecmp", + "strcasestr", + "strcoll", # LC_COLLATE + #"strerror", + "strfmon", + "strftime", # LC_TIME + "strncasecmp", + "strptime", + "strtod", # LC_NUMERIC + "strtof", + "strtoimax", + "strtol", # LC_NUMERIC + "strtold", + "strtoll", + "strtoq", + "strtoul", # LC_NUMERIC + "strtoull", + "strtoumax", + "strtouq", + "strxfrm", # LC_COLLATE + "swprintf", + "to_lower", # boost::locale::to_lower + "to_title", # boost::locale::to_title + "to_upper", # boost::locale::to_upper + "tolower", # LC_CTYPE + "toupper", # LC_CTYPE + "towctrans", + "towlower", # LC_CTYPE + "towupper", # LC_CTYPE + "trim", # boost::algorithm::trim + "trim_left", # boost::algorithm::trim_left + "trim_right", # boost::algorithm::trim_right + "ungetwc", + "vasprintf", + "vdprintf", + "versionsort", + "vfprintf", + "vfscanf", + "vfwprintf", + "vprintf", + "vscanf", + "vsnprintf", + "vsprintf", + "vsscanf", + "vswprintf", + "vwprintf", + "wcrtomb", + "wcscasecmp", + "wcscoll", # LC_COLLATE + "wcsftime", # LC_TIME + "wcsncasecmp", + "wcsnrtombs", + "wcsrtombs", + "wcstod", # LC_NUMERIC + "wcstof", + "wcstoimax", + "wcstol", # LC_NUMERIC + "wcstold", + "wcstoll", + "wcstombs", # LC_CTYPE + "wcstoul", # LC_NUMERIC + "wcstoull", + "wcstoumax", + "wcswidth", + "wcsxfrm", # LC_COLLATE + "wctob", + "wctomb", # LC_CTYPE + "wctrans", + "wctype", + "wcwidth", + "wprintf" +] + + +def find_locale_dependent_function_uses(): + regexp_locale_dependent_functions = "|".join(LOCALE_DEPENDENT_FUNCTIONS) + exclude_args = [":(exclude)" + excl for excl in REGEXP_EXTERNAL_DEPENDENCIES_EXCLUSIONS] + git_grep_command = ["git", "grep", "-E", "[^a-zA-Z0-9_\\`'\"<>](" + regexp_locale_dependent_functions + "(_r|_s)?)[^a-zA-Z0-9_\\`'\"<>]", "--", "*.cpp", "*.h"] + exclude_args + git_grep_output = list() + + try: + git_grep_output = check_output(git_grep_command, universal_newlines=True, encoding="utf8").splitlines() + except CalledProcessError as e: + if e.returncode > 1: + raise e + + return git_grep_output + + +def main(): + exit_code = 0 + + regexp_ignore_known_violations = "|".join(KNOWN_VIOLATIONS) + git_grep_output = find_locale_dependent_function_uses() + + for locale_dependent_function in LOCALE_DEPENDENT_FUNCTIONS: + matches = [line for line in git_grep_output + if re.search("[^a-zA-Z0-9_\\`'\"<>]" + locale_dependent_function + "(_r|_s)?[^a-zA-Z0-9_\\`'\"<>]", line) + and not re.search("\\.(c|cpp|h):\\s*(//|\\*|/\\*|\").*" + locale_dependent_function, line) + and not re.search(regexp_ignore_known_violations, line)] + if matches: + print(f"The locale dependent function {locale_dependent_function}(...) appears to be used:") + for match in matches: + print(match) + print("") + exit_code = 1 + + if exit_code == 1: + print("Unnecessary locale depedence can cause bugs that are very tricky to isolate and fix. Please avoid using locale dependent functions if possible.\n") + print(f"Advice not applicable in this specific case? Add an exception by updating the ignore list in {sys.argv[0]}") + + sys.exit(exit_code) + + +if __name__ == "__main__": + main() diff --git a/test/lint/lint-locale-dependence.sh b/test/lint/lint-locale-dependence.sh deleted file mode 100755 index 7d608eed6a..0000000000 --- a/test/lint/lint-locale-dependence.sh +++ /dev/null @@ -1,241 +0,0 @@ -#!/usr/bin/env bash -# Copyright (c) 2018-2021 The Bitcoin Core developers -# Distributed under the MIT software license, see the accompanying -# file COPYING or http://www.opensource.org/licenses/mit-license.php. - -export LC_ALL=C - -# Be aware that bitcoind and bitcoin-qt differ in terms of localization: Qt -# opts in to POSIX localization by running setlocale(LC_ALL, "") on startup, -# whereas no such call is made in bitcoind. -# -# Qt runs setlocale(LC_ALL, "") on initialization. This installs the locale -# specified by the user's LC_ALL (or LC_*) environment variable as the new -# C locale. -# -# In contrast, bitcoind does not opt in to localization -- no call to -# setlocale(LC_ALL, "") is made and the environment variables LC_* are -# thus ignored. -# -# This results in situations where bitcoind is guaranteed to be running -# with the classic locale ("C") whereas the locale of bitcoin-qt will vary -# depending on the user's environment variables. -# -# An example: Assuming the environment variable LC_ALL=de_DE then the -# call std::to_string(1.23) will return "1.230000" in bitcoind but -# "1,230000" in bitcoin-qt. -# -# From the Qt documentation: -# "On Unix/Linux Qt is configured to use the system locale settings by default. -# This can cause a conflict when using POSIX functions, for instance, when -# converting between data types such as floats and strings, since the notation -# may differ between locales. To get around this problem, call the POSIX function -# setlocale(LC_NUMERIC,"C") right after initializing QApplication, QGuiApplication -# or QCoreApplication to reset the locale that is used for number formatting to -# "C"-locale." -# -# See https://doc.qt.io/qt-5/qcoreapplication.html#locale-settings and -# https://stackoverflow.com/a/34878283 for more details. - -# TODO: Reduce KNOWN_VIOLATIONS by replacing uses of locale dependent snprintf with strprintf. -KNOWN_VIOLATIONS=( - "src/dbwrapper.cpp:.*vsnprintf" - "src/test/dbwrapper_tests.cpp:.*snprintf" - "src/test/fuzz/locale.cpp" - "src/test/fuzz/string.cpp" - "src/test/util_tests.cpp" -) - -REGEXP_IGNORE_EXTERNAL_DEPENDENCIES="^src/(crypto/ctaes/|leveldb/|secp256k1/|minisketch/|tinyformat.h|univalue/)" - -LOCALE_DEPENDENT_FUNCTIONS=( - alphasort # LC_COLLATE (via strcoll) - asctime # LC_TIME (directly) - asprintf # (via vasprintf) - atof # LC_NUMERIC (via strtod) - atoi # LC_NUMERIC (via strtol) - atol # LC_NUMERIC (via strtol) - atoll # (via strtoll) - atoq - btowc # LC_CTYPE (directly) - ctime # (via asctime or localtime) - dprintf # (via vdprintf) - fgetwc - fgetws - fold_case # boost::locale::fold_case - fprintf # (via vfprintf) - fputwc - fputws - fscanf # (via __vfscanf) - fwprintf # (via __vfwprintf) - getdate # via __getdate_r => isspace // __localtime_r - getwc - getwchar - is_digit # boost::algorithm::is_digit - is_space # boost::algorithm::is_space - isalnum # LC_CTYPE - isalpha # LC_CTYPE - isblank # LC_CTYPE - iscntrl # LC_CTYPE - isctype # LC_CTYPE - isdigit # LC_CTYPE - isgraph # LC_CTYPE - islower # LC_CTYPE - isprint # LC_CTYPE - ispunct # LC_CTYPE - isspace # LC_CTYPE - isupper # LC_CTYPE - iswalnum # LC_CTYPE - iswalpha # LC_CTYPE - iswblank # LC_CTYPE - iswcntrl # LC_CTYPE - iswctype # LC_CTYPE - iswdigit # LC_CTYPE - iswgraph # LC_CTYPE - iswlower # LC_CTYPE - iswprint # LC_CTYPE - iswpunct # LC_CTYPE - iswspace # LC_CTYPE - iswupper # LC_CTYPE - iswxdigit # LC_CTYPE - isxdigit # LC_CTYPE - localeconv # LC_NUMERIC + LC_MONETARY - mblen # LC_CTYPE - mbrlen - mbrtowc - mbsinit - mbsnrtowcs - mbsrtowcs - mbstowcs # LC_CTYPE - mbtowc # LC_CTYPE - mktime - normalize # boost::locale::normalize - printf # LC_NUMERIC - putwc - putwchar - scanf # LC_NUMERIC - setlocale - snprintf - sprintf - sscanf - std::locale::global - std::to_string - stod - stof - stoi - stol - stold - stoll - stoul - stoull - strcasecmp - strcasestr - strcoll # LC_COLLATE -# strerror - strfmon - strftime # LC_TIME - strncasecmp - strptime - strtod # LC_NUMERIC - strtof - strtoimax - strtol # LC_NUMERIC - strtold - strtoll - strtoq - strtoul # LC_NUMERIC - strtoull - strtoumax - strtouq - strxfrm # LC_COLLATE - swprintf - to_lower # boost::locale::to_lower - to_title # boost::locale::to_title - to_upper # boost::locale::to_upper - tolower # LC_CTYPE - toupper # LC_CTYPE - towctrans - towlower # LC_CTYPE - towupper # LC_CTYPE - trim # boost::algorithm::trim - trim_left # boost::algorithm::trim_left - trim_right # boost::algorithm::trim_right - ungetwc - vasprintf - vdprintf - versionsort - vfprintf - vfscanf - vfwprintf - vprintf - vscanf - vsnprintf - vsprintf - vsscanf - vswprintf - vwprintf - wcrtomb - wcscasecmp - wcscoll # LC_COLLATE - wcsftime # LC_TIME - wcsncasecmp - wcsnrtombs - wcsrtombs - wcstod # LC_NUMERIC - wcstof - wcstoimax - wcstol # LC_NUMERIC - wcstold - wcstoll - wcstombs # LC_CTYPE - wcstoul # LC_NUMERIC - wcstoull - wcstoumax - wcswidth - wcsxfrm # LC_COLLATE - wctob - wctomb # LC_CTYPE - wctrans - wctype - wcwidth - wprintf -) - -function join_array { - local IFS="$1" - shift - echo "$*" -} - -REGEXP_IGNORE_KNOWN_VIOLATIONS=$(join_array "|" "${KNOWN_VIOLATIONS[@]}") - -# Invoke "git grep" only once in order to minimize run-time -REGEXP_LOCALE_DEPENDENT_FUNCTIONS=$(join_array "|" "${LOCALE_DEPENDENT_FUNCTIONS[@]}") -GIT_GREP_OUTPUT=$(git grep -E "[^a-zA-Z0-9_\`'\"<>](${REGEXP_LOCALE_DEPENDENT_FUNCTIONS}(_r|_s)?)[^a-zA-Z0-9_\`'\"<>]" -- "*.cpp" "*.h") - -EXIT_CODE=0 -for LOCALE_DEPENDENT_FUNCTION in "${LOCALE_DEPENDENT_FUNCTIONS[@]}"; do - MATCHES=$(grep -E "[^a-zA-Z0-9_\`'\"<>]${LOCALE_DEPENDENT_FUNCTION}(_r|_s)?[^a-zA-Z0-9_\`'\"<>]" <<< "${GIT_GREP_OUTPUT}" | \ - grep -vE "\.(c|cpp|h):\s*(//|\*|/\*|\").*${LOCALE_DEPENDENT_FUNCTION}") - if [[ ${REGEXP_IGNORE_EXTERNAL_DEPENDENCIES} != "" ]]; then - MATCHES=$(grep -vE "${REGEXP_IGNORE_EXTERNAL_DEPENDENCIES}" <<< "${MATCHES}") - fi - if [[ ${REGEXP_IGNORE_KNOWN_VIOLATIONS} != "" ]]; then - MATCHES=$(grep -vE "${REGEXP_IGNORE_KNOWN_VIOLATIONS}" <<< "${MATCHES}") - fi - if [[ ${MATCHES} != "" ]]; then - echo "The locale dependent function ${LOCALE_DEPENDENT_FUNCTION}(...) appears to be used:" - echo "${MATCHES}" - echo - EXIT_CODE=1 - fi -done -if [[ ${EXIT_CODE} != 0 ]]; then - echo "Unnecessary locale dependence can cause bugs that are very" - echo "tricky to isolate and fix. Please avoid using locale dependent" - echo "functions if possible." - echo - echo "Advice not applicable in this specific case? Add an exception" - echo "by updating the ignore list in $0" -fi -exit ${EXIT_CODE} diff --git a/test/lint/lint-python-utf8-encoding.py b/test/lint/lint-python-utf8-encoding.py new file mode 100755 index 0000000000..62fdc34d50 --- /dev/null +++ b/test/lint/lint-python-utf8-encoding.py @@ -0,0 +1,73 @@ +#!/usr/bin/env python3 +# +# Copyright (c) 2018-2022 The Bitcoin Core developers +# Distributed under the MIT software license, see the accompanying +# file COPYING or http://www.opensource.org/licenses/mit-license.php. +# +# Make sure we explicitly open all text files using UTF-8 (or ASCII) encoding to +# avoid potential issues on the BSDs where the locale is not always set. + +import sys +import re + +from subprocess import check_output, CalledProcessError + +EXCLUDED_DIRS = ["src/crc32c/"] + + +def get_exclude_args(): + return [":(exclude)" + dir for dir in EXCLUDED_DIRS] + + +def check_fileopens(): + fileopens = list() + + try: + fileopens = check_output(["git", "grep", r" open(", "--", "*.py"] + get_exclude_args(), universal_newlines=True, encoding="utf8").splitlines() + except CalledProcessError as e: + if e.returncode > 1: + raise e + + filtered_fileopens = [fileopen for fileopen in fileopens if not re.search(r"encoding=.(ascii|utf8|utf-8).|open\([^,]*, ['\"][^'\"]*b[^'\"]*['\"]", fileopen)] + + return filtered_fileopens + + +def check_checked_outputs(): + checked_outputs = list() + + try: + checked_outputs = check_output(["git", "grep", "check_output(", "--", "*.py"] + get_exclude_args(), universal_newlines=True, encoding="utf8").splitlines() + except CalledProcessError as e: + if e.returncode > 1: + raise e + + filtered_checked_outputs = [checked_output for checked_output in checked_outputs if re.search(r"universal_newlines=True", checked_output) and not re.search(r"encoding=.(ascii|utf8|utf-8).", checked_output)] + + return filtered_checked_outputs + + +def main(): + exit_code = 0 + + nonexplicit_utf8_fileopens = check_fileopens() + if nonexplicit_utf8_fileopens: + print("Python's open(...) seems to be used to open text files without explicitly specifying encoding='utf8':\n") + for fileopen in nonexplicit_utf8_fileopens: + print(fileopen) + exit_code = 1 + + nonexplicit_utf8_checked_outputs = check_checked_outputs() + if nonexplicit_utf8_checked_outputs: + if nonexplicit_utf8_fileopens: + print("\n") + print("Python's check_output(...) seems to be used to get program outputs without explicitly specifying encoding='utf8':\n") + for checked_output in nonexplicit_utf8_checked_outputs: + print(checked_output) + exit_code = 1 + + sys.exit(exit_code) + + +if __name__ == "__main__": + main() diff --git a/test/lint/lint-python-utf8-encoding.sh b/test/lint/lint-python-utf8-encoding.sh deleted file mode 100755 index 6e5b18fc23..0000000000 --- a/test/lint/lint-python-utf8-encoding.sh +++ /dev/null @@ -1,28 +0,0 @@ -#!/usr/bin/env bash -# -# Copyright (c) 2018-2020 The Bitcoin Core developers -# Distributed under the MIT software license, see the accompanying -# file COPYING or http://www.opensource.org/licenses/mit-license.php. -# -# Make sure we explicitly open all text files using UTF-8 (or ASCII) encoding to -# avoid potential issues on the BSDs where the locale is not always set. - -export LC_ALL=C -EXIT_CODE=0 -OUTPUT=$(git grep " open(" -- "*.py" ":(exclude)src/crc32c/" | grep -vE "encoding=.(ascii|utf8|utf-8)." | grep -vE "open\([^,]*, ['\"][^'\"]*b[^'\"]*['\"]") -if [[ ${OUTPUT} != "" ]]; then - echo "Python's open(...) seems to be used to open text files without explicitly" - echo "specifying encoding=\"utf8\":" - echo - echo "${OUTPUT}" - EXIT_CODE=1 -fi -OUTPUT=$(git grep "check_output(" -- "*.py" ":(exclude)src/crc32c/"| grep "universal_newlines=True" | grep -vE "encoding=.(ascii|utf8|utf-8).") -if [[ ${OUTPUT} != "" ]]; then - echo "Python's check_output(...) seems to be used to get program outputs without explicitly" - echo "specifying encoding=\"utf8\":" - echo - echo "${OUTPUT}" - EXIT_CODE=1 -fi -exit ${EXIT_CODE} diff --git a/test/lint/lint-submodule.py b/test/lint/lint-submodule.py new file mode 100755 index 0000000000..89d4c80f55 --- /dev/null +++ b/test/lint/lint-submodule.py @@ -0,0 +1,23 @@ +#!/usr/bin/env python3 +# +# Copyright (c) 2022 The Bitcoin Core developers +# Distributed under the MIT software license, see the accompanying +# file COPYING or http://www.opensource.org/licenses/mit-license.php. + +""" +This script checks for git modules +""" + +import subprocess +import sys + +def main(): + submodules_list = subprocess.check_output(['git', 'submodule', 'status', '--recursive'], + universal_newlines = True, encoding = 'utf8').rstrip('\n') + if submodules_list: + print("These submodules were found, delete them:\n", submodules_list) + sys.exit(1) + sys.exit(0) + +if __name__ == '__main__': + main() diff --git a/test/lint/lint-submodule.sh b/test/lint/lint-submodule.sh deleted file mode 100755 index d9aa021df7..0000000000 --- a/test/lint/lint-submodule.sh +++ /dev/null @@ -1,20 +0,0 @@ -#!/usr/bin/env bash -# -# Copyright (c) 2020 The Bitcoin Core developers -# Distributed under the MIT software license, see the accompanying -# file COPYING or http://www.opensource.org/licenses/mit-license.php. -# -# This script checks for git modules -export LC_ALL=C -EXIT_CODE=0 - -CMD=$(git submodule status --recursive) -if test -n "$CMD"; -then - echo These submodules were found, delete them: - echo "$CMD" - EXIT_CODE=1 -fi - -exit $EXIT_CODE - |