Python inspector.year_range函数代码示例

def run(options):
  year_range = inspector.year_range(options, archive)

  # Pull the audit reports
  for year in year_range:
    if year < 2006:  # This is the oldest year for these reports
      year = 2006
    url = AUDIT_REPORTS_BASE_URL.format(year)
    doc = beautifulsoup_from_url(url)
    results = []
    results.extend(doc.select("tr.ms-rteTableOddRow-default"))
    results.extend(doc.select("tr.ms-rteTableEvenRow-default"))
    for result in results:
      report = audit_report_from(result, url, year_range)
      if report:
        inspector.save_report(report)

  for report_type, url in OTHER_URLS.items():
    doc = beautifulsoup_from_url(url)
    results = doc.select("#ctl00_PlaceHolderMain_ctl05_ctl01__ControlWrapper_RichHtmlField > p > a")
    for result in results:
      report = report_from(result, url, report_type, year_range)
      if report:
        inspector.save_report(report)

  doc = beautifulsoup_from_url(SEMIANNUAL_REPORTS_URL)
  results = doc.select("#ctl00_PlaceHolderMain_ctl05_ctl01__ControlWrapper_RichHtmlField > p > a")
  for result in results:
    report = semiannual_report_from(result, SEMIANNUAL_REPORTS_URL, year_range)
    if report:
      inspector.save_report(report)

def run(options):
  year_range = inspector.year_range(options, archive)

  component = options.get('component')
  if component:
    components = [component]
  else:
    components = list(COMPONENTS.keys())

  report_id = options.get('report_id')

  limit = int(options.get('limit', 0))

  all_audit_reports = {}

  for component in components:
    logging.info("## Fetching reports for component %s" % component)
    url = url_for(options, component)
    body = utils.download(url)

    doc = BeautifulSoup(body)

    results = doc.select("table.contentpaneopen table[border=1] tr")
    # accept only trs that look like body tr's (no 'align' attribute)
    #   note: HTML is very inconsistent. cannot rely on thead or tbody
    results = [x for x in results if x.get('align') is None]
    if not results:
      raise inspector.NoReportsFoundError("DHS (%s)" % component)

    count = 0
    for result in results:
      report = report_from(result, component, url)
      if not report:
        continue

      if report_id and (report_id != report['report_id']):
        continue

      if inspector.year_from(report) not in year_range:
        # logging.info("[%s] Skipping, not in requested range." % report['report_id'])
        continue

      key = (report["report_id"], report["title"])
      if key in all_audit_reports:
        all_audit_reports[key]["agency"] = all_audit_reports[key]["agency"] + \
                ", " + report["agency"]
        all_audit_reports[key]["agency_name"] = \
                all_audit_reports[key]["agency_name"] + ", " + \
                report["agency_name"]
      else:
        all_audit_reports[key] = report

      count += 1
      if limit and (count >= limit):
        break

    logging.info("## Fetched %i reports for component %s\n\n" % (count, component))

  for report in all_audit_reports.values():
    inspector.save_report(report)

def run(options):
  year_range = inspector.year_range(options, archive)

  doc = BeautifulSoup(utils.download(REPORTS_URL))

  # Pull the semiannual reports
  semiannul_results = doc.select("#AnnualManagementReports select")[0]
  for result in semiannul_results.select("option"):
    report = semiannual_report_from(result, year_range)
    if report:
      inspector.save_report(report)

  # Pull the special reports
  special_report_table = doc.find("table", attrs={"bordercolor": "#808080"})
  for index, result in enumerate(special_report_table.select("tr")):
    if not index:
      # Skip the header row
      continue
    report = report_from(result, REPORTS_URL, report_type='other', year_range=year_range)
    if report:
      inspector.save_report(report)

  # Pull the audit reports
  for year in year_range:
    if year < 2001:  # The oldest fiscal year page available
      continue
    year_url = AUDIT_REPORTS_URL.format(year=year)
    doc = BeautifulSoup(utils.download(year_url))
    for index, result in enumerate(doc.select("#main table tr")):
      if not index:
        # Skip the header row
        continue
      report = report_from(result, year_url, report_type='audit', year_range=year_range)
      if report:
        inspector.save_report(report)

def run(options):
  year_range = inspector.year_range(options, archive)
  if datetime.datetime.now().month >= 10:
    # October, November, and December fall into the next fiscal year
    # Add next year to year_range to compensate
    year_range.append(max(year_range) + 1)

  # Pull the audit reports
  for year in year_range:
    url = audit_report_url(year)
    if url:
      parse_result_from_js_url(url, "auditreports", year, year_range, report_type='audit')
    url = inspection_report_url(year)
    if url:
      parse_result_from_js_url(url, "iereports", year, year_range, report_type='inspection')

  # Pull the congressional testimony
  doc = utils.beautifulsoup_from_url(CONGRESSIONAL_TESTIMONY_REPORTS_URL)
  results = doc.findAll("ul", type='disc')[0].select("li")
  for result in results:
    report = congressional_testimony_report_from(result, year_range)
    if report:
      inspector.save_report(report)

  # Pull the semiannual reports
  doc = utils.beautifulsoup_from_url(SEMIANNUAL_REPORTS_URL)
  results = doc.findAll("ul", type='disc')[0].select("li")
  for result in results:
    report = semiannual_report_from(result, year_range)
    if report:
      inspector.save_report(report)

def run(options):
    year_range = inspector.year_range(options, archive)

    # Pull the audit reports
    for year in year_range:
        if year < 2002:  # The oldest page for audit reports
            continue
        doc = BeautifulSoup(utils.download(AUDIT_REPORTS_URL.format(year=year)))
        results = doc.select("div.content table tr")
        for index, result in enumerate(results):
            if not index:
                # Skip the header row
                continue
            report = report_from(result, report_type="audit", year_range=year_range)
            if report:
                inspector.save_report(report)

    # Pull the FOIA reports
    doc = BeautifulSoup(utils.download(FOIA_REPORTS_URL))
    results = doc.select("div.content table tr")
    for index, result in enumerate(results):
        if not index:
            # Skip the header row
            continue
        report = report_from(result, report_type="other", year_range=year_range)
        if report:
            inspector.save_report(report)

    # Pull the semiannual reports
    doc = BeautifulSoup(utils.download(SEMIANNUAL_REPORTS_URL))
    results = doc.select("div.content a")
    for result in results:
        report = semiannual_report_from(result, year_range)
        if report:
            inspector.save_report(report)

def run(options):
  year_range = inspector.year_range(options, archive)

  # Pull the audit reports
  for year in year_range:
    if year < 2005:  # This is the earliest audits go back
      continue
    url = AUDIT_REPORTS_URL.format(year=year)
    doc = BeautifulSoup(utils.download(url))
    results = doc.select("div.content")
    if not results:
      raise inspector.NoReportsFoundError("Tennessee Valley Authority (%d)" % year)
    for result in results:
      report = audit_report_from(result, url, year_range)
      if report:
        inspector.save_report(report)

  # Pull the semiannual reports
  doc = BeautifulSoup(utils.download(SEMIANNUAL_REPORTS_URL))
  results = doc.select("report")
  if not results:
    raise inspector.NoReportsFoundError("Tennessee Valley Authority (semiannual reports)")
  for result in results:
    report = semiannual_report_from(result, year_range)
    if report:
      inspector.save_report(report)

def run(options):
  year_range = inspector.year_range(options)
  pages = options.get('pages', ALL_PAGES)

  max_page = None
  for page in range(1, (int(pages) + 1)):
    if max_page and (page > max_page):
      print("End of pages!")
      break

    print("## Downloading page %i" % page)
    url = url_for(options, page)
    body = utils.download(url)
    doc = BeautifulSoup(body)
    max_page = last_page_for(doc)

    results = doc.select(".views-row")

    for result in results:
      report = report_from(result)

      # inefficient enforcement of --year arg, USPS doesn't support it server-side
      # TODO: change to published_on.year once it's a datetime
      if inspector.year_from(report) not in year_range:
        print("[%s] Skipping report, not in requested range." % report['report_id'])
        continue

      inspector.save_report(report)

def run(options):
    year_range = inspector.year_range(options, archive)

    # Find the number of pages to iterate
    doc = BeautifulSoup(utils.download(REPORTS_URL))
    page_count_text = doc.select("div.AspNet-GridView-Pagination")[0].text
    page_count = int(re.search("Page 1 of (\d+)", page_count_text).groups()[0])

    # Iterate over those pages
    for page in range(1, page_count + 1):
        response = utils.scraper.post(
            REPORTS_URL,
            data={
                "__EVENTTARGET": "ctl00$ctl00$MainContent$NavTreeSubContent$sv$GridViewSummary",
                "__EVENTARGUMENT": "Page${page_number}".format(page_number=page),
            },
            cookies=COOKIES,
        )
        doc = BeautifulSoup(response.content)
        results = doc.select("div.AspNet-GridView table tr")
        if not results:
            break
        for index, result in enumerate(results):
            if not index:
                # Skip the header row
                continue
            report = report_from(result, year_range)
            if report:
                inspector.save_report(report)

def run(options):
  year_range = inspector.year_range(options)

  # Pull the audit reports
  for year in year_range:
    url = audit_report_url(year)
    if url:
      parse_result_from_js_url(url, "auditreports", year, year_range)
    url = inspection_report_url(year)
    if url:
      parse_result_from_js_url(url, "iereports", year, year_range)

  # Pull the congressional testimony
  doc = BeautifulSoup(utils.download(CONGRESSIONAL_TESTIMONY_REPORTS_URL))
  results = doc.findAll("ul", type='disc')[0].select("li")
  for result in results:
    report = congressional_testimony_report_from(result, year_range)
    if report:
      inspector.save_report(report)

  # Pull the semiannual reports
  doc = BeautifulSoup(utils.download(SEMIANNUAL_REPORTS_URL))
  results = doc.findAll("ul", type='disc')[0].select("li")
  for result in results:
    report = semiannual_report_from(result, year_range)
    if report:
      inspector.save_report(report)

def run(options):
  year_range = inspector.year_range(options, archive)
  results_flag = False

  # Pull the audit reports
  for year in year_range:
    if year < 2002:  # The oldest page for audit reports
      continue
    if year == 2018:
      doc = utils.beautifulsoup_from_url(LATEST_AUDIT_REPORTS_URL)
    else:
      doc = utils.beautifulsoup_from_url(AUDIT_REPORTS_URL.format(year=year))

    if doc is None:
      # Next year's audit page may not be published yet
      continue

    results = doc.select("div.mainCenter table tr")
    if results:
      results_flag = True
    for index, result in enumerate(results):
      if not index:
        # Skip the header row
        continue
      report = report_from(result, report_type='audit', year_range=year_range)
      if report:
        inspector.save_report(report)

  if not results_flag:
    raise inspector.NoReportsFoundError("NCUA (audit reports)")

  # Pull the other reports
  doc = utils.beautifulsoup_from_url(OTHER_REPORTS_URL)
  results = doc.select("div.mainCenter p")
  if not results:
    raise inspector.NoReportsFoundError("NCUA (other)")
  for result in results:
    report = other_report_from(result, year_range=year_range)
    if report:
      inspector.save_report(report)

  # Pull the semiannual reports
  doc = utils.beautifulsoup_from_url(SEMIANNUAL_REPORTS_URL)
  results = doc.select("div#mainColumns div.mainCenter a")
  if not results:
    raise inspector.NoReportsFoundError("NCUA (semiannual reports)")
  for result in results:
    report = semiannual_report_from(result, year_range)
    if report:
      inspector.save_report(report)

  # Pull the performance and strategic plans
  doc = utils.beautifulsoup_from_url(PLANS_URL)
  results = doc.select("div.mainCenter p")
  if not results:
    raise inspector.NoReportsFoundError("NCUA (performance/strategic plans)")
  for result in results:
    report = plan_from(result, year_range)
    if report:
      inspector.save_report(report)

def run(options):
  year_range = inspector.year_range(options)
  only_id = options.get('report_id')

  print("## Downloading reports from %i to %i" % (year_range[0], year_range[-1]))

  url = url_for()
  body = utils.download(url)

  doc = BeautifulSoup(body)
  results = doc.select("section")

  for result in results:
    try:
      year = int(result.get("title"))
      # check that the fetched year is in the range
      if year not in year_range:
        continue
      print("## Downloading year %i " % year)
    except ValueError:
      continue

    # gets each table entry and sends generates a report from it
    listings = result.div.table.tbody.contents
    for item in listings:
      if type(item) is not bs4.element.Tag:
        continue
      report = report_from(item)

      # can limit it to just one report, for debugging convenience
      if only_id and only_id != report['report_id']:
        continue

      inspector.save_report(report)

def scrape_restricted_reports(options):
  """Restricted Products.

  A single HTML page lists unreleased reports since 2014, with no links."""

  # These reports are unreleased -- we could make this the text?
  """The following products have been determined to contain either
classified information or controlled unclassified information by the audited
agencies and cannot be publicly released.

Members of Congress or congressional staff who wish to obtain one or more of
these products should call or e-mail the Congressional Relations Office.
All others who wish to obtain one or more of these products should follow the
instructions found on Requesting Restricted Products."""

  REPORTS_URL = 'http://www.gao.gov/restricted/restricted_reports'
  archive = 2014

  year_range = inspector.year_range(options, archive)
  doc = utils.beautifulsoup_from_url(REPORTS_URL)
  results = doc.select("div.listing")
  for result in results:
    report = process_restricted_report(result, year_range, REPORTS_URL)
    if report:
      inspector.save_report(report)

def run(options):
  year_range = inspector.year_range(options, archive)

  # # Pull the RSS feed
  doc = BeautifulSoup(utils.download(RSS_URL))
  results = doc.select("item")
  for result in results:
    report = rss_report_from(result, year_range)
    if report:
      inspector.save_report(report)

  # # Pull the recent audit reports.
  doc = BeautifulSoup(utils.download(RECENT_AUDITS_URL))
  results = doc.select("div.block > a")
  for result in results:
    report = report_from(result, year_range)
    if report:
      inspector.save_report(report)

  # Pull the archive audit reports
  doc = BeautifulSoup(utils.download(AUDIT_ARCHIVE_URL))
  results = doc.select("div.block a")
  for result in results:
    report = report_from(result, year_range)
    if report:
      inspector.save_report(report)

  # Pull the other reports
  doc = BeautifulSoup(utils.download(OTHER_REPORTS_URl))
  results = doc.select("div.block > a")
  for result in results:
    report = report_from(result, year_range)
    if report:
      inspector.save_report(report)

def scrape_reports(options):
  """Pull reports from "Reports and Testimonies - Browse by date" web page."""

  REPORTS_URL = 'http://www.gao.gov/browse/date/custom?adv_begin_date=01/01/' +\
    '%s&adv_end_date=12/31/%s&rows=50&o=%s' # % (year, year, offset)
  archive = 1970
  # Amazingly, reports go back to 1940, though those are unlikely to be
  # legible enough to OCR. Also very cool, even 1950s-era reports seem to have
  # a highlightable embedded text layer in them. Of course, it was the
  # General Accounting Office back then and less oversighty.

  year_range = inspector.year_range(options, archive)
  for year in year_range:
    is_next_page = True
    offset = 0
    while is_next_page:
      doc = utils.beautifulsoup_from_url(
        REPORTS_URL % (year, year, offset))
      results = doc.select("div.listing")
      for result in results:
        report = process_report(result, year_range)
        if report:
          inspector.save_report(report)
      page_links = doc.select("a.non-current_page")
      if len(page_links) and page_links[-1].text.startswith('Next'):
        offset += 50
      else:
        is_next_page = False

def run(options):
  year_range = inspector.year_range(options)
  max_pages = int(options.get('pages', 1))
  for year in year_range:
    page = 1
    done = False
    while not done:
      url = url_for(options, page, year)
      body = utils.download(url)

      doc = BeautifulSoup(body)

      next_page = page + 1
      found_next_page = False
      page_links = doc.select("li.pager-item a.active")
      for page_link in page_links:
        if page_link.text == str(next_page):
          found_next_page = True
          break
      if not found_next_page:
        done = True
      if next_page > max_pages:
        done = True

      results = doc.select("table.views-table > tbody > tr")
      for result in results:
        report = report_from(result)
        inspector.save_report(report)

      page = next_page
      if not done:
        print('Moving to next page (%d)' % page)