From 83a2be6ec51f793a1ac300ecf2c940f7a6327476 Mon Sep 17 00:00:00 2001
From: Sterling Harper <sterlingmharper@gmail.com>
Date: Tue, 21 Jul 2015 22:10:15 -0600
Subject: [PATCH] Make source code style checking script

---
 tests/check_source.py | 235 ++++++++++++++++++++++++++++++++++++++++++
 tests/travis.sh       |   1 +
 2 files changed, 236 insertions(+)
 create mode 100755 tests/check_source.py

diff --git a/tests/check_source.py b/tests/check_source.py
new file mode 100755
index 0000000000..490eb9c7c3
--- /dev/null
+++ b/tests/check_source.py
@@ -0,0 +1,235 @@
+#!/usr/bin/env python
+
+from __future__ import print_function
+
+import glob
+from string import whitespace
+from sys import exit
+from textwrap import TextWrapper
+
+
+################################################################################
+# Fortran reading/parsing backend.
+################################################################################
+
+
+class LineOfCode(object):
+    """Contains and provides basic info about a line of Fortran 90 code."""
+    def __init__(self, number, content, is_continuation=False,
+                 string_terminator=None):
+        # Initialize member variables.
+        self.number = number
+        self.content = content
+        self.is_continuation = is_continuation
+        self.is_continued = False
+        assert (string_terminator == None or
+                string_terminator == "'" or
+                string_terminator == '"')
+        self.initial_string_terminator = string_terminator
+        self.final_string_terminator = None
+        self.is_comment_only = False
+
+        # Parse the string.
+        self.parse()
+
+    def __repr__(self):
+        rep = 'LineOfCode: line number = {0:d}\n'.format(self.number)
+        rep += '\tis_continuation = ' + str(self.is_continuation) + '\n'
+        rep += '\tis_continued = ' + str(self.is_continued) + '\n'
+        rep += ('\tinitial_string_terminator = '
+                + str(self.initial_string_terminator) + '\n')
+        rep += ('\tfinal_string_terminator = '
+                + str(self.final_string_terminator) + '\n')
+        rep += '\tis_comment_only = ' + str(self.is_comment_only) + '\n'
+        rep += '\tContent:\n' + self.content
+        return rep
+
+    def __str__(self):
+        return repr(self)
+
+    def parse(self):
+        # Initialize the string variables.
+        if self.initial_string_terminator == "'":
+            in_a_single_quote = True
+            in_a_double_quote = False
+        elif self.initial_string_terminator == '"':
+            in_a_single_quote = False
+            in_a_double_quote = True
+        else:
+            in_a_single_quote = False
+            in_a_double_quote = False
+
+        # Initialize other variables.
+        in_a_comment = False
+        ends_with_ampersand = False
+        has_real_content = False
+
+        # Parse through the line.
+        for char in self.content:
+            # Check for the start or end of strings.
+            if char == "'" and in_a_single_quote:
+                in_a_single_quote = False
+            elif char == "'" and not in_a_double_quote:
+                in_a_single_quote = True
+            elif char == '"' and in_a_double_quote:
+                in_a_double_quote = False
+            elif char == '"' and not in_a_single_quote:
+                in_a_double_quote = True
+
+            # Check for the start of a comment.
+            if (char == "!" and not in_a_single_quote
+                 and not in_a_double_quote):
+                in_a_comment = True
+                break # Don't care about anything in a comment.
+
+            # Check for a continuation ampersand.
+            if char == "&":
+                ends_with_ampersand = True
+            elif all([char != w for w in whitespace]):
+                ends_with_ampersand = False
+
+            # Check to see if there is any real content (non-whitespace, not in
+            # a comment)
+            if (not in_a_comment) and all([char != w for w in whitespace]):
+                has_real_content = True
+
+            # Make sure nothing went terribly wrong.
+            assert not (in_a_single_quote and in_a_double_quote)
+            assert not (in_a_single_quote and in_a_comment)
+            assert not (in_a_double_quote and in_a_comment)
+
+        # Is this line continued onto the next?
+        if ends_with_ampersand:
+            self.is_continued = True
+
+        # Is a multiline string continued on the next line?
+        if in_a_single_quote:
+            assert self.is_continued
+            self.final_string_terminator = "'"
+        if in_a_double_quote:
+            assert self.is_continued
+            self.final_string_terminator = '"'
+
+        # Is this line a comment-only line?
+        if (not has_real_content) and in_a_comment:
+            self.is_comment_only = True
+
+
+    def get_indent(self):
+        """Return the number of indentation spaces prefixing this line."""
+        return len(self.content) - len(self.content.lstrip(' '))
+
+
+    def contains_whitespace_only(self):
+        """Return True/False if all characters in the line are whitespace."""
+        if len(self.content.strip('\n')) == 0: return False
+        is_ws = [ any([char == ws for ws in whitespace])
+                  for char in self.content.strip('\n') ]
+        return all(is_ws)
+
+
+    def has_trailing_whitespace(self):
+        """Return True/False if this line ends with a whitespace character."""
+        stripped = self.content.strip('\n')
+        if len(stripped) == 0: return False
+        return any([stripped[-1] == ws for ws in whitespace])
+
+    def contains_tab(self):
+        """Return True/False if a tab character appears in the line."""
+        return '\t' in self.content
+
+
+def read_lines_of_code(fname):
+    line_num = 0
+    cont = False
+    str_term = None
+    with open(fname) as fin:
+        for line in fin:
+            loc = LineOfCode(line_num, line, is_continuation=cont,
+                             string_terminator=str_term)
+            cont = loc.is_continued
+            str_term = loc.final_string_terminator
+            line_num += 1
+            yield loc
+
+
+################################################################################
+# Error checking.
+################################################################################
+
+
+def print_error(fname, line_number, err_msg):
+    header = "Error in file {0}, line {1}:".format(fname, line_number + 1)
+
+    tw = TextWrapper(width=80, initial_indent=' '*4, subsequent_indent=' '*4)
+    body = '\n'.join(tw.wrap(err_msg))
+
+    print(header + '\n' + body + '\n')
+
+
+def check_source(fname):
+    """Make sure the given Fortran source file meets OpenMC standards.
+
+    If errors are found, messages will be printed to the screen describing the
+    error.  The function will return True if no errors were found or False
+    otherwise.
+    """
+    good_code = True
+    base_indent = 0
+
+    for loc in read_lines_of_code(fname):
+        # Check for extra whitespace errors.
+        if loc.contains_whitespace_only():
+            good_code = False
+            print_error(fname, loc.number, "Line contains whitespace " \
+                 "characters but no content.  Please remove whitespace.")
+        elif loc.has_trailing_whitespace():
+            good_code = False
+            print_error(fname, loc.number, "Line has trailing whitespace after"\
+                 " the content.  Please remove trailing whitespace.")
+        if loc.contains_tab():
+            good_code = False
+            print_error(fname, loc.number, "Line contains a tab character.  " \
+                 "Please replace with single whitespace characters.")
+
+        # Check indentation.
+        current_indent = loc.get_indent()
+        if ((not loc.is_continuation) and (not loc.is_comment_only) and
+              (not loc.contains_whitespace_only()) and current_indent % 2 != 0):
+            good_code = False
+            print_error(fname, loc.number, "Line is indented an odd number of "\
+                 "spaces.  All non-continuation lines should be indented an "\
+                 "even number of spaces.")
+        if loc.is_continuation and current_indent < base_indent + 5:
+            good_code = False
+            print_error(fname, loc.number, "Continuation lines must be "\
+                 "indented by at least 5 spaces, but this line is indented {0}"\
+                 " spaces.".format(current_indent - base_indent))
+
+        # Set base indentation for next lines.
+        if not loc.is_continuation:
+            base_indent = current_indent
+
+    return good_code
+
+
+################################################################################
+# Main loop.
+################################################################################
+
+
+if __name__ == '__main__':
+    # Get a list of the F90 source files.
+    f90_files = glob.glob('../src/*.F90')
+
+    # Make sure we found the source files.
+    assert len(f90_files) != 0, 'No .F90 source files found.'
+
+    # Make sure all the F90 source files meet our standards.
+    good_code = [check_source(fname) for fname in f90_files]
+    if not all(good_code):
+        print("ERROR: The Fortran source code does not meet OpenMC's standards")
+        exit(-1)
+    else:
+        print("SUCCESS! Looks like the Fortran source meets our standards")
+        exit(0)
diff --git a/tests/travis.sh b/tests/travis.sh
index d591428686..fac8d793a6 100755
--- a/tests/travis.sh
+++ b/tests/travis.sh
@@ -3,6 +3,7 @@
 set -ev
 
 # Run all debug tests
+./check_source.py
 if [ "$TRAVIS_PULL_REQUEST" != "false" ]; then
   ./run_tests.py -C "^basic-debug$|^hdf5-debug$|^mpi-omp-debug$|^phdf5-omp-debug$" -j 2 -s
 else