Python django.conf.settings 模块，DATA_DIR 实例源码

def handle(self, *args, **options):
        processed_data_path = os.path.join(
            settings.DATA_DIR, 'processed_berkeley_faculty.csv')
        fieldnames = ['first', 'last', 'department', 'year',
                      'title_category', 'title', 'gross_salary']
        with open(processed_data_path, 'w') as processed_data_file:
            writer = csv.DictWriter(
                processed_data_file, fieldnames=fieldnames)
            writer.writeheader()
            for person in Person.objects.exclude(
                    directory_record__department_obj=None):
                row = {}
                row['first'] = person.first
                row['last'] = person.last
                row['department'] = person.directory_record.department_obj\
                    .canonical
                for salaryrecord in person.salaryrecord_set.all():
                    row['year'] = salaryrecord.year
                    row['title_category'] = salaryrecord.title_category
                    row['title'] = salaryrecord.title
                    row['gross_salary'] = salaryrecord.gross
                    writer.writerow(row)

def remove_research_files(sender, instance, using, **kwargs):
    if instance.sealed:
        backup = Research()
        backup.researcher = instance.researcher
        backup.status = 5
        backup.sealed = True
        backup.start_time = instance.start_time
        backup.finish_time = instance.finish_time
        backup.script_name = instance.script_name
        backup.save()
        os.rename(
            instance.get_folder(), os.path.join(
                settings.DATA_DIR, "research", str(
                    backup.id)))
    else:
        try:
            rmtree(instance.get_folder())
        except BaseException:
            pass

def handle(self, *args, **kwargs):
        with open(join(settings.DATA_DIR, 'config.yaml'), 'w') as f:
            yaml.dump(settings.CONFIG, f, width=80, indent=4,
                      default_flow_style=False)

def build_titles(self):
        """
        Takes a CSV of titles that determines whether they correspond
        to a faculty position that we're interested in analyzing.
        """
        titles = {}
        titles_file_path = os.path.join(
            settings.DATA_DIR, 'directory', 'titles.csv')
        with open(titles_file_path, 'r') as titles_file:
            reader = csv.DictReader(titles_file)
            for row in reader:
                if row['keep'] == '1':
                    titles[row['db_title']] = row
        return titles

def handle(self, *args, **options):
        titles_dictionary = self.build_titles()
        clean_file_path = os.path.join(
            settings.DATA_DIR, 'berkeley_faculty.csv')
        with open(clean_file_path, 'w') as clean_file:
            writer = csv.DictWriter(
                clean_file, fieldnames=self.get_fieldnames())
            writer.writeheader()

            merged_file_path = os.path.join(settings.DATA_DIR, 'merged.csv')
            with open(merged_file_path, 'r') as raw_file:
                reader = csv.DictReader(raw_file)
                for row in reader:
                    # Only Berkeley records
                    if 'BERKELEY' not in row['location'].upper():
                        continue

                    # Only positions we care about
                    category = titles_dictionary.get(row['title'], None)
                    if category:
                        row['title_category'] = category['type']
                        row['title_qualifier'] = category['qualifier']
                        row['title_year_code'] = category['year_code']
                    else:
                        continue

                    writer.writerow(row)

def handle(self, *args, **options):
        SalaryRecord.objects.all().delete()
        clean_file_path = os.path.join(
            settings.DATA_DIR, 'berkeley_faculty.csv')

        fields = [field.name for field in SalaryRecord._meta.get_fields()
                  if field.name != 'id']
        mapping = {field: field for field in fields}

        c = CopyMapping(
            SalaryRecord,
            clean_file_path,
            mapping
        )
        c.save()

def build_cache(self):
        """
        Returns dict with {searched_name: row}
        """
        cache = {}
        cache_path = os.path.join(
            settings.DATA_DIR, 'directory', 'directory.csv')
        with open(cache_path, 'r') as cache_file:
            reader = csv.DictReader(cache_file)
            for row in reader:
                cache[row['searched_name']] = row
        return cache

def handle(self, *args, **options):
        print('Delete Existing Data.')

        Eat.objects.all().delete()
        Restaurant.objects.all().delete()
        CongressMember.objects.all().delete()

        print('Load Data.')

        geocodes_file_path = path.join(settings.DATA_DIR, 'geocodes.json')
        data_file_path = path.join(settings.DATA_DIR, 'data.json')

        geocodes = json.load(open(geocodes_file_path))
        data = json.load(open(data_file_path))

        print('Save Objects to Database.')

        for man_name, party, restaurant_name, address, kind, price, memo in data:
            if geocodes[address]:
                lat, lng = geocodes[address]
            else:
                lat, lng = None, None

            member, _ = CongressMember.objects.get_or_create(
                name=man_name,
                party=party,
            )

            restaurant, _ = Restaurant.objects.get_or_create(
                name=restaurant_name,
                address=address,
                lat=lat,
                lng=lng,
            )

            eat = Eat(
                member=member,
                restaurant=restaurant,
                price=price,
            )
            eat.save()

def get_folder(self):
        path = os.path.join(
            settings.DATA_DIR,
            "datasets",
            self.dataset.text_id,
            "models",
            self.text_id)
        if not os.path.exists(path):
            os.makedirs(path)
        return path

def get_visual_folder(self):
        path = os.path.join(
            settings.DATA_DIR,
            "datasets",
            self.dataset.text_id,
            "models",
            self.text_id,
            "visual")
        if not os.path.exists(path):
            os.makedirs(path)
        return path

def get_dist_folder(self):
        path = os.path.join(
            settings.DATA_DIR,
            "datasets",
            self.dataset.text_id,
            "models",
            self.text_id,
            "dist")
        if not os.path.exists(path):
            os.makedirs(path)
        return path

def get_folder(self):
        if self.type == "segmentation":
            path = os.path.join(
                settings.DATA_DIR,
                "datasets",
                self.dataset.text_id,
                "segmentation")
        else:
            path = os.path.join(settings.DATA_DIR, "assessment", str(self.id))

        if not os.path.exists(path):
            os.makedirs(path)
        return path

def read_log(self):
        try:
            log_file_name = os.path.join(
                settings.DATA_DIR, "datasets", self.text_id, "log.txt")
            with open(log_file_name, "r") as f:
                return f.read()
        except BaseException:
            return "Datased is reloading"

def get_folder(self):
        path = os.path.join(settings.DATA_DIR, "datasets", self.text_id)
        if not os.path.exists(path):
            os.makedirs(path)
        return path

def get_folder(self):
        path = os.path.join(settings.DATA_DIR, "research", str(self.id))
        if not os.path.exists(path):
            os.makedirs(path)
        return path

def get_pic_folder(self):
        path = os.path.join(settings.DATA_DIR, "research", str(self.id), "pic")
        if not os.path.exists(path):
            os.makedirs(path)
        return path

def fix_space_foreign_keys(apps, schema_editor):
    models = ('Area', 'Stair', 'LineObstacle', 'Obstacle')
    Space = apps.get_model('mapdata', 'Space')
    MAP_PATH = os.path.join(settings.DATA_DIR, 'map')
    PACKAGE_PATHS = [os.path.join(MAP_PATH, dirname) for dirname in os.listdir(MAP_PATH)]
    PACKAGE_PATHS = [path for path in PACKAGE_PATHS if os.path.isdir(path)]
    spaces = {}
    for space in Space.objects.all():
        spaces.setdefault(space.section.name + '_' + space.level, []).append(space)
    for model_name in models:
        model = apps.get_model('mapdata', model_name)
        orig_dir_names = [model._meta.default_related_name.lower()]
        if model.__name__ == 'Area':
            orig_dir_names = ['arealocations', 'stuffedareas']

        orig_objects = []
        for package_path in PACKAGE_PATHS:
            for orig_dir_name in orig_dir_names:
                dir_name = os.path.join(package_path, orig_dir_name)
                if not os.path.isdir(dir_name):
                    continue
                for filename in os.listdir(dir_name):
                    abs_filename = os.path.join(dir_name, filename)
                    if not filename.endswith('.json') or not os.path.isfile(abs_filename):
                        continue
                    obj = json.load(open(abs_filename))
                    obj['name'] = filename[:-5]
                    obj['geometry'] = shape(obj['geometry'])
                    orig_objects.append(obj)
        matches = {}
        for obj in model.objects.all().order_by('id' if hasattr(model, 'id') else 'locationslug_ptr_id'):
            for i, orig_obj in enumerate(orig_objects):
                if obj.geometry.almost_equals(orig_obj['geometry']):
                    matches.setdefault(i, []).append(obj)
                    break

        for orig_i, objects in matches.items():
            orig_obj = orig_objects[orig_i]

            if '-' in orig_obj['level']:
                splitted = orig_obj['level'].split('-')
                possible_spaces = spaces[splitted[0] + '_upper'] + spaces[splitted[1] + '_lower']
            else:
                possible_spaces = spaces[orig_obj['level'] + '_']

            possible_spaces = [space for space in possible_spaces
                               if space.geometry.intersects(orig_obj['geometry'])]
            if len(objects) == len(possible_spaces):
                pass  # nice
            elif len(objects) > len(possible_spaces):
                pass  # well, whatever
            else:
                def compare(space):
                    return space.geometry.intersection(orig_obj['geometry']).area

                possible_spaces.sort(key=compare, reverse=True)

            for i, obj in enumerate(objects[:len(possible_spaces)]):
                obj.space = possible_spaces[i]
                obj.save()

def handle(self, *args, **options):
        merged_file_path = os.path.join(settings.DATA_DIR, 'merged.csv')
        with open(merged_file_path, 'w') as merged_file:
            writer = csv.DictWriter(merged_file,
                                    fieldnames=self.get_fieldnames())
            writer.writeheader()

            for year in range(2006, 2016):
                print('Processing {}'.format(year))
                raw_file_path = os.path.join(settings.DATA_DIR,
                                             'salary',
                                             'salary_{}.csv'.format(year))
                with open(raw_file_path, 'r') as raw_file:
                    reader = csv.DictReader(raw_file)
                    for row in reader:
                        # First name and last name are distinct fields for
                        # 2013 onwards
                        if year > 2012:
                            # Don't want starred names
                            if '*' in row['last']:
                                continue
                        else:
                            # Don't want starred names
                            if '*' in row['name'] or '---' in row['name']:
                                continue
                            names = row['name'].split(',')
                            row['last'] = names[0]
                            row['first'] = ' '.join(names[1:])

                        row['year'] = str(year)
                        row = self.clean_row(row)

                        # Attempt to deal with middle names
                        first_names = [name.replace('.', '')
                                       for name in row['first'].split(' ')]
                        if len(first_names) > 1:
                            row['first'] = first_names[0]
                            row['middle'] = ' '.join(first_names[1:])

                        row['name'] = '{}, {}'.format(
                            row['last'],
                            row['first'])

                        writer.writerow(row)

def reload(self):
        self.prepare_log()
        self.log("Loading dataset " + self.text_id + "...")

        Term.objects.filter(dataset=self).delete()
        Document.objects.filter(dataset=self).delete()
        Modality.objects.filter(dataset=self).delete()
        from models.models import ArtmModel
        ArtmModel.objects.filter(dataset=self).delete()

        try:
            meta_file = os.path.join(self.get_folder(), "meta", "meta.json")
            with open(meta_file) as f:
                self.docs_info = json.load(f)
        except BaseException as ex:
            self.log("WARNING! Wasn't able to load meta.json")
            self.log(str(ex))
            self.time_provided = False
            self.docs_info = {}

        try:
            preprocessing_params = json.loads(self.preprocessing_params)
            self.log("Preprocessing params:" + str(preprocessing_params))
        except BaseException:
            preprocessing_params = {}
            self.log("Warning! Failed to load preprocessing parameters.")

        # Preprocessing
        custom_vocab = False
        if "parse" in preprocessing_params:
            self.preprocess_parse(preprocessing_params["parse"])
        if "filter" in preprocessing_params:
            self.preprocess_filter(preprocessing_params["filter"])
            custom_vocab = True
        if "custom_vocab" in preprocessing_params and preprocessing_params[
                "custom_vocab"]:
            self.log("Will use custom vocab.txt")
            custom_vocab = True

        self.create_batches()
        self.gather_dictionary(custom_vocab=custom_vocab)
        self.load_documents()

        self.log("Loaded " + str(self.documents_count) + " documents.")

        # Creating folder for models
        model_path = os.path.join(
            settings.DATA_DIR, "datasets", self.text_id, "models")
        if not os.path.exists(model_path):
            os.makedirs(model_path)

        self.log("Dataset " + self.text_id + " loaded.")
        self.creation_time = datetime.now()
        self.status = 0
        self.save()