Merge #13448: Add linter: Make sure we explicitly open all text files using UTF-8 encoding in Python

c8176b3cc7556d7bcec39a55ae4d6ba16453baaa Add linter: Make sure we explicitly open all text files using UTF-8 or ASCII encoding in Python (practicalswift)
634bd970013eca90f4b4c1f9044eec8c97ba62c2 Explicitly specify encoding when opening text files in Python code (practicalswift)

Pull request description:

  Add linter: Make sure we explicitly open all text files using UTF-8 encoding in Python.

  As requested by @laanwj in #13440.

Tree-SHA512: 1651c00fe220ceb273324abd6703aee504029b96c7ef0e3029145901762c733c9b9d24927da281394fd4681a5bff774336c04eed01fafea997bb32192c334c06
Signed-off-by: pasta <pasta@dashboost.org>

# Conflicts:
#	contrib/devtools/circular-dependencies.py
#	contrib/linearize/linearize-data.py
#	contrib/linearize/linearize-hashes.py
#	contrib/seeds/generate-seeds.py
#	contrib/verify-commits/verify-commits.py
#	test/functional/multiwallet.py
#	test/functional/notifications.py
#	test/functional/test_runner.py
#	test/util/rpcauth-test.py

contrib/auto_gdb/log_size.py

@@ -23,7 +23,7 @@ class LogSizeCommand (gdb.Command):
         try:
             args = gdb.string_to_argv(arg)
             obj = gdb.parse_and_eval(args[0])
-            logfile = open(args[1], 'a')
+            logfile = open(args[1], 'a', encoding="utf8")
             size = common_helpers.get_instance_size(obj)
             logfile.write("%s %s: %d\n" % (str(datetime.datetime.now()), args[0], size))
             logfile.close()

contrib/devtools/check-rpc-mappings.py

@@ -49,7 +49,7 @@ def process_commands(fname):
     """Find and parse dispatch table in implementation file `fname`."""
     cmds = []
     in_rpcs = False
-    with open(fname, "r") as f:
+    with open(fname, "r", encoding="utf8") as f:
         for line in f:
             line = line.rstrip()
             if not in_rpcs:
@@ -75,7 +75,7 @@ def process_mapping(fname):
     """Find and parse conversion table in implementation file `fname`."""
     cmds = []
     in_rpcs = False
-    with open(fname, "r") as f:
+    with open(fname, "r", encoding="utf8") as f:
         for line in f:
             line = line.rstrip()
             if not in_rpcs:

contrib/devtools/clang-format-diff.py

@@ -152,7 +152,7 @@ def main():
       sys.exit(p.returncode)
 
     if not args.i:
-      with open(filename) as f:
+      with open(filename, encoding="utf8") as f:
         code = f.readlines()
       formatted_code = io.StringIO(stdout).readlines()
       diff = difflib.unified_diff(code, formatted_code,

contrib/devtools/copyright_header.py

@@ -148,7 +148,7 @@ def file_has_without_c_style_copyright_for_holder(contents, holder_name):
 ################################################################################
 
 def read_file(filename):
-    return open(os.path.abspath(filename), 'r').read()
+    return open(os.path.abspath(filename), 'r', encoding="utf8").read()
 
 def gather_file_info(filename):
     info = {}
@@ -327,13 +327,13 @@ def get_most_recent_git_change_year(filename):
 ################################################################################
 
 def read_file_lines(filename):
-    f = open(os.path.abspath(filename), 'r')
+    f = open(os.path.abspath(filename), 'r', encoding="utf8")
     file_lines = f.readlines()
     f.close()
     return file_lines
 
 def write_file_lines(filename, file_lines):
-    f = open(os.path.abspath(filename), 'w')
+    f = open(os.path.abspath(filename), 'w', encoding="utf8")
     f.write(''.join(file_lines))
     f.close()
 

contrib/devtools/github-merge.py

@@ -190,7 +190,7 @@ def main():
     merge_branch = 'pull/'+pull+'/merge'
     local_merge_branch = 'pull/'+pull+'/local-merge'
 
-    devnull = open(os.devnull,'w')
+    devnull = open(os.devnull, 'w', encoding="utf8")
     try:
         subprocess.check_call([GIT,'checkout','-q',branch])
     except subprocess.CalledProcessError as e:

contrib/devtools/test-security-check.py

@@ -9,7 +9,7 @@ import subprocess
 import unittest
 
 def write_testcode(filename):
-    with open(filename, 'w') as f:
+    with open(filename, 'w', encoding="utf8") as f:
         f.write('''
     #include <stdio.h>
     int main()

contrib/filter-lcov.py

@@ -13,8 +13,8 @@ pattern = args.pattern
 outfile = args.outfile
 
 in_remove = False
-with open(tracefile, 'r') as f:
+with open(tracefile, 'r', encoding="utf8") as f:
-    with open(outfile, 'w') as wf:
+    with open(outfile, 'w', encoding="utf8") as wf:
         for line in f:
             for p in pattern:
                 if line.startswith("SF:") and p in line:

contrib/linearize/linearize-data.py

@@ -78,7 +78,7 @@ def get_blk_dt(blk_hdr):
 # When getting the list of block hashes, undo any byte reversals.
 def get_block_hashes(settings):
 	blkindex = []
-	f = open(settings['hashlist'], "r")
+	f = open(settings['hashlist'], "r", encoding="utf8")
 	for line in f:
 		line = line.rstrip()
 		if settings['rev_hash_bytes'] == 'true':
@@ -264,7 +264,7 @@ if __name__ == '__main__':
 		print("Usage: linearize-data.py CONFIG-FILE")
 		sys.exit(1)
 
-	f = open(sys.argv[1])
+	f = open(sys.argv[1], encoding="utf8")
 	for line in f:
 		# skip comment lines
 		m = re.search('^\s*#', line)

contrib/linearize/linearize-hashes.py

@@ -97,7 +97,7 @@ def get_block_hashes(settings, max_blocks_per_call=10000):
 
 def get_rpc_cookie():
 	# Open the cookie file
-	with open(os.path.join(os.path.expanduser(settings['datadir']), '.cookie'), 'r') as f:
+	with open(os.path.join(os.path.expanduser(settings['datadir']), '.cookie'), 'r', encoding="ascii") as f:
 		combined = f.readline()
 		combined_split = combined.split(":")
 		settings['rpcuser'] = combined_split[0]
@@ -108,7 +108,7 @@ if __name__ == '__main__':
 		print("Usage: linearize-hashes.py CONFIG-FILE")
 		sys.exit(1)
 
-	f = open(sys.argv[1])
+	f = open(sys.argv[1], encoding="utf8")
 	for line in f:
 		# skip comment lines
 		m = re.search('^\s*#', line)

contrib/seeds/generate-seeds.py

@@ -11,7 +11,7 @@ argument:
     nodes_main.txt
     nodes_test.txt
 
-These files must consist of lines in the format 
+These files must consist of lines in the format
 
     <ip>
     <ip>:<port>
@@ -126,13 +126,13 @@ def main():
     g.write(' * Each line contains a 16-byte IPv6 address and a port.\n')
     g.write(' * IPv4 as well as onion addresses are wrapped inside an IPv6 address accordingly.\n')
     g.write(' */\n')
-    with open(os.path.join(indir,'nodes_main.txt'),'r') as f:
+    with open(os.path.join(indir,'nodes_main.txt'), 'r', encoding="utf8") as f:
         process_nodes(g, f, 'pnSeed6_main', 9999)
     g.write('\n')
-    with open(os.path.join(indir,'nodes_test.txt'),'r') as f:
+    with open(os.path.join(indir,'nodes_test.txt'), 'r', encoding="utf8") as f:
         process_nodes(g, f, 'pnSeed6_test', 19999)
     g.write('#endif // DASH_CHAINPARAMSSEEDS_H\n')
-            
+
 if __name__ == '__main__':
     main()
 

contrib/seeds/makeseeds.py

@@ -139,7 +139,7 @@ def filterbyasn(ips, max_per_asn, max_total):
 def main():
     # This expects a json as outputted by "protx list valid 1"
     if len(sys.argv) > 1:
-        with open(sys.argv[1], 'r') as f:
+        with open(sys.argv[1], 'r', encoding="utf8") as f:
             mns = json.load(f)
     else:
         mns = json.load(sys.stdin)

share/qt/extract_strings_qt.py

@@ -63,7 +63,7 @@ child = Popen([XGETTEXT,'--output=-','-n','--keyword=_'] + files, stdout=PIPE)
 
 messages = parse_po(out.decode('utf-8'))
 
-f = open(OUT_CPP, 'w')
+f = open(OUT_CPP, 'w', encoding="utf8")
 f.write("""
 
 #include <QtGlobal>

test/functional/multiwallet.py

@@ -81,7 +81,7 @@ class MultiWalletTest(BitcoinTestFramework):
         self.assert_start_raises_init_error(0, ['-walletdir=bad'], 'Error: Specified -walletdir "bad" does not exist')
         # should not initialize if the specified walletdir is not a directory
         not_a_dir = wallet_dir('notadir')
-        open(not_a_dir, 'a').close()
+        open(not_a_dir, 'a', encoding="utf8").close()
         self.assert_start_raises_init_error(0, ['-walletdir=' + not_a_dir], 'Error: Specified -walletdir "' + not_a_dir + '" is not a directory')
 
         # if wallets/ doesn't exist, datadir should be the default wallet dir

test/functional/notifications.py

@@ -36,7 +36,7 @@ class NotificationsTest(BitcoinTestFramework):
         wait_until(lambda: os.path.isfile(self.block_filename) and os.stat(self.block_filename).st_size >= (block_count * 65), timeout=10)
 
         # file content should equal the generated blocks hashes
-        with open(self.block_filename, 'r') as f:
+        with open(self.block_filename, 'r', encoding="utf-8") as f:
             assert_equal(sorted(blocks), sorted(f.read().splitlines()))
 
         self.log.info("test -walletnotify")
@@ -45,7 +45,7 @@ class NotificationsTest(BitcoinTestFramework):
 
         # file content should equal the generated transaction hashes
         txids_rpc = list(map(lambda t: t['txid'], self.nodes[1].listtransactions("*", block_count)))
-        with open(self.tx_filename, 'r') as f:
+        with open(self.tx_filename, 'r', encoding="ascii") as f:
             assert_equal(sorted(txids_rpc), sorted(f.read().splitlines()))
         os.remove(self.tx_filename)
 
@@ -58,7 +58,7 @@ class NotificationsTest(BitcoinTestFramework):
 
         # file content should equal the generated transaction hashes
         txids_rpc = list(map(lambda t: t['txid'], self.nodes[1].listtransactions("*", block_count)))
-        with open(self.tx_filename, 'r') as f:
+        with open(self.tx_filename, 'r', encoding="ascii") as f:
             assert_equal(sorted(txids_rpc), sorted(f.read().splitlines()))
 
         # Mine another 41 up-version blocks. -alertnotify should trigger on the 51st.

test/functional/rpc_getblockstats.py

@@ -80,11 +80,11 @@ class GetblockstatsTest(BitcoinTestFramework):
             'mocktime': int(self.mocktime),
             'stats': self.expected_stats,
         }
-        with open(filename, 'w') as f:
+        with open(filename, 'w', encoding="utf8") as f:
             json.dump(to_dump, f, sort_keys=True, indent=2)
 
     def load_test_data(self, filename):
-        with open(filename, 'r') as f:
+        with open(filename, 'r', encoding="utf8") as f:
             d = json.load(f)
             blocks = d['blocks']
             self.mocktime = d['mocktime']

test/functional/test_framework/util.py

@@ -325,7 +325,7 @@ def get_auth_cookie(datadir):
                     assert password is None  # Ensure that there is only one rpcpassword line
                     password = line.split("=")[1].strip("\n")
     if os.path.isfile(os.path.join(datadir, "regtest", ".cookie")):
-        with open(os.path.join(datadir, "regtest", ".cookie"), 'r') as f:
+        with open(os.path.join(datadir, "regtest", ".cookie"), 'r', encoding="ascii") as f:
             userpass = f.read()
             split_userpass = userpass.split(':')
             user = split_userpass[0]

test/functional/test_runner.py

@@ -222,7 +222,7 @@ def main():
     # Read config generated by configure.
     config = configparser.ConfigParser()
     configfile = os.path.abspath(os.path.dirname(__file__)) + "/../config.ini"
-    config.read_file(open(configfile))
+    config.read_file(open(configfile, encoding="utf8"))
 
     passon_args.append("--configfile=%s" % configfile)
 
@@ -596,7 +596,7 @@ class RPCCoverage():
         if not os.path.isfile(coverage_ref_filename):
             raise RuntimeError("No coverage reference found")
 
-        with open(coverage_ref_filename, 'r') as file:
+        with open(coverage_ref_filename, 'r', encoding="utf8") as file:
             all_cmds.update([line.strip() for line in file.readlines()])
 
         for root, dirs, files in os.walk(self.dir):
@@ -605,7 +605,7 @@ class RPCCoverage():
                     coverage_filenames.add(os.path.join(root, filename))
 
         for filename in coverage_filenames:
-            with open(filename, 'r') as file:
+            with open(filename, 'r', encoding="utf8") as file:
                 covered_cmds.update([line.strip() for line in file.readlines()])
 
         return all_cmds - covered_cmds

test/lint/lint-python-utf8-encoding.sh

@@ -0,0 +1,19 @@
+#!/bin/bash
+#
+# Copyright (c) 2018 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.
+
+EXIT_CODE=0
+OUTPUT=$(git grep " open(" -- "*.py" | 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
+exit ${EXIT_CODE}

test/util/bitcoin-util-test.py

@@ -28,7 +28,7 @@ import sys
 def main():
     config = configparser.ConfigParser()
     config.optionxform = str
-    config.readfp(open(os.path.join(os.path.dirname(__file__), "../config.ini")))
+    config.readfp(open(os.path.join(os.path.dirname(__file__), "../config.ini"), encoding="utf8"))
     env_conf = dict(config.items('environment'))
 
     parser = argparse.ArgumentParser(description=__doc__)
@@ -49,7 +49,7 @@ def main():
 def bctester(testDir, input_basename, buildenv):
     """ Loads and parses the input file, runs all tests and reports results"""
     input_filename = os.path.join(testDir, input_basename)
-    raw_data = open(input_filename).read()
+    raw_data = open(input_filename, encoding="utf8").read()
     input_data = json.loads(raw_data)
 
     failed_testcases = []
@@ -86,7 +86,7 @@ def bctest(testDir, testObj, buildenv):
     inputData = None
     if "input" in testObj:
         filename = os.path.join(testDir, testObj["input"])
-        inputData = open(filename).read()
+        inputData = open(filename, encoding="utf8").read()
         stdinCfg = subprocess.PIPE
 
     # Read the expected output data (if there is any)
@@ -97,7 +97,7 @@ def bctest(testDir, testObj, buildenv):
         outputFn = testObj['output_cmp']
         outputType = os.path.splitext(outputFn)[1][1:]  # output type from file extension (determines how to compare)
         try:
-            outputData = open(os.path.join(testDir, outputFn)).read()
+            outputData = open(os.path.join(testDir, outputFn), encoding="utf8").read()
         except:
             logging.error("Output file " + outputFn + " can not be opened")
             raise