''' Copyright 2011 by the Massachusetts Institute of Technology. All Rights Reserved. Export of this software from the United States of America may require a specific license from the United States Government. It is the responsibility of any person or organization contemplating export to obtain such a license before exporting. WITHIN THAT CONSTRAINT, permission to use, copy, modify, and distribute this software and its documentation for any purpose and without fee is hereby granted, provided that the above copyright notice appear in all copies and that both that copyright notice and this permission notice appear in supporting documentation, and that the name of M.I.T. not be used in advertising or publicity pertaining to distribution of the software without specific, written prior permission. Furthermore if you modify this software you must label your software as modified software and not distribute it in such a fashion that it might be confused with the original M.I.T. software. M.I.T. makes no representations about the suitability of this software for any purpose. It is provided "as is" without express or implied warranty. ''' import sys import re from collections import defaultdict from xml.sax import make_parser from xml.sax.handler import ContentHandler from docmodel import * exclude_funcs = ['krb5_free_octet_data'] class DocNode(object): """ Represents the structure of xml node. """ def __init__(self, name): """ @param node: name - the name of a node. @param attributes: a dictionary populated with attributes of a node @param children: a dictionary with lists of children nodes. Nodes in lists are ordered as they appear in a document. @param content: a content of xml node represented as a list of tuples [(type,value)] with type = ['char'|'element']. If type is 'char' then the value is a character string otherwise it is a reference to a child node. """ self.name = name self.content = list() self.attributes = dict() self.children = defaultdict(list) def walk(self, decorators, sub_ws, stack=[]): result = list() decorator = decorators.get(self.name, decorators['default']) stack.append(decorators['default']) decorators['default'] = decorator for (obj_type,obj) in self.content: if obj_type == 'char': if obj != '': result.append(obj) else: partial = obj.walk(decorators,1, stack) if partial is not None: result.append(' %s ' % partial) decorators['default'] = stack.pop() result = decorator(self, ''.join(result)) if result is not None: if sub_ws == 1: result = re.sub(r'[ ]+', r' ', result) else: result = result.strip() return result def getContent(self): decorators = {'default': lambda node,value: value} result = self.walk(decorators, 1) if len(result) == 0: result = None return result def __repr__(self): result = ['Content: %s' % self.content] for (key,value) in self.attributes.iteritems(): result.append('Attr: %s = %s' % (key,value)) for (key,value) in self.children.iteritems(): result.append('Child: %s,%i' % (key,len(value))) return '\n'.join(result) class DoxyContenHandler(object, ContentHandler): def __init__(self, builder): self.builder = builder self.counters = defaultdict(int) self._nodes = None self._current = None def startDocument(self): pass def endDocument(self): import sys def startElement(self, name, attrs): if name == self.builder.toplevel: self._nodes = [] if name == 'memberdef': kind = attrs.get('kind') if kind is None: raise ValueError('Kind is not defined') self.counters[kind] += 1 if self._nodes is None: return node = DocNode(name) for (key,value) in attrs.items(): node.attributes[key] = value if self._current is not None: self._current.children[name].append(node) self._nodes.append(self._current) self._current = node def characters(self, content): if self._current is not None: self._current.content.append(('char',content.strip())) def endElement(self, name): if name == self.builder.toplevel: assert(len(self._nodes) == 0) self._nodes = None self.builder.document.append(self._current) self._current = None else: if self._nodes is not None: node = self._current self._current = self._nodes.pop() self._current.content.append(('element',node)) class XML2AST(object): """ Translates XML document into Abstract Syntax Tree like representation The content of document is stored in self.document """ def __init__(self, xmlpath, toplevel='doxygen'): self.document = list() self.toplevel = toplevel self.parser = make_parser() handler = DoxyContenHandler(self) self.parser.setContentHandler(handler) filename = 'krb5_8hin.xml' filepath = '%s/%s' % (xmlpath,filename) self.parser.parse(open(filepath,'r')) class DoxyFuncs(XML2AST): def __init__(self, path): super(DoxyFuncs, self).__init__(path,toplevel='memberdef') self.objects = list() def run(self): for node in self.document: self.process(node) def process(self, node): node_type = node.attributes['kind'] if node_type == 'function': data = self._process_function_node(node) else: return if 'name' in data and data['name'] in exclude_funcs: return self.objects.append(DocModel(**data)) def save(self, templates, target_dir): for obj in self.objects: template_path = templates[obj.category] outpath = '%s/%s.rst' % (target_dir,obj.name) obj.save(outpath, template_path) def _process_function_node(self, node): f_name = node.children['name'][0].getContent() f_Id = node.attributes['id'] f_ret_type = self._process_type_node(node.children['type'][0]) f_brief = node.children['briefdescription'][0].getContent() f_detailed = node.children['detaileddescription'][0] detailed_description = self._process_description_node(f_detailed) return_value_description = self._process_return_value_description(f_detailed) retval_description = self._process_retval_description(f_detailed) warning_description = self._process_warning_description(f_detailed) seealso_description = self._process_seealso_description(f_detailed) notes_description = self._process_notes_description(f_detailed) f_version = self._process_version_description(f_detailed) deprecated_description = self._process_deprecated_description(f_detailed) param_description_map = self.process_parameter_description(f_detailed) f_definition = node.children['definition'][0].getContent() f_argsstring = node.children['argsstring'][0].getContent() function_descr = {'category': 'function', 'name': f_name, 'Id': f_Id, 'return_type': f_ret_type[1], 'return_description': return_value_description, 'retval_description': retval_description, 'sa_description': seealso_description, 'warn_description': warning_description, 'notes_description': notes_description, 'short_description': f_brief, 'version_num': f_version, 'long_description': detailed_description, 'deprecated_description': deprecated_description, 'parameters': list()} parameters = function_descr['parameters'] for (i,p) in enumerate(node.children['param']): type_node = p.children['type'][0] p_type = self._process_type_node(type_node) if p_type[1].find('...') > -1 : p_name = '' else: p_name = None p_name_node = p.children.get('declname') if p_name_node is not None: p_name = p_name_node[0].getContent() (p_direction,p_descr) = param_description_map.get(p_name,(None,None)) param_descr = {'seqno': i, 'name': p_name, 'direction': p_direction, 'type': p_type[1], 'typeId': p_type[0], 'description': p_descr} parameters.append(param_descr) result = Function(**function_descr) print >> self.tmp, result return function_descr def _process_type_node(self, type_node): """ Type node has form type_string for build in types and 'type_name' postfix (ex. *, **m, etc.) for user defined types. """ type_ref_node = type_node.children.get('ref') if type_ref_node is not None: p_type_id = type_ref_node[0].attributes['refid'] else: p_type_id = None p_type = type_node.getContent() # remove some macros p_type = re.sub('KRB5_ATTR_DEPRECATED', '', p_type) p_type = re.sub('KRB5_CALLCONV_C', '', p_type) p_type = re.sub('KRB5_CALLCONV_WRONG', '', p_type) p_type = re.sub('KRB5_CALLCONV', '', p_type) p_type = p_type.strip() return (p_type_id, p_type) def _process_description_node(self, node): """ Description node is comprised of ... sections """ para = node.children.get('para') result = list() if para is not None: decorators = {'default': self.paragraph_content_decorator} for e in para: result.append(str(e.walk(decorators, 1))) result.append('\n') result = '\n'.join(result) return result def return_value_description_decorator(self, node, value): if node.name == 'simplesect': if node.attributes['kind'] == 'return': cont = set() cont = node.getContent() return value else: return None def paragraph_content_decorator(self, node, value): if node.name == 'para': return value + '\n' elif node.name == 'simplesect': if node.attributes['kind'] == 'return': return None elif node.name == 'ref': if value.find('()') >= 0: # functions return ':c:func:' + '`' + value + '`' else: # macro's return ':data:' + '`' + value + '`' elif node.name == 'emphasis': return '*' + value + '*' elif node.name == 'itemizedlist': return '\n' + value elif node.name == 'listitem': return '\n\t - ' + value + '\n' elif node.name == 'computeroutput': return '**' + value + '**' else: return None def parameter_name_decorator(self, node, value): if node.name == 'parametername': direction = node.attributes.get('direction') if direction is not None: value = '%s:%s' % (value,direction) return value elif node.name == 'parameterdescription': return None else: return value def parameter_description_decorator(self, node, value): if node.name == 'parameterdescription': return value elif node.name == 'parametername': return None else: return value def process_parameter_description(self, node): """ Parameter descriptions reside inside detailed description section. """ para = node.children.get('para') result = dict() if para is not None: for e in para: param_list = e.children.get('parameterlist') if param_list is None: continue param_items = param_list[0].children.get('parameteritem') if param_items is None: continue for it in param_items: decorators = {'default': self.parameter_name_decorator} direction = None name = it.walk(decorators,0).split(':') if len(name) == 2: direction = name[1] decorators = {'default': self.parameter_description_decorator, 'para': self.paragraph_content_decorator} description = it.walk(decorators, 0) result[name[0]] = (direction,description) return result def _process_return_value_description(self, node): result = None ret = list() para = node.children.get('para') if para is not None: for p in para: simplesect_list = p.children.get('simplesect') if simplesect_list is None: continue for it in simplesect_list: decorators = {'default': self.return_value_description_decorator, 'para': self.parameter_name_decorator} result = it.walk(decorators, 1) if result is not None: ret.append(result) return ret def _process_retval_description(self, node): """ retval descriptions reside inside detailed description section. """ para = node.children.get('para') result = None ret = list() if para is not None: for e in para: param_list = e.children.get('parameterlist') if param_list is None: continue for p in param_list: kind = p.attributes['kind'] if kind == 'retval': param_items = p.children.get('parameteritem') if param_items is None: continue for it in param_items: param_descr = it.children.get('parameterdescription') if param_descr is not None: val = param_descr[0].children.get('para') if val is not None: val_descr = val[0].getContent() else: val_descr ='' decorators = {'default': self.parameter_name_decorator} name = it.walk(decorators, 1).split(':') val = name[0] result = " %s %s" % (val, val_descr) ret.append (result) return ret def return_warning_decorator(self, node, value): if node.name == 'simplesect': if node.attributes['kind'] == 'warning': return value else: return None def _process_warning_description(self, node): result = None para = node.children.get('para') if para is not None: for p in para: simplesect_list = p.children.get('simplesect') if simplesect_list is None: continue for it in simplesect_list: decorators = {'default': self.return_warning_decorator, 'para': self.paragraph_content_decorator} result = it.walk(decorators, 1) # Assuming that only one Warning per function if result is not None: return result return result def return_seealso_decorator(self, node, value): if node.name == 'simplesect': if node.attributes['kind'] == 'see': return value else: return None def _process_seealso_description(self, node): result = None para = node.children.get('para') if para is not None: for p in para: simplesect_list = p.children.get('simplesect') if simplesect_list is None: continue for it in simplesect_list: decorators = {'default': self.return_seealso_decorator, 'para': self.paragraph_content_decorator} result = it.walk(decorators, 1) return result def return_version_decorator(self, node, value): if node.name == 'simplesect': if node.attributes['kind'] == 'version': return value else: return None def _process_version_description(self, node): result = None para = node.children.get('para') if para is not None: for p in para: simplesect_list = p.children.get('simplesect') if simplesect_list is None: continue for it in simplesect_list: decorators = {'default': self.return_version_decorator, 'para': self.paragraph_content_decorator} result = it.walk(decorators, 1) if result is not None: return result return result def return_notes_decorator(self, node, value): if node.name == 'simplesect': if node.attributes['kind'] == 'note': # We indent notes with an extra tab. Do it for all paragraphs. return value.replace("\n ", "\n\n\t "); else: return None def _process_notes_description(self, node): result = None para = node.children.get('para') if para is not None: for p in para: simplesect_list = p.children.get('simplesect') if simplesect_list is None: continue for it in simplesect_list: decorators = {'default': self.return_notes_decorator, 'para': self.paragraph_content_decorator} result = it.walk(decorators, 1) if result is not None: return result return result def return_deprecated_decorator(self, node, value): if node.name == 'xrefsect': if node.attributes['id'].find('deprecated_') > -1: xreftitle = node.children.get('xreftitle') if xreftitle[0] is not None: xrefdescr = node.children.get('xrefdescription') deprecated_descr = "DEPRECATED %s" % xrefdescr[0].getContent() return deprecated_descr else: return None def _process_deprecated_description(self, node): result = None para = node.children.get('para') if para is not None: for p in para: xrefsect_list = p.children.get('xrefsect') if xrefsect_list is None: continue for it in xrefsect_list: decorators = {'default': self.return_deprecated_decorator, 'para': self.paragraph_content_decorator} result = it.walk(decorators, 1) if result is not None: return result return result def break_into_lines(self, value, linelen=82): breaks = range(0,len(value),linelen) + [len(value)] result = list() for (start,end) in zip(breaks[:-1],breaks[1:]): result.append(value[start:end]) result = '\n'.join(result) return result def _save(self, table, path = None): if path is None: f = sys.stdout else: f = open(path, 'w') for l in table: f.write('%s\n' % ','.join(l)) if path is not None: f.close() class DoxyBuilderFuncs(DoxyFuncs): def __init__(self, xmlpath, rstpath): super(DoxyBuilderFuncs,self).__init__(xmlpath) self.target_dir = rstpath outfile = '%s/%s' % (self.target_dir, 'out.txt') self.tmp = open(outfile, 'w') def run_all(self): self.run() templates = {'function': 'func_document.tmpl'} self.save(templates, self.target_dir) def test_run(self): self.run() if __name__ == '__main__': builder = DoxyBuilderFuncs(xmlpath, rstpath) builder.run_all()