STY: cdaweb methods streamline

[jklenzing]

Some of the target parsing in the cdaweb methods for recognizing file names could be improved.

Alternatively, better incorporation of the official API may replace these routines entirely.

[jklenzing]

It looks like using the get_original_filenames function in cdasws will handle this in a much more streamlined way. It won't eliminate the problem, as not all datasets are supported here.

[jklenzing]

cdasws was incorporated in #145, solving this for most datasets.

To close, I think that the following lines need to be simplified.

pysatNASA/pysatNASA/instruments/methods/cdaweb.py

Lines 769 to 788 in b0003c1

	try:
	for top_url in url_list:
	for level in range(n_layers + 1):
	for directory in remote_dirs[level]:
	temp_url = '/'.join((top_url.strip('/'), directory))
	soup = BeautifulSoup(requests.get(temp_url).content,
	"lxml")
	links = soup.find_all('a', href=True)
	for link in links:
	# If there is room to go down, look for directories
	if link['href'].count('/') == 1:
	remote_dirs[level + 1].append(link['href'])
	else:
	# If at the endpoint, add matching files to list
	add_file = True
	for target in targets:
	if link['href'].count(target) == 0:
	add_file = False
	if add_file:
	full_files.append(link['href'])

Probably need to incorporate a directory structure as part of supported_tags so that the layers loop can be removed.

[jklenzing]

Work on after merge of #200