def main():
from optparse import OptionParser
parser = OptionParser(option_class=magicdate.MagicDateOption,
usage='%prog [options]',
version='%prog ' + __version__)
parser.add_option('-s',
'--start-date',
type='magicdate',
default=None,
help='Force a start time (magicdate)')
parser.add_option(
'-e',
'--end-date',
type='magicdate',
default=magicdate.magicdate('today'),
help='Force an end time (magicdate) [default: %default]')
parser.add_option(
'-l',
'--log-format',
default='%Y%m%d.log',
help=
'datetime strftime format string to create the log file names to search for [default: %default]'
)
(options, args) = parser.parse_args()
assert options.start_date is not None
for d in date_generator(options.start_date, options.end_date):
#print d.strftime(options.log_format)
filename = d.strftime(options.log_format)
try:
s = os.stat(filename)
except OSError:
print 'missing', filename
def reports_for_date(date_string):
try:
date_object = magicdate(date_string)
output_list = parser.earnings_reports_on_date(date_object)
return jsonify(earnings_reports=output_list)
except:
print('Could not parse date')
return jsonify(earning_events=None)
def do_once(self):
cx = self.cx
cu = self.cu
v = self.verbose
tzoffset = datetime.timedelta(
seconds=time.timezone) # Always work in utc. Magicdate is local
if self.track_start is not None:
track_start = mark_utc(
magicdate.magicdate(self.track_start) + tzoffset)
if v:
cu.execute(
'SELECT COUNT(*) FROM track_lines WHERE update_timestamp < %s;',
(track_start, ))
print 'Deleting from track_lines:', cu.fetchone()[0]
cu.execute('DELETE FROM track_lines WHERE update_timestamp < %s;',
(track_start, ))
# if the points are too old to be in a track_line, then delete them
if v:
cu.execute(
'SELECT COUNT(*) FROM position WHERE cg_timestamp < %s;',
(track_start, ))
print 'Deleting from position:', cu.fetchone()[0]
cu.execute('DELETE FROM position WHERE cg_timestamp < %s;',
(track_start, ))
if self.last_position_start is not None:
last_position_start = mark_utc(
magicdate.magicdate(self.last_position_start) + tzoffset)
if v:
cu.execute(
'SELECT COUNT(*) FROM last_position WHERE cg_timestamp < %s;',
(track_start, ))
print 'Deleting from last_position:', cu.fetchone()[0]
cu.execute('DELETE FROM last_position WHERE cg_timestamp < %s;',
(last_position_start, ))
if self.last_position_start is not None or self.track_start is not None:
cx.commit()
def get_container_by_channel(slug,
number=10,
depth=1,
include_children=True,
**kwargs):
box = None
magic_date = kwargs.pop('magic_date', False)
date = timezone.now()
if magic_date:
try:
date = magicdate(magic_date)
except Exception:
pass
if include_children:
try:
kwargs['channel_long_slug__in'] = cache.get(
'get_container_by_channel-{}'.format(slug))
if not kwargs['channel_long_slug__in']:
base_channel = Channel.objects.get(long_slug=slug)
kwargs['channel_long_slug__in'] = [base_channel.long_slug]
def _append_recursivelly(channel, current_level=0):
# Depth test
if current_level >= depth:
return
elif current_level < depth:
current_level += 1
for children in channel.get_children():
kwargs['channel_long_slug__in'].append(
children.long_slug)
# Recursion
_channel = Channel.objects.get(
long_slug=children.long_slug)
_append_recursivelly(_channel, current_level)
_append_recursivelly(base_channel)
cache.set('get_container_by_channel-{}'.format(slug),
kwargs['channel_long_slug__in'],
settings.OPPS_CACHE_EXPIRE)
except Channel.DoesNotExist:
kwargs['channel_long_slug__in'] = []
try:
kwargs['site'] = settings.SITE_ID
if settings.OPPS_CONTAINERS_SITE_ID:
kwargs['site'] = settings.OPPS_CONTAINERS_SITE_ID
kwargs['show_on_root_channel'] = include_children
kwargs['date_available__lte'] = date
kwargs['published'] = True
box = Container.objects.distinct().filter(
**kwargs).order_by('-date_available')[:number]
except:
pass
return box
def do_process(self, input_value):
if isinstance(input_value, tuple):
input_value = input_value[0]
try:
return (int(input_value),)
except ValueError:
try:
datetime = magicdate.magicdate(input_value)
return (datetime.year,)
except:
return None
def do_process(self, input_value):
if isinstance(input_value, tuple):
input_value = input_value[0]
try:
return (int(input_value), )
except ValueError:
try:
datetime = magicdate.magicdate(input_value)
return (datetime.year, )
except:
return None
def get_container_by_channel(slug, number=10, depth=1,
include_children=True, **kwargs):
box = None
magic_date = kwargs.pop('magic_date', False)
date = timezone.now()
if magic_date:
try:
date = magicdate(magic_date)
except Exception:
pass
# __in split treatment
splited = dict([
(key, value.split(','))
for key, value
in kwargs.items()
if key.endswith('__in') and type(value) is not list])
kwargs.update(splited)
if include_children:
k = 'channel_id__in'
kwargs[k] = cache.get(
'get_container_by_channel-{0}'.format(slug))
if not kwargs[k]:
try:
channel = Channel.objects.get(long_slug=slug)
qs = channel.get_descendants(include_self=True)
qs = qs.filter(level__lte=channel.level + depth)
kwargs[k] = \
qs.values_list("id", flat=True)
cache.set(
'get_container_by_channel-{0}'.format(slug),
kwargs[k],
settings.OPPS_CACHE_EXPIRE)
except Channel.DoesNotExist:
kwargs[k] = []
try:
kwargs['site'] = settings.SITE_ID
if settings.OPPS_CONTAINERS_SITE_ID:
kwargs['site'] = settings.OPPS_CONTAINERS_SITE_ID
kwargs['show_on_root_channel'] = include_children
kwargs['date_available__lte'] = date
kwargs['published'] = True
box = Container.objects.distinct().filter(
**kwargs).order_by('-date_available')[:number]
except:
pass
return box
def get_container_by_channel(slug, number=10, depth=1,
include_children=True, **kwargs):
box = None
magic_date = kwargs.pop('magic_date', False)
date = timezone.now()
if magic_date:
try:
date = magicdate(magic_date)
except Exception:
pass
if include_children:
try:
kwargs['channel_long_slug__in'] = cache.get(
'get_container_by_channel-{}'.format(slug))
if not kwargs['channel_long_slug__in']:
base_channel = Channel.objects.get(long_slug=slug)
kwargs['channel_long_slug__in'] = [base_channel.long_slug]
def _append_recursivelly(channel, current_level=0):
# Depth test
if current_level >= depth:
return
elif current_level < depth:
current_level += 1
for children in channel.get_children():
kwargs['channel_long_slug__in'].append(
children.long_slug)
# Recursion
_channel = Channel.objects.get(
long_slug=children.long_slug)
_append_recursivelly(_channel, current_level)
_append_recursivelly(base_channel)
cache.set('get_container_by_channel-{}'.format(slug),
kwargs['channel_long_slug__in'],
settings.OPPS_CACHE_EXPIRE)
except Channel.DoesNotExist:
kwargs['channel_long_slug__in'] = []
try:
kwargs['site'] = settings.SITE_ID
if settings.OPPS_CONTAINERS_SITE_ID:
kwargs['site'] = settings.OPPS_CONTAINERS_SITE_ID
kwargs['show_on_root_channel'] = include_children
kwargs['date_available__lte'] = date
kwargs['published'] = True
box = Container.objects.distinct().filter(
**kwargs).order_by('-date_available')[:number]
except:
pass
return box
def guess(s, parse=True, is_gmt=False, set_gmt=False,
try_iso=True, try_num=True, try_en=True):
"""Guess the format, and optionally parse, the input string.
If 'is_gmt' is True, assume timezone is GMT when not given.
Otherwise, assume localtime.
If 'set_gmt' is True then set the timezone to GMT, otherwise
set it to localtime.
The answer is a pair containing the guessed format and, if the 'parse'
flag was given, the parsed value as seconds since the epoch, otherwise None.
The format is a constant defined in this module:
UNKNOWN - Cannot guess the format (associated value is None)
ISO8601 - This is a prefix of the ISO8601 format accepted by completeISO()
ENGLISH - This is an natural English-language format accepted by makeISO()
SECONDS - This is seconds since the UNIX epoch (Midnight on 1970/1/1).
"""
if not(s):
return UNKNOWN, None
sec = None
s = s.strip()
# try ISO8601
if try_iso:
m = ISO_DATE_PARTS.match(s)
if m and m.start() == 0 and m.end() == len(s):
if parse:
if s[-1] == 'Z':
# explicit timezone overrides option
is_gmt = True
iso_s = completeISO(s, is_gmt=is_gmt, set_gmt=set_gmt)
sec = parseISO(iso_s)
return ISO8601, sec
# try number
if try_num:
m = NUMBER_DATE.match(s)
if m and m.start() == 0 and m.end() == len(s):
if parse:
sec = float(s)
return SECONDS, sec
# try natural language
if try_en:
try:
d = magicdate.magicdate(s)
except Exception, E:
d = None
if d is not None:
if parse:
partial_iso = d.isoformat()
iso = completeISO(partial_iso, is_gmt=False, set_gmt=set_gmt)
sec = parseISO(iso)
return ENGLISH, sec
def get_container_by_channel(slug,
number=10,
depth=1,
include_children=True,
**kwargs):
box = None
magic_date = kwargs.pop('magic_date', False)
date = timezone.now()
if magic_date:
try:
date = magicdate(magic_date)
except Exception:
pass
# __in split treatment
splited = dict([(key, value.split(',')) for key, value in kwargs.items()
if key.endswith('__in') and type(value) is not list])
kwargs.update(splited)
if include_children:
k = 'channel_id__in'
kwargs[k] = cache.get('get_container_by_channel-{0}'.format(slug))
if not kwargs[k]:
try:
channel = Channel.objects.get(long_slug=slug)
qs = channel.get_descendants(include_self=True)
qs = qs.filter(level__lte=channel.level + depth)
kwargs[k] = \
qs.values_list("id", flat=True)
cache.set('get_container_by_channel-{0}'.format(slug),
kwargs[k], settings.OPPS_CACHE_EXPIRE)
except Channel.DoesNotExist:
kwargs[k] = []
try:
kwargs['site'] = settings.SITE_ID
if settings.OPPS_CONTAINERS_SITE_ID:
kwargs['site'] = settings.OPPS_CONTAINERS_SITE_ID
kwargs['show_on_root_channel'] = include_children
kwargs['date_available__lte'] = date
kwargs['published'] = True
box = Container.objects.distinct().filter(
**kwargs).order_by('-date_available')[:number]
except:
pass
return box
def do_once(self):
cx = self.cx
cu = self.cu
v = self.verbose
tzoffset = datetime.timedelta(seconds=time.timezone) # Always work in utc. Magicdate is local
if self.track_start is not None:
track_start = mark_utc(magicdate.magicdate(self.track_start)+tzoffset )
if v:
cu.execute('SELECT COUNT(*) FROM track_lines WHERE update_timestamp < %s;', (track_start,))
print 'Deleting from track_lines:',cu.fetchone()[0]
cu.execute('DELETE FROM track_lines WHERE update_timestamp < %s;', (track_start,))
# if the points are too old to be in a track_line, then delete them
if v:
cu.execute('SELECT COUNT(*) FROM position WHERE cg_timestamp < %s;', (track_start,))
print 'Deleting from position:',cu.fetchone()[0]
cu.execute('DELETE FROM position WHERE cg_timestamp < %s;', (track_start,))
if self.last_position_start is not None:
last_position_start = mark_utc(magicdate.magicdate(self.last_position_start)+tzoffset )
if v:
cu.execute('SELECT COUNT(*) FROM last_position WHERE cg_timestamp < %s;', (track_start,))
print 'Deleting from last_position:',cu.fetchone()[0]
cu.execute('DELETE FROM last_position WHERE cg_timestamp < %s;', (last_position_start,))
if self.last_position_start is not None or self.track_start is not None:
cx.commit()
def main():
from optparse import OptionParser
parser = OptionParser(option_class=magicdate.MagicDateOption,
usage='%prog [options]',
version='%prog '+__version__)
parser.add_option('-s', '--start-date', type='magicdate', default=None, help='Force a start time (magicdate)')
parser.add_option('-e', '--end-date', type='magicdate', default=magicdate.magicdate('today'), help='Force an end time (magicdate) [default: %default]')
parser.add_option('-l', '--log-format', default='%Y%m%d.log', help='datetime strftime format string to create the log file names to search for [default: %default]')
(options,args) = parser.parse_args()
assert options.start_date is not None
for d in date_generator(options.start_date, options.end_date):
#print d.strftime(options.log_format)
filename = d.strftime(options.log_format)
try:
s = os.stat(filename)
except OSError:
print 'missing',filename
def makeISO(value, is_gmt=False, set_gmt=False):
"""If value is a tuple, assume it is the one returned by time.gmtime() or time.localtime()
Otherwise, assume value is an English language description (for partial ISO
strings, use completeISO() instead).
Return an ISO8601 string, with timezone set to GMT or localtime.
"""
tz_str = 'Z' # assume GMT
if isinstance(value,tuple) or isinstance(value,list):
fmt = ("%04d","-%02d","-%02d","T%02d",":%02d", ":%02d")
s = ''.join([f % v for f,v in zip(fmt,value)])
if not gmt:
tz_str = getLocaltimeISO(value)
iso = s + tz_str
else:
try:
d = magicdate.magicdate(value)
except Exception, E:
raise ValueError("magicdate cannot parse '%s'" % value)
partial_iso = d.isoformat()
iso = completeISO(partial_iso, is_gmt=is_gmt, set_gmt=set_gmt)
from boto.s3.key import Key
import magicdate
import gzip
import json
import glob
import logging
from filechunkio import FileChunkIO
import math
import shutil
from pprint import pprint
import subprocess
print "Daily file tody up. Pre S3 push"
# we are bundling the previous days data.
yesterday = magicdate.magicdate('yesterday')
dir_file = gzip.open("/home/TfL_feeds/directory_data.json.gz")
dir_json = json.load(dir_file)
os.chdir(dir_json['home_directory'] + "/data/")
# Define folder name within local directory
folder_name = str(yesterday)
message = "Tarring and compressing " + str(folder_name)
print message
folder_exists = os.path.isdir(folder_name)
folder = (folder_name + ".tar.gz")
print "Working on, ", str(folder)
def __setitem__(self, key, value) :
new_value = value
if key == 'Date' :
new_value = magicdate(value)
self._dict.__setitem__(key, new_value)
def bag2kmlbbox(in_name,
out_file,
title=None,
kml_complete=False,
verbose=False,
placemark=False):
v = verbose
f = h5py.File(in_name) #'H11302_OLS_OSS/H11302_2m_1.bag')
#o = file(out_name,'w')
# FIX: if out_file is a string, then open
o = out_file
#print f.listobjects()
#print f.listitems()
bag_root = f['/BAG_root']
metadata_xml = ''.join(bag_root['metadata'])
#o = file('metadata.xml','w')
#o.write(metadata_xml)
#del o
#root = etree.parse(StringIO(metadata_xml)).getroot()
#root = etree.parse(StringIO(metadata_xml.replace('smXML:',''))).getroot()
root = etree.XML(metadata_xml.replace('smXML:', ''))
xmin = float(root.xpath('//*/westBoundLongitude')[0].text)
xmax = float(root.xpath('//*/eastBoundLongitude')[0].text)
ymin = float(root.xpath('//*/southBoundLatitude')[0].text)
ymax = float(root.xpath('//*/northBoundLatitude')[0].text)
# WARNING: This date does not relate to the dates the survey was collected!
date = root.xpath('//*/CI_Date/date')[0].text
abstract = root.xpath('//*/abstract')[0].text
timestamp = '' # No timestamp if we can't handle it
try:
import datetime, magicdate
adate = magicdate.magicdate(date)
timestamp = '<TimeStamp>' + adate.strftime(
iso8601_timeformat) + '</TimeStamp>'
except:
print 'WARNING: Unable to handle timestamp:', date
if v:
print xmin, xmax, '->', ymin, ymax
print 'date:', date
print 'abstract:', abstract
#import subprocess
#p = subprocess.Popen(
# ['source-highlight','-s', 'xml', '--out-format=html'],
# stdin=subprocess.PIPE,
# stdout=subprocess.PIPE
# )
#metadata_html = p.communicate(input=etree.tostring(root, pretty_print=True ) ) [0]
metadata_html = etree.tostring(root, pretty_print=True).replace(
'</', ' ').replace('<', ' ').replace('>',
' ') #.replace('\n','<br/>\n')
if v: print metadata_html
if not title:
title = '%s : %s' % (abstract, date)
kml_data = {
'title': title,
'x': (xmin + xmax) / 2.,
'y': (ymin + ymax) / 2.,
'xmin': xmin,
'xmax': xmax,
'ymin': ymin,
'ymax': ymax,
'metadata': metadata_html,
'timestamp': timestamp,
}
#o = file('out.kml','w')
if kml_complete:
o.write('''<?xml version="1.0" encoding="UTF-8"?>
<kml xmlns="http://www.opengis.net/kml/2.2" xmlns:gx="http://www.google.com/kml/ext/2.2" xmlns:kml="http://www.opengis.net/kml/2.2" xmlns:atom="http://www.w3.org/2005/Atom">
<Document>''')
if placemark:
o.write('''
<Placemark>
<name>{title}</name>
{timestamp}
<description><![CDATA[
<pre>
{metadata}
</pre>
]]>
</description>
<Point> <coordinates> {x},{y},0 </coordinates> </Point>
</Placemark>
''')
o.write('''
<Placemark>
<name>{title}</name>
{timestamp}
<LineString>
<coordinates>
{xmin},{ymin},0
{xmax},{ymin},0
{xmax},{ymax},0
{xmin},{ymax},0
{xmin},{ymin},0
</coordinates>
</LineString>
</Placemark>
'''.format(**kml_data))
if kml_complete:
o.write('''</Document>
</kml>
''')
return
# Invalid record, skip it.
continue
else:
output_list.append(tmp_dict)
except Exception, e:
continue
return output_list
def earnings_report_date_for_symbol(symbol):
url_string = 'http://biz.yahoo.com/rr/?s=' + symbol + '&d=research%2Fearncal'
print url_string
try:
page = urllib2.urlopen(url_string)
soup = BeautifulSoup(page, 'html.parser')
except Exception, e:
print e
return None
date_element = soup.find_all("font", attrs={'face': 'arial', 'size': '+1'})[0]
date_text = date_element.find('b').string.replace('\n', ' ')
sanitized_date_text = date_text.replace('US Earnings Calendar for ', '')
output_date = str(magicdate(sanitized_date_text))
return output_date
def add_bag_to_db(cx, infile_name, survey, filename_base, verbose=False, write_xml=False):
# filename_base - without .bag or path
#print ('file:',infile_name, file=sys.stderr)
v = verbose
#if v:
# print ('opening:',infile_name,os.path.getsize(infile_name))
f = h5py.File(infile_name) #'H11302_OLS_OSS/H11302_2m_1.bag')
#o = file('foo.out','w')
bag_root = f['/BAG_root']
metadata_xml = ''.join(bag_root['metadata'])
#o = file('metadata.xml','w')
#o.write(metadata_xml)
#del o
#root = etree.parse(StringIO(metadata_xml)).getroot()
#root = etree.parse(StringIO(metadata_xml.replace('smXML:',''))).getroot()
try:
root = etree.XML(metadata_xml.replace('smXML:',''))
except:
print ('bad_metadata:',infile_name) # What can we do?
return # ouch... better if we could try to fix it somehow
x_min_metadata = float(root.xpath('//*/westBoundLongitude')[0].text)
x_max_metadata = float(root.xpath('//*/eastBoundLongitude')[0].text)
y_min_metadata = float(root.xpath('//*/southBoundLatitude')[0].text)
y_max_metadata = float(root.xpath('//*/northBoundLatitude')[0].text)
software = root.xpath('//*/BAG_ProcessStep/description')[0].text
#print ('software:',software)
utm_zone = int(root.xpath('//*/zone')[0].text)
# The WGS84 geographic is often foulded up.
utm_coords = root.xpath('//*/gml:coordinates', namespaces={'gml':"http://www.opengis.net/gml"})[0].text
#print ('\t',utm_coords)
utm_coords = utm_coords.split()
utm_x_min,utm_y_min = [float(coord) for coord in utm_coords[0].split(',')]
utm_x_max,utm_y_max = [float(coord) for coord in utm_coords[1].split(',')]
params = {'proj':'utm', 'zone':utm_zone}
proj = Proj(params)
x_min,y_min = proj(utm_x_min,utm_y_min, inverse=True)
x_max,y_max = proj(utm_x_max,utm_y_max, inverse=True)
#print ('\t',utm_x_min,utm_y_min, utm_x_max,utm_y_max)
#print ('\t\t',x_min,y_min,x_max,y_max)
#print ('\t\t',x_min_metadata,y_min_metadata,x_max_metadata,y_max_metadata)
if abs(x_max - x_max_metadata) > 0.05 or abs(y_max - y_max_metadata) > 0.05:
print ('%s: %.4f %.4f %.4f %.4f' % (filename_base,
x_min - x_min_metadata,y_min - y_min_metadata,
x_max - x_max_metadata,y_max - y_max_metadata)
)
vdatum = None
datums = [entry.text.strip() for entry in root.xpath('//*/datum/RS_Identifier/code')]
if len(datums)==0:
pass
elif 'MLLW' in datums: vdatum = 'MLLW'
else:
vdatum = datums[-1] # just guess that it is the last one
print('datums:',datums,'->',vdatum,filename_base)
axes = (root.xpath('//*/axisDimensionProperties'))
dx = dy = None
width = height = None
for axis in axes:
#print(etree.tostring(axis, pretty_print=True))
dim_name = axis.xpath('*/dimensionName')[0].text
dim_size = int(axis.xpath('*/dimensionSize')[0].text)
delta = float(axis.xpath('*/*/*/value')[0].text)
#print ('dim_name: "%s"' % (dim_name,))
if 'row' == dim_name:
dy = delta
height = dim_size
elif 'column' == dim_name:
dx = delta
width = dim_size
else:
print ('ERROR: unable to handle dim',dim_name)
assert False
# WARNING: This date does not relate to the dates the survey was collected!
date = root.xpath('//*/CI_Date/date')[0].text
abstract = root.xpath('//*/abstract')[0].text
title = root.xpath('//*/title')[0].text
timestamp = '' # No timestamp if we can't handle it
try:
import datetime, magicdate
#timestamp = magicdate.magicdate(date)
creation = magicdate.magicdate(date)
#timestamp = adate.strftime(iso8601_timeformat)
except:
print ('WARNING: Unable to handle timestamp:',date)
creation = None
# if v:
# print (x_min,x_max,'->',y_min,y_max)
# print ('date:',date)
# print ('abstract:',abstract)
metadata_txt = etree.tostring(root, pretty_print=True ).replace('</',' ').replace('<',' ').replace('>',' ') #.replace('\n','<br/>\n')
# FIX: base url must change based on the number of the survey
base_url = 'http://surveys.ngdc.noaa.gov/mgg/NOS/coast/H10001-H12000/'
dr_url = base_url + survey + '/DR/' + survey + '.pdf'
bag_url = base_url + survey + '/BAG/' + filename_base + '.bag.gz'
sql_field_names = ('file', 'survey', 'title','abstract', 'survey', 'creation', 'x_min', 'y_min', 'x_max', 'y_max', 'width', 'height', 'dx', 'dy', 'vdatum', 'utm_zone', 'dr_url', 'bag_url', 'metadata_txt','metadata_xml', 'utm_x_min','utm_y_min' , 'utm_x_max' ,'utm_y_max', 'software')
file = filename_base
# check for errors
#for field in sql_field_names:
# print('%s:' % (field,) ,locals()[field])
sql_insert = 'INSERT INTO bag (' + ','.join(sql_field_names) + ') VALUES (' + ', '.join([':%s' %(field,) for field in sql_field_names ]) + ');'
#print (bag_data)
#print (sql_insert)
cx.execute(sql_insert,locals()) # Passing locals sees crazy
cx.commit()
def bag2kmlbbox(in_name, out_file, title=None, kml_complete=False, verbose=False, placemark=False):
v = verbose
f = h5py.File(in_name) #'H11302_OLS_OSS/H11302_2m_1.bag')
# o = file(out_name,'w')
# FIX: if out_file is a string, then open
o = out_file
# print f.listobjects()
# print f.listitems()
bag_root = f["/BAG_root"]
metadata_xml = "".join(bag_root["metadata"])
# o = file('metadata.xml','w')
# o.write(metadata_xml)
# del o
# root = etree.parse(StringIO(metadata_xml)).getroot()
# root = etree.parse(StringIO(metadata_xml.replace('smXML:',''))).getroot()
root = etree.XML(metadata_xml.replace("smXML:", ""))
xmin = float(root.xpath("//*/westBoundLongitude")[0].text)
xmax = float(root.xpath("//*/eastBoundLongitude")[0].text)
ymin = float(root.xpath("//*/southBoundLatitude")[0].text)
ymax = float(root.xpath("//*/northBoundLatitude")[0].text)
# WARNING: This date does not relate to the dates the survey was collected!
date = root.xpath("//*/CI_Date/date")[0].text
abstract = root.xpath("//*/abstract")[0].text
timestamp = "" # No timestamp if we can't handle it
try:
import datetime, magicdate
adate = magicdate.magicdate(date)
timestamp = "<TimeStamp>" + adate.strftime(iso8601_timeformat) + "</TimeStamp>"
except:
print "WARNING: Unable to handle timestamp:", date
if v:
print xmin, xmax, "->", ymin, ymax
print "date:", date
print "abstract:", abstract
# import subprocess
# p = subprocess.Popen(
# ['source-highlight','-s', 'xml', '--out-format=html'],
# stdin=subprocess.PIPE,
# stdout=subprocess.PIPE
# )
# metadata_html = p.communicate(input=etree.tostring(root, pretty_print=True ) ) [0]
metadata_html = (
etree.tostring(root, pretty_print=True).replace("</", " ").replace("<", " ").replace(">", " ")
) # .replace('\n','<br/>\n')
if v:
print metadata_html
if not title:
title = "%s : %s" % (abstract, date)
kml_data = {
"title": title,
"x": (xmin + xmax) / 2.0,
"y": (ymin + ymax) / 2.0,
"xmin": xmin,
"xmax": xmax,
"ymin": ymin,
"ymax": ymax,
"metadata": metadata_html,
"timestamp": timestamp,
}
# o = file('out.kml','w')
if kml_complete:
o.write(
"""<?xml version="1.0" encoding="UTF-8"?>
<kml xmlns="http://www.opengis.net/kml/2.2" xmlns:gx="http://www.google.com/kml/ext/2.2" xmlns:kml="http://www.opengis.net/kml/2.2" xmlns:atom="http://www.w3.org/2005/Atom">
<Document>"""
)
if placemark:
o.write(
"""
<Placemark>
<name>{title}</name>
{timestamp}
<description><![CDATA[
<pre>
{metadata}
</pre>
]]>
</description>
<Point> <coordinates> {x},{y},0 </coordinates> </Point>
</Placemark>
"""
)
o.write(
"""
<Placemark>
<name>{title}</name>
{timestamp}
<LineString>
<coordinates>
{xmin},{ymin},0
{xmax},{ymin},0
{xmax},{ymax},0
{xmin},{ymax},0
{xmin},{ymin},0
</coordinates>
</LineString>
</Placemark>
""".format(
**kml_data
)
)
if kml_complete:
o.write(
"""</Document>
</kml>
"""
)
return
def add_bag_to_db(cx,
infile_name,
survey,
filename_base,
verbose=False,
write_xml=False):
# filename_base - without .bag or path
#print ('file:',infile_name, file=sys.stderr)
v = verbose
#if v:
# print ('opening:',infile_name,os.path.getsize(infile_name))
f = h5py.File(infile_name) #'H11302_OLS_OSS/H11302_2m_1.bag')
#o = file('foo.out','w')
bag_root = f['/BAG_root']
metadata_xml = ''.join(bag_root['metadata'])
#o = file('metadata.xml','w')
#o.write(metadata_xml)
#del o
#root = etree.parse(StringIO(metadata_xml)).getroot()
#root = etree.parse(StringIO(metadata_xml.replace('smXML:',''))).getroot()
try:
root = etree.XML(metadata_xml.replace('smXML:', ''))
except:
print('bad_metadata:', infile_name) # What can we do?
return # ouch... better if we could try to fix it somehow
x_min_metadata = float(root.xpath('//*/westBoundLongitude')[0].text)
x_max_metadata = float(root.xpath('//*/eastBoundLongitude')[0].text)
y_min_metadata = float(root.xpath('//*/southBoundLatitude')[0].text)
y_max_metadata = float(root.xpath('//*/northBoundLatitude')[0].text)
software = root.xpath('//*/BAG_ProcessStep/description')[0].text
#print ('software:',software)
utm_zone = int(root.xpath('//*/zone')[0].text)
# The WGS84 geographic is often foulded up.
utm_coords = root.xpath('//*/gml:coordinates',
namespaces={'gml':
"http://www.opengis.net/gml"})[0].text
#print ('\t',utm_coords)
utm_coords = utm_coords.split()
utm_x_min, utm_y_min = [float(coord) for coord in utm_coords[0].split(',')]
utm_x_max, utm_y_max = [float(coord) for coord in utm_coords[1].split(',')]
params = {'proj': 'utm', 'zone': utm_zone}
proj = Proj(params)
x_min, y_min = proj(utm_x_min, utm_y_min, inverse=True)
x_max, y_max = proj(utm_x_max, utm_y_max, inverse=True)
#print ('\t',utm_x_min,utm_y_min, utm_x_max,utm_y_max)
#print ('\t\t',x_min,y_min,x_max,y_max)
#print ('\t\t',x_min_metadata,y_min_metadata,x_max_metadata,y_max_metadata)
if abs(x_max - x_max_metadata) > 0.05 or abs(y_max -
y_max_metadata) > 0.05:
print('%s: %.4f %.4f %.4f %.4f' %
(filename_base, x_min - x_min_metadata, y_min - y_min_metadata,
x_max - x_max_metadata, y_max - y_max_metadata))
vdatum = None
datums = [
entry.text.strip()
for entry in root.xpath('//*/datum/RS_Identifier/code')
]
if len(datums) == 0:
pass
elif 'MLLW' in datums:
vdatum = 'MLLW'
else:
vdatum = datums[-1] # just guess that it is the last one
print('datums:', datums, '->', vdatum, filename_base)
axes = (root.xpath('//*/axisDimensionProperties'))
dx = dy = None
width = height = None
for axis in axes:
#print(etree.tostring(axis, pretty_print=True))
dim_name = axis.xpath('*/dimensionName')[0].text
dim_size = int(axis.xpath('*/dimensionSize')[0].text)
delta = float(axis.xpath('*/*/*/value')[0].text)
#print ('dim_name: "%s"' % (dim_name,))
if 'row' == dim_name:
dy = delta
height = dim_size
elif 'column' == dim_name:
dx = delta
width = dim_size
else:
print('ERROR: unable to handle dim', dim_name)
assert False
# WARNING: This date does not relate to the dates the survey was collected!
date = root.xpath('//*/CI_Date/date')[0].text
abstract = root.xpath('//*/abstract')[0].text
title = root.xpath('//*/title')[0].text
timestamp = '' # No timestamp if we can't handle it
try:
import datetime, magicdate
#timestamp = magicdate.magicdate(date)
creation = magicdate.magicdate(date)
#timestamp = adate.strftime(iso8601_timeformat)
except:
print('WARNING: Unable to handle timestamp:', date)
creation = None
# if v:
# print (x_min,x_max,'->',y_min,y_max)
# print ('date:',date)
# print ('abstract:',abstract)
metadata_txt = etree.tostring(root, pretty_print=True).replace(
'</', ' ').replace('<', ' ').replace('>',
' ') #.replace('\n','<br/>\n')
# FIX: base url must change based on the number of the survey
base_url = 'http://surveys.ngdc.noaa.gov/mgg/NOS/coast/H10001-H12000/'
dr_url = base_url + survey + '/DR/' + survey + '.pdf'
bag_url = base_url + survey + '/BAG/' + filename_base + '.bag.gz'
sql_field_names = ('file', 'survey', 'title', 'abstract', 'survey',
'creation', 'x_min', 'y_min', 'x_max', 'y_max', 'width',
'height', 'dx', 'dy', 'vdatum', 'utm_zone', 'dr_url',
'bag_url', 'metadata_txt', 'metadata_xml', 'utm_x_min',
'utm_y_min', 'utm_x_max', 'utm_y_max', 'software')
file = filename_base
# check for errors
#for field in sql_field_names:
# print('%s:' % (field,) ,locals()[field])
sql_insert = 'INSERT INTO bag (' + ','.join(
sql_field_names) + ') VALUES (' + ', '.join(
[':%s' % (field, ) for field in sql_field_names]) + ');'
#print (bag_data)
#print (sql_insert)
cx.execute(sql_insert, locals()) # Passing locals sees crazy
cx.commit()
