def _runSingleProcess(self, urlGen, outputFile):
'''
Single process implementation. Iterates through urls in the generator
and builds result dataframe in memory. Suitable for smaller data
pulls.
'''
parser = FileParser()
lineItems = []
disclosures = []
for (i, url) in enumerate(urlGen, 1):
res, dd = parser.fileUrl2Result(url)
lineItems.append(res._toDict())
disclosures.append(dd._asdict())
if (i % 25 == 0):
edgarScraperLog.info("Scraped {} total files".format(i))
if (i % 100 == 0):
self._recordResults(lineItems, disclosures, outputFile)
lineItems = []
disclosures = []
edgarScraperLog.info("Job Finished {} total files".format(i))
if lineItems:
self._recordResults(lineItems, disclosures, outputFile)
if not outputFile:
lineItemDf = pd.concat(self.lineItemBuffer)
disclosureDf = pd.concat(self.disclosureBuffer)
return (lineItemDf, disclosureDf)
def test_xbrl(self):
url = self.baseUrl + 'data/730255/0001206774-15-003149.txt'
testIndexResult = IndexResult(0, 'test', 'test', 'test', url)
parser = FileParser()
result, dd = parser.fileUrl2Result(testIndexResult)
self.assertIsNotNone(result.lineItems)
self.assertIsInstance(result, ResultSet)
def test_text(self):
url = self.baseUrl + 'data/1005406/0000898430-96-001816.txt'
testIndexResult = IndexResult(2, 'test', 'test', 'test', url)
parser = FileParser()
result, dd = parser.fileUrl2Result(testIndexResult)
self.assertIsNotNone(result.lineItems)
self.assertIsInstance(result, ResultSet)
self.assertIsNone(dd.TEXT)
def test_html(self):
url = self.baseUrl + 'data/1144215/0001193125-10-001623.txt'
testIndexResult = IndexResult(1, 'test', 'test', 'test', url)
parser = FileParser()
result, dd = parser.fileUrl2Result(testIndexResult)
self.assertIsNotNone(result.lineItems)
self.assertIsInstance(result, ResultSet)
self.assertIsNone(dd.TEXT)
def _mpJob(urlGen, file):
'''
Multiprocessing task must be defined at top-level. Each worker process
iterates over a slice of the url-generator. Urls are processed and
results are returned to the main processes for aggregation.
'''
parser = FileParser()
resDictList = []
ddList = []
for url in urlGen:
(res, dd) = parser.fileUrl2Result(url)
resDictList.append(res._toDict())
ddList.append(dd._asdict())
if url.seqNo % 100 == 0:
edgarScraperLog.info('finished consuming file {}'.format(url.seqNo))
return (resDictList, ddList)
def setUp(self):
self.baseUrl = 'https://www.sec.gov/Archives/edgar/'
self.parser = FileParser()
self.he = HTMLExtractor()