Add shellcheck tools and modify copyright hook (#27722)

gongweibao · web-flow · commit 4237fefeb4bc · 2020-10-13T10:11:15.000+08:00
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -48,5 +48,12 @@ repos:
         name: copyright_checker
         entry: python ./tools/codestyle/copyright.hook
         language: system
-        files: \.(c|cc|cxx|cpp|cu|h|hpp|hxx|proto|py)$
+        files: \.(c|cc|cxx|cpp|cu|h|hpp|hxx|proto|py|sh)$
         exclude: (?!.*third_party)^.*$ | (?!.*book)^.*$
+-   repo: local
+    hooks:
+    -   id: shellcheck
+        name: shellcheck
+        entry: shellcheck
+        language: system
+        files: .sh$
diff --git a/paddle/fluid/train/demo/clean.sh b/paddle/fluid/train/demo/clean.sh
@@ -1,4 +1,20 @@
+#!/bin/bash
+
+# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
+# 
+# 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.
+
 set -x
-cd `dirname $0`
+cd "$(dirname "$0")"
 rm -rf build/
 set +x
diff --git a/tools/codestyle/copyright.hook b/tools/codestyle/copyright.hook
@@ -1,15 +1,29 @@
+# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
+#
+# 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 absolute_import
 from __future__ import print_function
 from __future__ import unicode_literals
 
 import argparse
-import io, re
-import sys, os
-import subprocess
-import platform
+import io
+import re
+import sys
+import os
+import datetime
 
-COPYRIGHT = '''
-Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved.
+COPYRIGHT = '''Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved.
 
 Licensed under the Apache License, Version 2.0 (the "License");
 you may not use this file except in compliance with the License.
@@ -21,74 +35,80 @@ 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.
-'''
+limitations under the License.'''
 
-LANG_COMMENT_MARK = None
+def _generate_copyright(comment_mark):
+    copyright=COPYRIGHT.split(os.linesep)
+    header = copyright[0].rstrip()
 
-NEW_LINE_MARK = None
+    p = re.search('(\d{4})', header).group(0)
+    now = datetime.datetime.now()
 
-COPYRIGHT_HEADER = None
+    header = header.replace(p,str(now.year))
 
-if platform.system() == "Windows":
-    NEW_LINE_MARK = "\r\n"
-else:
-    NEW_LINE_MARK = '\n'
-    COPYRIGHT_HEADER = COPYRIGHT.split(NEW_LINE_MARK)[1]
-    p = re.search('(\d{4})', COPYRIGHT_HEADER).group(0)
-    process = subprocess.Popen(["date", "+%Y"], stdout=subprocess.PIPE)
-    date, err = process.communicate()
-    date = date.decode("utf-8").rstrip("\n")
-    COPYRIGHT_HEADER = COPYRIGHT_HEADER.replace(p, date)
+    ans=[comment_mark + " " + header + os.linesep]
+    for idx, line in enumerate(copyright[1:]):
+        ans.append(comment_mark + " " + line.rstrip() + os.linesep)
 
+    return ans
 
-def generate_copyright(template, lang='C'):
-    if lang == 'Python':
-        LANG_COMMENT_MARK = '#'
-    else:
-        LANG_COMMENT_MARK = "//"
-
-    lines = template.split(NEW_LINE_MARK)
-    BLANK = " "
-    ans = LANG_COMMENT_MARK + BLANK + COPYRIGHT_HEADER + NEW_LINE_MARK
-    for lino, line in enumerate(lines):
-        if lino == 0 or lino == 1 or lino == len(lines) - 1: continue
-        if len(line)  == 0:
-            BLANK = ""
-        else:
-            BLANK = " "
-        ans += LANG_COMMENT_MARK + BLANK + line + NEW_LINE_MARK
-
-    return ans + "\n"
-
-
-def lang_type(filename):
-    if filename.endswith(".py"):
-        return "Python"
-    elif filename.endswith(".h"):
-        return "C"
-    elif filename.endswith(".c"):
-        return "C"
-    elif filename.endswith(".hpp"):
-        return "C"
-    elif filename.endswith(".cc"):
-        return "C"
-    elif filename.endswith(".cpp"):
-        return "C"
-    elif filename.endswith(".cu"):
-        return "C"
-    elif filename.endswith(".cuh"):
-        return "C"
-    elif filename.endswith(".go"):
-        return "C"
-    elif filename.endswith(".proto"):
-        return "C"
+def _get_comment_mark(path):
+    lang_type=re.compile(r"\.(py|sh)$")
+    if lang_type.search(path) is not None:
+        return "#"
+
+    lang_type=re.compile(r"\.(h|c|hpp|cc|cpp|cu|go|cuh|proto)$")
+    if lang_type.search(path) is not None:
+        return "//"
+
+    return None
+
+
+RE_ENCODE = re.compile(r"^[ \t\v]*#.*?coding[:=]", re.IGNORECASE)
+RE_COPYRIGHT = re.compile(r".*Copyright \(c\) \d{4}", re.IGNORECASE)
+RE_SHEBANG = re.compile(r"^[ \t\v]*#[ \t]?\!")
+
+def _check_copyright(path):
+    head=[]
+    try:
+        with open(path) as f:
+            head = [next(f) for x in range(4)]
+    except StopIteration:
+        pass
+
+    for idx, line in enumerate(head):
+        if RE_COPYRIGHT.search(line) is not None:
+            return True
+
+    return False
+
+def generate_copyright(path, comment_mark):
+    original_contents = io.open(path, encoding="utf-8").readlines()
+    head = original_contents[0:4]
+
+    insert_line_no=0
+    for i, line in enumerate(head):
+        if RE_ENCODE.search(line) or RE_SHEBANG.search(line):
+            insert_line_no=i+1
+
+    copyright = _generate_copyright(comment_mark)
+    if insert_line_no == 0:
+        new_contents = copyright
+        if len(original_contents) > 0 and len(original_contents[0].strip()) != 0:
+            new_contents.append(os.linesep)
+        new_contents.extend(original_contents)
     else:
-        print("Unsupported filetype %s", filename)
-        exit(0)
+        new_contents=original_contents[0:insert_line_no]
+        new_contents.append(os.linesep)
+        new_contents.extend(copyright)
+        if len(original_contents) > insert_line_no and len(original_contents[insert_line_no].strip()) != 0:
+            new_contents.append(os.linesep)
+        new_contents.extend(original_contents[insert_line_no:])
+    new_contents="".join(new_contents)
 
+    with io.open(path, 'w') as output_file:
+        output_file.write(new_contents)
 
-PYTHON_ENCODE = re.compile("^[ \t\v]*#.*?coding[:=][ \t]*([-_.a-zA-Z0-9]+)")
 
 
 def main(argv=None):
@@ -98,23 +118,16 @@ def main(argv=None):
     args = parser.parse_args(argv)
 
     retv = 0
-    for filename in args.filenames:
-        fd = io.open(filename, encoding="utf-8")
-        first_line = fd.readline()
-        second_line = fd.readline()
-        if "COPYRIGHT (C)" in first_line.upper(): continue
-        if first_line.startswith("#!") or PYTHON_ENCODE.match(
-                second_line) != None or PYTHON_ENCODE.match(first_line) != None:
+    for path in args.filenames:
+        comment_mark = _get_comment_mark(path)
+        if comment_mark is None:
+            print("warning:Unsupported file", path, file=sys.stderr)
             continue
-        original_contents = io.open(filename, encoding="utf-8").read()
-        new_contents = generate_copyright(
-            COPYRIGHT, lang_type(filename)) + original_contents
-        print('Auto Insert Copyright Header {}'.format(filename))
-        retv = 1
-        with io.open(filename, 'w') as output_file:
-            output_file.write(new_contents)
-
-    return retv
+
+        if _check_copyright(path):
+            continue
+
+        generate_copyright(path, comment_mark)
 
 
 if __name__ == '__main__':
diff --git a/tools/dockerfile/Dockerfile.ubuntu b/tools/dockerfile/Dockerfile.ubuntu
@@ -29,7 +29,7 @@ RUN apt-get update && \
     python-matplotlib \
     automake locales clang-format swig  \
     liblapack-dev liblapacke-dev \
-    net-tools libtool module-init-tools && \
+    net-tools libtool module-init-tools shellcheck && \
     apt-get clean -y
 
 # Downgrade gcc&&g++
diff --git a/tools/manylinux1/Dockerfile.cuda10_cudnn7_gcc8_ubuntu16 b/tools/manylinux1/Dockerfile.cuda10_cudnn7_gcc8_ubuntu16
@@ -26,7 +26,7 @@ RUN rm -rf /temp_gcc82 && rm -rf /gcc-8.2.0.tar.xz && rm -rf /gcc-8.2.0
 RUN apt-get update && \
     apt-get install -y make build-essential libssl-dev zlib1g-dev libbz2-dev \
     libreadline-dev libsqlite3-dev wget curl llvm libncurses5-dev libncursesw5-dev \
-    xz-utils tk-dev libffi-dev liblzma-dev openmpi-bin openmpi-doc libopenmpi-dev
+    xz-utils tk-dev libffi-dev liblzma-dev openmpi-bin openmpi-doc libopenmpi-dev shellcheck
 
 # gcc8.2
 RUN wget -q https://paddle-docker-tar.bj.bcebos.com/home/users/tianshuo/bce-python-sdk-0.8.27/gcc-8.2.0.tar.xz && \
