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

c8176b3cc7 Add linter: Make sure we explicitly open all text files using UTF-8 or ASCII encoding in Python (practicalswift)
634bd97001 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

test/functional/feature_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(l.strip() for l in 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(l.strip() for l in 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(l.strip() for l in f.read().splitlines()))
 
         # Mine another 41 up-version blocks. -alertnotify should trigger on the 51st.

test/functional/rpc_getblockstats.py

@@ -81,11 +81,11 @@ class GetblockstatsTest(BitcoinTestFramework):
             'mocktime': int(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']
             mocktime = d['mocktime']

test/functional/test_framework/util.py

@@ -327,7 +327,7 @@ 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

@@ -213,7 +213,7 @@ 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)
 
@@ -590,7 +590,7 @@ class RPCCoverage():
         if not os.path.isfile(coverage_ref_filename):
             raise RuntimeError("No coverage reference found")
 
-        with open(coverage_ref_filename, 'r') as coverage_ref_file:
+        with open(coverage_ref_filename, 'r', encoding="utf8") as coverage_ref_file:
             all_cmds.update([line.strip() for line in coverage_ref_file.readlines()])
 
         for root, dirs, files in os.walk(self.dir):
@@ -599,7 +599,7 @@ class RPCCoverage():
                     coverage_filenames.add(os.path.join(root, filename))
 
         for filename in coverage_filenames:
-            with open(filename, 'r') as coverage_file:
+            with open(filename, 'r', encoding="utf8") as coverage_file:
                 covered_cmds.update([line.strip() for line in coverage_file.readlines()])
 
         return all_cmds - covered_cmds

test/functional/wallet_multiwallet.py

@@ -88,7 +88,7 @@ class MultiWalletTest(BitcoinTestFramework):
         self.nodes[0].assert_start_raises_init_error(['-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.nodes[0].assert_start_raises_init_error(['-walletdir=' + not_a_dir], 'Error: Specified -walletdir "' + not_a_dir + '" is not a directory')
 
         self.log.info("Do not allow -zapwallettxes with multiwallet")

test/lint/check-rpc-mappings.py

@@ -44,7 +44,7 @@ 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:
@@ -70,7 +70,7 @@ 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:

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

test/util/rpcauth-test.py

@@ -18,7 +18,7 @@ class TestRPCAuth(unittest.TestCase):
         config_path = os.path.abspath(
             os.path.join(os.sep, os.path.abspath(os.path.dirname(__file__)),
             "../config.ini"))
-        with open(config_path) as config_file:
+        with open(config_path, encoding="utf8") as config_file:
             config.read_file(config_file)
         sys.path.insert(0, os.path.dirname(config['environment']['RPCAUTH']))
         self.rpcauth = importlib.import_module('rpcauth')