Finishing basic repo stats script

thundergolfer-two · thundergolfer-two · commit 71f8bf968717 · 2016-09-03T17:02:12.000+10:00
diff --git a/housekeeping/repo_health_checkup.py b/housekeeping/repo_health_checkup.py
@@ -6,18 +6,37 @@
 import sys, token, tokenize
 
 def main():
-    for folder, subfolders, files in os.walk(os.getcwd(), topdown=False):
+    search_dir = os.path.dirname(os.getcwd()) # we want the parent directory
+    for folder, subfolders, files in os.walk(search_dir, topdown=False):
         for f in files:
             fullpath = os.path.join(folder,f)
-            if '.git' in fullpath: continue # ignore .git hidden folder files
+            if not fullpath.endswith('.py') and not fullpath.endswith('.md'): continue # ignore .git hidden folder files
             print(fullpath) # just to check where we're walking
-            # HEALTH CHECK FLOW
-            #
-            # if: README for a problem check for -> formatting, spelling, and a code output example
-
-            # if: problem solution source code file then check for -> tags, unit tests, spelling
-
-            # and... any other file check for -> spelling, "TODO" markers, amount of commenting
+            with open(fullpath, 'r') as f:
+                try:
+                    lines = f.readlines()
+                except UnicodeDecodeError: continue
+                file_health = fullpath + '\n'
+                # HEALTH CHECK FLOW
+                #
+                # if: README for a problem check for -> formatting, spelling, and a code output example
+                if fullpath[-2:] == 'md':
+                    file_health += "" if check_for_code_output_example(lines) else "missing output ex.\n"
+                    # TODO: Add the rest of the checks
+                # if: problem solution source code file then check for -> tags, unit tests, spelling
+                elif fullpath.endswith('-answer.py'):
+                    file_health += "" if check_for_tags(lines) else "missing tags!\n"
+                    file_health += "" if check_for_unittests(lines) else "missing unittests!\n"
+                    # TODO: Add spelling check
+                # and... any other file check for -> spelling, "TODO" markers, amount of commenting
+                else:
+                    has_todos, _ = check_for_todos(lines)
+                    if has_todos: file_health += "has TODOs to complete.\n"
+                    comment_lines, total_lines = check_amount_of_commenting( f.readline )
+                    if comment_lines / total_lines < 0.15: # arbitrary limit
+                        file_health += "low commenting! " + str(comment_lines / total_lines) + "\n"
+            if file_health != fullpath + '\n':
+                print(file_health)
 
 def check_markdown( lines ):
     """
@@ -26,11 +45,25 @@ def check_markdown( lines ):
     """
     return True # TODO
 
+def check_for_todos( lines ): # TODO: It is inefficient to re-read through lines to check for different things
+    """ Check for presence of 'TODO' markers in files. """
+    todos = []
+    for line in lines:
+        if 'TODO' in line:
+            todos.append(line[line.index('TODO'):])
+    if not todos:
+        return False, None
+    else:
+        return True, todos
+
+
 def check_amount_of_commenting( readable ):
     """
     Check lines of file to see whether it has a lot of commenting in it,
     which we take to mean the presence of code explanation.
     """
+    # TODO: I have hacked this code together and it could be silly and buggy.
+    #       It does seem to get things roughly correct, which is good enough for the moment
     prev_toktype = token.INDENT
     first_line = None
     last_lineno = -1
@@ -46,21 +79,20 @@ def check_amount_of_commenting( readable ):
                 ))
         if slineno > last_lineno:
             last_col = 0
-        if scol > last_col:
-            comment_lines += scol - last_col
         if toktype == token.STRING and prev_toktype == token.INDENT:
             # Docstring
-            mod.write("#--")
+            comment_lines += 1
         elif toktype == tokenize.COMMENT:
             # Comment
             comment_lines += 1
         prev_toktype = toktype
         last_col = ecol
         last_lineno = elineno
-    # TODO // fix this function
+    return comment_lines, last_lineno # roughly num of comment lines divided by num of lines
 
 def check_for_tags( lines ):
     """ Check if a problems answer file has tags in it. True, if yes. """
+    lines = ''.join(lines)
     match = re.search(r'##\$\$##(\s\S+,)+(\s\S+)', lines )
     return True if match else False
 
@@ -69,9 +101,14 @@ def check_for_code_output_example( lines ):
     Check a Markdown problem description file for the presence of a code input to
     output example. ie. file should have something like: double_this(4) # > 8
     """
+    lines = ''.join(lines)
     match = re.search(r'`{3}.+\s`{3}', lines, re.DOTALL)
     return True if match else False
 
+def check_for_unittests( lines ):
+    """ Check python source code for unittests. """
+    return True # TODO: Implement this
+
 def check_spelling( lines ):
     """ Check the spelling of all words in the lines provided. """
     lines_to_check = []
diff --git a/housekeeping/repo_stats.py b/housekeeping/repo_stats.py
@@ -6,7 +6,28 @@
 from os.path import isfile, isdir, join, splitext
 
 def main():
-    pass
+    """ Run through all relevant files in the projet and print to console some stats. """
+    nloc_in_project, num_problems, num_questions = 0,0,0
+    search_dir = os.path.dirname(os.getcwd()) # we want the parent directory
+    for folder, subfolders, files in os.walk(search_dir, topdown=False):
+        for f in files:
+            fullpath = os.path.join(folder,f)
+            if (not fullpath.endswith('.py')
+                and not fullpath.endswith('.ipynb')
+                and not fullpath.endswith('.md')): continue # ignore .git hidden folder files and others
+            with open(fullpath, 'r') as f:
+                try:
+                    lines = f.readlines()
+                except UnicodeDecodeError: continue
+                if fullpath.endswith('.py'): nloc_in_project += len(lines)
+                if fullpath.endswith('.py') and '-answer' in fullpath: num_problems += 1
+                if fullpath.endswith('.ipynb'): num_questions += 1
+
+    print("Lines of code in project: ", nloc_in_project)
+    print("Number of Code Exercises: ", num_problems)
+    print("Number of Worded Questions: ", num_questions )
+
+
 
 def line_count_file(file_path, flags=None):
     """ Counts lines for given file in file_name """
