def get(self, parameter: JobField, soup: BeautifulSoup) -> Any:
"""Get a single job attribute from a soup object by JobField
NOTE: priority is all the same.
"""
if parameter == JobField.KEY_ID:
# TODO: is there a way to combine these calls?
# NOTE: do not use 'data-m_impr_j_jobid' as this is duplicated
return soup.find('h2', attrs={
'class': 'title'
}).find('a').get('data-m_impr_j_postingid')
elif parameter == JobField.TITLE:
return soup.find('h2', attrs={'class': 'title'}).text.strip()
elif parameter == JobField.COMPANY:
return soup.find('div', attrs={'class': 'company'}).text.strip()
elif parameter == JobField.LOCATION:
return soup.find('div', attrs={'class': 'location'}).text.strip()
elif parameter == JobField.POST_DATE:
return calc_post_date_from_relative_str(
soup.find('time').text.strip())
elif parameter == JobField.URL:
# NOTE: seems that it is a bit hard to view these links? getting 503
return str(
soup.find('a', attrs={
'data-bypass': '******'
}).get('href'))
else:
raise NotImplementedError(f"Cannot get {parameter.name}")
def get(self, parameter: JobField, soup: BeautifulSoup) -> Any:
"""Get a single job attribute from a soup object by JobField
"""
if parameter == JobField.TITLE:
return soup.find(
'a', attrs={'data-tn-element': 'jobTitle'}
).text.strip()
elif parameter == JobField.COMPANY:
return soup.find('span', attrs={'class': 'company'}).text.strip()
elif parameter == JobField.LOCATION:
return soup.find('span', attrs={'class': 'location'}).text.strip()
elif parameter == JobField.TAGS:
# tags may not be on page and that's ok.
table_soup = soup.find(
'table', attrs={'class': 'jobCardShelfContainer'}
)
if table_soup:
return [
td.text.strip() for td in table_soup.find_all(
'td', attrs={'class': 'jobCardShelfItem'}
)
]
else:
return []
elif parameter == JobField.REMOTENESS:
remote_field = soup.find('span', attrs={'class': 'remote'})
if remote_field:
remoteness_str = remote_field.text.strip().lower()
if remoteness_str in REMOTENESS_STR_MAP:
return REMOTENESS_STR_MAP[remoteness_str]
return Remoteness.UNKNOWN
elif parameter == JobField.WAGE:
# We may not be able to obtain a wage
potential = soup.find('span', attrs={'class': 'salaryText'})
if potential:
return potential.text.strip()
else:
return ''
elif parameter == JobField.POST_DATE:
return calc_post_date_from_relative_str(
soup.find('span', attrs={'class': 'date'}).text.strip()
)
elif parameter == JobField.KEY_ID:
return ID_REGEX.findall(
str(
soup.find(
'a', attrs={'class': 'sl resultLink save-job-link'}
)
)
)[0]
else:
raise NotImplementedError(f"Cannot get {parameter.name}")
def get(self, parameter: JobField, soup: BeautifulSoup) -> Any:
"""Get a single job attribute from a soup object by JobField
TODO: impl div class=compactStars value somewhere.
"""
if parameter == JobField.TITLE:
# TODO: we should instead get what user sees in the <span>
return soup.get('data-normalize-job-title')
elif parameter == JobField.COMPANY:
return soup.find(
'div', attrs={'class', 'jobInfoItem jobEmpolyerName'}
).text.strip()
elif parameter == JobField.LOCATION:
return soup.get('data-job-loc')
# FIXME: impl.
# elif parameter == JobField.TAGS:
# labels = soup.find_all('div', attrs={'class', 'jobLabel'})
# if labels:
# return [
# l.text.strip() for l in labels if l.text.strip() != 'New'
# ]
# else:
# return []
# FIXME: impl JobField.REMOTE
elif parameter == JobField.POST_DATE:
return calc_post_date_from_relative_str(
soup.find(
'div', attrs={
'class': 'd-flex align-items-end pl-std css-mi55ob'
}
).text.strip()
)
elif parameter == JobField.WAGE:
# NOTE: most jobs don't have this so we wont raise a warning here
# and will fail silently instead
wage = soup.find('span', attrs={'class': 'gray salary'})
if wage is not None:
return wage.text.strip()
else:
return ''
elif parameter == JobField.KEY_ID:
return soup.get('data-id')
elif parameter == JobField.URL:
part_url = soup.find(
'div', attrs={'class', 'logoWrap'}
).find('a').get('href')
return (
f'https://www.glassdoor.{self.config.search_config.domain}'
f'{part_url}'
)
else:
raise NotImplementedError(f"Cannot get {parameter.name}")
