JobSpy/src/jobspy/jobs/__init__.py

from typing import Optional
from datetime import date
from enum import Enum
from pydantic import BaseModel


class JobType(Enum):
    FULL_TIME = (
        "fulltime",
        "períodointegral",
        "estágio/trainee",
        "cunormăîntreagă",
        "tiempocompleto",
        "vollzeit",
        "voltijds",
        "tempointegral",
        "全职",
        "plnýúvazek",
        "fuldtid",
        "دوامكامل",
        "kokopäivätyö",
        "tempsplein",
        "vollzeit",
        "πλήρηςαπασχόληση",
        "teljesmunkaidő",
        "tempopieno",
        "tempsplein",
        "heltid",
        "jornadacompleta",
        "pełnyetat",
        "정규직",
        "100%",
        "全職",
        "งานประจำ",
        "tamzamanlı",
        "повназайнятість",
        "toànthờigian",
    )
    PART_TIME = ("parttime", "teilzeit", "částečnýúvazek", "deltid")
    CONTRACT = ("contract", "contractor")
    TEMPORARY = ("temporary",)
    INTERNSHIP = (
        "internship",
        "prácticas",
        "ojt(onthejobtraining)",
        "praktikum",
        "praktik",
    )

    PER_DIEM = ("perdiem",)
    NIGHTS = ("nights",)
    OTHER = ("other",)
    SUMMER = ("summer",)
    VOLUNTEER = ("volunteer",)


class Country(Enum):
    """
    Gets the subdomain for Indeed and Glassdoor.
    The second item in the tuple is the subdomain for Indeed
    The third item in the tuple is the subdomain (and tld if there's a ':' separator) for Glassdoor
    """

    ARGENTINA = ("argentina", "ar", "com.ar")
    AUSTRALIA = ("australia", "au", "com.au")
    AUSTRIA = ("austria", "at", "at")
    BAHRAIN = ("bahrain", "bh")
    BELGIUM = ("belgium", "be", "fr:be")
    BRAZIL = ("brazil", "br", "com.br")
    CANADA = ("canada", "ca", "ca")
    CHILE = ("chile", "cl")
    CHINA = ("china", "cn")
    COLOMBIA = ("colombia", "co")
    COSTARICA = ("costa rica", "cr")
    CZECHREPUBLIC = ("czech republic,czechia", "cz")
    DENMARK = ("denmark", "dk")
    ECUADOR = ("ecuador", "ec")
    EGYPT = ("egypt", "eg")
    FINLAND = ("finland", "fi")
    FRANCE = ("france", "fr", "fr")
    GERMANY = ("germany", "de", "de")
    GREECE = ("greece", "gr")
    HONGKONG = ("hong kong", "hk", "com.hk")
    HUNGARY = ("hungary", "hu")
    INDIA = ("india", "in", "co.in")
    INDONESIA = ("indonesia", "id")
    IRELAND = ("ireland", "ie", "ie")
    ISRAEL = ("israel", "il")
    ITALY = ("italy", "it", "it")
    JAPAN = ("japan", "jp")
    KUWAIT = ("kuwait", "kw")
    LUXEMBOURG = ("luxembourg", "lu")
    MALAYSIA = ("malaysia", "malaysia")
    MEXICO = ("mexico", "mx", "com.mx")
    MOROCCO = ("morocco", "ma")
    NETHERLANDS = ("netherlands", "nl", "nl")
    NEWZEALAND = ("new zealand", "nz", "co.nz")
    NIGERIA = ("nigeria", "ng")
    NORWAY = ("norway", "no")
    OMAN = ("oman", "om")
    PAKISTAN = ("pakistan", "pk")
    PANAMA = ("panama", "pa")
    PERU = ("peru", "pe")
    PHILIPPINES = ("philippines", "ph")
    POLAND = ("poland", "pl")
    PORTUGAL = ("portugal", "pt")
    QATAR = ("qatar", "qa")
    ROMANIA = ("romania", "ro")
    SAUDIARABIA = ("saudi arabia", "sa")
    SINGAPORE = ("singapore", "sg", "sg")
    SOUTHAFRICA = ("south africa", "za")
    SOUTHKOREA = ("south korea", "kr")
    SPAIN = ("spain", "es", "es")
    SWEDEN = ("sweden", "se")
    SWITZERLAND = ("switzerland", "ch", "de:ch")
    TAIWAN = ("taiwan", "tw")
    THAILAND = ("thailand", "th")
    TURKEY = ("turkey", "tr")
    UKRAINE = ("ukraine", "ua")
    UNITEDARABEMIRATES = ("united arab emirates", "ae")
    UK = ("uk,united kingdom", "uk", "co.uk")
    USA = ("usa,us,united states", "www", "com")
    URUGUAY = ("uruguay", "uy")
    VENEZUELA = ("venezuela", "ve")
    VIETNAM = ("vietnam", "vn")

    # internal for ziprecruiter
    US_CANADA = ("usa/ca", "www")

    # internal for linkedin
    WORLDWIDE = ("worldwide", "www")

    @property
    def indeed_domain_value(self):
        return self.value[1]

    @property
    def glassdoor_domain_value(self):
        if len(self.value) == 3:
            subdomain, _, domain = self.value[2].partition(":")
            if subdomain and domain:
                return f"{subdomain}.glassdoor.{domain}"
            else:
                return f"www.glassdoor.{self.value[2]}"
        else:
            raise Exception(f"Glassdoor is not available for {self.name}")

    def get_url(self):
        return f"https://{self.glassdoor_domain_value}/"

    @classmethod
    def from_string(cls, country_str: str):
        """Convert a string to the corresponding Country enum."""
        country_str = country_str.strip().lower()
        for country in cls:
            country_names = country.value[0].split(',')
            if country_str in country_names:
                return country
        valid_countries = [country.value for country in cls]
        raise ValueError(
            f"Invalid country string: '{country_str}'. Valid countries are: {', '.join([country[0] for country in valid_countries])}"
        )


class Location(BaseModel):
    country: Country | None = None
    city: Optional[str] = None
    state: Optional[str] = None

    def display_location(self) -> str:
        location_parts = []
        if self.city:
            location_parts.append(self.city)
        if self.state:
            location_parts.append(self.state)
        if self.country and self.country not in (Country.US_CANADA, Country.WORLDWIDE):
            country_name = self.country.value[0]
            if "," in country_name:
                country_name = country_name.split(",")[0]
            if country_name in ("usa", "uk"):
                location_parts.append(country_name.upper())
            else:
                location_parts.append(country_name.title())
        return ", ".join(location_parts)


class CompensationInterval(Enum):
    YEARLY = "yearly"
    MONTHLY = "monthly"
    WEEKLY = "weekly"
    DAILY = "daily"
    HOURLY = "hourly"

    @classmethod
    def get_interval(cls, pay_period):
        return cls[pay_period].value if pay_period in cls.__members__ else None


class Compensation(BaseModel):
    interval: Optional[CompensationInterval] = None
    min_amount: int | None = None
    max_amount: int | None = None
    currency: Optional[str] = "USD"


class JobPost(BaseModel):
    title: str
    company_name: str
    job_url: str
    location: Optional[Location]

    description: str | None = None
    company_url: str | None = None

    job_type: list[JobType] | None = None
    compensation: Compensation | None = None
    date_posted: date | None = None
    benefits: str | None = None
    emails: list[str] | None = None
    num_urgent_words: int | None = None
    is_remote: bool | None = None
    # company_industry: str | None = None


class JobResponse(BaseModel):
    jobs: list[JobPost] = []