Refactor to use new protocol module and sipo.
[bertos.git] / wizard / bertos_utils.py
index 9cd3211593f21c4d72d41459811ed83f91ca3261..7be67fe10cdfcafb24b897cc4dc0ae8d36c22935 100644 (file)
@@ -1,10 +1,33 @@
 #!/usr/bin/env python
 # encoding: utf-8
 #
+# This file is part of BeRTOS.
+#
+# Bertos is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+#
+# As a special exception, you may use this file as part of a free software
+# library without restriction.  Specifically, if other files instantiate
+# templates or use macros or inline functions from this file, or you compile
+# this file and link it with other files to produce an executable, this
+# file does not by itself cause the resulting executable to be covered by
+# the GNU General Public License.  This exception does not however
+# invalidate any other reasons why the executable file might be covered by
+# the GNU General Public License.
+#
 # Copyright 2008 Develer S.r.l. (http://www.develer.com/)
-# All rights reserved.
 #
-# $Id:$
 #
 # Author: Lorenzo Berni <duplo@develer.com>
 #
@@ -14,8 +37,27 @@ import fnmatch
 import glob
 import re
 import shutil
+# Use custom copytree function
+import copytree
+import relpath
+import pickle
 
 import const
+import plugins
+import DefineException
+
+from _wizard_version import WIZARD_VERSION
+
+from LoadException import VersionException, ToolchainException
+
+def _cmp(x, y):
+    result = cmp(x["info"].get('ord', 0), y["info"].get('ord', 0))
+    if result == 0:
+        result = cmp(
+            x["info"].get("name", x["info"]["filename"]).lower(),
+            y["info"].get("name", y["info"]["filename"]).lower()
+        )
+    return result
 
 def isBertosDir(directory):
    return os.path.exists(directory + "/VERSION")
@@ -23,84 +65,297 @@ def isBertosDir(directory):
 def bertosVersion(directory):
    return open(directory + "/VERSION").readline().strip()
 
-def createBertosProject(projectInfos):
-    directory = projectInfos.info("PROJECT_PATH")
-    sourcesDir = projectInfos.info("SOURCES_PATH")
-    if not os.path.isdir(directory):
-        os.mkdir(directory)
-    f = open(directory + "/project.bertos", "w")
-    f.write(repr(projectInfos))
-    f.close()
-    ## Destination source dir
-    srcdir = directory + "/bertos"
+def enabledModules(project_info):
+    enabled_modules = []
+    for name, module in project_info.info("MODULES").items():
+        if module["enabled"]:
+            enabled_modules.append(name)
+    return enabled_modules
+
+def presetList(directory):
+    """
+    Return the list of the preset found in the selected BeRTOS Version.
+    """
+    abspath = os.path.join(directory, const.PREDEFINED_BOARDS_DIR)
+    preset_list = dict([
+        (os.path.join(abspath, preset_dir), presetInfo(os.path.join(abspath, preset_dir)))
+        for preset_dir in os.listdir(os.path.join(directory, const.PREDEFINED_BOARDS_DIR))
+    ])
+    return preset_list
+
+def presetInfo(preset_dir):
+    """
+    Return the preset-relevant info contined into the project_file.
+    """
+    preset_info = pickle.loads(open(os.path.join(preset_dir, "project.bertos"), "r").read())
+    try:
+        description = open(os.path.join(preset_dir, "description"), "r").read()
+    except IOError:
+        # No description file found.
+        description = ""
+    relevant_info = {
+        "CPU_NAME": preset_info.get("CPU_NAME"),
+        "SELECTED_FREQ": preset_info.get("SELECTED_FREQ"),
+        "WIZARD_VERSION": preset_info.get("WIZARD_VERSION", None),
+        "PRESET_NAME": preset_info.get("PROJECT_NAME"),
+        "PRESET_DESCRIPTION": description.decode("utf-8"),
+    }
+    return relevant_info
+
+def mergeSources(srcdir, new_sources, old_sources):
+    # The current mergeSources function provide only a raw copy of the sources in the
+    # created project.
+    #
+    # TODO: implement the three way merge algorithm
+    #
     shutil.rmtree(srcdir, True)
-    shutil.copytree(sourcesDir + "/bertos", srcdir)
-    ## Destination makefile
-    makefile = directory + "/Makefile"
-    if os.path.exists(makefile):
-        os.remove(makefile)
-    makefile = open("mktemplates/Makefile").read()
-    makefile = makefileGenerator(projectInfos, makefile)
-    open(directory + "/Makefile", "w").write(makefile)
-    ## Destination project dir
-    prjdir = directory + "/" + os.path.basename(directory)
-    shutil.rmtree(prjdir, True)
-    os.mkdir(prjdir)
-    ## Destination configurations files
-    cfgdir = prjdir + "/cfg"
-    shutil.rmtree(cfgdir, True)
-    os.mkdir(cfgdir)
-    for key, value in projectInfos.info("CONFIGURATIONS").items():
-        string = open(sourcesDir + "/" + key, "r").read()
-        for parameter, infos in value.items():
-            value = infos["value"]
-            if "long" in infos["informations"].keys() and infos["informations"]["long"]:
-                value += "L"
-            string = sub(string, parameter, value)
-        f = open(cfgdir + "/" + os.path.basename(key), "w")
-        f.write(string)
-        f.close()
-    ## Destinatio mk file
-    makefile = open("mktemplates/template.mk", "r").read()
-    makefile = mkGenerator(projectInfos, makefile)
-    open(prjdir + "/" + "project.mk", "w").write(makefile)
-
-def mkGenerator(projectInfos, makefile):
+    copytree.copytree(os.path.join(new_sources, "bertos"), srcdir, ignore_list=const.IGNORE_LIST)
+
+def projectFileGenerator(project_info):
+    directory = project_info.info("PROJECT_PATH")
+    project_data = {}
+    enabled_modules = []
+    for module, information in project_info.info("MODULES").items():
+        if information["enabled"]:
+            enabled_modules.append(module)
+    project_data["ENABLED_MODULES"] = sorted(enabled_modules)
+    if project_info.info("PRESET"):
+        # For presets save again the BERTOS_PATH into project file
+        project_data["PRESET"] = True
+        project_data["BERTOS_PATH"] = relpath.relpath(project_info.info("BERTOS_PATH"), directory)
+    elif project_info.edit:
+        # If in editing mode the BERTOS_PATH is maintained
+        project_data["BERTOS_PATH"] = relpath.relpath(project_info.info("BERTOS_PATH"), directory)
+    else:
+        # Use the local BeRTOS version instead of the original one
+        # project_data["BERTOS_PATH"] = project_info.info("BERTOS_PATH")
+        project_data["BERTOS_PATH"] = "."
+    project_data["PROJECT_NAME"] = project_info.info("PROJECT_NAME", os.path.basename(directory))
+    project_src_relpath = relpath.relpath(project_info.info("PROJECT_SRC_PATH"), directory)
+    project_data["PROJECT_SRC_PATH"] = project_src_relpath
+    project_data["PROJECT_SRC_PATH_FROM_MAKEFILE"] = project_info.info("PROJECT_SRC_PATH_FROM_MAKEFILE")
+    project_data["TOOLCHAIN"] = {'path': project_info.info("TOOLCHAIN")['path']}
+    project_data["CPU_NAME"] = project_info.info("CPU_NAME")
+    project_data["SELECTED_FREQ"] = project_info.info("SELECTED_FREQ")
+    project_data["OUTPUT"] = project_info.info("OUTPUT")
+    project_data["WIZARD_VERSION"] = WIZARD_VERSION
+    project_data["PRESET"] = project_info.info("PRESET")
+    project_data["PROJECT_HW_PATH"] = relpath.relpath(project_info.info("PROJECT_HW_PATH"), directory)
+    project_data["PROJECT_HW_PATH_FROM_MAKEFILE"] = project_info.info("PROJECT_HW_PATH_FROM_MAKEFILE")
+    return pickle.dumps(project_data)
+
+def loadPlugin(plugin):
+    """
+    Returns the given plugin module.
+    """
+    return getattr(__import__("plugins", {}, {}, [plugin]), plugin)
+
+def versionFileGenerator(project_info, version_file):
+    version = bertosVersion(project_info.info("BERTOS_PATH"))
+    return version_file.replace('$version', version)
+
+def userMkGeneratorFromPreset(project_info):
+    project_name = project_info.info("PROJECT_NAME")
+    preset_path = project_info.info("PRESET_PATH")
+    preset_name = project_info.info("PRESET_NAME")
+    preset_src_dir = project_info.info("PRESET_SRC_PATH")
+    makefile = open(os.path.join(preset_path, preset_src_dir, "%s_user.mk" %preset_name), 'r').read()
+    destination = os.path.join(project_info.prjdir, "%s_user.mk" %project_info.info("PROJECT_NAME"))
+    # Temporary code.
+    # TODO: write it using regular expressions to secure this function
+    makefile = makefile.replace(preset_name + "_", project_name + "_")
+    open(destination, "w").write(makefile)
+
+def userMkGenerator(project_info):
+    makefile = open(os.path.join(const.DATA_DIR, "mktemplates/template_user.mk"), "r").read()
+    destination = os.path.join(project_info.prjdir, os.path.basename(project_info.prjdir) + "_user.mk")
+    # Deadly performances loss was here :(
+    mk_data = {}
+    mk_data["$pname"] = os.path.basename(project_info.info("PROJECT_PATH"))
+    mk_data["$ppath"] = project_info.info("PROJECT_SRC_PATH_FROM_MAKEFILE")
+    mk_data["$main"] = "/".join(["$(%s_SRC_PATH)" %project_info.info("PROJECT_NAME"), "main.c"])
+    for key in mk_data:
+        makefile = makefile.replace(key, mk_data[key])
+    open(destination, "w").write(makefile)
+
+def mkGenerator(project_info):
     """
     Generates the mk file for the current project.
     """
-    mkData = {}
-    mkData["pname"] = os.path.basename(projectInfos.info("PROJECT_PATH"))
-    mkData["cpuname"] = projectInfos.info("CPU_INFOS")["CPU_NAME"]
-    mkData["cflags"] = " ".join(projectInfos.info("CPU_INFOS")["C_FLAGS"])
-    mkData["ldflags"] = " ".join(projectInfos.info("CPU_INFOS")["LD_FLAGS"])
-    for key in mkData:
-        while makefile.find(key) != -1:
-            makefile = makefile.replace(key, mkData[key])
-    return makefile
-
-def makefileGenerator(projectInfos, makefile):
+    makefile = open(os.path.join(const.DATA_DIR, "mktemplates/template.mk"), "r").read()
+    prjdir = os.path.abspath(project_info.prjdir)
+    destination = os.path.join(prjdir, os.path.basename(prjdir) + ".mk")
+    mk_data = {}
+    mk_data["$pname"] = project_info.info("PROJECT_NAME")
+    mk_data["$ppath"] = project_info.info("PROJECT_SRC_PATH_FROM_MAKEFILE")
+    mk_data["$cpuclockfreq"] = project_info.info("SELECTED_FREQ")
+    cpu_mk_parameters = []
+    for key, value in project_info.info("CPU_INFOS").items():
+        if key.startswith(const.MK_PARAM_ID):
+            cpu_mk_parameters.append("%s = %s" %(key.replace("MK", mk_data["$pname"]), value))
+    mk_data["$cpuparameters"] = "\n".join(cpu_mk_parameters)
+    mk_data["$csrc"], mk_data["$pcsrc"], mk_data["$cppasrc"], mk_data["$cxxsrc"], mk_data["$asrc"], mk_data["$constants"] = csrcGenerator(project_info)
+    mk_data["$prefix"] = replaceSeparators(project_info.info("TOOLCHAIN")["path"].rsplit("gcc", 1)[0])
+    mk_data["$suffix"] = replaceSeparators(project_info.info("TOOLCHAIN")["path"].rsplit("gcc", 1)[1])
+    mk_data["$hwpath"] = project_info.info("PROJECT_HW_PATH_FROM_MAKEFILE")
+    for key in mk_data:
+        makefile = makefile.replace(key, mk_data[key])
+    open(destination, "w").write(makefile)
+
+def makefileGenerator(project_info):
     """
     Generate the Makefile for the current project.
     """
-    # TODO: write a general function that works for both the mk file and the Makefile
-    while makefile.find("project_name") != -1:
-        makefile = makefile.replace("project_name", os.path.basename(projectInfos.info("PROJECT_PATH")))
-    return makefile
+    makefile = open(os.path.join(const.DATA_DIR, "mktemplates/Makefile"), "r").read()
+    destination = os.path.join(project_info.maindir, "Makefile")
+    # TODO write a general function that works for both the mk file and the Makefile
+    mk_data = {}
+    mk_data["$pname"] = project_info.info("PROJECT_NAME")
+    mk_data["$ppath"] = project_info.info("PROJECT_SRC_PATH_FROM_MAKEFILE")
+    for key in mk_data:
+        makefile = makefile.replace(key, mk_data[key])
+    open(destination, "w").write(makefile)
+
+def csrcGenerator(project_info):
+    modules = project_info.info("MODULES")
+    files = project_info.info("FILES")
+    if "harvard" in project_info.info("CPU_INFOS")["CPU_TAGS"]:
+        harvard = True
+    else:
+        harvard = False
+    # file to be included in CSRC variable
+    csrc = []
+    # file to be included in PCSRC variable
+    pcsrc = []
+    # files to be included in CPPASRC variable
+    cppasrc = []
+    # files to be included in CXXSRC variable
+    cxxsrc = []
+    # files to be included in ASRC variable
+    asrc = []
+    # constants to be included at the beginning of the makefile
+    constants = {}
+    for module, information in modules.items():
+        module_files = set([])
+        dependency_files = set([])
+        # assembly sources
+        asm_files = set([])
+        hwdir = os.path.basename(project_info.info("PROJECT_PATH")) + "/hw"
+        if information["enabled"]:
+            if "constants" in information:
+                constants.update(information["constants"])
+            cfiles, sfiles = findModuleFiles(module, project_info)
+            module_files |= set(cfiles)
+            asm_files |= set(sfiles)
+            for file in information["hw"]:
+                if file.endswith(".c"):
+                    module_files |= set([hwdir + "/" + os.path.basename(file)])
+            for file_dependency in information["depends"] + tuple(files.keys()):
+                    dependencyCFiles, dependencySFiles = findModuleFiles(file_dependency, project_info)
+                    dependency_files |= set(dependencyCFiles)
+                    asm_files |= set(dependencySFiles)
+            for file in module_files:
+                if not harvard or information.get("harvard", "both") == "both":
+                    csrc.append(file)
+                if harvard and "harvard" in information:
+                    pcsrc.append(file)
+            for file in dependency_files:
+                csrc.append(file)
+            for file in project_info.info("CPU_INFOS")["C_SRC"]:
+                csrc.append(file)
+            for file in project_info.info("CPU_INFOS")["PC_SRC"]:
+                pcsrc.append(file)
+            for file in asm_files:
+                cppasrc.append(file)
+    for file in project_info.info("CPU_INFOS")["CPPA_SRC"]:
+        cppasrc.append(file)
+    for file in project_info.info("CPU_INFOS")["CXX_SRC"]:
+        cxxsrc.append(file)
+    for file in project_info.info("CPU_INFOS")["ASRC"]:
+        asrc.append(file)
+    csrc = sorted(set(csrc))
+    csrc = " \\\n\t".join(csrc) + " \\"
+    pcsrc = sorted(set(pcsrc))
+    pcsrc = " \\\n\t".join(pcsrc) + " \\"
+    cppasrc = sorted(set(cppasrc))
+    cppasrc = " \\\n\t".join(cppasrc) + " \\"
+    cxxsrc = sorted(set(cxxsrc))
+    cxxsrc = " \\\n\t".join(cxxsrc) + " \\"
+    asrc = sorted(set(asrc))
+    asrc = " \\\n\t".join(asrc) + " \\"
+    constants = "\n".join([project_info.info("PROJECT_NAME") + "_" + key + " = " + unicode(value) for key, value in constants.items()])
+    return csrc, pcsrc, cppasrc, cxxsrc, asrc, constants
+
+def findModuleFiles(module, project_info):
+    # Find the files related to the selected module
+    cfiles = []
+    sfiles = []
+    # .c files related to the module and the cpu architecture
+    for filename, path in project_info.searchFiles(module + ".c"):
+        path = path.replace(project_info.info("BERTOS_PATH") + os.sep, "")
+        path = replaceSeparators(path)
+        cfiles.append(path + "/" + filename)
+    # .s files related to the module and the cpu architecture
+    for filename, path in project_info.searchFiles(module + ".s") + \
+            project_info.searchFiles(module + ".S"):
+        path = path.replace(project_info.info("BERTOS_PATH") + os.sep, "")
+        path = replaceSeparators(path)
+        sfiles.append(path + "/" + filename)
+    # .c and .s files related to the module and the cpu tags
+    tags = project_info.info("CPU_INFOS")["CPU_TAGS"]
+
+    # Awful, but secure check for version
+    # TODO: split me in a method/function
+    try:
+        version_string = bertosVersion(project_info.info("BERTOS_PATH"))
+        version_list = [int(i) for i in version_string.split()[1].split('.')]
+    except ValueError:
+        # If the version file hasn't a valid version number assume it's an older
+        # project.
+        version_list = [0, 0]
+    if version_list < [2, 5]:
+        # For older versions of BeRTOS add the toolchain to the tags
+        tags.append(project_info.info("CPU_INFOS")["TOOLCHAIN"])
+
+    for tag in tags:
+        for filename, path in project_info.searchFiles(module + "_" + tag + ".c"):
+            path = path.replace(project_info.info("BERTOS_PATH") + os.sep, "")
+            if os.sep != "/":
+                path = replaceSeparators(path)
+            cfiles.append(path + "/" + filename)
+        for filename, path in project_info.searchFiles(module + "_" + tag + ".s") + \
+                project_info.searchFiles(module + "_" + tag + ".S"):
+            path = path.replace(project_info.info("BERTOS_PATH") + os.sep, "")
+            path = replaceSeparators(path)
+            sfiles.append(path + "/" + filename)
+    return cfiles, sfiles
+
+def replaceSeparators(path):
+    """
+    Replace the separators in the given path with unix standard separator.
+    """
+    if os.sep != "/":
+        path = path.replace(os.sep, "/")
+    return path
 
 def getSystemPath():
     path = os.environ["PATH"]
-    if os.name == "nt":
-        path = path.split(";")
-    else:
-        path = path.split(":")
+    path = path.split(os.pathsep)
     return path
 
-def findToolchains(pathList):
+def findInPath(file, path=None):
+    if path is None:
+        path = os.environ.get('PATH', '')
+    if type(path) is type(''):
+        path = path.split(os.pathsep)
+    return filter(os.path.exists, map(lambda dir, file=file: os.path.join(dir, file), path))
+
+
+def findToolchains(path_list):
     toolchains = []
-    for element in pathList:
+    for element in path_list:
         for toolchain in glob.glob(element+ "/" + const.GCC_NAME):
-            if not os.path.islink(toolchain):
+            if not os.path.isdir(toolchain):
                 toolchains.append(toolchain)
     return list(set(toolchains))
 
@@ -130,18 +385,18 @@ def getToolchainInfo(output):
         info["thread"] = thread[0].split("Thread model: ")[1]
     return info
 
-def findDefinitions(ftype, path):
-    L = os.walk(path)
-    for element in L:
-        for filename in element[2]:
-            if fnmatch.fnmatch(filename, ftype):
-                yield (filename, element[0])
+def getToolchainName(toolchain_info):
+    name = "GCC " + toolchain_info["version"] + " - " + toolchain_info["target"].strip()
+    return name
+
+def getTagSet(cpu_info):
+    tag_set = set([])
+    for cpu in cpu_info:
+        tag_set |= set([cpu["CPU_NAME"]])
+        tag_set |= set(cpu["CPU_TAGS"])
+        tag_set |= set([cpu["TOOLCHAIN"]])
+    return tag_set
 
-def loadCpuInfos(path):
-    cpuInfos = []
-    for definition in findDefinitions(const.CPU_DEFINITION, path):
-        cpuInfos.append(getInfos(definition))
-    return cpuInfos
 
 def getInfos(definition):
     D = {}
@@ -155,16 +410,146 @@ def getInfos(definition):
     del D["include"]
     return D
 
+def getCommentList(string):
+    comment_list = re.findall(r"/\*{2}\s*([^*]*\*(?:[^/*][^*]*\*+)*)/", string)
+    comment_list = [re.findall(r"^\s*\* *(.*?)$", comment, re.MULTILINE) for comment in comment_list]
+    return comment_list
+
+def loadModuleDefinition(first_comment):
+    to_be_parsed = False
+    module_definition = {}
+    for num, line in enumerate(first_comment):
+        index = line.find("$WIZ$")
+        if index != -1:
+            to_be_parsed = True
+            try:
+                exec line[index + len("$WIZ$ "):] in {}, module_definition
+            except:
+                raise ParseError(num, line[index:])
+        elif line.find("\\brief") != -1:
+            module_definition["module_description"] = line[line.find("\\brief") + len("\\brief "):]
+    module_dict = {}
+    if "module_name" in module_definition:
+        module_name = module_definition[const.MODULE_DEFINITION["module_name"]]
+        del module_definition[const.MODULE_DEFINITION["module_name"]]
+        module_dict[module_name] = {}
+        if const.MODULE_DEFINITION["module_depends"] in module_definition:
+            depends = module_definition[const.MODULE_DEFINITION["module_depends"]]
+            del module_definition[const.MODULE_DEFINITION["module_depends"]]
+            if type(depends) == str:
+                depends = (depends,)
+            module_dict[module_name]["depends"] = depends
+        else:
+            module_dict[module_name]["depends"] = ()
+        if const.MODULE_DEFINITION["module_configuration"] in module_definition:
+            module_dict[module_name]["configuration"] = module_definition[const.MODULE_DEFINITION["module_configuration"]]
+            del module_definition[const.MODULE_DEFINITION["module_configuration"]]
+        else:
+            module_dict[module_name]["configuration"] = ""
+        if "module_description" in module_definition:
+            module_dict[module_name]["description"] = module_definition["module_description"]
+            del module_definition["module_description"]
+        if const.MODULE_DEFINITION["module_harvard"] in module_definition:
+            harvard = module_definition[const.MODULE_DEFINITION["module_harvard"]]
+            module_dict[module_name]["harvard"] = harvard
+            del module_definition[const.MODULE_DEFINITION["module_harvard"]]
+        if const.MODULE_DEFINITION["module_hw"] in module_definition:
+            hw = module_definition[const.MODULE_DEFINITION["module_hw"]]
+            del module_definition[const.MODULE_DEFINITION["module_hw"]]
+            if type(hw) == str:
+                hw = (hw, )
+            module_dict[module_name]["hw"] = hw
+        else:
+            module_dict[module_name]["hw"] = ()
+        if const.MODULE_DEFINITION["module_supports"] in module_definition:
+            supports = module_definition[const.MODULE_DEFINITION["module_supports"]]
+            del module_definition[const.MODULE_DEFINITION["module_supports"]]
+            module_dict[module_name]["supports"] = supports
+        module_dict[module_name]["constants"] = module_definition
+        module_dict[module_name]["enabled"] = False
+    return to_be_parsed, module_dict
+
+def isSupported(project, module=None, property_id=None):
+    if not module and property_id:
+        item = project.info("CONFIGURATIONS")[property_id[0]][property_id[1]]["informations"]
+    else:
+        item = project.info("MODULES")[module]
+    tag_dict = project.info("ALL_CPU_TAGS")
+    if "supports" in item:
+        support_string = item["supports"]
+        supported = {}
+        try:
+            exec "supported = " + support_string in tag_dict, supported
+        except:
+            raise SupportedException(support_string)
+        return supported["supported"]
+    else:
+        return True
+
+def loadDefineLists(comment_list):
+    define_list = {}
+    for comment in comment_list:
+        for num, line in enumerate(comment):
+            index = line.find("$WIZ$")
+            if index != -1:
+                try:
+                    exec line[index + len("$WIZ$ "):] in {}, define_list
+                except:
+                    raise ParseError(num, line[index:])
+    for key, value in define_list.items():
+        if type(value) == str:
+            define_list[key] = (value,)
+    return define_list
+
+def getDescriptionInformations(comment):
+    """
+    Take the doxygen comment and strip the wizard informations, returning the tuple
+    (comment, wizard_information)
+    """
+    brief = ""
+    description = ""
+    information = {}
+    for num, line in enumerate(comment):
+        index = line.find("$WIZ$")
+        if index != -1:
+            if len(brief) == 0:
+                brief += line[:index].strip()
+            else:
+                description += " " + line[:index]
+            try:
+                exec line[index + len("$WIZ$ "):] in {}, information
+            except:
+                raise ParseError(num, line[index:])
+        else:
+            if len(brief) == 0:
+                brief += line.strip()
+            else:
+                description += " " + line
+                description = description.strip()
+    return brief.strip(), description.strip(), information
+
 def getDefinitionBlocks(text):
     """
     Take a text and return a list of tuple (description, name-value).
     """
     block = []
-    block_tmp = re.findall(r"/\*{2}\s*([^*]*\*(?:[^/*][^*]*\*+)*)/\s*#define\s+((?:[^/]*?/?)+)\s*?(?:/{2,3}[^<].*?)?$", text, re.MULTILINE)
-    for comment, define in block_tmp:
-        block.append((" ".join(re.findall(r"^\s*\*?\s*(.*?)\s*?(?:/{2}.*?)?$", comment, re.MULTILINE)).strip(), define))
-    block += re.findall(r"/{3}\s*([^<].*?)\s*#define\s+((?:[^/]*?/?)+)\s*?(?:/{2,3}[^<].*?)?$", text, re.MULTILINE)
-    block += [(comment, define) for define, comment in re.findall(r"#define\s*(.*?)\s*/{3}<\s*(.+?)\s*?(?:/{2,3}[^<].*?)?$", text, re.MULTILINE)]
+    block_tmp = re.finditer(r"/\*{2}\s*([^*]*\*(?:[^/*][^*]*\*+)*)/\s*#define\s+((?:[^/]*?/?)+)\s*?(?:/{2,3}[^<].*?)?$", text, re.MULTILINE)
+    for match in block_tmp:
+        # Only the first element is needed
+        comment = match.group(1)
+        define = match.group(2)
+        start = match.start()
+        block.append(([re.findall(r"^\s*\* *(.*?)$", line, re.MULTILINE)[0] for line in comment.splitlines()], define, start))
+    for match in re.finditer(r"/{3}\s*([^<].*?)\s*#define\s+((?:[^/]*?/?)+)\s*?(?:/{2,3}[^<].*?)?$", text, re.MULTILINE):
+        comment = match.group(1)
+        define = match.group(2)
+        start = match.start()
+        block.append(([comment], define, start))
+    for match in re.finditer(r"#define\s*(.*?)\s*/{3}<\s*(.+?)\s*?(?:/{2,3}[^<].*?)?$", text, re.MULTILINE):
+        comment = match.group(2)
+        define = match.group(1)
+        start = match.start()
+        block.append(([comment], define, start))
     return block
 
 def formatParamNameValue(text):
@@ -175,19 +560,6 @@ def formatParamNameValue(text):
     block = re.findall("\s*([^\s]+)\s*(.+?)\s*$", text, re.MULTILINE)
     return block[0]
 
-def getDescriptionInformations(text): 
-    """ 
-    Take the doxygen comment and strip the wizard informations, returning the tuple 
-    (comment, wizard_informations) 
-    """ 
-    index = text.find("$WIZARD") 
-    if index != -1: 
-        exec(text[index + 1:]) 
-        informations = WIZARD 
-        return text[:index].strip(), informations
-    else:
-        return text.strip(), {}
-
 def loadConfigurationInfos(path):
     """
     Return the module configurations found in the given file as a dict with the
@@ -199,79 +571,104 @@ def loadConfigurationInfos(path):
             "min": the minimum value for integer parameters
             "max": the maximum value for integer parameters
             "long": boolean indicating if the num is a long
+            "unsigned": boolean indicating if the num is an unsigned
             "value_list": the name of the enum for enum parameters
+            "conditional_deps": the list of conditional dependencies for boolean parameters
     """
-    configurationInfos = {}
-    for comment, define in getDefinitionBlocks(open(path, "r").read()):
+    configuration_infos = {}
+    configuration_infos["paramlist"] = []
+    for comment, define, start in getDefinitionBlocks(open(path, "r").read()):
         name, value = formatParamNameValue(define)
-        description, informations = getDescriptionInformations(comment)
-        configurationInfos[name] = {}
-        configurationInfos[name]["value"] = value
-        configurationInfos[name]["informations"] = informations
-        configurationInfos[name]["description"] = description
-    return configurationInfos
-
-def loadModuleInfos(path):
+        brief, description, informations = getDescriptionInformations(comment)
+        configuration_infos["paramlist"].append((start, name))
+        configuration_infos[name] = {}
+        configuration_infos[name]["value"] = value
+        configuration_infos[name]["informations"] = informations
+        if not "type" in configuration_infos[name]["informations"]:
+            configuration_infos[name]["informations"]["type"] = findParameterType(configuration_infos[name])
+        if ("type" in configuration_infos[name]["informations"] and
+                configuration_infos[name]["informations"]["type"] == "int" and
+                configuration_infos[name]["value"].find("L") != -1):
+            configuration_infos[name]["informations"]["long"] = True
+            configuration_infos[name]["value"] = configuration_infos[name]["value"].replace("L", "")
+        if ("type" in configuration_infos[name]["informations"] and
+                configuration_infos[name]["informations"]["type"] == "int" and
+                configuration_infos[name]["value"].find("U") != -1):
+            configuration_infos[name]["informations"]["unsigned"] = True
+            configuration_infos[name]["value"] = configuration_infos[name]["value"].replace("U", "")
+        if "conditional_deps" in configuration_infos[name]["informations"]:
+            if (type(configuration_infos[name]["informations"]["conditional_deps"]) == str or
+                    type(configuration_infos[name]["informations"]["conditional_deps"]) == unicode):
+                configuration_infos[name]["informations"]["conditional_deps"] = (configuration_infos[name]["informations"]["conditional_deps"], )
+            elif type(configuration_infos[name]["informations"]["conditional_deps"]) == tuple:
+                pass
+            else:
+                configuration_infos[name]["informations"]["conditional_deps"] = ()
+        configuration_infos[name]["description"] = description
+        configuration_infos[name]["brief"] = brief
+    return configuration_infos
+
+def updateConfigurationValues(def_conf, user_conf):
+    for param in def_conf["paramlist"]:
+        if param[1] in user_conf and "value" in user_conf[param[1]]:
+            def_conf[param[1]]["value"] = user_conf[param[1]]["value"]
+    return def_conf
+
+def findParameterType(parameter):
+    if "value_list" in parameter["informations"]:
+        return "enum"
+    if "min" in parameter["informations"] or "max" in parameter["informations"] or re.match(r"^\d+U?L?$", parameter["value"]) != None:
+        return "int"
+
+def sub(string, parameter, value):
     """
-    Return the module infos found in the given file as a dict with the module
-    name as key and a dict containig the fields above as value or an empty dict
-    if the given file is not a BeRTOS module:
-        "depends": a list of modules needed by this module
-        "configuration": the cfg_*.h with the module configurations
-        "description": a string containing the brief description of doxygen
-        "enabled": contains False but the wizard will change if the user select
-        the module
+    Substitute the given value at the given parameter define in the given string
     """
-    moduleInfos = {}
-    string = open(path, "r").read()
-    commentList = re.findall(r"/\*{2}\s*([^*]*\*(?:[^/*][^*]*\*+)*)/", string)
-    commentList = [" ".join(re.findall(r"^\s*\*?\s*(.*?)\s*?(?:/{2}.*?)?$", comment, re.MULTILINE)).strip() for comment in commentList]
-    for comment in commentList:
-        index = comment.find("$WIZARD_MODULE")
-        if index != -1:
-            exec(comment[index + 1:])
-            moduleInfos[WIZARD_MODULE["name"]] = {"depends": WIZARD_MODULE["depends"],
-                                                    "configuration": WIZARD_MODULE["configuration"],
-                                                    "description": "",
-                                                    "enabled": False}
-            return moduleInfos
-    return {}
-
-def loadModuleInfosDict(path):
+    return re.sub(r"(?P<define>#define\s+" + parameter + r"\s+)([^\s]+)", r"\g<define>" + value, string)
+
+def isInt(informations):
     """
-    Return the dict containig all the modules
+    Return True if the value is a simple int.
     """
-    moduleInfosDict = {}
-    for filename, path in findDefinitions("*.h", path):
-        moduleInfosDict.update(loadModuleInfos(path + "/" + filename))
-    return moduleInfosDict
+    if ("long" not in informatios or not informations["long"]) and ("unsigned" not in informations or informations["unsigned"]):
+        return True
+    else:
+        return False
 
-def loadDefineLists(path):
+def isLong(informations):
     """
-    Return a dict with the name of the list as key and a list of string as value
+    Return True if the value is a long.
     """
-    string = open(path, "r").read()
-    commentList = re.findall(r"/\*{2}\s*([^*]*\*(?:[^/*][^*]*\*+)*)/", string)
-    commentList = [" ".join(re.findall(r"^\s*\*?\s*(.*?)\s*?(?:/{2}.*?)?$", comment, re.MULTILINE)).strip() for comment in commentList]
-    listDict = {}
-    for comment in commentList:
-        index = comment.find("$WIZARD_LIST")
-        if index != -1:
-            exec(comment[index + 1:])
-            listDict.update(WIZARD_LIST)
-    return listDict
+    if "long" in informations and informations["long"] and "unsigned" not in informations:
+        return True
+    else:
+        return False
 
-def loadDefineListsDict(path):
+def isUnsigned(informations):
     """
-    Return the dict containing all the define lists
+    Return True if the value is an unsigned.
     """
-    defineListsDict = {}
-    for filename, path in findDefinitions("*.h", path):
-        defineListsDict.update(loadDefineLists(path + "/" + filename))
-    return defineListsDict
+    if "unsigned" in informations and informations["unsigned"] and "long" not in informations:
+        return True
+    else:
+        return False
 
-def sub(string, parameter, value):
+def isUnsignedLong(informations):
     """
-    Substitute the given value at the given parameter define in the given string
+    Return True if the value is an unsigned long.
     """
-    return re.sub(r"(?P<define>#define\s+" + parameter + r"\s+)([^\s]+)", r"\g<define>" + value, string)
+    if "unsigned" in informations and "long" in informations and informations["unsigned"] and informations["long"]:
+        return True
+    else:
+        return False
+
+class ParseError(Exception):
+    def __init__(self, line_number, line):
+        Exception.__init__(self)
+        self.line_number = line_number
+        self.line = line
+
+class SupportedException(Exception):
+    def __init__(self, support_string):
+        Exception.__init__(self)
+        self.support_string = support_string