check_typed_item_calls.py: add --consecutive flag

Add a check to point out where consecutive items have the same filter
but different labels.  Quite a few of these look like bugs.

Also, make some REs raw strings, as identified as an issue in
https://gitlab.com/wireshark/wireshark/-/merge_requests/346

3 files changed, 59 insertions, 16 deletions

diff --git a/tools/check_spelling.py b/tools/check_spelling.py
index 1f910148d4..f8b1309273 100755
--- a/tools/check_spelling.py
+++ b/tools/check_spelling.py
@@ -53,7 +53,7 @@ missing_words = []
 
 # Split camelCase string into separate words.
 def camelCaseSplit(identifier):
-    matches = re.finditer('.+?(?:(?<=[a-z])(?=[A-Z])|(?<=[A-Z])(?=[A-Z][a-z])|$)', identifier)
+    matches = re.finditer(r'.+?(?:(?<=[a-z])(?=[A-Z])|(?<=[A-Z])(?=[A-Z][a-z])|$)', identifier)
     return [m.group(0) for m in matches]
 
 
@@ -191,9 +191,9 @@ class File:
                     missing_words.append(word)
 
 def removeComments(code_string):
-    code_string = re.sub(re.compile("/\*.*?\*/",re.DOTALL ) ,"" ,code_string) # C-style comment
+    code_string = re.sub(re.compile(r"/\*.*?\*/",re.DOTALL ) ,"" ,code_string) # C-style comment
     # Remove this for now as can get tripped up if see htpps://www.... within a string!
-    #code_string = re.sub(re.compile("//.*?\n" ) ,"" ,code_string)             # C++-style comment
+    #code_string = re.sub(re.compile(r"//.*?\n" ) ,"" ,code_string)             # C++-style comment
     return code_string
 
 def removeSingleQuotes(code_string):
@@ -364,7 +364,7 @@ else:
 
 # If scanning a subset of files, list them here.
 print('Examining:')
-if args.file or args.commits or args.open:
+if args.file or args.folder or args.commits or args.open:
     if files:
         print(' '.join(files), '\n')
     else:
diff --git a/tools/check_tfs.py b/tools/check_tfs.py
index 89581d00f0..5d8a06d740 100755
--- a/tools/check_tfs.py
+++ b/tools/check_tfs.py
@@ -60,8 +60,8 @@ class TFS:
 
 
 def removeComments(code_string):
-    code_string = re.sub(re.compile("/\*.*?\*/",re.DOTALL ) ,"" ,code_string) # C-style comment
-    code_string = re.sub(re.compile("//.*?\n" ) ,"" ,code_string)             # C++-style comment
+    code_string = re.sub(re.compile(r"/\*.*?\*/",re.DOTALL ) ,"" ,code_string) # C-style comment
+    code_string = re.sub(re.compile(r"//.*?\n" ) ,"" ,code_string)             # C++-style comment
     return code_string
 
 
@@ -89,7 +89,7 @@ def findItems(filename):
 
 
 def is_dissector_file(filename):
-    p = re.compile('.*packet-.*\.c')
+    p = re.compile(r'.*packet-.*\.c')
     return p.match(filename)
 
 def findDissectorFilesInFolder(folder):
diff --git a/tools/check_typed_item_calls.py b/tools/check_typed_item_calls.py
index 3bc7a6ed54..122d46d775 100755
--- a/tools/check_typed_item_calls.py
+++ b/tools/check_typed_item_calls.py
@@ -113,11 +113,22 @@ field_widths = {
 
 # The relevant parts of an hf item.  Used as value in dict where hf variable name is key.
 class Item:
-    def __init__(self, filename, filter, label, item_type, mask=None, check_mask=False, check_label=False):
+
+    previousItem = None
+
+    def __init__(self, filename, filter, label, item_type, mask=None, check_mask=False, check_label=False, check_consecutive=False):
         self.filename = filename
         self.filter = filter
         self.label = label
 
+        if check_consecutive:
+            if Item.previousItem and Item.previousItem.filter == filter:
+                if label != Item.previousItem.label:
+                    print('Warn: ' + filename + ': - filter "' + filter +
+                          '" appears consecutively - labels are "' + Item.previousItem.label + '" and "' + label + '"')
+            Item.previousItem = self
+
+
         # Optionally check label.
         if check_label:
             if label.startswith(' ') or label.endswith(' '):
@@ -241,12 +252,39 @@ apiChecks.append(APICheck('proto_tree_add_none_format', { 'FT_NONE'}))
 
 
 def removeComments(code_string):
-    code_string = re.sub(re.compile("/\*.*?\*/",re.DOTALL ) ,"" ,code_string) # C-style comment
-    code_string = re.sub(re.compile("//.*?\n" ) ,"" ,code_string)             # C++-style comment
+    code_string = re.sub(re.compile(r"/\*.*?\*/",re.DOTALL ) ,"" , code_string) # C-style comment
+    code_string = re.sub(re.compile(r"//.*?\n" ) ,"" , code_string)             # C++-style comment
     return code_string
 
+# Test for whether the given file was automatically generated.
+def isGeneratedFile(filename):
+    # Open file
+    f_read = open(os.path.join(filename), 'r')
+    lines_tested = 0
+    for line in f_read:
+        # The comment to say that its generated is near the top, so give up once
+        # get a few lines down.
+        if lines_tested > 10:
+            f_read.close()
+            return False
+        if (line.find('Generated automatically') != -1 or
+            line.find('Autogenerated from') != -1 or
+            line.find('is autogenerated') != -1 or
+            line.find('automatically generated by Pidl') != -1 or
+            line.find('Created by: The Qt Meta Object Compiler') != -1 or
+            line.find('This file was generated') != -1):
+
+            f_read.close()
+            return True
+        lines_tested = lines_tested + 1
+
+    # OK, looks like a hand-written file!
+    f_read.close()
+    return False
+
 # Look for hf items in a dissector file.
-def find_items(filename, check_mask=False, check_label=False):
+def find_items(filename, check_mask=False, check_label=False, check_consecutive=False):
+    is_generated = isGeneratedFile(filename)
     items = {}
     with open(filename, 'r') as f:
         contents = f.read()
@@ -257,13 +295,15 @@ def find_items(filename, check_mask=False, check_label=False):
             # Store this item.
             hf = m.group(1)
             items[hf] = Item(filename, filter=m.group(3), label=m.group(2), item_type=m.group(4), mask=m.group(5),
-                             check_mask=check_mask, check_label=check_label)
+                             check_mask=check_mask,
+                             check_label=check_label,
+                             check_consecutive=(not is_generated and check_consecutive))
     return items
 
 
 
 def is_dissector_file(filename):
-    p = re.compile('.*packet-.*\.c')
+    p = re.compile(r'.*packet-.*\.c')
     return p.match(filename)
 
 def findDissectorFilesInFolder(folder, dissector_files=[], recursive=False):
@@ -287,9 +327,9 @@ def findDissectorFilesInFolder(folder, dissector_files=[], recursive=False):
 
 
 # Check the given dissector file.
-def checkFile(filename, check_mask=False, check_label=False):
+def checkFile(filename, check_mask=False, check_label=False, check_consecutive=False):
     # Find important parts of items.
-    items = find_items(filename, check_mask, check_label)
+    items = find_items(filename, check_mask, check_label, check_consecutive)
 
     # Check each API
     for c in apiChecks:
@@ -313,6 +353,9 @@ parser.add_argument('--mask', action='store_true',
                    help='when set, check mask field too')
 parser.add_argument('--label', action='store_true',
                    help='when set, check label field too')
+parser.add_argument('--consecutive', action='store_true',
+                    help='when set, copy copy/paste errors between consecutive items')
+
 
 
 args = parser.parse_args()
@@ -372,7 +415,7 @@ else:
 for f in files:
     if should_exit:
         exit(1)
-    checkFile(f, check_mask=args.mask, check_label=args.label)
+    checkFile(f, check_mask=args.mask, check_label=args.label, check_consecutive=args.consecutive)
 
 # Show summary.
 print(issues_found, 'issues found')