Formato

Introducción

La aplicación de formato describe cómo se debe ingerir un archivo de datos: qué columnas considerar, qué variables contienen, el formato de fecha y hora, etc. Un resumen de los modelos involucrados se puede ver en el siguiente diagrama:

Figure 1: Diagrama UML de los modelos de la aplicación de formato.

Componentes básicos

Extension

Extension of the data file.

It is mostly used to choose the tool to be employed to ingest the data. While it can take any value, there is currently explicit support only for xlsx and xlx. Anything else will be interpreted as a text file and loaded using pandas.read_csv.

Attributes:

Name	Type	Description
extension_id	AutoField	Primary key.
value	CharField	The extension value. eg. xlsx, xlx, txt.

Functions

__str__()

Return the string representation of the object.

Source code in formatting/models.py

def __str__(self) -> str:
    """Return the string representation of the object."""
    return str(self.value)

get_absolute_url()

Get the absolute URL of the object.

Source code in formatting/models.py

def get_absolute_url(self) -> str:
    """Get the absolute URL of the object."""
    return reverse("formatting:extension_detail", kwargs={"pk": self.pk})

Delimiter

Delimiter between columns in the data file.

One or more characters that separate columns in a text file. The most common values are ,, ;, and \t (tab).

Attributes:

Name	Type	Description
delimiter_id	AutoField	Primary key.
name	CharField	The name of the delimiter. eg. comma, semicolon, tab.
character	CharField	The character used as a delimiter. eg. ,, ;, \t.

Functions

__str__()

Return the string representation of the object.

Source code in formatting/models.py

def __str__(self) -> str:
    """Return the string representation of the object."""
    return str(self.name)

get_absolute_url()

Get the absolute URL of the object.

Source code in formatting/models.py

def get_absolute_url(self) -> str:
    """Get the absolute URL of the object."""
    return reverse("formatting:delimiter_detail", kwargs={"pk": self.pk})

Date

Date format.

Format string for the date column. It is used to parse the date column in the data file. The format string must be compatible with the datetime module in Python. See the datetime documentation for more information on valid format codes.

Attributes:

Name	Type	Description
date_id	AutoField	Primary key.
date_format	CharField	The format string for the date column in human readable form, eg. DD-MM-YYYY.
code	CharField	The code used to parse the date column, eg. %d-%m-%Y.

Functions

__str__()

Return the string representation of the object.

Source code in formatting/models.py

def __str__(self) -> str:
    """Return the string representation of the object."""
    return str(self.date_format)

get_absolute_url()

Get the absolute URL of the object.

Source code in formatting/models.py

def get_absolute_url(self) -> str:
    """Get the absolute URL of the object."""
    return reverse("formatting:date_detail", kwargs={"pk": self.pk})

Time

Time format.

Format string for the time column. It is used to parse the time column in the data file. The format string must be compatible with the datetime module in Python. See the datetime documentation for more information on valid format codes.

Attributes:

Name	Type	Description
date_id	AutoField	Primary key.
date_format	CharField	The format string for the date column in human readable form, eg. HH:MM:SS 24H.
code	CharField	The code used to parse the date column, eg. %H:%M:%S.

Functions

__str__()

Return the string representation of the object.

Source code in formatting/models.py

def __str__(self) -> str:
    """Return the string representation of the object."""
    return str(self.time_format)

get_absolute_url()

Get the absolute URL of the object.

Source code in formatting/models.py

def get_absolute_url(self) -> str:
    """Get the absolute URL of the object."""
    return reverse("formatting:time_detail", kwargs={"pk": self.pk})

Componentes principales

Format

Details of the data file format, describing how to read the file.

It combines several properties, such as the file extension, the delimiter, the date and time formats, and the column indices for the date and time columns, instructing how to read the data file and parse the dates. It is mostly used to ingest data from text files, like CSV. For Thingsboard imports, only the name, description and thingsboard fields are applicable.

Attributes:

Name	Type	Description
format_id	AutoField	Primary key.
name	CharField	Short name of the format entry.
description	TextField	Description of the format.
extension	ForeignKey	The extension of the data file.
delimiter	ForeignKey	The delimiter between columns in the data file. Only required for text files.
first_row	PositiveSmallIntegerField	Index of the first row with data, starting in 0.
footer_rows	PositiveSmallIntegerField	Number of footer rows to be ignored at the end.
date	ForeignKey	Format for the date column. Only required for text files.
date_column	PositiveSmallIntegerField	Index of the date column, starting in 0.
time	ForeignKey	Format for the time column. Only required for text files.
time_column	PositiveSmallIntegerField	Index of the time column, starting in 0.
thingsboard	BooleanField	Whether the data is being imported from Thingsboard.

Attributes

datetime_format property

Obtain the datetime format string.

Functions

__str__()

Return the string representation of the object.

Source code in formatting/models.py

def __str__(self) -> str:
    """Return the string representation of the object."""
    return str(self.name)

clean()

Validate the model instance.

Checks that the required fields for non-Thingsboard data are provided.

Source code in formatting/models.py

def clean(self) -> None:
    """Validate the model instance.

    Checks that the required fields for non-Thingsboard data are provided.
    """
    super().clean()
    errors = {}
    if not self.thingsboard:
        required_fields = (
            "extension",
            "first_row",
            "footer_rows",
            "date_column",
            "time_column",
        )
        for field in required_fields:
            if getattr(self, field) is None:
                errors[field] = "Field is required for non-Thingsboard data."

    if errors:
        raise ValidationError(errors)

datetime_columns(delimiter)

Column indices that correspond to the date and time columns in the dataset.

Parameters:

Name	Type	Description	Default
delimiter	str	The delimiter used to split the date and time codes.	required

Returns:

Type	Description
list[int]	list[int]: A list of column indices.

Source code in formatting/models.py

def datetime_columns(self, delimiter: str) -> list[int]:
    """Column indices that correspond to the date and time columns in the dataset.

    Args:
        delimiter (str): The delimiter used to split the date and time codes.

    Returns:
        list[int]: A list of column indices.
    """
    date_items = self.date.code.split(delimiter)
    date_cols = list(range(self.date_column, self.date_column + len(date_items)))
    time_items = self.time.code.split(delimiter)
    time_cols = list(range(self.time_column, self.time_column + len(time_items)))
    return date_cols + time_cols

get_absolute_url()

Get the absolute URL of the object.

Source code in formatting/models.py

def get_absolute_url(self) -> str:
    """Get the absolute URL of the object."""
    return reverse("formatting:format_detail", kwargs={"pk": self.pk})

Classification

Contains instructions on how to classify the data into a specific variable.

In particular, it links a format to a variable, and provides the column indices for the value, maximum, and minimum columns, as well as the validator columns. It also contains information on whether the data is accumulated, incremental, and the resolution of the data. For Thingsboard imports, only the format, variable, accumulate, resolution and incremental fields are applicable.

Attributes:

Name	Type	Description
cls_id	AutoField	Primary key.
format	ForeignKey	The format of the data file.
variable	ForeignKey	The variable to which the data belongs.
value	PositiveSmallIntegerField	Index of the value column, starting in 0.
maximum	PositiveSmallIntegerField	Index of the maximum value column, starting in 0.
minimum	PositiveSmallIntegerField	Index of the minimum value column, starting in 0.
value_validator_column	PositiveSmallIntegerField	Index of the value validator column, starting in 0.
value_validator_text	CharField	Value validator text.
maximum_validator_column	PositiveSmallIntegerField	Index of the maximum value validator column, starting in 0.
maximum_validator_text	CharField	Maximum value validator text.
minimum_validator_column	PositiveSmallIntegerField	Index of the minimum value validator column, starting in 0.
minimum_validator_text	CharField	Minimum value validator text.
accumulate	PositiveSmallIntegerField	If set to a number of minutes, the data will be accumulated over that period.
resolution	DecimalField	Resolution of the data. Only used if it is to be accumulated.
incremental	BooleanField	Whether the data is an incremental counter. If it is, any value below the previous one will be removed.
decimal_comma	BooleanField	Whether the data uses a comma as a decimal separator.

Functions

__str__()

Return the string representation of the object.

Source code in formatting/models.py

def __str__(self) -> str:
    """Return the string representation of the object."""
    return str(self.cls_id)

clean()

Validate the model instance.

It checks that the column indices are different, and that the accumulation period is greater than zero if it is set; the resolution is set if the data is accumulated; and that the value column is set if the import is not from Thingsboard.

Source code in formatting/models.py

def clean(self) -> None:
    """Validate the model instance.

    It checks that the column indices are different, and that the accumulation
    period is greater than zero if it is set; the resolution is set if the data is
    accumulated; and that the value column is set if the import is not from
    Thingsboard.
    """
    if self.accumulate and self.resolution is None:
        raise ValidationError(
            {"resolution": "The resolution must be set if the data is accumulated."}
        )

    col_names = [
        "value",
        "maximum",
        "minimum",
        "value_validator_column",
        "maximum_validator_column",
        "minimum_validator_column",
    ]
    unique = defaultdict(list)
    for name in col_names:
        if getattr(self, name) is not None:
            unique[getattr(self, name)].append(name)
    for _, names in unique.items():
        if len(names) != 1:
            msg = "The columns must be different."
            raise ValidationError({field: msg for field in names})

    # for non-Thingsboard classifications
    if not self.format.thingsboard and self.value is None:
        raise ValidationError(
            {
                "value": (
                    "A value column must be specified for non-Thingsboard formats."
                )
            }
        )

get_absolute_url()

Get the absolute URL of the object.

Source code in formatting/models.py

def get_absolute_url(self) -> str:
    """Get the absolute URL of the object."""
    return reverse("formatting:classification_detail", kwargs={"pk": self.pk})