def upload(params):
Debug = DebugManager.DebugManager();
Debug.start();
Debug.trace('start');
dbManager = SharedMemoryManager.getInstance();
db = dbManager.query();
date = fn.getNestedElement(params, 'date');
path = fn.getNestedElement(params, 'path');
# url = fn.getNestedElement(params, 'callback_url'); # required params to handle callback_url
paths, should_reset = ModelUpload.getPath(params);
for idx in range(0, len(paths)):
p = paths[idx];
processed_filename = File.converExcelFileToCsv(p, ignore_index=True);
Logger.v('processed_filename', processed_filename);
Debug.trace('convert to json : path {0}'.format( processed_filename ) );
if idx == 0 and should_reset: #reset once at the beginning
Logger.v('Reset Database.');
reset(date); #reset stock_issue collection
ModelSIIntegrity.reset(date); #reset stock_issue_datalog by date given
File.readCsvFileInChunks(processed_filename, save, params, chunksize=chunksize);
Debug.trace('uploaded to mongo.');
generateIndex();
ModelSIIntegrity.generateIndex();
Debug.trace('indexing mongo collection.');
saveIssueOption();
Debug.trace('save option to json.');
trigger_params = copy.deepcopy(params);
trigger_params['result'] = 'data count: {0}'.format(params['data_count'][path]);
# Logger.v('trigger_params', trigger_params);
dbManager.executeBulkOperations(None); # Insert all the remaining job at once.
ReportStock.triggerOnComplete(trigger_params);
Debug.trace('trigger api on complete.');
Debug.end();
Debug.show('Stock.upload');
def stockIntegrity(params, data):
Debug = DebugManager.DebugManager()
Debug.start()
global msia_tz, date_retrieve_limit
result = []
today = DateTime.now(tzinfo=msia_tz)
start_date = DateTime.getDaysAgo(date_retrieve_limit, datefrom=today)
durations = DateTime.getBetween([start_date, today],
element='date',
offset=24)['order']
# offset 24 hour to include today
state_data = fn.getNestedElement(data, 'state')
facility_data_by_state = fn.getNestedElement(data, 'state_facility')
check_data = combinedFacilityList(data=facility_data_by_state)
result = getIntegrity(params={
'durations': durations,
},
data={
'facility': facility_data_by_state,
'state': state_data,
'to_update': result,
'check_data': check_data,
})
updateStateData(result)
result = list(sorted(result, key=lambda k: k['name'], reverse=False))
Debug.end()
Debug.show('Model.Structure.stockIntegrity')
return result
def createSchedules(args={}): #upid, page_type
global filter_page_type
Debug = DebugManager.DebugManager()
Debug.start()
dbManager = SharedMemoryManager.getInstance()
# crawl_duration = fn.getNestedElement(fn.config,'CRAWL_DURATION', 12);
incomplete_task, incomplete_task_count = checkRemaining()
new_queue_count = 0
# Logger.v(incomplete_task_count, incomplete_task, filter_page_type);
extra_params = {
'crawl_comment': fn.getNestedElement(args, 'crawl_comment', None)
}
extra_params = {k: v for k, v in extra_params.items() if v is not None}
for platform in filter_page_type:
if args and not platform in fn.getNestedElement(
args, 'page_type', platform).split(','):
Logger.v('Skip Platform:%s' % (platform))
continue
# skip when page_type appear and not same
pages = fn.getNestedElement(args, 'pages.{0}'.format(platform), [])
Logger.v('platform', platform)
# Logger.v('page', args['pages']['budget']);
for page in pages: #Create queue for each
# Logger.v('page', page);
Queue.create(page,
extra_params=extra_params,
priority=fn.getNestedElement(args, 'priority',
'daily'),
batch=True)
new_queue_count += 1
Logger.v('new_queue_count', new_queue_count)
# Debug.trace();
Logger.v('Incomplete:%s, New Queue: %s' %
(incomplete_task_count, new_queue_count))
if incomplete_task_count > (new_queue_count *
int(fn.config['DEBUG_CRAWL_WARNING']) /
100) or incomplete_task_count > int(
fn.config['DEBUG_CRAWL_WARNING']):
# Mail.send('[%s]Incomplete Crawl [%s], Current Schedule: [%s]'%(DateTime.getReadableDate(DateTime.now()),
# incomplete_task_count, new_queue_count),
# fn.dumps(incomplete_task, encode=False)
# );
pass
result = {
'pending_count': new_queue_count,
'incomplete_count': incomplete_task_count
}
dbManager.executeBulkOperations(None)
# Debug.show('Create Schedule');
return Params.generate(True, result)
def getResult2(action, params):
Debug = DebugManager.DebugManager()
Debug.start()
Debug.trace('start')
if action == 'procurement':
result = ModelProcurement.get(params=params)
# result = Budget.get(params=params);
elif action == 'budget':
result = ModelBudget.get(params=params)
# result = Budget.get(params=params);
Debug.end()
Debug.show('Report.getResult2')
return result
def check(params):
Debug = DebugManager.DebugManager()
Debug.start()
Debug.trace('start')
global process_order
global key_to_join
global group_by
group_by_list = fn.getNestedElement(params, 'group_by', [])
filter_quantity = fn.getNestedElement(params, 'filter.quantity', [])
export = fn.getNestedElement(params, 'export', None)
custom_params = copy.deepcopy(params)
result = {}
# filter and read mongo db
data = ModelStockIssue.get(params)
Debug.trace('read mongo')
# filtering by group
if group_by_list:
group_by = group_by_list[-1]['id']
else:
group_by = 'state'
custom_params['group_by_key'] = group_by
custom_params['process_order'] = process_order[group_by]
custom_params['key_to_join'] = key_to_join[group_by]
custom_params['item_key_to_show'] = item_key_to_show[group_by]
# processing data
if data:
temp_result = ModelStockIssue.calculateData(params=custom_params,
data=data)
Debug.trace('calculate data')
result = toOutputStructure(params=custom_params, data=temp_result)
Debug.trace('structure data')
if result == {}:
result = []
Debug.end()
Debug.show('StockIssue.run')
if export:
export_result = generateExcelStructure(params=custom_params,
data=result)
return Params.generate(True, export_result)
else:
return Params.generate(True, result)
def get(params):
Debug = DebugManager.DebugManager()
Debug.start()
Debug.trace('start')
report_name = Report.getReportName(params)
Logger.v('-----Start Getting Data-----')
data = getData(params=params)
Debug.trace('get data')
Logger.v('-----Start Calculate Data-----')
calculated_data = calculateData(params=params, data=data)
Debug.trace('calculate data')
Logger.v('-----Start Restructure Data-----')
result = toOutputStructure(params=params,
data={
'data': data,
'calculated_data': calculated_data
})
Debug.trace('structure data')
Debug.end()
Debug.show('Model.Procurement.get')
return result
def get(action, params):
Debug = DebugManager.DebugManager()
Debug.start()
if action == 'whitelist':
result = getWhitelist(params)
elif action == 'procurement':
Logger.v('action:', action)
# result = getResult(action, params); # XXX
result = getResult2(action, params)
# XXX
elif action == 'budget':
Logger.v('action:', action)
# result = getResult(action, params); # XXX
result = getResult2(action, params)
# XXX
elif action == 'stockcheck':
Logger.v('action', action)
result = Stock.run(params)
Debug.end()
Debug.show('Report.get')
return Params.generate(True, result)
def run(params):
Mail.send(
'{0} Crawl - Store starts'.format(
DateTime.getReadableDate(DateTime.now())),
'Start at: {0}'.format(DateTime.now(tzinfo=msia_tz)))
recordTime(key='create_schedule_start_time')
Debug = DebugManager.DebugManager()
Debug.start()
start_crawl = fn.getNestedElement(params, 'schedule_params.start_crawl',
False)
check_empty = fn.getNestedElement(params, 'schedule_params.check_empty',
False)
Logger.v('Creating schedule:')
updateDropdownOptions(params=params)
crawl_params = generateCrawlParam(params)
Debug.trace('Generate Crawl Params')
createSchedules({'pages': crawl_params})
Debug.trace('Create Schedule')
if start_crawl:
Crawl.start(params)
Debug.trace('crawling')
recordTime(key='create_schedule_end_time')
Debug.show('Run')
def extractSearch():
from lib import File, Logger, DebugManager
from Report import Item, Stock
Debug = DebugManager.DebugManager()
Debug.start()
Debug.trace('start')
# filename = '2020-03-30.xlsx';
filename = 'upload/desk.csv'
# fn.ensureDirectory('tests');
data = Item.upload(filename)
filename = 'upload/desh.csv'
# fn.ensureDirectory('tests');
data = Item.upload(filename, new=False)
# Stock.upload({
# 'action': 'upload_stock',
# 'id': 1,
# 'date': '2020-03-30',
# 'path': filename, # staging filepath
# });
# Item.save(data);
#
result = Item.find({
'search_text': 'oxygen',
# 'item_code': 'Y1690580002.00,Y1690130024.00'
# 'group_by' : 'sub_group_desc'
})
fn.show(result)
# for d in result['data']:
# Logger.v(d);
# File.writeJSONFile('search.json',data);
Debug.trace('end')
Debug.end()
Debug.show('Daniel.py')
def toOutputStructure(params, data):
Debug = DebugManager.DebugManager()
Debug.start()
Debug.trace('start')
new_data = fn.getNestedElement(data, 'data')
calculated_data = fn.getNestedElement(data, 'calculated_data')
report_name = Report.getReportName(params)
min_purchase_amount = fn.getNestedElement(params,
'filter.min_purchase_amount', 0)
if type(min_purchase_amount) == str:
min_purchase_amount = int(min_purchase_amount)
total_keys = global_report_key_update[report_name]
result = {}
all_keys = global_group_order[report_name]
keys = {}
for idx in range(0, len(all_keys)):
ak = all_keys[idx]
i = str(idx)
keys['key' + i] = ak
new_data_0 = copy.deepcopy(new_data)
portion_0 = []
unique_values0 = sorted(
list(
set(obj_[global_group_order_kepmap[keys['key0']]]
for obj_ in new_data_0)))
for uv0 in unique_values0:
new_data_1 = list(
filter(lambda d: d[global_group_order_kepmap[keys['key0']]] == uv0,
new_data_0))
portion_1 = []
unique_values1 = sorted(
list(
set(obj_[global_group_order_kepmap[keys['key1']]]
for obj_ in new_data_1)))
join_key_1 = '_'.join([uv0])
for uv1 in unique_values1:
new_data_2 = list(
filter(
lambda d: d[global_group_order_kepmap[keys['key1']]] ==
uv1, new_data_1))
portion_2 = []
unique_values2 = sorted(
list(
set(obj_[global_group_order_kepmap[keys['key2']]]
for obj_ in new_data_2)))
join_key_2 = '_'.join([uv0, uv1])
if report_name == 'budget':
for uv2 in unique_values2:
new_data_3 = list(
filter(
lambda d: d[global_group_order_kepmap[keys['key2']]
] == uv2, new_data_2))
portion_3 = []
unique_values3 = sorted(
list(
set(obj_[global_group_order_kepmap[keys['key3']]]
for obj_ in new_data_3)))
join_key_3 = '_'.join([uv0, uv1, uv2])
for uv3 in unique_values3:
new_data_4 = list(
filter(
lambda d: d[global_group_order_kepmap[keys[
'key3']]] == uv3, new_data_3))
portion_4 = []
unique_values4 = sorted(
list(
set(obj_[global_group_order_kepmap[
keys['key4']]] for obj_ in new_data_4)))
join_key_4 = '_'.join([uv0, uv1, uv2, uv3])
for uv4 in unique_values4: # last key
new_data_5 = list(
filter(
lambda d: d[global_group_order_kepmap[keys[
'key4']]] == uv4, new_data_4))
join_key_5 = '_'.join([uv0, uv1, uv2, uv3, uv4])
children = generateChildren(params=params,
data=new_data_5)
if children:
portion_4.append({
'id':
fn.getNestedElement(
calculated_data,
'{0}.id'.format(join_key_5)),
'name':
fn.getNestedElement(
calculated_data,
'{0}.name'.format(join_key_5)),
'code':
fn.getNestedElement(
calculated_data,
'{0}.code'.format(join_key_5)),
'total':
generateSummary(params=params,
data=children),
'children':
children,
})
if portion_4:
portion_3.append({
'id':
fn.getNestedElement(
calculated_data,
'{0}.id'.format(join_key_4)),
'name':
fn.getNestedElement(
calculated_data,
'{0}.name'.format(join_key_4)),
'code':
fn.getNestedElement(
calculated_data,
'{0}.code'.format(join_key_4)),
'total':
generateSummary(params=params, data=portion_4),
keys['key4']:
portion_4,
})
if portion_3:
portion_2.append({
'id':
fn.getNestedElement(calculated_data,
'{0}.id'.format(join_key_3)),
'name':
fn.getNestedElement(calculated_data,
'{0}.name'.format(join_key_3)),
'code':
fn.getNestedElement(calculated_data,
'{0}.code'.format(join_key_3)),
'total':
generateSummary(params=params, data=portion_3),
keys['key3']:
portion_3,
})
if portion_2:
portion_1.append({
'id':
fn.getNestedElement(calculated_data,
'{0}.id'.format(join_key_2)),
'name':
fn.getNestedElement(calculated_data,
'{0}.name'.format(join_key_2)),
'code':
fn.getNestedElement(calculated_data,
'{0}.code'.format(join_key_2)),
'total':
generateSummary(params=params, data=portion_2),
keys['key2']:
portion_2,
})
elif report_name == 'procurement':
for uv2 in unique_values2:
new_data_3 = list(
filter(
lambda d: d[global_group_order_kepmap[keys['key2']]
] == uv2, new_data_2))
join_key_3 = '_'.join([uv0, uv1, uv2])
children = generateChildren(params=params, data=new_data_3)
if children:
portion_2.append({
'id':
fn.getNestedElement(calculated_data,
'{0}.id'.format(join_key_3)),
'name':
fn.getNestedElement(calculated_data,
'{0}.name'.format(join_key_3)),
'code':
fn.getNestedElement(calculated_data,
'{0}.code'.format(join_key_3)),
'total':
generateSummary(params=params, data=children),
'children':
children,
})
if portion_2:
portion_1.append({
'id':
fn.getNestedElement(calculated_data,
'{0}.id'.format(join_key_2)),
'name':
fn.getNestedElement(calculated_data,
'{0}.name'.format(join_key_2)),
'code':
fn.getNestedElement(calculated_data,
'{0}.code'.format(join_key_2)),
'total':
generateSummary(params=params, data=portion_2),
keys['key2']:
portion_2,
})
if portion_1:
portion_0.append({
'id':
fn.getNestedElement(calculated_data,
'{0}.id'.format(join_key_1)),
'name':
fn.getNestedElement(calculated_data,
'{0}.name'.format(join_key_1)),
'code':
fn.getNestedElement(calculated_data,
'{0}.code'.format(join_key_1)),
'total':
generateSummary(params=params, data=portion_1),
keys['key1']:
portion_1,
})
if portion_0:
result[report_name] = {
'group_order': global_group_order[report_name],
'total': generateSummary(params=params, data=portion_0),
keys['key0']: portion_0,
}
Debug.end()
Debug.show('toOutputStructure')
return result
def calculateData(params, data):
Debug = DebugManager.DebugManager()
Debug.start()
Debug.trace('start')
custom_params = copy.deepcopy(params)
report_name = Report.getReportName(params)
group_order = global_group_order[report_name]
result = {}
unique_keys = generateUniqueKeys(params, data)
Debug.trace('get unique_keys')
for uk in unique_keys:
# Logger.v('uk', uk);
split_uk = uk.split('_')
# Logger.v('split_uk', split_uk);
# Logger.v('lenght split_uk', len(split_uk));
custom_params['split_uk'] = split_uk
functions = generateFilterFunction(params=custom_params)
function_name = 'function_{0}'.format(len(split_uk))
# Logger.v('functions', functions);
filtered_data = list(filter(functions[function_name], data))
# Logger.v('filtered_data', filtered_data);
# Logger.v('sum of first allocation', sum(list(obj_['first_allocation'] for obj_ in filtered_data)));
total_keys = global_report_key_update[report_name]
# Debug.trace('get filtered_data');
if filtered_data:
item = generateItemDetail(params=custom_params, data=filtered_data)
if uk not in result:
result[uk] = {}
result[uk] = {
'id': item['id'],
'name': item['name'],
'code': item['code'],
'total': {},
}
for tk in total_keys:
try:
result[uk]['total'][tk] = sum(
list(obj_[tk] for obj_ in filtered_data))
except Exception as KeyError:
# Logger.v('Report.calculateData: {0} not found, sum up after data cleaning process.'.format(tk));
if tk == 'total_allocation':
result[uk]['total'][tk] = result[uk]['total'][
'first_allocation'] + result[uk]['total'][
'additional_allocation']
elif tk == 'balance_amount':
result[uk]['total'][tk] = result[uk]['total'][
'first_allocation'] + result[uk]['total'][
'additional_allocation'] - result[uk]['total'][
'pending_amount'] - result[uk]['total'][
'liablity_amount'] - result[uk][
'total']['utilized_amount']
# Logger.v('tk', tk);
# Debug.trace('calculating data');
# Logger.v('result', result);
Debug.end()
Debug.show('Report.calculateData')
return result
def generateCrawlParam(params):
Debug = DebugManager.DebugManager()
Debug.start()
global pass_month_quantity
dbManager = SharedMemoryManager.getInstance()
db = dbManager.query()
crawl_params = {}
limit_for_test = 10
report_keys = fn.getNestedElement(params, 'keys.report',
['budget', 'procurement'])
interval = fn.getNestedElement(params, 'interval', 1)
filter_facility_code = fn.getNestedElement(params, 'filter.facility_code',
True)
check_empty = fn.getNestedElement(params, 'schedule_params.check_empty',
False)
today = fn.getNestedElement(
params, 'schedule_params.today',
DateTime.toString(DateTime.now(tzinfo=msia_tz)))
# Logger.v('filter_facility_code', filter_facility_code);
if check_empty:
# past_dates = DateTime.getPastDate(count=pass_month_quantity, duration=interval);
past_dates = DateTime.getPastDate(
count=pass_month_quantity,
duration=interval,
end=DateTime.convertDateTimeFromString(today))
# Logger.v('past_dates', past_dates);
# exit();
else:
past_dates = DateTime.getPastDate(count=pass_month_quantity,
duration=interval)
# Logger.v('past_dates', past_dates);
state_codes = retrieveOption(collection_name='state',
show_keys=['state_code'],
hide_keys=['_id'])
state_code = extractListByKey(data=state_codes, key='state_code')
facility_codes = retrieveOption(collection_name='facility',
show_keys=['facility_code'],
hide_keys=['_id'])
facility_code = extractListByKey(data=facility_codes, key='facility_code')
for key in report_keys:
# Logger.v('collection', key, past_dates[0]);
Debug.trace()
if key not in crawl_params:
crawl_params[key] = []
mongo_data = list(db[key].find({}, {}))
if len(mongo_data) == 0:
dates = past_dates[0][:]
else:
dates = past_dates[0][:1]
year = extractYear(data=dates)
# Logger.v('year', year);
# Logger.v('filter_facility_code', filter_facility_code);
if key == 'budget':
if not filter_facility_code:
iteration = 0
total = len(year) * len(state_code)
# fn.printProgressBar(iteration=iteration, total=total);
for y in year:
for sc in state_code:
obj_ = {
'financial_year': y,
'state_code': sc,
'page_type': key,
'upid': '_'.join([sc, y]),
'url': api_links[key].format(sc, y, ''),
'start_date': today,
'end_date': today,
}
if obj_ not in crawl_params[key]:
crawl_params[key].append(obj_)
# Logger.v('len(crawl_param])', len(crawl_params[key]));
iteration += 1
# fn.printProgressBar(iteration=iteration, total=total);
else:
iteration = 0
total = len(year) * len(state_code) * len(
facility_code[:limit_for_test])
# fn.printProgressBar(iteration=iteration, total=total);
for y in year:
for sc in state_code:
for fc in facility_code[:limit_for_test]:
obj_ = {
'financial_year': y,
'state_code': sc,
'page_type': key,
'upid': '_'.join([sc, y, fc]),
'facility_code': fc,
'url': api_links[key].format(sc, y, fc),
'start_date': today,
'end_date': today,
}
if obj_ not in crawl_params[key]:
crawl_params[key].append(obj_)
# Logger.v('len(crawl_param])', len(crawl_params[key]));
iteration += 1
# fn.printProgressBar(iteration=iteration, total=total);
elif key == 'procurement':
if not filter_facility_code:
for past_duration in dates:
start_date = DateTime.toString(
DateTime.getDaysAgo(days_to_crawl=-1,
datefrom=past_duration[0]))
end_date = DateTime.toString(
DateTime.getDaysAgo(days_to_crawl=1,
datefrom=past_duration[1]))
for sc in state_code:
obj_ = {
'state_code':
sc,
'start_date':
start_date,
'end_date':
end_date,
'page_type':
key,
'upid':
'_'.join([sc, start_date, end_date]),
'url':
api_links[key].format(sc,
start_date.replace('-', ''),
end_date.replace('-', ''),
''),
}
if obj_ not in crawl_params[key]:
crawl_params[key].append(obj_)
# Logger.v('len(crawl_param])', len(crawl_params[key]));
else:
for past_duration in dates:
start_date = DateTime.toString(
DateTime.getDaysAgo(days_to_crawl=-1,
datefrom=past_duration[0]))
end_date = DateTime.toString(
DateTime.getDaysAgo(days_to_crawl=1,
datefrom=past_duration[1]))
for sc in state_code:
for fc in facility_code[:limit_for_test]:
obj_ = {
'state_code':
sc,
'start_date':
start_date,
'end_date':
end_date,
'page_type':
key,
'facility_code':
fc,
'upid':
'_'.join([sc, start_date, end_date, fc]),
'url':
api_links[key].format(
sc, start_date.replace('-', ''),
end_date.replace('-', ''), fc)
}
if obj_ not in crawl_params[key]:
crawl_params[key].append(obj_)
# Logger.v('len(crawl_param])', len(crawl_params[key]));
for c in crawl_params:
# Logger.v('crawl_params', c, len(crawl_params[c]));
fn.writeExcelFile(filename='{0}/{1}'.format(test_folder, c),
data=crawl_params[c])
Logger.v('crawl_params', len(crawl_params))
Debug.show('Generate Crawl Params')
return crawl_params
def sourceKlsescreener(params):
Debug = DebugManager.DebugManager()
Debug.start()
Debug.trace('start')
stock_list = pd.read_csv('reference/stock_info/stock_list.csv')
stock_codes = stock_list['code'].unique().tolist()
Debug.trace('read stock list')
summary = []
error = []
for code in stock_codes:
result = {}
url = 'https://www.klsescreener.com/v2/stocks/view/{0}'.format(code)
Logger.v('crawling:', url)
page = ''
error_count = 0
while page == '':
try:
page = requests.get(url)
error_count = 0
break
except:
if error_count > 3:
page = 'None'
error_count = 0
break
print("Connection refused by the server..")
print("Let me sleep for 5 seconds")
print("ZZzzzz...")
time.sleep(5)
print("Was a nice sleep, now let me continue...")
error_count += 1
continue
if not page == '':
time.sleep(3)
soup = BeautifulSoup(page.text, 'html.parser')
# print(soup.prettify());
# print(soup.find_all('div', id='quarter_reports'));
# Logger.v('Quarter Report');
quarter_reports_div = soup.find_all('div', id='quarter_reports')
result['quarter_reports'] = getQuarterData(
data=quarter_reports_div)
# Logger.v('Annual Report'); # th in tbody
annual_report_div = soup.find_all('div', id='annual')
result['annual'] = getAnnualData(data=annual_report_div)
# Logger.v('Dividend');
dividends_div = soup.find_all('div', id='dividends')
result['dividends'] = getQuarterData(data=dividends_div)
# Logger.v('Capital Changes');
capital_changes_div = soup.find_all('div', id='capital_changes')
result['capital_changes'] = getQuarterData(
data=capital_changes_div)
# Logger.v('Warrants'); # every row got new tbody
warrants_div = soup.find_all('div', id='warrants')
result['warrants'] = getQuarterData(data=warrants_div)
# Logger.v('Shareholding Changes'); # every row got new tbody
shareholding_changes_div = soup.find_all('div',
id='shareholding_changes')
result['shareholding_changes'] = getQuarterData(
data=shareholding_changes_div)
Debug.trace('crawl report')
save_dir = 'crawled_data/reports'
fn.ensureDirectory(save_dir)
save_file = '{0}/{1}.json'.format(save_dir, code)
fn.writeJSONFile(save_file, result)
summary.append({
'code':
code,
'quarter_reports':
len(result['quarter_reports']),
'annual':
len(result['annual']),
'dividends':
len(result['dividends']),
'capital_changes':
len(result['capital_changes']),
'warrants':
len(result['warrants']),
'shareholding_changes':
len(result['shareholding_changes']),
})
Debug.trace('save report')
elif page == 'None':
error.append({'code': code})
# Logger.v('summary', summary);
summary_file = '{0}/summary/klsescreener_report'.format(save_dir)
fn.writeExcelFile(filename=summary_file, data=summary)
Debug.trace('save summary')
error_file = '{0}/error/klsescreener_report'.format(save_dir)
fn.writeExcelFile(filename=error_file, data=error)
Debug.trace('save error')
Debug.end()
Debug.show('sourceKlsescreener')
return True
def sourceBursaMalaysia(params):
Debug = DebugManager.DebugManager()
Debug.start()
Debug.trace('start')
url = 'https://www.bursamalaysia.com/market_information/equities_prices'
page = requests.get(url)
soup = BeautifulSoup(page.text, 'html.parser')
total_page = int(soup.find('li', id='total_page').get('data-val'))
limit_page = None
Logger.v('total pages', total_page)
# print(soup.find_all('h2'));
date = ''
time = ''
for div in soup.find_all('div'):
if div.get('class') == [
'col', 'bg-white', 'mr-1', 'mb-0', 'h5', 'bold', 'p-2'
] and date == '':
# print(div.get_text().strip());
date = div.get_text().strip()
if div.get('class') == [
'col', 'bg-white', 'h5', 'mb-0', 'bold', 'p-2'
] and time == '':
# print(div.get_text().strip());
time = div.get_text().strip()
Logger.v('date', date, 'time', time)
result = []
Debug.trace('get detail')
for page in range(1, total_page + 1)[:limit_page]:
fn.printProgressBar(iteration=page,
total=total_page + 1,
prefix='Page:{0}'.format(page))
url_with_page = 'https://www.bursamalaysia.com/market_information/equities_prices?page={0}'.format(
page)
page = requests.get(url_with_page)
soup = BeautifulSoup(page.text, 'html.parser')
tables = soup.find_all('table')
for table in tables:
if table.get('class') == [
'table', 'datatable-striped', 'text-center',
'equity_prices_table', 'datatable-with-sneak-peek',
'js-anchor-price-table'
]: # only take first table (price)
headers = getTableHeader(data=table)
headers += ['date', 'time']
content = getTableContent(data=table)
for row in content:
row += [date, time]
result.append(dict(zip(headers, row)))
# Logger.v('headers', headers)
# Logger.v('result', len(result), result);
Debug.trace('crawl price')
save_dir = 'crawled_data/price'
filename = '{0}_{1}_price.csv'.format(date, time)
fn.ensureDirectory(save_dir)
save_file = '{0}/{1}'.format(save_dir, filename)
df = pd.DataFrame(result)
df.to_csv(save_file, index=False)
Logger.v('saved {0}'.format(save_file))
Debug.trace('save price')
saveStockList(data=result)
Debug.trace('save stock list')
Debug.end()
Debug.show('sourceBursaMalaysia')
return True
def calculateData(params, data):
Debug = DebugManager.DebugManager()
Debug.start()
Debug.trace('start')
global naming_keymap, crawl_folder
item_key_to_show = fn.getNestedElement(params, 'item_key_to_show')
process_order = fn.getNestedElement(params, 'process_order')
custom_params = copy.deepcopy(params)
result = {}
main_po = {}
df = pd.DataFrame(data[:]).astype(str)
df = preprocessDataframe(params=custom_params, data=df)
summation_df = groupDataframe(params=custom_params, data=df)
# to check summation result
# output_file = '{0}/output/find_result.xlsx'.format(crawl_folder);
# df.to_excel(output_file);
Debug.trace('dataframe process')
for idx in range(0, len(process_order)):
main_po[idx] = {
'po': process_order[idx],
'group_po': [],
}
for idx1 in range(0, idx + 1):
main_po[idx]['group_po'].append(process_order[idx1])
for idx in range(0, len(process_order)):
po = main_po[idx]['po']
group_po = main_po[idx]['group_po']
custom_params['po'] = {
'po': po,
'naming_keymap': naming_keymap,
}
grouped_df = df.groupby(group_po).groups
Logger.v('len', idx, 'th', len(grouped_df.keys()))
if idx == 0:
for gk in grouped_df.keys():
code = gk.split('|')[-1]
name = df[df[po] == gk][naming_keymap[po]].unique().tolist()[0]
result[gk] = {
'id': fn.convertToSnakecase(gk),
'name': name,
'code': code,
}
else:
for gk in grouped_df.keys():
level = gk[-2]
code = gk[-1]
if idx == 1:
temp_result = result
elif idx == 2:
# Logger.v('gk0', gk[0], main_po[idx-1]['po'], 'level', level, 'code', code);
temp_result = fn.getNestedElement(
result, '{0}.{1}'.format(gk[0],
main_po[idx - 1]['po']))
elif idx == 3:
# Logger.v('gk0', gk[0], main_po[idx-2]['po'], gk[1], main_po[idx-1]['po'], 'level', level, 'code', code);
temp_result = fn.getNestedElement(
result,
'{0}.{1}.{2}.{3}'.format(gk[0], main_po[idx - 2]['po'],
gk[1],
main_po[idx - 1]['po']))
if po not in temp_result[level]:
temp_result[level][po] = {}
if code not in temp_result[level][po]:
temp_result[level][po][code] = {}
custom_params['po'].update({
'gk': code,
})
# when this is the last element in process_order
if process_order[-1] == po:
last_child_data = insertNthChild(params=custom_params,
data=summation_df,
is_last=True)
info = last_child_data['info']
temp_result[level][po][code] = last_child_data['obj_']
# add extra info by group_by
for ik in item_key_to_show:
temp_result[level][po][code].update({
ik:
info[ik].values[0],
})
else:
last_child_data = insertNthChild(params=custom_params,
data=summation_df)
temp_result[level][po][code] = last_child_data['obj_']
Debug.trace('{0}th'.format(idx))
Debug.end()
Debug.show('Model.Stock.calculateData')
return result
