]> git.ipfire.org Git - thirdparty/libvirt.git/commitdiff
* Makefile.am configure.in docs/apibuild.py docs/libxen-api.xml
authorDaniel Veillard <veillard@redhat.com>
Thu, 1 Dec 2005 17:34:21 +0000 (17:34 +0000)
committerDaniel Veillard <veillard@redhat.com>
Thu, 1 Dec 2005 17:34:21 +0000 (17:34 +0000)
  docs/libxen-refs.xml docs/structures.fig: starting to add docs
  and the extraction tool
* src/libxen.c: fixed comments error raised by apibuild
Daniel

ChangeLog
Makefile.am
configure.in
docs/apibuild.py [new file with mode: 0755]
docs/libxen-api.xml [new file with mode: 0644]
docs/libxen-refs.xml [new file with mode: 0644]
docs/structures.fig [new file with mode: 0644]
src/libxen.c

index 4a4ce8931ae203b5bf9faae0544f4c105e384e5d..02149f7093dcccd39411bab947223ba1451774bd 100644 (file)
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,10 @@
+Thu Dec  1 18:32:43 CET 2005 Daniel Veillard <veillard@redhat.com>
+
+       * Makefile.am configure.in docs/apibuild.py docs/libxen-api.xml
+         docs/libxen-refs.xml docs/structures.fig: starting to add docs
+         and the extraction tool
+       * src/libxen.c: fixed comments error raised by apibuild
+
 Thu Dec  1 17:34:23 CET 2005 Daniel Veillard <veillard@redhat.com>
 
        * include/libxen.h src/libxen.c src/libxen_sym.version: add read-only
index 0c6d1b577b272cda548b2f03c14ef2d2937e0297..9c6812c7c9a12837a63c714c4419890488b7d502 100644 (file)
@@ -1,6 +1,6 @@
 ## Process this file with automake to produce Makefile.in
 
-SUBDIRS = src include #docs
+SUBDIRS = src include docs
 EXTRA_DIST = libxen.spec.in libxen.spec COPYING.LIB \
              libxen.pc.in libxen.pc TODO AUTHORS ChangeLog \
             NEWS README 
index 11e9c96611c7ded117f14ad74b630debaf7498a7..6e049c8c78db0f53550c04fd15121d8cabea99c9 100644 (file)
@@ -51,4 +51,4 @@ dnl search for the low level Xen library
 AC_SEARCH_LIBS(xc_domain_create, [xenctrl], [], [AC_MSG_ERROR([Xen control library not found])])
 AC_SEARCH_LIBS(xs_read, [xenstore], [], [AC_MSG_ERROR([Xen store library not found])])
 
-AC_OUTPUT(Makefile src/Makefile include/Makefile libxen.pc libxen.spec)
+AC_OUTPUT(Makefile src/Makefile include/Makefile docs/Makefile libxen.pc libxen.spec)
diff --git a/docs/apibuild.py b/docs/apibuild.py
new file mode 100755 (executable)
index 0000000..b7efe1e
--- /dev/null
@@ -0,0 +1,2076 @@
+#!/usr/bin/python -u
+#
+# This is the API builder, it parses the C sources and build the
+# API formal description in XML.
+#
+# See Copyright for the status of this software.
+#
+# daniel@veillard.com
+#
+import os, sys
+import string
+import glob
+
+debug=0
+debugsym=None
+
+#
+# C parser analysis code
+#
+ignored_files = {
+  "xensh.c": "testing tool",
+  "hash.c": "internal hash table stuff",
+  "hash.h": "internal hash table stuff",
+}
+
+ignored_words = {
+  "ATTRIBUTE_UNUSED": (0, "macro keyword"),
+}
+
+def escape(raw):
+    raw = string.replace(raw, '&', '&amp;')
+    raw = string.replace(raw, '<', '&lt;')
+    raw = string.replace(raw, '>', '&gt;')
+    raw = string.replace(raw, "'", '&apos;')
+    raw = string.replace(raw, '"', '&quot;')
+    return raw
+
+def uniq(items):
+    d = {}
+    for item in items:
+        d[item]=1
+    return d.keys()
+
+class identifier:
+    def __init__(self, name, header=None, module=None, type=None, lineno = 0,
+                 info=None, extra=None, conditionals = None):
+        self.name = name
+       self.header = header
+       self.module = module
+       self.type = type
+       self.info = info
+       self.extra = extra
+       self.lineno = lineno
+       self.static = 0
+       if conditionals == None or len(conditionals) == 0:
+           self.conditionals = None
+       else:
+           self.conditionals = conditionals[:]
+       if self.name == debugsym:
+           print "=> define %s : %s" % (debugsym, (module, type, info,
+                                        extra, conditionals))
+
+    def __repr__(self):
+        r = "%s %s:" % (self.type, self.name)
+       if self.static:
+           r = r + " static"
+       if self.module != None:
+           r = r + " from %s" % (self.module)
+       if self.info != None:
+           r = r + " " +  `self.info`
+       if self.extra != None:
+           r = r + " " + `self.extra`
+       if self.conditionals != None:
+           r = r + " " + `self.conditionals`
+       return r
+
+
+    def set_header(self, header):
+        self.header = header
+    def set_module(self, module):
+        self.module = module
+    def set_type(self, type):
+        self.type = type
+    def set_info(self, info):
+        self.info = info
+    def set_extra(self, extra):
+        self.extra = extra
+    def set_lineno(self, lineno):
+        self.lineno = lineno
+    def set_static(self, static):
+        self.static = static
+    def set_conditionals(self, conditionals):
+       if conditionals == None or len(conditionals) == 0:
+           self.conditionals = None
+       else:
+           self.conditionals = conditionals[:]
+
+    def get_name(self):
+        return self.name
+    def get_header(self):
+        return self.module
+    def get_module(self):
+        return self.module
+    def get_type(self):
+        return self.type
+    def get_info(self):
+        return self.info
+    def get_lineno(self):
+        return self.lineno
+    def get_extra(self):
+        return self.extra
+    def get_static(self):
+        return self.static
+    def get_conditionals(self):
+        return self.conditionals
+
+    def update(self, header, module, type = None, info = None, extra=None,
+               conditionals=None):
+       if self.name == debugsym:
+           print "=> update %s : %s" % (debugsym, (module, type, info,
+                                        extra, conditionals))
+        if header != None and self.header == None:
+           self.set_header(module)
+        if module != None and (self.module == None or self.header == self.module):
+           self.set_module(module)
+        if type != None and self.type == None:
+           self.set_type(type)
+        if info != None:
+           self.set_info(info)
+        if extra != None:
+           self.set_extra(extra)
+        if conditionals != None:
+           self.set_conditionals(conditionals)
+
+class index:
+    def __init__(self, name = "noname"):
+        self.name = name
+        self.identifiers = {}
+        self.functions = {}
+       self.variables = {}
+       self.includes = {}
+       self.structs = {}
+       self.enums = {}
+       self.typedefs = {}
+       self.macros = {}
+       self.references = {}
+       self.info = {}
+
+    def add_ref(self, name, header, module, static, type, lineno, info=None, extra=None, conditionals = None):
+        if name[0:2] == '__':
+           return None
+        d = None
+        try:
+          d = self.identifiers[name]
+          d.update(header, module, type, lineno, info, extra, conditionals)
+       except:
+          d = identifier(name, header, module, type, lineno, info, extra, conditionals)
+          self.identifiers[name] = d
+
+       if d != None and static == 1:
+           d.set_static(1)
+
+       if d != None and name != None and type != None:
+           self.references[name] = d
+
+       if name == debugsym:
+           print "New ref: %s" % (d)
+
+       return d
+
+    def add(self, name, header, module, static, type, lineno, info=None, extra=None, conditionals = None):
+        if name[0:2] == '__':
+           return None
+        d = None
+        try:
+          d = self.identifiers[name]
+          d.update(header, module, type, lineno, info, extra, conditionals)
+       except:
+          d = identifier(name, header, module, type, lineno, info, extra, conditionals)
+          self.identifiers[name] = d
+
+       if d != None and static == 1:
+           d.set_static(1)
+
+       if d != None and name != None and type != None:
+           if type == "function":
+               self.functions[name] = d
+           elif type == "functype":
+               self.functions[name] = d
+           elif type == "variable":
+               self.variables[name] = d
+           elif type == "include":
+               self.includes[name] = d
+           elif type == "struct":
+               self.structs[name] = d
+           elif type == "enum":
+               self.enums[name] = d
+           elif type == "typedef":
+               self.typedefs[name] = d
+           elif type == "macro":
+               self.macros[name] = d
+           else:
+               print "Unable to register type ", type
+
+       if name == debugsym:
+           print "New symbol: %s" % (d)
+
+       return d
+
+    def merge(self, idx):
+        for id in idx.functions.keys():
+              #
+              # macro might be used to override functions or variables
+              # definitions
+              #
+            if self.macros.has_key(id):
+                del self.macros[id]
+            if self.functions.has_key(id):
+                print "function %s from %s redeclared in %s" % (
+                   id, self.functions[id].header, idx.functions[id].header)
+            else:
+                self.functions[id] = idx.functions[id]
+                self.identifiers[id] = idx.functions[id]
+        for id in idx.variables.keys():
+              #
+              # macro might be used to override functions or variables
+              # definitions
+              #
+            if self.macros.has_key(id):
+                del self.macros[id]
+            if self.variables.has_key(id):
+                print "variable %s from %s redeclared in %s" % (
+                   id, self.variables[id].header, idx.variables[id].header)
+            else:
+                self.variables[id] = idx.variables[id]
+                self.identifiers[id] = idx.variables[id]
+        for id in idx.structs.keys():
+            if self.structs.has_key(id):
+                print "struct %s from %s redeclared in %s" % (
+                   id, self.structs[id].header, idx.structs[id].header)
+            else:
+                self.structs[id] = idx.structs[id]
+                self.identifiers[id] = idx.structs[id]
+        for id in idx.typedefs.keys():
+            if self.typedefs.has_key(id):
+                print "typedef %s from %s redeclared in %s" % (
+                   id, self.typedefs[id].header, idx.typedefs[id].header)
+            else:
+                self.typedefs[id] = idx.typedefs[id]
+                self.identifiers[id] = idx.typedefs[id]
+        for id in idx.macros.keys():
+              #
+              # macro might be used to override functions or variables
+              # definitions
+              #
+             if self.variables.has_key(id):
+                 continue
+             if self.functions.has_key(id):
+                 continue
+             if self.enums.has_key(id):
+                 continue
+            if self.macros.has_key(id):
+                print "macro %s from %s redeclared in %s" % (
+                   id, self.macros[id].header, idx.macros[id].header)
+            else:
+                self.macros[id] = idx.macros[id]
+                self.identifiers[id] = idx.macros[id]
+        for id in idx.enums.keys():
+            if self.enums.has_key(id):
+                print "enum %s from %s redeclared in %s" % (
+                   id, self.enums[id].header, idx.enums[id].header)
+            else:
+                self.enums[id] = idx.enums[id]
+                self.identifiers[id] = idx.enums[id]
+
+    def merge_public(self, idx):
+        for id in idx.functions.keys():
+            if self.functions.has_key(id):
+                # check that function condition agrees with header
+                if idx.functions[id].conditionals != \
+                   self.functions[id].conditionals:
+                    print "Header condition differs from Function for %s:" \
+                       % id
+                    print "  H: %s" % self.functions[id].conditionals
+                    print "  C: %s" % idx.functions[id].conditionals
+                up = idx.functions[id]
+                self.functions[id].update(None, up.module, up.type, up.info, up.extra)
+        #     else:
+        #         print "Function %s from %s is not declared in headers" % (
+        #              id, idx.functions[id].module)
+        # TODO: do the same for variables.
+
+    def analyze_dict(self, type, dict):
+        count = 0
+       public = 0
+        for name in dict.keys():
+           id = dict[name]
+           count = count + 1
+           if id.static == 0:
+               public = public + 1
+        if count != public:
+           print "  %d %s , %d public" % (count, type, public)
+       elif count != 0:
+           print "  %d public %s" % (count, type)
+
+
+    def analyze(self):
+       self.analyze_dict("functions", self.functions)
+       self.analyze_dict("variables", self.variables)
+       self.analyze_dict("structs", self.structs)
+       self.analyze_dict("typedefs", self.typedefs)
+       self.analyze_dict("macros", self.macros)
+         
+class CLexer:
+    """A lexer for the C language, tokenize the input by reading and
+       analyzing it line by line"""
+    def __init__(self, input):
+        self.input = input
+       self.tokens = []
+       self.line = ""
+       self.lineno = 0
+
+    def getline(self):
+        line = ''
+       while line == '':
+           line = self.input.readline()
+           if not line:
+               return None
+           self.lineno = self.lineno + 1
+           line = string.lstrip(line)
+           line = string.rstrip(line)
+           if line == '':
+               continue
+           while line[-1] == '\\':
+               line = line[:-1]
+               n = self.input.readline()
+               self.lineno = self.lineno + 1
+               n = string.lstrip(n)
+               n = string.rstrip(n)
+               if not n:
+                   break
+               else:
+                   line = line + n
+        return line
+        
+    def getlineno(self):
+        return self.lineno
+
+    def push(self, token):
+        self.tokens.insert(0, token);
+
+    def debug(self):
+        print "Last token: ", self.last
+       print "Token queue: ", self.tokens
+       print "Line %d end: " % (self.lineno), self.line
+
+    def token(self):
+        while self.tokens == []:
+           if self.line == "":
+               line = self.getline()
+           else:
+               line = self.line
+               self.line = ""
+           if line == None:
+               return None
+
+           if line[0] == '#':
+               self.tokens = map((lambda x: ('preproc', x)),
+                                 string.split(line))
+               break;
+           l = len(line)
+           if line[0] == '"' or line[0] == "'":
+               end = line[0]
+               line = line[1:]
+               found = 0
+               tok = ""
+               while found == 0:
+                   i = 0
+                   l = len(line)
+                   while i < l:
+                       if line[i] == end:
+                           self.line = line[i+1:]
+                           line = line[:i]
+                           l = i
+                           found = 1
+                           break
+                       if line[i] == '\\':
+                           i = i + 1
+                       i = i + 1
+                   tok = tok + line
+                   if found == 0:
+                       line = self.getline()
+                       if line == None:
+                           return None
+               self.last = ('string', tok)
+               return self.last
+
+           if l >= 2 and line[0] == '/' and line[1] == '*':
+               line = line[2:]
+               found = 0
+               tok = ""
+               while found == 0:
+                   i = 0
+                   l = len(line)
+                   while i < l:
+                       if line[i] == '*' and i+1 < l and line[i+1] == '/':
+                           self.line = line[i+2:]
+                           line = line[:i-1]
+                           l = i
+                           found = 1
+                           break
+                       i = i + 1
+                   if tok != "":
+                       tok = tok + "\n"
+                   tok = tok + line
+                   if found == 0:
+                       line = self.getline()
+                       if line == None:
+                           return None
+               self.last = ('comment', tok)
+               return self.last
+           if l >= 2 and line[0] == '/' and line[1] == '/':
+               line = line[2:]
+               self.last = ('comment', line)
+               return self.last
+           i = 0
+           while i < l:
+               if line[i] == '/' and i+1 < l and line[i+1] == '/':
+                   self.line = line[i:]
+                   line = line[:i]
+                   break
+               if line[i] == '/' and i+1 < l and line[i+1] == '*':
+                   self.line = line[i:]
+                   line = line[:i]
+                   break
+               if line[i] == '"' or line[i] == "'":
+                   self.line = line[i:]
+                   line = line[:i]
+                   break
+               i = i + 1
+           l = len(line)
+           i = 0
+           while i < l:
+               if line[i] == ' ' or line[i] == '\t':
+                   i = i + 1
+                   continue
+               o = ord(line[i])
+               if (o >= 97 and o <= 122) or (o >= 65 and o <= 90) or \
+                  (o >= 48 and o <= 57):
+                   s = i
+                   while i < l:
+                       o = ord(line[i])
+                       if (o >= 97 and o <= 122) or (o >= 65 and o <= 90) or \
+                          (o >= 48 and o <= 57) or string.find(
+                              " \t(){}:;,+-*/%&!|[]=><", line[i]) == -1:
+                           i = i + 1
+                       else:
+                           break
+                   self.tokens.append(('name', line[s:i]))
+                   continue
+               if string.find("(){}:;,[]", line[i]) != -1:
+#                 if line[i] == '(' or line[i] == ')' or line[i] == '{' or \
+#                  line[i] == '}' or line[i] == ':' or line[i] == ';' or \
+#                  line[i] == ',' or line[i] == '[' or line[i] == ']':
+                   self.tokens.append(('sep', line[i]))
+                   i = i + 1
+                   continue
+               if string.find("+-*><=/%&!|.", line[i]) != -1:
+#                 if line[i] == '+' or line[i] == '-' or line[i] == '*' or \
+#                  line[i] == '>' or line[i] == '<' or line[i] == '=' or \
+#                  line[i] == '/' or line[i] == '%' or line[i] == '&' or \
+#                  line[i] == '!' or line[i] == '|' or line[i] == '.':
+                   if line[i] == '.' and  i + 2 < l and \
+                      line[i+1] == '.' and line[i+2] == '.':
+                       self.tokens.append(('name', '...'))
+                       i = i + 3
+                       continue
+
+                   j = i + 1
+                   if j < l and (
+                      string.find("+-*><=/%&!|", line[j]) != -1):
+#                      line[j] == '+' or line[j] == '-' or line[j] == '*' or \
+#                      line[j] == '>' or line[j] == '<' or line[j] == '=' or \
+#                      line[j] == '/' or line[j] == '%' or line[j] == '&' or \
+#                      line[j] == '!' or line[j] == '|'):
+                       self.tokens.append(('op', line[i:j+1]))
+                       i = j + 1
+                   else:
+                       self.tokens.append(('op', line[i]))
+                       i = i + 1
+                   continue
+               s = i
+               while i < l:
+                   o = ord(line[i])
+                   if (o >= 97 and o <= 122) or (o >= 65 and o <= 90) or \
+                      (o >= 48 and o <= 57) or (
+                       string.find(" \t(){}:;,+-*/%&!|[]=><", line[i]) == -1):
+#                       line[i] != ' ' and line[i] != '\t' and
+#                       line[i] != '(' and line[i] != ')' and
+#                       line[i] != '{'  and line[i] != '}' and
+#                       line[i] != ':' and line[i] != ';' and
+#                       line[i] != ',' and line[i] != '+' and
+#                       line[i] != '-' and line[i] != '*' and
+#                       line[i] != '/' and line[i] != '%' and
+#                       line[i] != '&' and line[i] != '!' and
+#                       line[i] != '|' and line[i] != '[' and
+#                       line[i] != ']' and line[i] != '=' and
+#                       line[i] != '*' and line[i] != '>' and
+#                       line[i] != '<'):
+                       i = i + 1
+                   else:
+                       break
+               self.tokens.append(('name', line[s:i]))
+             
+       tok = self.tokens[0]
+       self.tokens = self.tokens[1:]
+       self.last = tok
+       return tok
+      
+class CParser:
+    """The C module parser"""
+    def __init__(self, filename, idx = None):
+        self.filename = filename
+       if len(filename) > 2 and filename[-2:] == '.h':
+           self.is_header = 1
+       else:
+           self.is_header = 0
+        self.input = open(filename)
+       self.lexer = CLexer(self.input)
+       if idx == None:
+           self.index = index()
+       else:
+           self.index = idx
+       self.top_comment = ""
+       self.last_comment = ""
+       self.comment = None
+       self.collect_ref = 0
+       self.no_error = 0
+       self.conditionals = []
+       self.defines = []
+
+    def collect_references(self):
+        self.collect_ref = 1
+
+    def stop_error(self):
+        self.no_error = 1
+
+    def start_error(self):
+        self.no_error = 0
+
+    def lineno(self):
+        return self.lexer.getlineno()
+
+    def index_add(self, name, module, static, type, info=None, extra = None):
+       if self.is_header == 1:
+           self.index.add(name, module, module, static, type, self.lineno(),
+                          info, extra, self.conditionals)
+       else:
+           self.index.add(name, None, module, static, type, self.lineno(),
+                          info, extra, self.conditionals)
+
+    def index_add_ref(self, name, module, static, type, info=None,
+                      extra = None):
+       if self.is_header == 1:
+           self.index.add_ref(name, module, module, static, type,
+                              self.lineno(), info, extra, self.conditionals)
+       else:
+           self.index.add_ref(name, None, module, static, type, self.lineno(),
+                              info, extra, self.conditionals)
+
+    def warning(self, msg):
+        if self.no_error:
+           return
+       print msg
+
+    def error(self, msg, token=-1):
+        if self.no_error:
+           return
+
+        print "Parse Error: " + msg
+       if token != -1:
+           print "Got token ", token
+       self.lexer.debug()
+       sys.exit(1)
+
+    def debug(self, msg, token=-1):
+        print "Debug: " + msg
+       if token != -1:
+           print "Got token ", token
+       self.lexer.debug()
+
+    def parseTopComment(self, comment):
+       res = {}
+       lines = string.split(comment, "\n")
+       item = None
+       for line in lines:
+           while line != "" and (line[0] == ' ' or line[0] == '\t'):
+               line = line[1:]
+           while line != "" and line[0] == '*':
+               line = line[1:]
+           while line != "" and (line[0] == ' ' or line[0] == '\t'):
+               line = line[1:]
+           try:
+               (it, line) = string.split(line, ":", 1)
+               item = it
+               while line != "" and (line[0] == ' ' or line[0] == '\t'):
+                   line = line[1:]
+               if res.has_key(item):
+                   res[item] = res[item] + " " + line
+               else:
+                   res[item] = line
+           except:
+               if item != None:
+                   if res.has_key(item):
+                       res[item] = res[item] + " " + line
+                   else:
+                       res[item] = line
+       self.index.info = res
+
+    def parseComment(self, token):
+        if self.top_comment == "":
+           self.top_comment = token[1]
+       if self.comment == None or token[1][0] == '*':
+           self.comment = token[1];
+       else:
+           self.comment = self.comment + token[1]
+       token = self.lexer.token()
+
+        if string.find(self.comment, "DOC_DISABLE") != -1:
+           self.stop_error()
+
+        if string.find(self.comment, "DOC_ENABLE") != -1:
+           self.start_error()
+
+       return token
+
+    #
+    # Parse a comment block associate to a typedef
+    #
+    def parseTypeComment(self, name, quiet = 0):
+        if name[0:2] == '__':
+           quiet = 1
+
+        args = []
+       desc = ""
+
+        if self.comment == None:
+           if not quiet:
+               self.warning("Missing comment for type %s" % (name))
+           return((args, desc))
+        if self.comment[0] != '*':
+           if not quiet:
+               self.warning("Missing * in type comment for %s" % (name))
+           return((args, desc))
+       lines = string.split(self.comment, '\n')
+       if lines[0] == '*':
+           del lines[0]
+       if lines[0] != "* %s:" % (name):
+           if not quiet:
+               self.warning("Misformatted type comment for %s" % (name))
+               self.warning("  Expecting '* %s:' got '%s'" % (name, lines[0]))
+           return((args, desc))
+       del lines[0]
+       while len(lines) > 0 and lines[0] == '*':
+           del lines[0]
+       desc = ""
+       while len(lines) > 0:
+           l = lines[0]
+           while len(l) > 0 and l[0] == '*':
+               l = l[1:]
+           l = string.strip(l)
+           desc = desc + " " + l
+           del lines[0]
+                    
+       desc = string.strip(desc)
+
+       if quiet == 0:
+           if desc == "":
+               self.warning("Type comment for %s lack description of the macro" % (name))
+
+       return(desc)
+    #
+    # Parse a comment block associate to a macro
+    #
+    def parseMacroComment(self, name, quiet = 0):
+        if name[0:2] == '__':
+           quiet = 1
+
+        args = []
+       desc = ""
+
+        if self.comment == None:
+           if not quiet:
+               self.warning("Missing comment for macro %s" % (name))
+           return((args, desc))
+        if self.comment[0] != '*':
+           if not quiet:
+               self.warning("Missing * in macro comment for %s" % (name))
+           return((args, desc))
+       lines = string.split(self.comment, '\n')
+       if lines[0] == '*':
+           del lines[0]
+       if lines[0] != "* %s:" % (name):
+           if not quiet:
+               self.warning("Misformatted macro comment for %s" % (name))
+               self.warning("  Expecting '* %s:' got '%s'" % (name, lines[0]))
+           return((args, desc))
+       del lines[0]
+       while lines[0] == '*':
+           del lines[0]
+       while len(lines) > 0 and lines[0][0:3] == '* @':
+           l = lines[0][3:]
+           try:
+               (arg, desc) = string.split(l, ':', 1)
+               desc=string.strip(desc)
+               arg=string.strip(arg)
+            except:
+               if not quiet:
+                   self.warning("Misformatted macro comment for %s" % (name))
+                   self.warning("  problem with '%s'" % (lines[0]))
+               del lines[0]
+               continue
+           del lines[0]
+           l = string.strip(lines[0])
+           while len(l) > 2 and l[0:3] != '* @':
+               while l[0] == '*':
+                   l = l[1:]
+               desc = desc + ' ' + string.strip(l)
+               del lines[0]
+               if len(lines) == 0:
+                   break
+               l = lines[0]
+            args.append((arg, desc))
+       while len(lines) > 0 and lines[0] == '*':
+           del lines[0]
+       desc = ""
+       while len(lines) > 0:
+           l = lines[0]
+           while len(l) > 0 and l[0] == '*':
+               l = l[1:]
+           l = string.strip(l)
+           desc = desc + " " + l
+           del lines[0]
+                    
+       desc = string.strip(desc)
+
+       if quiet == 0:
+           if desc == "":
+               self.warning("Macro comment for %s lack description of the macro" % (name))
+
+       return((args, desc))
+
+     #
+     # Parse a comment block and merge the informations found in the
+     # parameters descriptions, finally returns a block as complete
+     # as possible
+     #
+    def mergeFunctionComment(self, name, description, quiet = 0):
+        if name == 'main':
+           quiet = 1
+        if name[0:2] == '__':
+           quiet = 1
+
+       (ret, args) = description
+       desc = ""
+       retdesc = ""
+
+        if self.comment == None:
+           if not quiet:
+               self.warning("Missing comment for function %s" % (name))
+           return(((ret[0], retdesc), args, desc))
+        if self.comment[0] != '*':
+           if not quiet:
+               self.warning("Missing * in function comment for %s" % (name))
+           return(((ret[0], retdesc), args, desc))
+       lines = string.split(self.comment, '\n')
+       if lines[0] == '*':
+           del lines[0]
+       if lines[0] != "* %s:" % (name):
+           if not quiet:
+               self.warning("Misformatted function comment for %s" % (name))
+               self.warning("  Expecting '* %s:' got '%s'" % (name, lines[0]))
+           return(((ret[0], retdesc), args, desc))
+       del lines[0]
+       while lines[0] == '*':
+           del lines[0]
+       nbargs = len(args)
+       while len(lines) > 0 and lines[0][0:3] == '* @':
+           l = lines[0][3:]
+           try:
+               (arg, desc) = string.split(l, ':', 1)
+               desc=string.strip(desc)
+               arg=string.strip(arg)
+            except:
+               if not quiet:
+                   self.warning("Misformatted function comment for %s" % (name))
+                   self.warning("  problem with '%s'" % (lines[0]))
+               del lines[0]
+               continue
+           del lines[0]
+           l = string.strip(lines[0])
+           while len(l) > 2 and l[0:3] != '* @':
+               while l[0] == '*':
+                   l = l[1:]
+               desc = desc + ' ' + string.strip(l)
+               del lines[0]
+               if len(lines) == 0:
+                   break
+               l = lines[0]
+           i = 0
+           while i < nbargs:
+               if args[i][1] == arg:
+                   args[i] = (args[i][0], arg, desc)
+                   break;
+               i = i + 1
+           if i >= nbargs:
+               if not quiet:
+                   self.warning("Unable to find arg %s from function comment for %s" % (
+                      arg, name))
+       while len(lines) > 0 and lines[0] == '*':
+           del lines[0]
+       desc = ""
+       while len(lines) > 0:
+           l = lines[0]
+           while len(l) > 0 and l[0] == '*':
+               l = l[1:]
+           l = string.strip(l)
+           if len(l) >= 6 and  l[0:6] == "return" or l[0:6] == "Return":
+               try:
+                   l = string.split(l, ' ', 1)[1]
+               except:
+                   l = ""
+               retdesc = string.strip(l)
+               del lines[0]
+               while len(lines) > 0:
+                   l = lines[0]
+                   while len(l) > 0 and l[0] == '*':
+                       l = l[1:]
+                   l = string.strip(l)
+                   retdesc = retdesc + " " + l
+                   del lines[0]
+           else:
+               desc = desc + " " + l
+               del lines[0]
+                    
+       retdesc = string.strip(retdesc)
+       desc = string.strip(desc)
+
+       if quiet == 0:
+            #
+            # report missing comments
+            #
+           i = 0
+           while i < nbargs:
+               if args[i][2] == None and args[i][0] != "void" and args[i][1] != None:
+                   self.warning("Function comment for %s lacks description of arg %s" % (name, args[i][1]))
+               i = i + 1
+           if retdesc == "" and ret[0] != "void":
+               self.warning("Function comment for %s lacks description of return value" % (name))
+           if desc == "":
+               self.warning("Function comment for %s lacks description of the function" % (name))
+
+
+       return(((ret[0], retdesc), args, desc))
+
+    def parsePreproc(self, token):
+       if debug:
+           print "=> preproc ", token, self.lexer.tokens
+        name = token[1]
+       if name == "#include":
+           token = self.lexer.token()
+           if token == None:
+               return None
+           if token[0] == 'preproc':
+               self.index_add(token[1], self.filename, not self.is_header,
+                               "include")
+               return self.lexer.token()
+           return token
+       if name == "#define":
+           token = self.lexer.token()
+           if token == None:
+               return None
+           if token[0] == 'preproc':
+                # TODO macros with arguments
+               name = token[1]
+               lst = []
+               token = self.lexer.token()
+               while token != None and token[0] == 'preproc' and \
+                     token[1][0] != '#':
+                   lst.append(token[1])
+                   token = self.lexer.token()
+                try:
+                   name = string.split(name, '(') [0]
+                except:
+                    pass
+                info = self.parseMacroComment(name, not self.is_header)
+               self.index_add(name, self.filename, not self.is_header,
+                               "macro", info)
+               return token
+
+       #
+       # Processing of conditionals modified by Bill 1/1/05
+       #
+       # We process conditionals (i.e. tokens from #ifdef, #ifndef,
+       # #if, #else and #endif) for headers and mainline code,
+       # store the ones from the header in libxml2-api.xml, and later
+       # (in the routine merge_public) verify that the two (header and
+       # mainline code) agree.
+       #
+       # There is a small problem with processing the headers. Some of
+       # the variables are not concerned with enabling / disabling of
+       # library functions (e.g. '__XML_PARSER_H__'), and we don't want
+       # them to be included in libxml2-api.xml, or involved in
+       # the check between the header and the mainline code.  To
+       # accomplish this, we ignore any conditional which doesn't include
+       # the string 'ENABLED'
+       #
+       if name == "#ifdef":
+           apstr = self.lexer.tokens[0][1]
+           try:
+               self.defines.append(apstr)
+               if string.find(apstr, 'ENABLED') != -1:
+                   self.conditionals.append("defined(%s)" % apstr)    
+           except:
+               pass
+       elif name == "#ifndef":
+           apstr = self.lexer.tokens[0][1]
+           try:
+               self.defines.append(apstr)
+               if string.find(apstr, 'ENABLED') != -1:
+                   self.conditionals.append("!defined(%s)" % apstr)    
+           except:
+               pass
+       elif name == "#if":
+           apstr = ""
+           for tok in self.lexer.tokens:
+               if apstr != "":
+                   apstr = apstr + " "
+               apstr = apstr + tok[1]
+           try:
+               self.defines.append(apstr)
+               if string.find(apstr, 'ENABLED') != -1:
+                   self.conditionals.append(apstr)
+           except:
+               pass
+       elif name == "#else":
+           if self.conditionals != [] and \
+              string.find(self.defines[-1], 'ENABLED') != -1:
+               self.conditionals[-1] = "!(%s)" % self.conditionals[-1]
+       elif name == "#endif":
+           if self.conditionals != [] and \
+              string.find(self.defines[-1], 'ENABLED') != -1:
+               self.conditionals = self.conditionals[:-1]
+           self.defines = self.defines[:-1]
+       token = self.lexer.token()
+       while token != None and token[0] == 'preproc' and \
+           token[1][0] != '#':
+           token = self.lexer.token()
+       return token
+
+     #
+     # token acquisition on top of the lexer, it handle internally
+     # preprocessor and comments since they are logically not part of
+     # the program structure.
+     #
+    def token(self):
+        global ignored_words
+
+        token = self.lexer.token()
+       while token != None:
+           if token[0] == 'comment':
+               token = self.parseComment(token)
+               continue
+           elif token[0] == 'preproc':
+               token = self.parsePreproc(token)
+               continue
+           elif token[0] == "name" and token[1] == "__const":
+               token = ("name", "const")
+               return token
+           elif token[0] == "name" and token[1] == "__attribute":
+               token = self.lexer.token()
+               while token != None and token[1] != ";":
+                   token = self.lexer.token()
+               return token
+           elif token[0] == "name" and ignored_words.has_key(token[1]):
+               (n, info) = ignored_words[token[1]]
+               i = 0
+               while i < n:
+                   token = self.lexer.token()
+                   i = i + 1
+               token = self.lexer.token()
+               continue
+           else:
+               if debug:
+                   print "=> ", token
+               return token
+       return None
+
+     #
+     # Parse a typedef, it records the type and its name.
+     #
+    def parseTypedef(self, token):
+        if token == None:
+           return None
+       token = self.parseType(token)
+       if token == None:
+           self.error("parsing typedef")
+           return None
+       base_type = self.type
+       type = base_type
+        #self.debug("end typedef type", token)
+       while token != None:
+           if token[0] == "name":
+               name = token[1]
+               signature = self.signature
+               if signature != None:
+                   type = string.split(type, '(')[0]
+                   d = self.mergeFunctionComment(name,
+                           ((type, None), signature), 1)
+                   self.index_add(name, self.filename, not self.is_header,
+                                   "functype", d)
+               else:
+                   if base_type == "struct":
+                       self.index_add(name, self.filename, not self.is_header,
+                                       "struct", type)
+                       base_type = "struct " + name
+                   else:
+                       # TODO report missing or misformatted comments
+                       info = self.parseTypeComment(name, 1)
+                       self.index_add(name, self.filename, not self.is_header,
+                                   "typedef", type, info)
+               token = self.token()
+           else:
+               self.error("parsing typedef: expecting a name")
+               return token
+            #self.debug("end typedef", token)
+           if token != None and token[0] == 'sep' and token[1] == ',':
+               type = base_type
+               token = self.token()
+               while token != None and token[0] == "op":
+                   type = type + token[1]
+                   token = self.token()
+           elif token != None and token[0] == 'sep' and token[1] == ';':
+               break;
+           elif token != None and token[0] == 'name':
+               type = base_type
+               continue;
+           else:
+               self.error("parsing typedef: expecting ';'", token)
+               return token
+       token = self.token()
+       return token
+            
+     #
+     # Parse a C code block, used for functions it parse till
+     # the balancing } included
+     #
+    def parseBlock(self, token):
+        while token != None:
+           if token[0] == "sep" and token[1] == "{":
+               token = self.token()
+               token = self.parseBlock(token)
+           elif token[0] == "sep" and token[1] == "}":
+               self.comment = None
+               token = self.token()
+               return token
+           else:
+               if self.collect_ref == 1:
+                   oldtok = token
+                   token = self.token()
+                   if oldtok[0] == "name" and oldtok[1][0:3] == "xen":
+                       if token[0] == "sep" and token[1] == "(":
+                           self.index_add_ref(oldtok[1], self.filename,
+                                               0, "function")
+                           token = self.token()
+                       elif token[0] == "name":
+                           token = self.token()
+                           if token[0] == "sep" and (token[1] == ";" or
+                              token[1] == "," or token[1] == "="):
+                               self.index_add_ref(oldtok[1], self.filename,
+                                                   0, "type")
+                   elif oldtok[0] == "name" and oldtok[1][0:4] == "XEN_":
+                       self.index_add_ref(oldtok[1], self.filename,
+                                           0, "typedef")
+                   elif oldtok[0] == "name" and oldtok[1][0:7] == "LIBXEN_":
+                       self.index_add_ref(oldtok[1], self.filename,
+                                           0, "typedef")
+                        
+               else:
+                   token = self.token()
+       return token
+
+     #
+     # Parse a C struct definition till the balancing }
+     #
+    def parseStruct(self, token):
+        fields = []
+        #self.debug("start parseStruct", token)
+        while token != None:
+           if token[0] == "sep" and token[1] == "{":
+               token = self.token()
+               token = self.parseTypeBlock(token)
+           elif token[0] == "sep" and token[1] == "}":
+               self.struct_fields = fields
+                #self.debug("end parseStruct", token)
+                #print fields
+               token = self.token()
+               return token
+           else:
+               base_type = self.type
+                #self.debug("before parseType", token)
+               token = self.parseType(token)
+                #self.debug("after parseType", token)
+               if token != None and token[0] == "name":
+                   fname = token[1]
+                   token = self.token()
+                   if token[0] == "sep" and token[1] == ";":
+                       self.comment = None
+                       token = self.token()
+                       fields.append((self.type, fname, self.comment))
+                       self.comment = None
+                   else:
+                       self.error("parseStruct: expecting ;", token)
+               elif token != None and token[0] == "sep" and token[1] == "{":
+                   token = self.token()
+                   token = self.parseTypeBlock(token)
+                   if token != None and token[0] == "name":
+                       token = self.token()
+                   if token != None and token[0] == "sep" and token[1] == ";":
+                       token = self.token()
+                   else:
+                       self.error("parseStruct: expecting ;", token)
+               else:
+                   self.error("parseStruct: name", token)
+                   token = self.token()
+               self.type = base_type;
+        self.struct_fields = fields
+        #self.debug("end parseStruct", token)
+        #print fields
+       return token
+
+     #
+     # Parse a C enum block, parse till the balancing }
+     #
+    def parseEnumBlock(self, token):
+        self.enums = []
+       name = None
+       self.comment = None
+       comment = ""
+       value = "0"
+        while token != None:
+           if token[0] == "sep" and token[1] == "{":
+               token = self.token()
+               token = self.parseTypeBlock(token)
+           elif token[0] == "sep" and token[1] == "}":
+               if name != None:
+                   if self.comment != None:
+                       comment = self.comment
+                       self.comment = None
+                   self.enums.append((name, value, comment))
+               token = self.token()
+               return token
+           elif token[0] == "name":
+                   if name != None:
+                       if self.comment != None:
+                           comment = string.strip(self.comment)
+                           self.comment = None
+                       self.enums.append((name, value, comment))
+                   name = token[1]
+                   comment = ""
+                   token = self.token()
+                   if token[0] == "op" and token[1][0] == "=":
+                       value = ""
+                       if len(token[1]) > 1:
+                           value = token[1][1:]
+                       token = self.token()
+                       while token[0] != "sep" or (token[1] != ',' and
+                             token[1] != '}'):
+                           value = value + token[1]
+                           token = self.token()
+                   else:
+                       try:
+                           value = "%d" % (int(value) + 1)
+                       except:
+                           self.warning("Failed to compute value of enum %s" % (name))
+                           value=""
+                   if token[0] == "sep" and token[1] == ",":
+                       token = self.token()
+           else:
+               token = self.token()
+       return token
+
+     #
+     # Parse a C definition block, used for structs it parse till
+     # the balancing }
+     #
+    def parseTypeBlock(self, token):
+        while token != None:
+           if token[0] == "sep" and token[1] == "{":
+               token = self.token()
+               token = self.parseTypeBlock(token)
+           elif token[0] == "sep" and token[1] == "}":
+               token = self.token()
+               return token
+           else:
+               token = self.token()
+       return token
+
+     #
+     # Parse a type: the fact that the type name can either occur after
+     #    the definition or within the definition makes it a little harder
+     #    if inside, the name token is pushed back before returning
+     #
+    def parseType(self, token):
+        self.type = ""
+       self.struct_fields = []
+        self.signature = None
+       if token == None:
+           return token
+
+       while token[0] == "name" and ( 
+             token[1] == "const" or \
+             token[1] == "unsigned" or \
+             token[1] == "signed"):
+           if self.type == "":
+               self.type = token[1]
+           else:
+               self.type = self.type + " " + token[1]
+           token = self.token()
+
+        if token[0] == "name" and (token[1] == "long" or token[1] == "short"):
+           if self.type == "":
+               self.type = token[1]
+           else:
+               self.type = self.type + " " + token[1]
+           if token[0] == "name" and token[1] == "int":
+               if self.type == "":
+                   self.type = tmp[1]
+               else:
+                   self.type = self.type + " " + tmp[1]
+            
+        elif token[0] == "name" and token[1] == "struct":
+           if self.type == "":
+               self.type = token[1]
+           else:
+               self.type = self.type + " " + token[1]
+           token = self.token()
+           nametok = None
+           if token[0] == "name":
+               nametok = token
+               token = self.token()
+           if token != None and token[0] == "sep" and token[1] == "{":
+               token = self.token()
+               token = self.parseStruct(token)
+           elif token != None and token[0] == "op" and token[1] == "*":
+               self.type = self.type + " " + nametok[1] + " *"
+               token = self.token()
+               while token != None and token[0] == "op" and token[1] == "*":
+                   self.type = self.type + " *"
+                   token = self.token()
+               if token[0] == "name":
+                   nametok = token
+                   token = self.token()
+               else:
+                   self.error("struct : expecting name", token)
+                   return token
+           elif token != None and token[0] == "name" and nametok != None:
+               self.type = self.type + " " + nametok[1]
+               return token
+
+           if nametok != None:
+               self.lexer.push(token)
+               token = nametok
+           return token
+
+        elif token[0] == "name" and token[1] == "enum":
+           if self.type == "":
+               self.type = token[1]
+           else:
+               self.type = self.type + " " + token[1]
+           self.enums = []
+           token = self.token()
+           if token != None and token[0] == "sep" and token[1] == "{":
+               token = self.token()
+               token = self.parseEnumBlock(token)
+           else:
+               self.error("parsing enum: expecting '{'", token)
+           enum_type = None
+           if token != None and token[0] != "name":
+               self.lexer.push(token)
+               token = ("name", "enum")
+           else:
+               enum_type = token[1]
+           for enum in self.enums:
+               self.index_add(enum[0], self.filename,
+                              not self.is_header, "enum",
+                              (enum[1], enum[2], enum_type))
+           return token
+
+       elif token[0] == "name":
+           if self.type == "":
+               self.type = token[1]
+           else:
+               self.type = self.type + " " + token[1]
+       else:
+           self.error("parsing type %s: expecting a name" % (self.type),
+                      token)
+           return token
+       token = self.token()
+        while token != None and (token[0] == "op" or
+             token[0] == "name" and token[1] == "const"):
+           self.type = self.type + " " + token[1]
+           token = self.token()
+
+        #
+        # if there is a parenthesis here, this means a function type
+        #
+       if token != None and token[0] == "sep" and token[1] == '(':
+           self.type = self.type + token[1]
+           token = self.token()
+           while token != None and token[0] == "op" and token[1] == '*':
+               self.type = self.type + token[1]
+               token = self.token()
+           if token == None or token[0] != "name" :
+               self.error("parsing function type, name expected", token);
+               return token
+           self.type = self.type + token[1]
+           nametok = token
+           token = self.token()
+           if token != None and token[0] == "sep" and token[1] == ')':
+               self.type = self.type + token[1]
+               token = self.token()
+               if token != None and token[0] == "sep" and token[1] == '(':
+                   token = self.token()
+                   type = self.type;
+                   token = self.parseSignature(token);
+                   self.type = type;
+               else:
+                   self.error("parsing function type, '(' expected", token);
+                   return token
+           else:
+               self.error("parsing function type, ')' expected", token);
+               return token
+           self.lexer.push(token)
+           token = nametok
+           return token
+
+         #
+        # do some lookahead for arrays
+        #
+       if token != None and token[0] == "name":
+           nametok = token
+           token = self.token()
+           if token != None and token[0] == "sep" and token[1] == '[':
+               self.type = self.type + nametok[1]
+               while token != None and token[0] == "sep" and token[1] == '[':
+                   self.type = self.type + token[1]
+                   token = self.token()
+                   while token != None and token[0] != 'sep' and \
+                         token[1] != ']' and token[1] != ';':
+                       self.type = self.type + token[1]
+                       token = self.token()
+               if token != None and token[0] == 'sep' and token[1] == ']':
+                   self.type = self.type + token[1]
+                   token = self.token()
+               else:
+                   self.error("parsing array type, ']' expected", token);
+                   return token
+           elif token != None and token[0] == "sep" and token[1] == ':':
+                # remove :12 in case it's a limited int size
+               token = self.token()
+               token = self.token()
+           self.lexer.push(token)
+           token = nametok
+
+       return token
+
+     #
+     # Parse a signature: '(' has been parsed and we scan the type definition
+     #    up to the ')' included
+    def parseSignature(self, token):
+        signature = []
+       if token != None and token[0] == "sep" and token[1] == ')':
+           self.signature = []
+           token = self.token()
+           return token
+       while token != None:
+           token = self.parseType(token)
+           if token != None and token[0] == "name":
+               signature.append((self.type, token[1], None))
+               token = self.token()
+           elif token != None and token[0] == "sep" and token[1] == ',':
+               token = self.token()
+               continue
+           elif token != None and token[0] == "sep" and token[1] == ')':
+                # only the type was provided
+               if self.type == "...":
+                   signature.append((self.type, "...", None))
+               else:
+                   signature.append((self.type, None, None))
+           if token != None and token[0] == "sep":
+               if token[1] == ',':
+                   token = self.token()
+                   continue
+               elif token[1] == ')':
+                   token = self.token()
+                   break
+       self.signature = signature
+       return token
+
+     #
+     # Parse a global definition, be it a type, variable or function
+     # the extern "C" blocks are a bit nasty and require it to recurse.
+     #
+    def parseGlobal(self, token):
+        static = 0
+        if token[1] == 'extern':
+           token = self.token()
+           if token == None:
+               return token
+           if token[0] == 'string':
+               if token[1] == 'C':
+                   token = self.token()
+                   if token == None:
+                       return token
+                   if token[0] == 'sep' and token[1] == "{":
+                       token = self.token()
+#                       print 'Entering extern "C line ', self.lineno()
+                       while token != None and (token[0] != 'sep' or
+                             token[1] != "}"):
+                           if token[0] == 'name':
+                               token = self.parseGlobal(token)
+                           else:
+                               self.error(
+                                "token %s %s unexpected at the top level" % (
+                                       token[0], token[1]))
+                               token = self.parseGlobal(token)
+#                       print 'Exiting extern "C" line', self.lineno()
+                       token = self.token()
+                       return token
+               else:
+                   return token
+       elif token[1] == 'static':
+           static = 1
+           token = self.token()
+           if token == None or  token[0] != 'name':
+               return token
+
+       if token[1] == 'typedef':
+           token = self.token()
+           return self.parseTypedef(token)
+       else:
+           token = self.parseType(token)
+           type_orig = self.type
+       if token == None or token[0] != "name":
+           return token
+       type = type_orig
+       self.name = token[1]
+       token = self.token()
+       while token != None and (token[0] == "sep" or token[0] == "op"):
+           if token[0] == "sep":
+               if token[1] == "[":
+                   type = type + token[1]
+                   token = self.token()
+                   while token != None and (token[0] != "sep" or \
+                         token[1] != ";"):
+                       type = type + token[1]
+                       token = self.token()
+
+           if token != None and token[0] == "op" and token[1] == "=":
+                #
+                # Skip the initialization of the variable
+                #
+               token = self.token()
+               if token[0] == 'sep' and token[1] == '{':
+                   token = self.token()
+                   token = self.parseBlock(token)
+               else:
+                   self.comment = None
+                   while token != None and (token[0] != "sep" or \
+                         (token[1] != ';' and token[1] != ',')):
+                           token = self.token()
+               self.comment = None
+               if token == None or token[0] != "sep" or (token[1] != ';' and
+                  token[1] != ','):
+                   self.error("missing ';' or ',' after value")
+
+           if token != None and token[0] == "sep":
+               if token[1] == ";":
+                   self.comment = None
+                   token = self.token()
+                   if type == "struct":
+                       self.index_add(self.name, self.filename,
+                            not self.is_header, "struct", self.struct_fields)
+                   else:
+                       self.index_add(self.name, self.filename,
+                            not self.is_header, "variable", type)
+                   break
+               elif token[1] == "(":
+                   token = self.token()
+                   token = self.parseSignature(token)
+                   if token == None:
+                       return None
+                   if token[0] == "sep" and token[1] == ";":
+                       d = self.mergeFunctionComment(self.name,
+                               ((type, None), self.signature), 1)
+                       self.index_add(self.name, self.filename, static,
+                                       "function", d)
+                       token = self.token()
+                   elif token[0] == "sep" and token[1] == "{":
+                       d = self.mergeFunctionComment(self.name,
+                               ((type, None), self.signature), static)
+                       self.index_add(self.name, self.filename, static,
+                                       "function", d)
+                       token = self.token()
+                       token = self.parseBlock(token);
+               elif token[1] == ',':
+                   self.comment = None
+                   self.index_add(self.name, self.filename, static,
+                                   "variable", type)
+                   type = type_orig
+                   token = self.token()
+                   while token != None and token[0] == "sep":
+                       type = type + token[1]
+                       token = self.token()
+                   if token != None and token[0] == "name":
+                       self.name = token[1]
+                       token = self.token()
+               else:
+                   break
+                   
+       return token
+
+    def parse(self):
+        self.warning("Parsing %s" % (self.filename))
+        token = self.token()
+       while token != None:
+            if token[0] == 'name':
+               token = self.parseGlobal(token)
+            else:
+               self.error("token %s %s unexpected at the top level" % (
+                      token[0], token[1]))
+               token = self.parseGlobal(token)
+               return
+       self.parseTopComment(self.top_comment)
+        return self.index
+                
+
+class docBuilder:
+    """A documentation builder"""
+    def __init__(self, name, directories=['.'], excludes=[]):
+        self.name = name
+        self.directories = directories
+       self.excludes = excludes + ignored_files.keys()
+       self.modules = {}
+       self.headers = {}
+       self.idx = index()
+        self.xref = {}
+       self.index = {}
+       self.basename = name
+
+    def indexString(self, id, str):
+       if str == None:
+           return
+       str = string.replace(str, "'", ' ')
+       str = string.replace(str, '"', ' ')
+       str = string.replace(str, "/", ' ')
+       str = string.replace(str, '*', ' ')
+       str = string.replace(str, "[", ' ')
+       str = string.replace(str, "]", ' ')
+       str = string.replace(str, "(", ' ')
+       str = string.replace(str, ")", ' ')
+       str = string.replace(str, "<", ' ')
+       str = string.replace(str, '>', ' ')
+       str = string.replace(str, "&", ' ')
+       str = string.replace(str, '#', ' ')
+       str = string.replace(str, ",", ' ')
+       str = string.replace(str, '.', ' ')
+       str = string.replace(str, ';', ' ')
+       tokens = string.split(str)
+       for token in tokens:
+           try:
+               c = token[0]
+               if string.find(string.letters, c) < 0:
+                   pass
+               elif len(token) < 3:
+                   pass
+               else:
+                   lower = string.lower(token)
+                   # TODO: generalize this a bit
+                   if lower == 'and' or lower == 'the':
+                       pass
+                   elif self.xref.has_key(token):
+                       self.xref[token].append(id)
+                   else:
+                       self.xref[token] = [id]
+           except:
+               pass
+
+    def analyze(self):
+        print "Project %s : %d headers, %d modules" % (self.name, len(self.headers.keys()), len(self.modules.keys()))
+       self.idx.analyze()
+
+    def scanHeaders(self):
+       for header in self.headers.keys():
+           parser = CParser(header)
+           idx = parser.parse()
+           self.headers[header] = idx;
+           self.idx.merge(idx)
+
+    def scanModules(self):
+       for module in self.modules.keys():
+           parser = CParser(module)
+           idx = parser.parse()
+           # idx.analyze()
+           self.modules[module] = idx
+           self.idx.merge_public(idx)
+
+    def scan(self):
+        for directory in self.directories:
+           files = glob.glob(directory + "/*.c")
+           for file in files:
+               skip = 0
+               for excl in self.excludes:
+                   if string.find(file, excl) != -1:
+                       skip = 1;
+                       break
+               if skip == 0:
+                   self.modules[file] = None;
+           files = glob.glob(directory + "/*.h")
+           for file in files:
+               skip = 0
+               for excl in self.excludes:
+                   if string.find(file, excl) != -1:
+                       skip = 1;
+                       break
+               if skip == 0:
+                   self.headers[file] = None;
+       self.scanHeaders()
+       self.scanModules()
+         
+    def modulename_file(self, file):
+        module = os.path.basename(file)
+       if module[-2:] == '.h':
+           module = module[:-2]
+       elif module[-2:] == '.c':
+           module = module[:-2]
+       return module
+
+    def serialize_enum(self, output, name):
+        id = self.idx.enums[name]
+        output.write("    <enum name='%s' file='%s'" % (name,
+                    self.modulename_file(id.header)))
+       if id.info != None:
+           info = id.info
+           if info[0] != None and info[0] != '':
+               try:
+                   val = eval(info[0])
+               except:
+                   val = info[0]
+               output.write(" value='%s'" % (val));
+           if info[2] != None and info[2] != '':
+               output.write(" type='%s'" % info[2]);
+           if info[1] != None and info[1] != '':
+               output.write(" info='%s'" % escape(info[1]));
+        output.write("/>\n")
+
+    def serialize_macro(self, output, name):
+        id = self.idx.macros[name]
+        output.write("    <macro name='%s' file='%s'>\n" % (name,
+                    self.modulename_file(id.header)))
+       if id.info != None:
+            try:
+               (args, desc) = id.info
+               if desc != None and desc != "":
+                   output.write("      <info>%s</info>\n" % (escape(desc)))
+                   self.indexString(name, desc)
+               for arg in args:
+                   (name, desc) = arg
+                   if desc != None and desc != "":
+                       output.write("      <arg name='%s' info='%s'/>\n" % (
+                                    name, escape(desc)))
+                       self.indexString(name, desc)
+                   else:
+                       output.write("      <arg name='%s'/>\n" % (name))
+            except:
+                pass
+        output.write("    </macro>\n")
+
+    def serialize_typedef(self, output, name):
+        id = self.idx.typedefs[name]
+       if id.info[0:7] == 'struct ':
+           output.write("    <struct name='%s' file='%s' type='%s'" % (
+                    name, self.modulename_file(id.header), id.info))
+           name = id.info[7:]
+           if self.idx.structs.has_key(name) and ( \
+              type(self.idx.structs[name].info) == type(()) or
+               type(self.idx.structs[name].info) == type([])):
+               output.write(">\n");
+               try:
+                   for field in self.idx.structs[name].info:
+                       desc = field[2]
+                       self.indexString(name, desc)
+                       if desc == None:
+                           desc = ''
+                       else:
+                           desc = escape(desc)
+                       output.write("      <field name='%s' type='%s' info='%s'/>\n" % (field[1] , field[0], desc))
+               except:
+                   print "Failed to serialize struct %s" % (name)
+               output.write("    </struct>\n")
+           else:
+               output.write("/>\n");
+       else :
+           output.write("    <typedef name='%s' file='%s' type='%s'" % (
+                        name, self.modulename_file(id.header), id.info))
+            try:
+               desc = id.extra
+               if desc != None and desc != "":
+                   output.write(">\n      <info>%s</info>\n" % (escape(desc)))
+                   output.write("    </typedef>\n")
+               else:
+                   output.write("/>\n")
+           except:
+               output.write("/>\n")
+
+    def serialize_variable(self, output, name):
+        id = self.idx.variables[name]
+       if id.info != None:
+           output.write("    <variable name='%s' file='%s' type='%s'/>\n" % (
+                   name, self.modulename_file(id.header), id.info))
+       else:
+           output.write("    <variable name='%s' file='%s'/>\n" % (
+                   name, self.modulename_file(id.header)))
+                     
+    def serialize_function(self, output, name):
+        id = self.idx.functions[name]
+       if name == debugsym:
+           print "=>", id
+
+        output.write("    <%s name='%s' file='%s' module='%s'>\n" % (id.type,
+                    name, self.modulename_file(id.header),
+                    self.modulename_file(id.module)))
+       #
+       # Processing of conditionals modified by Bill 1/1/05
+       #
+       if id.conditionals != None:
+           apstr = ""
+           for cond in id.conditionals:
+               if apstr != "":
+                   apstr = apstr + " &amp;&amp; "
+               apstr = apstr + cond
+           output.write("      <cond>%s</cond>\n"% (apstr));
+       try:
+           (ret, params, desc) = id.info
+           output.write("      <info>%s</info>\n" % (escape(desc)))
+           self.indexString(name, desc)
+           if ret[0] != None:
+               if ret[0] == "void":
+                   output.write("      <return type='void'/>\n")
+               else:
+                   output.write("      <return type='%s' info='%s'/>\n" % (
+                            ret[0], escape(ret[1])))
+                   self.indexString(name, ret[1])
+           for param in params:
+               if param[0] == 'void':
+                   continue
+               if param[2] == None:
+                   output.write("      <arg name='%s' type='%s' info=''/>\n" % (param[1], param[0]))
+               else:
+                   output.write("      <arg name='%s' type='%s' info='%s'/>\n" % (param[1], param[0], escape(param[2])))
+                   self.indexString(name, param[2])
+       except:
+           print "Failed to save function %s info: " % name, `id.info`
+        output.write("    </%s>\n" % (id.type))
+
+    def serialize_exports(self, output, file):
+        module = self.modulename_file(file)
+       output.write("    <file name='%s'>\n" % (module))
+       dict = self.headers[file]
+       if dict.info != None:
+           for data in ('Summary', 'Description', 'Author'):
+               try:
+                   output.write("     <%s>%s</%s>\n" % (
+                                string.lower(data),
+                                escape(dict.info[data]),
+                                string.lower(data)))
+               except:
+                   print "Header %s lacks a %s description" % (module, data)
+           if dict.info.has_key('Description'):
+               desc = dict.info['Description']
+               if string.find(desc, "DEPRECATED") != -1:
+                   output.write("     <deprecated/>\n")
+
+        ids = dict.macros.keys()
+       ids.sort()
+       for id in uniq(ids):
+           # Macros are sometime used to masquerade other types.
+           if dict.functions.has_key(id):
+               continue
+           if dict.variables.has_key(id):
+               continue
+           if dict.typedefs.has_key(id):
+               continue
+           if dict.structs.has_key(id):
+               continue
+           if dict.enums.has_key(id):
+               continue
+           output.write("     <exports symbol='%s' type='macro'/>\n" % (id))
+        ids = dict.enums.keys()
+       ids.sort()
+       for id in uniq(ids):
+           output.write("     <exports symbol='%s' type='enum'/>\n" % (id))
+        ids = dict.typedefs.keys()
+       ids.sort()
+       for id in uniq(ids):
+           output.write("     <exports symbol='%s' type='typedef'/>\n" % (id))
+        ids = dict.structs.keys()
+       ids.sort()
+       for id in uniq(ids):
+           output.write("     <exports symbol='%s' type='struct'/>\n" % (id))
+        ids = dict.variables.keys()
+       ids.sort()
+       for id in uniq(ids):
+           output.write("     <exports symbol='%s' type='variable'/>\n" % (id))
+        ids = dict.functions.keys()
+       ids.sort()
+       for id in uniq(ids):
+           output.write("     <exports symbol='%s' type='function'/>\n" % (id))
+       output.write("    </file>\n")
+
+    def serialize_xrefs_files(self, output):
+        headers = self.headers.keys()
+        headers.sort()
+        for file in headers:
+           module = self.modulename_file(file)
+           output.write("    <file name='%s'>\n" % (module))
+           dict = self.headers[file]
+           ids = uniq(dict.functions.keys() + dict.variables.keys() + \
+                 dict.macros.keys() + dict.typedefs.keys() + \
+                 dict.structs.keys() + dict.enums.keys())
+           ids.sort()
+           for id in ids:
+               output.write("      <ref name='%s'/>\n" % (id))
+           output.write("    </file>\n")
+        pass
+
+    def serialize_xrefs_functions(self, output):
+        funcs = {}
+       for name in self.idx.functions.keys():
+           id = self.idx.functions[name]
+           try:
+               (ret, params, desc) = id.info
+               for param in params:
+                   if param[0] == 'void':
+                       continue
+                   if funcs.has_key(param[0]):
+                       funcs[param[0]].append(name)
+                   else:
+                       funcs[param[0]] = [name]
+           except:
+               pass
+       typ = funcs.keys()
+       typ.sort()
+       for type in typ:
+           if type == '' or type == 'void' or type == "int" or \
+              type == "char *" or type == "const char *" :
+               continue
+           output.write("    <type name='%s'>\n" % (type))
+           ids = funcs[type]
+           ids.sort()
+           pid = ''    # not sure why we have dups, but get rid of them!
+           for id in ids:
+               if id != pid:
+                   output.write("      <ref name='%s'/>\n" % (id))
+                   pid = id
+           output.write("    </type>\n")
+
+    def serialize_xrefs_constructors(self, output):
+        funcs = {}
+       for name in self.idx.functions.keys():
+           id = self.idx.functions[name]
+           try:
+               (ret, params, desc) = id.info
+               if ret[0] == "void":
+                   continue
+               if funcs.has_key(ret[0]):
+                   funcs[ret[0]].append(name)
+               else:
+                   funcs[ret[0]] = [name]
+           except:
+               pass
+       typ = funcs.keys()
+       typ.sort()
+       for type in typ:
+           if type == '' or type == 'void' or type == "int" or \
+              type == "char *" or type == "const char *" :
+               continue
+           output.write("    <type name='%s'>\n" % (type))
+           ids = funcs[type]
+           ids.sort()
+           for id in ids:
+               output.write("      <ref name='%s'/>\n" % (id))
+           output.write("    </type>\n")
+
+    def serialize_xrefs_alpha(self, output):
+       letter = None
+       ids = self.idx.identifiers.keys()
+       ids.sort()
+       for id in ids:
+           if id[0] != letter:
+               if letter != None:
+                   output.write("    </letter>\n")
+               letter = id[0]
+               output.write("    <letter name='%s'>\n" % (letter))
+           output.write("      <ref name='%s'/>\n" % (id))
+       if letter != None:
+           output.write("    </letter>\n")
+
+    def serialize_xrefs_references(self, output):
+        typ = self.idx.identifiers.keys()
+       typ.sort()
+       for id in typ:
+           idf = self.idx.identifiers[id]
+           module = idf.header
+           output.write("    <reference name='%s' href='%s'/>\n" % (id,
+                        'html/' + self.basename + '-' +
+                        self.modulename_file(module) + '.html#' +
+                        id))
+
+    def serialize_xrefs_index(self, output):
+        index = self.xref
+       typ = index.keys()
+       typ.sort()
+       letter = None
+       count = 0
+       chunk = 0
+       chunks = []
+       for id in typ:
+           if len(index[id]) > 30:
+               continue
+           if id[0] != letter:
+               if letter == None or count > 200:
+                   if letter != None:
+                       output.write("      </letter>\n")
+                       output.write("    </chunk>\n")
+                       count = 0
+                       chunks.append(["chunk%s" % (chunk -1), first_letter, letter])
+                   output.write("    <chunk name='chunk%s'>\n" % (chunk))
+                   first_letter = id[0]
+                   chunk = chunk + 1
+               elif letter != None:
+                   output.write("      </letter>\n")
+               letter = id[0]
+               output.write("      <letter name='%s'>\n" % (letter))
+           output.write("        <word name='%s'>\n" % (id))
+           tokens = index[id];
+           tokens.sort()
+           tok = None
+           for token in tokens:
+               if tok == token:
+                   continue
+               tok = token
+               output.write("          <ref name='%s'/>\n" % (token))
+               count = count + 1
+           output.write("        </word>\n")
+       if letter != None:
+           output.write("      </letter>\n")
+           output.write("    </chunk>\n")
+           if count != 0:
+               chunks.append(["chunk%s" % (chunk -1), first_letter, letter])
+           output.write("    <chunks>\n")
+           for ch in chunks:
+               output.write("      <chunk name='%s' start='%s' end='%s'/>\n" % (
+                            ch[0], ch[1], ch[2]))
+           output.write("    </chunks>\n")
+
+    def serialize_xrefs(self, output):
+       output.write("  <references>\n")
+       self.serialize_xrefs_references(output)
+       output.write("  </references>\n")
+       output.write("  <alpha>\n")
+       self.serialize_xrefs_alpha(output)
+       output.write("  </alpha>\n")
+       output.write("  <constructors>\n")
+       self.serialize_xrefs_constructors(output)
+       output.write("  </constructors>\n")
+       output.write("  <functions>\n")
+       self.serialize_xrefs_functions(output)
+       output.write("  </functions>\n")
+       output.write("  <files>\n")
+       self.serialize_xrefs_files(output)
+       output.write("  </files>\n")
+       output.write("  <index>\n")
+       self.serialize_xrefs_index(output)
+       output.write("  </index>\n")
+
+    def serialize(self):
+        filename = "%s-api.xml" % self.name
+        print "Saving XML description %s" % (filename)
+        output = open(filename, "w")
+        output.write('<?xml version="1.0" encoding="ISO-8859-1"?>\n')
+        output.write("<api name='%s'>\n" % self.name)
+        output.write("  <files>\n")
+        headers = self.headers.keys()
+        headers.sort()
+        for file in headers:
+            self.serialize_exports(output, file)
+        output.write("  </files>\n")
+        output.write("  <symbols>\n")
+        macros = self.idx.macros.keys()
+        macros.sort()
+        for macro in macros:
+            self.serialize_macro(output, macro)
+        enums = self.idx.enums.keys()
+        enums.sort()
+        for enum in enums:
+            self.serialize_enum(output, enum)
+        typedefs = self.idx.typedefs.keys()
+        typedefs.sort()
+        for typedef in typedefs:
+            self.serialize_typedef(output, typedef)
+        variables = self.idx.variables.keys()
+        variables.sort()
+        for variable in variables:
+            self.serialize_variable(output, variable)
+        functions = self.idx.functions.keys()
+        functions.sort()
+        for function in functions:
+            self.serialize_function(output, function)
+        output.write("  </symbols>\n")
+        output.write("</api>\n")
+        output.close()
+
+        filename = "%s-refs.xml" % self.name
+        print "Saving XML Cross References %s" % (filename)
+        output = open(filename, "w")
+        output.write('<?xml version="1.0" encoding="ISO-8859-1"?>\n')
+        output.write("<apirefs name='%s'>\n" % self.name)
+        self.serialize_xrefs(output)
+        output.write("</apirefs>\n")
+        output.close()
+
+
+def rebuild():
+    builder = None
+    if glob.glob("../src/libxen.c") != [] :
+        print "Rebuilding API description for libxen"
+       builder = docBuilder("libxen", ["../src", "../include"],
+                            [])
+    elif glob.glob("src/parser.c") != [] :
+        print "Rebuilding API description for libxml2"
+       builder = docBuilder("libxen", ["src", "include"],
+                            [])
+    else:
+        print "rebuild() failed, unable to guess the module"
+       return None
+    builder.scan()
+    builder.analyze()
+    builder.serialize()
+    return builder
+
+#
+# for debugging the parser
+#
+def parse(filename):
+    parser = CParser(filename)
+    idx = parser.parse()
+    return idx
+
+if __name__ == "__main__":
+    if len(sys.argv) > 1:
+        debug = 1
+        parse(sys.argv[1])
+    else:
+       rebuild()
diff --git a/docs/libxen-api.xml b/docs/libxen-api.xml
new file mode 100644 (file)
index 0000000..896bcb9
--- /dev/null
@@ -0,0 +1,127 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<api name='libxen'>
+  <files>
+    <file name='libxen'>
+     <exports symbol='XEN_DOMAIN_NONE' type='enum'/>
+     <exports symbol='xenDomain' type='typedef'/>
+     <exports symbol='xenDomainFlags' type='typedef'/>
+     <exports symbol='xenConnectPtr' type='typedef'/>
+     <exports symbol='xenDomainPtr' type='typedef'/>
+     <exports symbol='xenConnect' type='typedef'/>
+     <exports symbol='xenDomainResume' type='function'/>
+     <exports symbol='xenDomainLookupByID' type='function'/>
+     <exports symbol='xenConnectOpenReadOnly' type='function'/>
+     <exports symbol='xenDomainGetName' type='function'/>
+     <exports symbol='xenDomainGetMaxMemory' type='function'/>
+     <exports symbol='xenDomainGetID' type='function'/>
+     <exports symbol='xenConnectOpen' type='function'/>
+     <exports symbol='xenDomainSetMaxMemory' type='function'/>
+     <exports symbol='xenDomainDestroy' type='function'/>
+     <exports symbol='xenDomainSuspend' type='function'/>
+     <exports symbol='xenConnectClose' type='function'/>
+     <exports symbol='xenDomainCreateLinux' type='function'/>
+     <exports symbol='xenDomainLookupByName' type='function'/>
+     <exports symbol='xenConnectGetVersion' type='function'/>
+    </file>
+    <file name='internal'>
+     <exports symbol='ATTRIBUTE_UNUSED' type='macro'/>
+     <exports symbol='TODO' type='macro'/>
+    </file>
+  </files>
+  <symbols>
+    <macro name='ATTRIBUTE_UNUSED' file='internal'>
+      <info>Macro to flag conciously unused parameters to functions</info>
+    </macro>
+    <macro name='TODO' file='internal'>
+      <info>macro to flag unimplemented blocks</info>
+    </macro>
+    <enum name='XEN_DOMAIN_NONE' file='libxen' value='0' type='xenDomainFlags'/>
+    <struct name='xenConnect' file='libxen' type='struct _xenConnect'/>
+    <typedef name='xenConnectPtr' file='libxen' type='xenConnect *'>
+      <info>a xenConnectPtr is pointer to a xenConnect private structure, this is the type used to reference a connection to the Xen Hypervisor in the API.</info>
+    </typedef>
+    <struct name='xenDomain' file='libxen' type='struct _xenDomain'/>
+    <typedef name='xenDomainFlags' file='libxen' type='enum'/>
+    <typedef name='xenDomainPtr' file='libxen' type='xenDomain *'>
+      <info>a xenDomainPtr is pointer to a xenDomain private structure, this is the type used to reference a Xen domain in the API.</info>
+    </typedef>
+    <function name='xenConnectClose' file='libxen' module='libxen'>
+      <info>This function closes the connection to the Hypervisor. This should not be called if further interaction with the Hypervisor are needed especially if there is running domain which need further monitoring by the application.</info>
+      <return type='int' info='0 in case of success or -1 in case of error.'/>
+      <arg name='conn' type='xenConnectPtr' info='pointer to the hypervisor connection'/>
+    </function>
+    <function name='xenConnectGetVersion' file='libxen' module='libxen'>
+      <info>Get the version level of the Hypervisor running.</info>
+      <return type='unsigned long' info='-1 in case of error or major * 10,000 + minor * 100 + rev otherwise'/>
+      <arg name='conn' type='xenConnectPtr' info='pointer to the hypervisor connection'/>
+    </function>
+    <function name='xenConnectOpen' file='libxen' module='libxen'>
+      <info>This function should be called first to get a connection to the Hypervisor and xen store</info>
+      <return type='xenConnectPtr' info='a pointer to the hypervisor connection or NULL in case of error'/>
+      <arg name='name' type='const char *' info='optional argument currently unused, pass NULL'/>
+    </function>
+    <function name='xenConnectOpenReadOnly' file='libxen' module='libxen'>
+      <info>This function should be called first to get a read-only connection to the xen store. The set of APIs usable are then restricted.</info>
+      <return type='xenConnectPtr' info='a pointer to the hypervisor connection or NULL in case of error'/>
+      <arg name='name' type='const char *' info='optional argument currently unused, pass NULL'/>
+    </function>
+    <function name='xenDomainCreateLinux' file='libxen' module='libxen'>
+      <info>Launch a new Linux guest domain</info>
+      <return type='xenDomainPtr' info='a new domain object or NULL in case of failure'/>
+      <arg name='conn' type='xenConnectPtr' info='pointer to the hypervisor connection'/>
+      <arg name='kernel_path' type='const char *' info='the file path to the kernel image'/>
+      <arg name='initrd_path' type='const char *' info='an optional file path to an initrd'/>
+      <arg name='cmdline' type='const char *' info='optional command line parameters for the kernel'/>
+      <arg name='memory' type='unsigned long' info='the memory size in kilobytes'/>
+      <arg name='flags' type='unsigned int' info='an optional set of xenDomainFlags'/>
+    </function>
+    <function name='xenDomainDestroy' file='libxen' module='libxen'>
+      <info>Destroy the domain object. The running instance is shutdown if not down already and all resources used by it are given back to the hypervisor.</info>
+      <return type='int' info='0 in case of success and -1 in case of failure.'/>
+      <arg name='domain' type='xenDomainPtr' info='a domain object'/>
+    </function>
+    <function name='xenDomainGetID' file='libxen' module='libxen'>
+      <info>Get the hypervisor ID number for the domain</info>
+      <return type='unsigned int' info='the domain ID number or (unsigned int) -1 in case of error'/>
+      <arg name='domain' type='xenDomainPtr' info='a domain object'/>
+    </function>
+    <function name='xenDomainGetMaxMemory' file='libxen' module='libxen'>
+      <info>Retrieve the maximum amount of physical memory allocated to a domain. If domain is NULL, then this get the amount of memory reserved to Domain0 i.e. the domain where the application runs.</info>
+      <return type='unsigned long' info='the memory size in kilobytes or 0 in case of error.'/>
+      <arg name='domain' type='xenDomainPtr' info='a domain object or NULL'/>
+    </function>
+    <function name='xenDomainGetName' file='libxen' module='libxen'>
+      <info>Get the public name for that domain</info>
+      <return type='const char *' info='a pointer to the name or NULL, the string need not be deallocated its lifetime will be the same as the domain object.'/>
+      <arg name='domain' type='xenDomainPtr' info='a domain object'/>
+    </function>
+    <function name='xenDomainLookupByID' file='libxen' module='libxen'>
+      <info>Try to find a domain based on the hypervisor ID number</info>
+      <return type='xenDomainPtr' info='a new domain object or NULL in case of failure'/>
+      <arg name='conn' type='xenConnectPtr' info='pointer to the hypervisor connection'/>
+      <arg name='id' type='int' info='the domain ID number'/>
+    </function>
+    <function name='xenDomainLookupByName' file='libxen' module='libxen'>
+      <info>Try to lookup a domain on the given hypervisor</info>
+      <return type='xenDomainPtr' info='a new domain object or NULL in case of failure'/>
+      <arg name='conn' type='xenConnectPtr' info='pointer to the hypervisor connection'/>
+      <arg name='name' type='const char *' info='name for the domain'/>
+    </function>
+    <function name='xenDomainResume' file='libxen' module='libxen'>
+      <info>Resume an suspended domain, the process is restarted from the state where it was frozen by calling xenSuspendDomain().</info>
+      <return type='int' info='0 in case of success and -1 in case of failure.'/>
+      <arg name='domain' type='xenDomainPtr' info='a domain object'/>
+    </function>
+    <function name='xenDomainSetMaxMemory' file='libxen' module='libxen'>
+      <info>Dynamically change the maximum amount of physical memory allocated to a domain. If domain is NULL, then this change the amount of memory reserved to Domain0 i.e. the domain where the application runs.</info>
+      <return type='int' info='0 in case of success and -1 in case of failure.'/>
+      <arg name='domain' type='xenDomainPtr' info='a domain object or NULL'/>
+      <arg name='memory' type='unsigned long' info='the memory size in kilobytes'/>
+    </function>
+    <function name='xenDomainSuspend' file='libxen' module='libxen'>
+      <info>Suspends an active domain, the process is frozen without further access to CPU resources and I/O but the memory used by the domain at the hypervisor level will stay allocated. Use xenDomainResume() to reactivate the domain.</info>
+      <return type='int' info='0 in case of success and -1 in case of failure.'/>
+      <arg name='domain' type='xenDomainPtr' info='a domain object'/>
+    </function>
+  </symbols>
+</api>
diff --git a/docs/libxen-refs.xml b/docs/libxen-refs.xml
new file mode 100644 (file)
index 0000000..bb3b351
--- /dev/null
@@ -0,0 +1,741 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<apirefs name='libxen'>
+  <references>
+    <reference name='ATTRIBUTE_UNUSED' href='html/libxen-internal.html#ATTRIBUTE_UNUSED'/>
+    <reference name='TODO' href='html/libxen-internal.html#TODO'/>
+    <reference name='XEN_DOMAIN_NONE' href='html/libxen-libxen.html#XEN_DOMAIN_NONE'/>
+    <reference name='xenConnect' href='html/libxen-libxen.html#xenConnect'/>
+    <reference name='xenConnectClose' href='html/libxen-libxen.html#xenConnectClose'/>
+    <reference name='xenConnectGetVersion' href='html/libxen-libxen.html#xenConnectGetVersion'/>
+    <reference name='xenConnectOpen' href='html/libxen-libxen.html#xenConnectOpen'/>
+    <reference name='xenConnectOpenReadOnly' href='html/libxen-libxen.html#xenConnectOpenReadOnly'/>
+    <reference name='xenConnectPtr' href='html/libxen-libxen.html#xenConnectPtr'/>
+    <reference name='xenDomain' href='html/libxen-libxen.html#xenDomain'/>
+    <reference name='xenDomainCreateLinux' href='html/libxen-libxen.html#xenDomainCreateLinux'/>
+    <reference name='xenDomainDestroy' href='html/libxen-libxen.html#xenDomainDestroy'/>
+    <reference name='xenDomainFlags' href='html/libxen-libxen.html#xenDomainFlags'/>
+    <reference name='xenDomainGetID' href='html/libxen-libxen.html#xenDomainGetID'/>
+    <reference name='xenDomainGetMaxMemory' href='html/libxen-libxen.html#xenDomainGetMaxMemory'/>
+    <reference name='xenDomainGetName' href='html/libxen-libxen.html#xenDomainGetName'/>
+    <reference name='xenDomainLookupByID' href='html/libxen-libxen.html#xenDomainLookupByID'/>
+    <reference name='xenDomainLookupByName' href='html/libxen-libxen.html#xenDomainLookupByName'/>
+    <reference name='xenDomainPtr' href='html/libxen-libxen.html#xenDomainPtr'/>
+    <reference name='xenDomainResume' href='html/libxen-libxen.html#xenDomainResume'/>
+    <reference name='xenDomainSetMaxMemory' href='html/libxen-libxen.html#xenDomainSetMaxMemory'/>
+    <reference name='xenDomainSuspend' href='html/libxen-libxen.html#xenDomainSuspend'/>
+  </references>
+  <alpha>
+    <letter name='A'>
+      <ref name='ATTRIBUTE_UNUSED'/>
+    </letter>
+    <letter name='T'>
+      <ref name='TODO'/>
+    </letter>
+    <letter name='X'>
+      <ref name='XEN_DOMAIN_NONE'/>
+    </letter>
+    <letter name='x'>
+      <ref name='xenConnect'/>
+      <ref name='xenConnectClose'/>
+      <ref name='xenConnectGetVersion'/>
+      <ref name='xenConnectOpen'/>
+      <ref name='xenConnectOpenReadOnly'/>
+      <ref name='xenConnectPtr'/>
+      <ref name='xenDomain'/>
+      <ref name='xenDomainCreateLinux'/>
+      <ref name='xenDomainDestroy'/>
+      <ref name='xenDomainFlags'/>
+      <ref name='xenDomainGetID'/>
+      <ref name='xenDomainGetMaxMemory'/>
+      <ref name='xenDomainGetName'/>
+      <ref name='xenDomainLookupByID'/>
+      <ref name='xenDomainLookupByName'/>
+      <ref name='xenDomainPtr'/>
+      <ref name='xenDomainResume'/>
+      <ref name='xenDomainSetMaxMemory'/>
+      <ref name='xenDomainSuspend'/>
+    </letter>
+  </alpha>
+  <constructors>
+    <type name='unsigned int'>
+      <ref name='xenDomainGetID'/>
+    </type>
+    <type name='unsigned long'>
+      <ref name='xenConnectGetVersion'/>
+      <ref name='xenDomainGetMaxMemory'/>
+    </type>
+    <type name='xenConnectPtr'>
+      <ref name='xenConnectOpen'/>
+      <ref name='xenConnectOpenReadOnly'/>
+    </type>
+    <type name='xenDomainPtr'>
+      <ref name='xenDomainCreateLinux'/>
+      <ref name='xenDomainLookupByID'/>
+      <ref name='xenDomainLookupByName'/>
+    </type>
+  </constructors>
+  <functions>
+    <type name='unsigned int'>
+      <ref name='xenDomainCreateLinux'/>
+    </type>
+    <type name='unsigned long'>
+      <ref name='xenDomainCreateLinux'/>
+      <ref name='xenDomainSetMaxMemory'/>
+    </type>
+    <type name='xenConnectPtr'>
+      <ref name='xenConnectClose'/>
+      <ref name='xenConnectGetVersion'/>
+      <ref name='xenDomainCreateLinux'/>
+      <ref name='xenDomainLookupByID'/>
+      <ref name='xenDomainLookupByName'/>
+    </type>
+    <type name='xenDomainPtr'>
+      <ref name='xenDomainDestroy'/>
+      <ref name='xenDomainGetID'/>
+      <ref name='xenDomainGetMaxMemory'/>
+      <ref name='xenDomainGetName'/>
+      <ref name='xenDomainResume'/>
+      <ref name='xenDomainSetMaxMemory'/>
+      <ref name='xenDomainSuspend'/>
+    </type>
+  </functions>
+  <files>
+    <file name='libxen'>
+      <ref name='XEN_DOMAIN_NONE'/>
+      <ref name='xenConnect'/>
+      <ref name='xenConnectClose'/>
+      <ref name='xenConnectGetVersion'/>
+      <ref name='xenConnectOpen'/>
+      <ref name='xenConnectOpenReadOnly'/>
+      <ref name='xenConnectPtr'/>
+      <ref name='xenDomain'/>
+      <ref name='xenDomainCreateLinux'/>
+      <ref name='xenDomainDestroy'/>
+      <ref name='xenDomainFlags'/>
+      <ref name='xenDomainGetID'/>
+      <ref name='xenDomainGetMaxMemory'/>
+      <ref name='xenDomainGetName'/>
+      <ref name='xenDomainLookupByID'/>
+      <ref name='xenDomainLookupByName'/>
+      <ref name='xenDomainPtr'/>
+      <ref name='xenDomainResume'/>
+      <ref name='xenDomainSetMaxMemory'/>
+      <ref name='xenDomainSuspend'/>
+    </file>
+    <file name='internal'>
+      <ref name='ATTRIBUTE_UNUSED'/>
+      <ref name='TODO'/>
+    </file>
+  </files>
+  <index>
+    <chunk name='chunk0'>
+      <letter name='A'>
+        <word name='APIs'>
+          <ref name='xenConnectOpenReadOnly'/>
+        </word>
+      </letter>
+      <letter name='C'>
+        <word name='CPU'>
+          <ref name='xenDomainSuspend'/>
+        </word>
+      </letter>
+      <letter name='D'>
+        <word name='Destroy'>
+          <ref name='xenDomainDestroy'/>
+        </word>
+        <word name='Domain0'>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainSetMaxMemory'/>
+        </word>
+        <word name='Dynamically'>
+          <ref name='xenDomainSetMaxMemory'/>
+        </word>
+      </letter>
+      <letter name='G'>
+        <word name='Get'>
+          <ref name='xenConnectGetVersion'/>
+          <ref name='xenDomainGetID'/>
+          <ref name='xenDomainGetName'/>
+        </word>
+      </letter>
+      <letter name='H'>
+        <word name='Hypervisor'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenConnectGetVersion'/>
+          <ref name='xenConnectOpen'/>
+        </word>
+      </letter>
+      <letter name='L'>
+        <word name='Launch'>
+          <ref name='xenDomainCreateLinux'/>
+        </word>
+        <word name='Linux'>
+          <ref name='xenDomainCreateLinux'/>
+        </word>
+      </letter>
+      <letter name='M'>
+        <word name='Macro'>
+          <ref name='ATTRIBUTE_UNUSED'/>
+        </word>
+      </letter>
+      <letter name='N'>
+        <word name='NULL'>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+          <ref name='xenDomainCreateLinux'/>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainGetName'/>
+          <ref name='xenDomainLookupByID'/>
+          <ref name='xenDomainLookupByName'/>
+          <ref name='xenDomainSetMaxMemory'/>
+        </word>
+      </letter>
+      <letter name='R'>
+        <word name='Resume'>
+          <ref name='xenDomainResume'/>
+        </word>
+        <word name='Retrieve'>
+          <ref name='xenDomainGetMaxMemory'/>
+        </word>
+      </letter>
+      <letter name='S'>
+        <word name='Suspends'>
+          <ref name='xenDomainSuspend'/>
+        </word>
+      </letter>
+      <letter name='T'>
+        <word name='This'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+        </word>
+        <word name='Try'>
+          <ref name='xenDomainLookupByID'/>
+          <ref name='xenDomainLookupByName'/>
+        </word>
+      </letter>
+      <letter name='U'>
+        <word name='Use'>
+          <ref name='xenDomainSuspend'/>
+        </word>
+      </letter>
+      <letter name='a'>
+        <word name='access'>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='active'>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='all'>
+          <ref name='xenDomainDestroy'/>
+        </word>
+        <word name='allocated'>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainSetMaxMemory'/>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='already'>
+          <ref name='xenDomainDestroy'/>
+        </word>
+        <word name='amount'>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainSetMaxMemory'/>
+        </word>
+        <word name='application'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainSetMaxMemory'/>
+        </word>
+        <word name='are'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenConnectOpenReadOnly'/>
+          <ref name='xenDomainDestroy'/>
+        </word>
+        <word name='argument'>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+        </word>
+      </letter>
+      <letter name='b'>
+        <word name='back'>
+          <ref name='xenDomainDestroy'/>
+        </word>
+        <word name='based'>
+          <ref name='xenDomainLookupByID'/>
+        </word>
+        <word name='blocks'>
+          <ref name='TODO'/>
+        </word>
+        <word name='but'>
+          <ref name='xenDomainSuspend'/>
+        </word>
+      </letter>
+      <letter name='c'>
+        <word name='called'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+        </word>
+        <word name='calling'>
+          <ref name='xenDomainResume'/>
+        </word>
+        <word name='case'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenConnectGetVersion'/>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+          <ref name='xenDomainCreateLinux'/>
+          <ref name='xenDomainDestroy'/>
+          <ref name='xenDomainGetID'/>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainLookupByID'/>
+          <ref name='xenDomainLookupByName'/>
+          <ref name='xenDomainResume'/>
+          <ref name='xenDomainSetMaxMemory'/>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='change'>
+          <ref name='xenDomainSetMaxMemory'/>
+        </word>
+        <word name='closes'>
+          <ref name='xenConnectClose'/>
+        </word>
+        <word name='command'>
+          <ref name='xenDomainCreateLinux'/>
+        </word>
+        <word name='conciously'>
+          <ref name='ATTRIBUTE_UNUSED'/>
+        </word>
+        <word name='connection'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenConnectGetVersion'/>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+          <ref name='xenDomainCreateLinux'/>
+          <ref name='xenDomainLookupByID'/>
+          <ref name='xenDomainLookupByName'/>
+        </word>
+        <word name='currently'>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+        </word>
+      </letter>
+      <letter name='d'>
+        <word name='deallocated'>
+          <ref name='xenDomainGetName'/>
+        </word>
+        <word name='down'>
+          <ref name='xenDomainDestroy'/>
+        </word>
+      </letter>
+      <letter name='e'>
+        <word name='error'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenConnectGetVersion'/>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+          <ref name='xenDomainGetID'/>
+          <ref name='xenDomainGetMaxMemory'/>
+        </word>
+        <word name='especially'>
+          <ref name='xenConnectClose'/>
+        </word>
+      </letter>
+      <letter name='f'>
+        <word name='failure'>
+          <ref name='xenDomainCreateLinux'/>
+          <ref name='xenDomainDestroy'/>
+          <ref name='xenDomainLookupByID'/>
+          <ref name='xenDomainLookupByName'/>
+          <ref name='xenDomainResume'/>
+          <ref name='xenDomainSetMaxMemory'/>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='file'>
+          <ref name='xenDomainCreateLinux'/>
+        </word>
+        <word name='find'>
+          <ref name='xenDomainLookupByID'/>
+        </word>
+        <word name='first'>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+        </word>
+        <word name='flag'>
+          <ref name='ATTRIBUTE_UNUSED'/>
+          <ref name='TODO'/>
+        </word>
+        <word name='for'>
+          <ref name='xenDomainCreateLinux'/>
+          <ref name='xenDomainGetID'/>
+          <ref name='xenDomainGetName'/>
+          <ref name='xenDomainLookupByName'/>
+        </word>
+        <word name='from'>
+          <ref name='xenDomainResume'/>
+        </word>
+        <word name='frozen'>
+          <ref name='xenDomainResume'/>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='function'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+        </word>
+        <word name='functions'>
+          <ref name='ATTRIBUTE_UNUSED'/>
+        </word>
+        <word name='further'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenDomainSuspend'/>
+        </word>
+      </letter>
+      <letter name='g'>
+        <word name='get'>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+          <ref name='xenDomainGetMaxMemory'/>
+        </word>
+        <word name='given'>
+          <ref name='xenDomainDestroy'/>
+          <ref name='xenDomainLookupByName'/>
+        </word>
+        <word name='guest'>
+          <ref name='xenDomainCreateLinux'/>
+        </word>
+      </letter>
+      <letter name='h'>
+        <word name='hypervisor'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenConnectGetVersion'/>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+          <ref name='xenDomainCreateLinux'/>
+          <ref name='xenDomainDestroy'/>
+          <ref name='xenDomainGetID'/>
+          <ref name='xenDomainLookupByID'/>
+          <ref name='xenDomainLookupByName'/>
+          <ref name='xenDomainSuspend'/>
+        </word>
+      </letter>
+      <letter name='i'>
+        <word name='image'>
+          <ref name='xenDomainCreateLinux'/>
+        </word>
+        <word name='initrd'>
+          <ref name='xenDomainCreateLinux'/>
+        </word>
+        <word name='instance'>
+          <ref name='xenDomainDestroy'/>
+        </word>
+        <word name='int'>
+          <ref name='xenDomainGetID'/>
+        </word>
+        <word name='interaction'>
+          <ref name='xenConnectClose'/>
+        </word>
+        <word name='its'>
+          <ref name='xenDomainGetName'/>
+        </word>
+      </letter>
+      <letter name='k'>
+        <word name='kernel'>
+          <ref name='xenDomainCreateLinux'/>
+        </word>
+        <word name='kilobytes'>
+          <ref name='xenDomainCreateLinux'/>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainSetMaxMemory'/>
+        </word>
+      </letter>
+      <letter name='l'>
+        <word name='level'>
+          <ref name='xenConnectGetVersion'/>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='lifetime'>
+          <ref name='xenDomainGetName'/>
+        </word>
+        <word name='line'>
+          <ref name='xenDomainCreateLinux'/>
+        </word>
+        <word name='lookup'>
+          <ref name='xenDomainLookupByName'/>
+        </word>
+      </letter>
+      <letter name='m'>
+        <word name='macro'>
+          <ref name='TODO'/>
+        </word>
+        <word name='major'>
+          <ref name='xenConnectGetVersion'/>
+        </word>
+        <word name='maximum'>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainSetMaxMemory'/>
+        </word>
+        <word name='memory'>
+          <ref name='xenDomainCreateLinux'/>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainSetMaxMemory'/>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='minor'>
+          <ref name='xenConnectGetVersion'/>
+        </word>
+        <word name='monitoring'>
+          <ref name='xenConnectClose'/>
+        </word>
+      </letter>
+      <letter name='n'>
+        <word name='name'>
+          <ref name='xenDomainGetName'/>
+          <ref name='xenDomainLookupByName'/>
+        </word>
+        <word name='need'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenDomainGetName'/>
+        </word>
+        <word name='needed'>
+          <ref name='xenConnectClose'/>
+        </word>
+        <word name='new'>
+          <ref name='xenDomainCreateLinux'/>
+          <ref name='xenDomainLookupByID'/>
+          <ref name='xenDomainLookupByName'/>
+        </word>
+        <word name='not'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenDomainDestroy'/>
+          <ref name='xenDomainGetName'/>
+        </word>
+        <word name='number'>
+          <ref name='xenDomainGetID'/>
+          <ref name='xenDomainLookupByID'/>
+        </word>
+      </letter>
+      <letter name='o'>
+        <word name='object'>
+          <ref name='xenDomainCreateLinux'/>
+          <ref name='xenDomainDestroy'/>
+          <ref name='xenDomainGetID'/>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainGetName'/>
+          <ref name='xenDomainLookupByID'/>
+          <ref name='xenDomainLookupByName'/>
+          <ref name='xenDomainResume'/>
+          <ref name='xenDomainSetMaxMemory'/>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='optional'>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+          <ref name='xenDomainCreateLinux'/>
+        </word>
+        <word name='otherwise'>
+          <ref name='xenConnectGetVersion'/>
+        </word>
+      </letter>
+      <letter name='p'>
+        <word name='parameters'>
+          <ref name='ATTRIBUTE_UNUSED'/>
+          <ref name='xenDomainCreateLinux'/>
+        </word>
+        <word name='pass'>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+        </word>
+        <word name='path'>
+          <ref name='xenDomainCreateLinux'/>
+        </word>
+        <word name='physical'>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainSetMaxMemory'/>
+        </word>
+        <word name='pointer'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenConnectGetVersion'/>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+          <ref name='xenDomainCreateLinux'/>
+          <ref name='xenDomainGetName'/>
+          <ref name='xenDomainLookupByID'/>
+          <ref name='xenDomainLookupByName'/>
+        </word>
+        <word name='process'>
+          <ref name='xenDomainResume'/>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='public'>
+          <ref name='xenDomainGetName'/>
+        </word>
+      </letter>
+    </chunk>
+    <chunk name='chunk1'>
+      <letter name='r'>
+        <word name='reactivate'>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='read-only'>
+          <ref name='xenConnectOpenReadOnly'/>
+        </word>
+        <word name='reserved'>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainSetMaxMemory'/>
+        </word>
+        <word name='resources'>
+          <ref name='xenDomainDestroy'/>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='restarted'>
+          <ref name='xenDomainResume'/>
+        </word>
+        <word name='restricted'>
+          <ref name='xenConnectOpenReadOnly'/>
+        </word>
+        <word name='rev'>
+          <ref name='xenConnectGetVersion'/>
+        </word>
+        <word name='running'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenConnectGetVersion'/>
+          <ref name='xenDomainDestroy'/>
+        </word>
+        <word name='runs'>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainSetMaxMemory'/>
+        </word>
+      </letter>
+      <letter name='s'>
+        <word name='same'>
+          <ref name='xenDomainGetName'/>
+        </word>
+        <word name='set'>
+          <ref name='xenConnectOpenReadOnly'/>
+          <ref name='xenDomainCreateLinux'/>
+        </word>
+        <word name='should'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+        </word>
+        <word name='shutdown'>
+          <ref name='xenDomainDestroy'/>
+        </word>
+        <word name='size'>
+          <ref name='xenDomainCreateLinux'/>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainSetMaxMemory'/>
+        </word>
+        <word name='state'>
+          <ref name='xenDomainResume'/>
+        </word>
+        <word name='stay'>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='store'>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+        </word>
+        <word name='string'>
+          <ref name='xenDomainGetName'/>
+        </word>
+        <word name='success'>
+          <ref name='xenConnectClose'/>
+          <ref name='xenDomainDestroy'/>
+          <ref name='xenDomainResume'/>
+          <ref name='xenDomainSetMaxMemory'/>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='suspended'>
+          <ref name='xenDomainResume'/>
+        </word>
+      </letter>
+      <letter name='t'>
+        <word name='that'>
+          <ref name='xenDomainGetName'/>
+        </word>
+        <word name='then'>
+          <ref name='xenConnectOpenReadOnly'/>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainSetMaxMemory'/>
+        </word>
+        <word name='there'>
+          <ref name='xenConnectClose'/>
+        </word>
+        <word name='this'>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainSetMaxMemory'/>
+        </word>
+      </letter>
+      <letter name='u'>
+        <word name='unimplemented'>
+          <ref name='TODO'/>
+        </word>
+        <word name='unsigned'>
+          <ref name='xenDomainGetID'/>
+        </word>
+        <word name='unused'>
+          <ref name='ATTRIBUTE_UNUSED'/>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+        </word>
+        <word name='usable'>
+          <ref name='xenConnectOpenReadOnly'/>
+        </word>
+        <word name='used'>
+          <ref name='xenDomainDestroy'/>
+          <ref name='xenDomainSuspend'/>
+        </word>
+      </letter>
+      <letter name='v'>
+        <word name='version'>
+          <ref name='xenConnectGetVersion'/>
+        </word>
+      </letter>
+      <letter name='w'>
+        <word name='was'>
+          <ref name='xenDomainResume'/>
+        </word>
+        <word name='where'>
+          <ref name='xenDomainGetMaxMemory'/>
+          <ref name='xenDomainResume'/>
+          <ref name='xenDomainSetMaxMemory'/>
+        </word>
+        <word name='which'>
+          <ref name='xenConnectClose'/>
+        </word>
+        <word name='will'>
+          <ref name='xenDomainGetName'/>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='with'>
+          <ref name='xenConnectClose'/>
+        </word>
+        <word name='without'>
+          <ref name='xenDomainSuspend'/>
+        </word>
+      </letter>
+      <letter name='x'>
+        <word name='xen'>
+          <ref name='xenConnectOpen'/>
+          <ref name='xenConnectOpenReadOnly'/>
+        </word>
+        <word name='xenDomainFlags'>
+          <ref name='xenDomainCreateLinux'/>
+        </word>
+        <word name='xenDomainResume'>
+          <ref name='xenDomainSuspend'/>
+        </word>
+        <word name='xenSuspendDomain'>
+          <ref name='xenDomainResume'/>
+        </word>
+      </letter>
+    </chunk>
+    <chunks>
+      <chunk name='chunk0' start='A' end='p'/>
+      <chunk name='chunk1' start='r' end='x'/>
+    </chunks>
+  </index>
+</apirefs>
diff --git a/docs/structures.fig b/docs/structures.fig
new file mode 100644 (file)
index 0000000..b6d3cce
--- /dev/null
@@ -0,0 +1,72 @@
+#FIG 3.2
+Landscape
+Center
+Inches
+Letter  
+100.00
+Single
+-2
+1200 2
+2 2 0 1 0 7 50 -1 -1 0.000 0 0 -1 0 0 5
+        1125 825 2400 825 2400 2850 1125 2850 1125 825
+2 1 0 1 0 7 50 -1 -1 4.000 0 0 -1 1 0 2
+       1 1 1.00 60.00 120.00
+        2400 1200 5175 1200
+2 1 0 1 0 7 50 -1 -1 4.000 0 0 -1 1 0 2
+       1 1 1.00 60.00 120.00
+        2400 1200 5175 2025
+2 4 0 1 0 7 50 -1 -1 4.000 0 0 7 0 0 5
+        5925 1500 5925 975 5175 975 5175 1500 5925 1500
+2 4 0 1 0 7 50 -1 -1 4.000 0 0 7 0 0 5
+        5925 2400 5925 1875 5175 1875 5175 2400 5925 2400
+2 4 0 1 0 7 50 -1 -1 4.000 0 0 7 0 0 5
+        2400 5250 2400 4500 1500 4500 1500 5250 2400 5250
+2 1 0 1 0 7 50 -1 -1 4.000 0 0 -1 1 0 2
+       1 1 1.00 60.00 120.00
+        1875 4725 1875 2850
+2 1 0 1 0 7 50 -1 -1 4.000 0 0 -1 1 0 2
+       1 1 1.00 60.00 120.00
+        5175 2175 2400 4500
+2 1 0 1 0 7 50 -1 -1 4.000 0 0 -1 1 0 2
+       1 1 1.00 60.00 120.00
+        5175 1350 2400 4500
+2 1 0 1 0 7 50 -1 -1 4.000 0 0 -1 1 0 2
+       1 1 1.00 60.00 120.00
+        5925 1200 6825 1200
+2 1 0 1 0 7 50 -1 -1 4.000 0 0 -1 1 0 2
+       1 1 1.00 60.00 120.00
+        5925 2100 6825 2100
+2 2 0 1 0 7 50 -1 -1 4.000 0 0 -1 0 0 5
+        3900 4650 6750 4650 6750 4875 3900 4875 3900 4650
+2 1 0 1 0 7 50 -1 -1 4.000 0 0 -1 0 0 2
+        4200 4650 4200 4875
+2 1 0 1 0 7 50 -1 -1 4.000 0 0 -1 0 0 2
+        4500 4650 4500 4875
+2 1 0 1 0 7 50 -1 -1 4.000 0 0 -1 0 0 2
+        4725 4650 4725 4875
+2 1 0 1 0 7 50 -1 -1 4.000 0 0 -1 0 0 2
+        6450 4650 6450 4875
+2 1 0 1 0 7 50 -1 -1 4.000 0 0 -1 0 0 2
+        6150 4650 6150 4875
+2 1 0 1 0 7 50 -1 -1 4.000 0 0 -1 1 0 2
+       1 1 1.00 60.00 120.00
+        4050 4725 5250 1500
+2 1 0 1 0 7 50 -1 -1 4.000 0 0 -1 1 0 2
+       1 1 1.00 60.00 120.00
+        4350 4725 5250 2400
+2 1 0 1 0 7 50 -1 -1 4.000 0 0 -1 1 0 2
+       1 1 1.00 60.00 120.00
+        2250 4800 3900 4800
+4 0 0 50 -1 0 18 0.0000 4 195 1005 1275 1125 Domains\001
+4 0 0 50 -1 0 18 0.0000 4 195 585 1275 1440 hash \001
+4 0 0 50 -1 0 18 0.0000 4 195 1650 4725 750 xenDomainPtr\001
+4 0 0 50 -1 0 18 0.0000 4 195 2055 975 5700 xenConnectionPtr\001
+4 0 0 50 -1 0 18 0.0000 4 195 945 900 4200 domains\001
+4 0 0 50 -1 0 18 0.0000 4 135 555 3450 2475 conn\001
+4 0 0 50 -1 0 18 0.0000 4 135 615 6075 1050 name\001
+4 0 0 50 -1 0 18 0.0000 4 195 1365 6975 1275 "Domain-0"\001
+4 0 0 50 -1 0 18 0.0000 4 195 585 6975 2175 "fc4"\001
+4 0 0 50 -1 0 18 0.0000 4 195 540 1275 1755 table\001
+4 0 0 50 -1 0 18 0.0000 4 135 615 1275 2385 name\001
+4 0 0 50 -1 0 18 0.0000 4 195 960 1275 2070 based on\001
+4 0 0 50 -1 0 18 0.0000 4 195 315 2850 5100 ids\001
index 32b7a1476a53989dcb9f50e6d0f2197cd02ddde3..7e4de49acd1744197b472b45ef69d55d2f288b00 100644 (file)
@@ -374,7 +374,7 @@ xenDomainDestroy(xenDomainPtr domain) {
  *
  * Suspends an active domain, the process is frozen without further access
  * to CPU resources and I/O but the memory used by the domain at the 
- * hypervisor level will stay allocated. Use xenResumeDomain() to reactivate
+ * hypervisor level will stay allocated. Use xenDomainResume() to reactivate
  * the domain.
  *
  * Returns 0 in case of success and -1 in case of failure.
@@ -388,7 +388,7 @@ xenDomainSuspend(xenDomainPtr domain) {
 }
 
 /**
- * xenResumeDomain:
+ * xenDomainResume:
  * @domain: a domain object
  *
  * Resume an suspended domain, the process is restarted from the state where