Add a checker for words which I often misspell.

scripts/misc_checks.py

@@ -21,18 +21,27 @@
 """Various small code checkers."""
 
 import os
+import re
 import sys
 import os.path
 import argparse
 import subprocess
 import tokenize
 import traceback
+import collections
 
 sys.path.insert(0, os.path.join(os.path.dirname(__file__), os.pardir))
 
 from scripts import utils
 
 
+def _py_files(target):
+    """Iterate over all python files and yield filenames."""
+    for (dirpath, _dirnames, filenames) in os.walk(target):
+        for name in (e for e in filenames if e.endswith('.py')):
+            yield os.path.join(dirpath, name)
+
+
 def check_git():
     """Check for uncommitted git files.."""
     if not os.path.isdir(".git"):
@@ -55,18 +64,49 @@ def check_git():
     return status
 
 
+def check_spelling(target):
+    """Check commonly misspelled words."""
+    # Words which I often misspell
+    words = {'behaviour', 'quitted', 'likelyhood', 'sucessfully',
+             'occur[^r .]', 'seperator', 'explicitely', 'resetted',
+             'auxillary', 'accidentaly', 'ambigious', 'loosly',
+             'initialis', 'convienence', 'similiar', 'uncommited',
+             'reproducable'}
+
+    # Words which look better when splitted, but might need some fine tuning.
+    words |= {'keystrings', 'webelements', 'mouseevent', 'keysequence',
+              'normalmode', 'eventloops', 'sizehint', 'statemachine',
+              'metaobject', 'logrecord', 'monkeypatch', 'filetype'}
+
+    seen = collections.defaultdict(list)
+    try:
+        ok = True
+        for fn in _py_files(target):
+            with tokenize.open(fn) as f:
+                if fn == os.path.join('scripts', 'misc_checks.py'):
+                    continue
+                for line in f:
+                    for w in words:
+                        if re.search(w, line) and fn not in seen[w]:
+                            print("Found '{}' in {}!".format(w, fn))
+                            seen[w].append(fn)
+        print()
+        return ok
+    except Exception:
+        traceback.print_exc()
+        return None
+
+
 def check_vcs_conflict(target):
     """Check VCS conflict markers."""
     try:
         ok = True
-        for (dirpath, _dirnames, filenames) in os.walk(target):
-            for name in (e for e in filenames if e.endswith('.py')):
-                fn = os.path.join(dirpath, name)
-                with tokenize.open(fn) as f:
-                    for line in f:
-                        if any(line.startswith(c * 7) for c in '<>=|'):
-                            print("Found conflict marker in {}".format(fn))
-                            ok = False
+        for fn in _py_files(target):
+            with tokenize.open(fn) as f:
+                for line in f:
+                    if any(line.startswith(c * 7) for c in '<>=|'):
+                        print("Found conflict marker in {}".format(fn))
+                        ok = False
         print()
         return ok
     except Exception:
@@ -77,7 +117,7 @@ def check_vcs_conflict(target):
 def main():
     """Main entry point."""
     parser = argparse.ArgumentParser()
-    parser.add_argument('checker', choices=('git', 'vcs'),
+    parser.add_argument('checker', choices=('git', 'vcs', 'spelling'),
                         help="Which checker to run.")
     parser.add_argument('target', help="What to check", nargs='*')
     args = parser.parse_args()
@@ -91,6 +131,13 @@ def main():
             if not ok:
                 is_ok = False
         return 0 if is_ok else 1
+    elif args.checker == 'spelling':
+        is_ok = True
+        for target in args.target:
+            ok = check_spelling(target)
+            if not ok:
+                is_ok = False
+        return 0 if is_ok else 1
 
 
 if __name__ == '__main__':

tox.ini

@@ -39,6 +39,7 @@ commands =
 commands =
     {envpython} scripts/misc_checks.py git
     {envpython} scripts/misc_checks.py vcs qutebrowser scripts
+    {envpython} scripts/misc_checks.py spelling qutebrowser scripts
 
 [testenv:pylint]
 skip_install = true