def download_filings(feedpath, args=None):
"""Go through all entries in the given EDGAR RSS feed and download any missing or new filings."""
logger.info("Processing RSS feed %s", feedpath)
dir = filings_dir(feedpath)
os.makedirs(dir, exist_ok=True)
filing_urls = []
for filing in feed_tools.read_feed(feedpath):
if args:
if args.company_re and not bool(
args.company_re.match(filing['companyName'])):
continue
if args.cik and args.cik != filing['cikNumber']:
continue
if args.sic and args.sic != filing['assignedSic']:
continue
if args.form_type and args.form_type != filing['formType']:
continue
if 'enclosureUrl' in filing and not exists_filing(
dir, filing['enclosureUrl'], filing['enclosureLength']):
filing_urls.append(filing['enclosureUrl'])
logger.info("Start downloading %d new filings", len(filing_urls))
with concurrent.futures.ThreadPoolExecutor(
max_workers=args.max_threads) as executor:
futures = [
executor.submit(download_filing, dir, url, args.max_retries)
for url in filing_urls
]
for future in concurrent.futures.as_completed(futures):
try:
future.result()
except Exception as e:
print(e)
def build_secdb(feeds):
# Setup python logging framework
setup_logging(args.log_file)
tickers = load_ticker_symbols()
# Setup up DB connection
global db_connect
db_connect = setup_db_connect(args.db_driver,args.db_name)
# Create all required DB tables
if args.create_tables:
create_db_tables()
create_db_indices()
insert_ticker_symbols(tickers)
# Process all filings in the given RSS feeds one month after another
for filepath in feeds:
# Load EDGAR filing metadata from RSS feed (and filter out all non 10-K/10-Q filings or companies without an assigned ticker symbol)
filings = {}
for filing in feed_tools.read_feed(filepath):
if args.cik is None or args.cik == filing['cikNumber']:
if filing['formType'] in ('10-K','10-K/A','10-Q','10-Q/A') and filing['cikNumber'] in tickers:
filing['ticker'] = tickers[filing['cikNumber']]
filings.setdefault(filing['cikNumber'],[]).append(filing)
# Process the selected XBRL filings
process_filings(filings)
def download_filings(feedpath, args=None):
"""Go through all entries in the given EDGAR RSS feed and download any missing or new filings."""
logger.info("Processing RSS feed %s", feedpath)
dir = filings_dir(feedpath)
os.makedirs(dir, exist_ok=True)
filing_urls = []
for filing in feed_tools.read_feed(feedpath):
if args:
if args.company_re and not bool(
args.company_re.match(filing['companyName'])):
continue
if args.cik and args.cik != filing['cikNumber']:
continue
if args.sic and args.sic != filing['assignedSic']:
continue
if args.form_type and args.form_type != filing['formType']:
continue
if 'enclosureUrl' in filing and not exists_filing(
dir, filing['enclosureUrl'], filing['enclosureLength']):
filing_urls.append(filing['enclosureUrl'])
logger.info("Start downloading %d new filings", len(filing_urls))
for url in filing_urls:
download_filing(dir, url, args.max_retries)
def generate_project(feedpath):
filings = feed_tools.read_feed(feedpath)
filings_by_company = {}
for filing in filings:
filings_by_company.setdefault(filing['companyName'], []).append(filing)
month = re.fullmatch(r'.*xbrlrss-(\d{4}-\d{2})\.xml',
os.path.basename(feedpath)).group(1)
dir = os.path.join(feed_tools.filings_dir, month)
file = os.path.join(dir, '%s.spp' % month)
print('Generating project file', file)
with open(file, 'w') as f:
f.write("""\
<?xml version="1.0" encoding="UTF-8"?>
<Project>
""")
f.write("""\
<Folder FolderName="Filings by name" ExtStr="xml">
""")
for filing in filings:
if filing['instanceUrl']:
f.write("""\
<File FilePath="%s" HomeFolder="Yes"/>
""" % filing['instanceUrl'][len('filings/YYYY-MM/'):].replace(
'%7Czip/', '|zip\\'))
f.write("""\
</Folder>
""")
f.write("""\
<Folder FolderName="Filings by company">
""")
for company in sorted(filings_by_company.keys()):
f.write("""\
<Folder FolderName="%s" ExtStr="xml">
""" % company.replace('&', '&').replace('<', '<'))
for filing in filings_by_company[company]:
if filing['instanceUrl']:
f.write("""\
<File FilePath="%s" HomeFolder="Yes"/>
""" % filing['instanceUrl'][len('filings/YYYY-MM/'):].replace(
'%7Czip/', '|zip\\'))
f.write("""\
</Folder>
""")
f.write("""\
</Folder>
""")
f.write("""\
</Project>
""")
def generate_project(feedpath):
filings = feed_tools.read_feed(feedpath)
filings_by_company = {}
for filing in filings:
filings_by_company.setdefault(filing['companyName'],[]).append(filing)
month = re.fullmatch(r'.*xbrlrss-(\d{4}-\d{2})\.xml',os.path.basename(feedpath)).group(1)
dir = os.path.join(feed_tools.filings_dir,month)
file = os.path.join(dir,'%s.spp'%month)
print('Generating project file',file)
with open(file,'w') as f:
f.write("""\
<?xml version="1.0" encoding="UTF-8"?>
<Project>
""")
f.write("""\
<Folder FolderName="Filings by name" ExtStr="xml">
""")
for filing in filings:
if filing['instanceUrl']:
f.write("""\
<File FilePath="%s" HomeFolder="Yes"/>
"""%filing['instanceUrl'][len('filings/YYYY-MM/'):].replace('%7Czip/','|zip\\'))
f.write("""\
</Folder>
""")
f.write("""\
<Folder FolderName="Filings by company">
""")
for company in sorted(filings_by_company.keys()):
f.write("""\
<Folder FolderName="%s" ExtStr="xml">
"""%company.replace('&','&').replace('<','<'))
for filing in filings_by_company[company]:
if filing['instanceUrl']:
f.write("""\
<File FilePath="%s" HomeFolder="Yes"/>
"""%filing['instanceUrl'][len('filings/YYYY-MM/'):].replace('%7Czip/','|zip\\'))
f.write("""\
</Folder>
""")
f.write("""\
</Folder>
""")
f.write("""\
</Project>
""")
def main():
# Parse script arguments
args = parse_args()
# Setup python logging framework
setup_logging(args)
# Validate all filings in the given RSS feeds one month after another
for filepath in collect_feeds(args):
# Load EDGAR filing metadata from RSS feed (and filter out all non 10-K/10-Q filings or companies without an assigned ticker symbol)
filings = []
for filing in feed_tools.read_feed(filepath):
# Google to Alphabet reorganization
if filing['cikNumber'] == 1288776:
filing['cikNumber'] = 1652044
if args.form_type is None or args.form_type == filing['formType']:
if args.sic is None or args.sic == filing['assignedSic']:
if args.cik is None or args.cik == filing['cikNumber']:
filings.append(filing)
# Validate the selected XBRL filings
validate_filings(filings[:100], args.max_threads)
def download_filings(feedpath,args=None):
"""Go through all entries in the given EDGAR RSS feed and download any missing or new filings."""
logger.info("Processing RSS feed %s",feedpath)
dir = filings_dir(feedpath)
os.makedirs(dir,exist_ok=True)
filing_urls = []
for filing in feed_tools.read_feed(feedpath):
if args:
if args.company_re and not bool(args.company_re.match(filing['companyName'])):
continue
if args.cik and args.cik != filing['cikNumber']:
continue
if args.sic and args.sic != filing['assignedSic']:
continue
if args.form_type and args.form_type != filing['formType']:
continue
if 'enclosureUrl' in filing and not exists_filing(dir,filing['enclosureUrl'],filing['enclosureLength']):
filing_urls.append(filing['enclosureUrl'])
logger.info("Start downloading %d new filings",len(filing_urls))
for url in filing_urls:
download_filing(dir,url,args.max_retries)
