Feature/osi doxygen

README

@@ -47,4 +47,16 @@ Version history:
     not moving member comments before the member but keeping it after
     the member instead
      * these changes lead into need of enabling JAVADOC_AUTOBRIEF
-  - added steps for enabling the filter in Doxygen in this file
+  - added steps for enabling the filter in Doxygen in this file
+--------------------
+ 0.7-beta (2018-04-19) OSI
+  - Include changes from University of California.
+  - Support for all OSI *.proto files.
+  - Separate statement and comments to treat both parts differently (remove bugs 
+    regarding string modifications).
+  - Remove "option" statements.
+  - Add support for "extend" statements.
+  - Change "repeat" from Template to standard member. --> Better collaboration 
+    diagrams.
+  - Fix problems with references of nested messages (replace "." with "::").
+  - Change mapping from C to C++.

doxyfile

@@ -32,7 +32,7 @@ PROJECT_NAME           = "Protocol Buffers demo project"
 # This could be handy for archiving the generated documentation or
 # if some version control system is used.
 
-PROJECT_NUMBER         = 0.6-beta
+PROJECT_NUMBER         = 0.7-beta
 
 # Using the PROJECT_BRIEF tag one can provide an optional one line description
 # for a project that appears at the top of each page and should give viewer
@@ -238,7 +238,7 @@ OPTIMIZE_OUTPUT_VHDL   = NO
 # (default is Fortran), use: inc=Fortran f=C. Note that for custom extensions
 # you also need to set FILE_PATTERNS otherwise the files are not read by doxygen.
 
-EXTENSION_MAPPING      = proto=C
+EXTENSION_MAPPING      = proto=C++
 
 # If you use STL classes (i.e. std::string, std::vector, etc.) but do not want
 # to include (a tag file for) the STL sources as input, then you should

proto2cpp.py

@@ -1,9 +1,10 @@
+#!/usr/bin/env python
 ##
 # Doxygen filter for Google Protocol Buffers .proto files.
 # This script converts .proto files into C++ style ones
 # and prints the output to standard output.
 #
-# version 0.6-beta
+# version 0.7-beta OSI
 #
 # How to enable this filter in Doxygen:
 #   1. Generate Doxygen configuration file with command 'doxygen -g <filename>'
@@ -12,15 +13,17 @@
 #        JAVADOC_AUTOBRIEF      = YES
 #   3. In the Doxygen configuration file, find FILE_PATTERNS and add *.proto
 #        FILE_PATTERNS          = *.proto
-#   4. In the Doxygen configuration file, find EXTENSION_MAPPING and add proto=C
+#   4. In the Doxygen configuration file, find EXTENSION_MAPPING and add proto=C++
 #        EXTENSION_MAPPING      = proto=C
 #   5. In the Doxygen configuration file, find INPUT_FILTER and add this script
 #        INPUT_FILTER           = "python proto2cpp.py"
 #   6. Run Doxygen with the modified configuration
 #        doxygen doxyfile
 #
 #
-# Copyright (C) 2012-2015 Timo Marjoniemi
+# Version 0.7 2018 Bugfix and extensions have been made by Open Simulation Interface (OSI) Carsten Kuebler https://github.com/OpenSimulationInterface
+# Copyright (C) 2016 Regents of the University of California https://github.com/vgteam/vg
+# Copyright (C) 2012-2015 Timo Marjoniemi https://sourceforge.net/p/proto2cpp/wiki/Home/
 # All rights reserved.
 #
 # This library is free software; you can redistribute it and/or
@@ -72,7 +75,7 @@ def __init__(self):
     self.logFile = "proto2cpp.log"
     ## Error log file name.
     self.errorLogFile = "proto2cpp.error.log"
-    ## Logging level.
+    ## Logging level. 
     self.logLevel = self.logNone
 
   ## Handles a file.
@@ -88,7 +91,7 @@ def handleFile(self, fileName):
       self.log('\nXXXXXXXXXX\nXX ' + filename + '\nXXXXXXXXXX\n\n')
       # Open the file. Use try to detect whether or not we have an actual file.
       try:
-        with open(filename, 'r') as inputFile:
+        with open(filename, 'r', encoding='utf8') as inputFile:
           self.parseFile(inputFile)
         pass
       except IOError as e:
@@ -97,7 +100,7 @@ def handleFile(self, fileName):
     elif not fnmatch.fnmatch(filename, os.path.basename(inspect.getfile(inspect.currentframe()))):
       self.log('\nXXXXXXXXXX\nXX ' + filename + '\nXXXXXXXXXX\n\n')
       try:
-        with open(filename, 'r') as theFile:
+        with open(filename, 'r', encoding='utf8') as theFile:
           output = ''
           for theLine in theFile:
             output += theLine
@@ -120,6 +123,7 @@ def handleFile(self, fileName):
   def parseFile(self, inputFile):
     # Go through the input file line by line.
     isEnum = False
+    inPackage = False
     # This variable is here as a workaround for not getting extra line breaks (each line
     # ends with a line separator and print() method will add another one).
     # We will be adding lines into this var and then print the var out at the end.
@@ -129,38 +133,87 @@ def parseFile(self, inputFile):
       # block to make Doxygen detect it.
       matchComment = re.search("//", line)
       # Search for semicolon and if one is found before comment, add a third slash character
-      # ("/") and a smaller than ("<") chracter to the comment to make Doxygen detect it.
+      # ("/") and a smaller than ("<") character to the comment to make Doxygen detect it.
       matchSemicolon = re.search(";", line)
       if matchSemicolon is not None and (matchComment is not None and matchSemicolon.start() < matchComment.start()):
-        line = line[:matchComment.start()] + "///<" + line[matchComment.end():]
+        comment = "///<" + line[matchComment.end():]
+        # Replace '.' in nested message references with '::'
+        # don't work for multi-nested references and generates problems with URLs and acronyms 
+        #comment = re.sub(r'\s(\w+)\.(\w+)\s', r' \1::\2 ', comment)
+        line = line[:matchComment.start()]
       elif matchComment is not None:
-        line = line[:matchComment.start()] + "///" + line[matchComment.end():]
+        comment = "///" + line[matchComment.end():]
+        # replace '.' in nested message references with '::'
+        # don't work for multi-nested references and generates problems with URLs and acronyms 
+        #comment = re.sub(r'\s(\w+)\.(\w+)\s', r' \1::\2 ', comment)
+        line = line[:matchComment.start()]
+      else:
+        comment = ""
+
+      # line = line.replace(".", "::") but not in quoted strings (Necessary for import statement)
+      line = re.sub(r'\.(?=(?:[^"]*"[^"]*")*[^"]*$)',r'::',line)
 
-      # Search for "enum" and if one is found before comment,
-      # start changing all semicolons (";") to commas (",").
-      matchEnum = re.search("enum", line)
-      if matchEnum is not None and (matchComment is None or matchEnum.start() < matchComment.start()):
+      # Search for " option ...;", remove it
+      line = re.sub(r'\boption\b[^;]+;', r'', line)
+
+      # Search for " package ", make a namespace
+      matchPackage = re.search(r"\bpackage\b", line)
+      if matchPackage is not None:
+        isPackage = True
+        # Convert to C++-style separator and block instead of statement
+        line = "namespace" + line[:matchPackage.start()] + line[matchPackage.end():].replace(";", " {")
+
+      # Search for " repeated " fields and make them ...
+      #matchRepeated = re.search(r"\brepeated\b", line)
+      #if matchRepeated is not None:
+      #  # Convert
+      #  line = re.sub(r'\brepeated\s+(\S+)', r' repeated \1', line)
+
+      # Search for "enum", start changing all semicolons (";") to commas (",").
+      matchEnum = re.search(r"\benum\b", line)
+      if matchEnum is not None:
         isEnum = True
+
       # Search again for semicolon if we have detected an enum, and replace semicolon with comma.
       if isEnum is True and re.search(";", line) is not None:
         matchSemicolon = re.search(";", line)
         line = line[:matchSemicolon.start()] + "," + line[matchSemicolon.end():]
+
       # Search for a closing brace.
-      matchClosingBrace = re.search("}", line[:matchComment.start()] if matchComment else line)
+      matchClosingBrace = re.search("}", line)
       if isEnum is True and matchClosingBrace is not None:
         line = line[:matchClosingBrace.start()] + "};" + line[matchClosingBrace.end():]
         isEnum = False
       elif isEnum is False and matchClosingBrace is not None:
         # Message (to be struct) ends => add semicolon so that it'll
         # be a proper C(++) struct and Doxygen will handle it correctly.
         line = line[:matchClosingBrace.start()] + "};" + line[matchClosingBrace.end():]
-      # Search for 'message' and replace it with 'struct' unless 'message' is behind a comment.
-      matchMsg = re.search("message", line)
-      if matchMsg is not None and (matchComment is None or matchMsg.start() < matchComment.start()):
-        output = "struct" + line[:matchMsg.start()] + line[matchMsg.end():]
-        theOutput += output
-      else:
-        theOutput += line
+
+      # Replacements change start of comment...
+      matchMsg = re.search(r"\bmessage\b", line)
+      if matchMsg is not None:
+        line = line[:matchMsg.start()] + "struct" + line[matchMsg.end():]
+
+      # Replacements change start of comment...
+      matchExt = re.search(r"\bextend\b", line)
+      if matchExt is not None:
+        a_extend = line[matchExt.end():]
+        matchName = re.search(r"\b\w[\S:]+\b", a_extend)
+        if matchName is not None:
+          name = a_extend[matchName.start():matchName.end()]
+          name = re.sub(r'\w+::',r'',name)
+          a_extend = a_extend[:matchName.start()] + name + ": public " + a_extend[matchName.start():]
+        else:
+          a_extend = "_Dummy: public " + a_extend;   
+        line = line[:matchExt.start()] + "struct " + a_extend
+
+      theOutput += line + comment
+
+    if isPackage:
+      # Close the package namespace
+      theOutput += "}"
+      isPackage = False
+
     # Now that we've got all lines in the string let's split the lines and print out
     # one by one.
     # This is a workaround to get rid of extra empty line at the end which print() method adds.