Refactored definition of supported languages

antiweb.py

@@ -94,18 +94,21 @@
 
 Read the documentation of the corresponding event file handler (:ref:`FileChangeHandler <label-filechangehandler>`).
 
+.. _label-supported_languages:
+
+*******************
+Supported Languages
+*******************
+
+#@include(supported_languages doc, antiweb_lib\readers\config.py)
+
+.. _label-add_language:
 
 ************************
 How to add new languages
 ************************
 
-New languages are added by writing a new Reader class
-and registering it in the readers dictionary (see readers).
-A simple Reader example is provides by :py:class:`CReader`
-a more advanced reader is :py:class:`PythonReader`.
-
-#@include(comments doc, antiweb_lib\readers\config.py)
-#@include(get_comment_markers doc, antiweb_lib\document.py)
+#@include(new_language doc, antiweb_lib\readers\config.py)
 
 *******
 Example
@@ -233,17 +236,21 @@
 from watchdog.observers import Observer
 from antiweb_lib.filechangehandler import FileChangeHandler
 
+from antiweb_lib.readers.config import is_file_supported
+
 #@rstart(management)
 
 #<<management>>
 #==============
 
 #@code
 
-__version__ = "0.9.1"
-
 logger = logging.getLogger('antiweb')
 
+def sys_exit(message):
+    logger.error(message)
+    sys.exit(1)
+
 #@cstart(parsing)
 def parsing():
 #@start(parsing doc)
@@ -282,7 +289,7 @@ def parsing():
     if not args:
         args.append(os.getcwd())
     # parsing() returns the selected options, arguments (the filepath/folderpath) and the parser
-    return (options, args, parser)
+    return options, args, parser
 #@(parsing)
 
 def main():
@@ -323,8 +330,7 @@ def main():
         #The program aborts if the directory does not exist or if the path refers to a file.
         #A file is not allowed here because the -r option requires a directory.
         if not os.path.isdir(directory):
-            logger.error("directory not found: %s", directory)
-            sys.exit(1)
+            sys_exit("directory not found: %s" % directory)
 
         os.chdir(directory)
 
@@ -335,22 +341,18 @@ def main():
 
 #@code
 
-        #Only files with the following extensions will be processed
-        rst_extension = ".rst"
-        ext_tuple = (".cs",".cpp",".py",".cc", rst_extension, ".xml")
-
         handled_files = []
 
         for root, dirs, files in os.walk(directory, topdown=False):
             for filename in files:
                 fname = os.path.join(root, filename)
 
-                if not (os.path.isfile(fname) and fname.endswith(ext_tuple)):
+                if not (os.path.isfile(fname) and is_file_supported(fname)):
                     continue
 
                 # rst files should be handled last as they might be a documentation file of a
                 # file that is not yet processed -> in this case the rst file will be ignored
-                if fname.endswith(rst_extension):
+                if fname.endswith(".rst"):
                     handled_files.append(fname)
                 else:
                     handled_files.insert(0, fname)
@@ -381,7 +383,7 @@ def main():
             try:
                 #observed directory => input directory
                 #recursive option is true in order to monitor all subdirectories
-                observer.schedule(FileChangeHandler(directory, ext_tuple, options, created_files), path=directory, recursive=True)
+                observer.schedule(FileChangeHandler(directory, options, created_files), path=directory, recursive=True)
 
                 print("\n------- starting daemon mode (exit with enter or ctrl+c) -------\n")
 
@@ -409,8 +411,10 @@ def main():
         #The program aborts if the file does not exist or if the path refers to a directory.
         #A directory is not allowed here because a directory can only be used with the -r option.
         if not os.path.isfile(absolute_file_path):
-            logger.error("file not found: %s", absolute_file_path)
-            sys.exit(1)
+            sys_exit("file not found: %s" % absolute_file_path)
+
+        if not is_file_supported(absolute_file_path):
+            sys_exit("file is not supported: %s" % absolute_file_path)
 
         directory = os.path.split(absolute_file_path)[0]
 

antiweb_lib/document.py

@@ -8,19 +8,10 @@
 import os
 import operator
 import logging
-import pygments.lexers as pm
 
-from antiweb_lib.readers.config import readers, comments
 
 from antiweb_lib.readers.Line import Line
-
-from antiweb_lib.readers.Reader import Reader
-from antiweb_lib.readers.CReader import CReader
-from antiweb_lib.readers.CSharpReader import CSharpReader
-from antiweb_lib.readers.ClojureReader import ClojureReader
-from antiweb_lib.readers.GenericReader import GenericReader
-from antiweb_lib.readers.RstReader import RstReader
-from antiweb_lib.readers.PythonReader import PythonReader
+from antiweb_lib.readers.config import get_reader_for_file
 
 
 logger = logging.getLogger('antiweb')
@@ -266,9 +257,7 @@ def insert_macros(subdoc):
 
         else:
             #parse the file
-            lexer = pm.get_lexer_for_filename(rpath)
-            single_comment_markers,  block_comment_markers = get_comment_markers(lexer.name)
-            reader = readers.get(lexer.name, Reader)(lexer, single_comment_markers,  block_comment_markers)
+            reader = get_reader_for_file(fpath)
 
             doc = Document(text, reader, rpath, self.tokens)
             doc.collect_blocks()
@@ -388,27 +377,4 @@ def find_next_directive(block):
         return block
     #@edoc
     #@rinclude(find_next_directive)
-    #@(Document.compile_block)
-
-#@cstart(get_comment_markers)
-
-def get_comment_markers(lexer_name):
-    #@start(get_comment_markers doc)
-    #From the map above the comment markers are retrieved via the following method:
-
-    """
-    ..  py:function:: get_comment_markers(lexer_name)
-
-        Retrieves the language specific comment markers from the comments map.
-        The comment markers of C serves as the default comment markers if the lexer name cannot be found.
-
-        :param string lexer_name: The name of the pygments lexer.
-        :return: The single and comment block markers defined by the language
-    """
-    #@indent 4
-    #@include(get_comment_markers)
-    #@(get_comment_markers doc)
-    comment_markers = comments.get(lexer_name, comments["C"])
-    single_comment_markers = comment_markers[0]
-    block_comment_markers = comment_markers[1]
-    return single_comment_markers,  block_comment_markers
+    #@(Document.compile_block)

antiweb_lib/filechangehandler.py

@@ -8,6 +8,7 @@
 from watchdog.events import FileSystemEventHandler
 from antiweb_lib.write import write
 from antiweb_lib.write import create_write_string
+from antiweb_lib.readers.config import is_file_supported
 import time
 
 #@start()
@@ -40,7 +41,6 @@ class FileChangeHandler(FileSystemEventHandler):
        This handler is responsible for handling changed file events in antiweb's daemon mode.
 
        :param string directory: absolute path to the monitored source directory
-       :param tuple<string> extensions: contains all handled file extensions ("*.cs", "*.py", etc)
        :param options: antiweb commandline options
        :param created_files: a set which contains the absolute paths of all previously created documentation files
 
@@ -49,11 +49,10 @@ class FileChangeHandler(FileSystemEventHandler):
 
     #@(FileChangeHandler doc)
 
-    def __init__(self, directory, extensions, options, created_files):
+    def __init__(self, directory, options, created_files):
         self._directory = directory
         #antiweb commandline options
         self._options = options
-        self._handled_extensions = extensions
         self._event_counter = 0
         self._created_files = set()
         self._created_files.update(created_files)
@@ -87,7 +86,7 @@ def process_event(self, event):
             #the file has been moved so it is now located in event.dest_path
             changed_file = event.dest_path
 
-        ignore_change = changed_file in self._created_files or not changed_file.endswith(self._handled_extensions) or \
+        ignore_change = changed_file in self._created_files or not is_file_supported(changed_file) or \
                   event.is_directory or event.event_type == "deleted"
 
         if not ignore_change:

antiweb_lib/readers/ClojureReader.py

@@ -50,6 +50,6 @@ def filter_output(self, lines):
                     #remove comments but not chapters
                     l.text = l.indented(stext[1:])
 
-        yield l
+            yield l
     #@edoc
     #@(ClojureReader)

antiweb_lib/readers/Language.py

@@ -0,0 +1,76 @@
+__author__ = "Michael Reithinger, Philipp Rathmanner, Lukas Tanner, Philipp Grandits, and Christian Eitner"
+__copyright__ = "Copyright 2017, antiweb team"
+__license__ = "GPL"
+__version__ = "0.9.1"
+__maintainer__ = "antiweb team"
+__email__ = "antiweb@freelists.org"
+
+import sys
+import pygments.lexers as pm
+from pygments.util import ClassNotFound
+
+import logging
+
+logger = logging.getLogger('antiweb')
+
+"""
+@start()
+.. _label-language:
+
+@include(Language doc)
+"""
+
+
+#@cstart(Language)
+class Language(object):
+    #@start(Language doc)
+    #Language
+    #========
+    """
+    .. py:class:: Language(name, reader, single_comments, block_comments)
+
+       This class represents a supported language of antiweb.
+    """
+
+    #@include(Language)
+    #@include(Language.__init__ doc)
+    #@include(Language.get_reader doc)
+    #@(Language doc)
+
+
+    #@cstart(Language.__init__)
+    def __init__(self, name, reader, single_comments, block_comments):
+        """
+        .. py:method:: __init__(name, reader, single_comments, block_comments)
+
+           The constructor.
+           The comment markers of a language have to be defined in the format:
+           ``[single_comment_tokens]`` and ``[start_block_token, end_block_token]``.
+           Multiple single and block comment markers can be defined.
+
+           :param string name: the name of a pygments lexer.
+           :param class reader: the reader class that should be used for the corresponding language.
+           :param list single_comments: a list of single comment characters supported by the language (e.g. ['#']).
+           :param list<tuple> block_comments: a list of block comment character tuples supported by the language (e.g. ["/*","*/"]).
+        """
+        try:
+            self.lexer = pm.get_lexer_by_name(name)
+        except ClassNotFound:
+            logger.error("\nError: No lexer for alias: '%s' found", name)
+            sys.exit(1)
+
+        self.reader = reader
+        self.single_comments = single_comments
+        self.block_comments = block_comments
+
+        #the lexer filenames have the format: ['*.cs', '*.cpp', ..]
+        self.supported_files = self.lexer.filenames
+
+    #@cstart(Language.get_reader)
+    def get_reader(self):
+        """
+        .. py:method:: get_reader()
+
+           returns a new instance of the :py:attr:`reader` class.
+        """
+        return self.reader(self.lexer, self.single_comments, self.block_comments)

antiweb_lib/readers/Reader.py

@@ -28,7 +28,6 @@
 @include(GenericReader doc, GenericReader.py)
 @include(RstReader doc, RstReader.py)
 @include(XmlReader doc, XmlReader.py)
-@include(reader_dictionary doc, config.py)
 """
 #@rstart(readers)
 #.. _readers: