1 |
#!/usr/bin/python |
---|
2 |
""" |
---|
3 |
Module to handle raw sodar data samples. |
---|
4 |
|
---|
5 |
Raw sodar data samples are collected into daily files. Each sample consists of a |
---|
6 |
header followed by an observation for each height. |
---|
7 |
|
---|
8 |
The daily file is split into a list (modeled by the class RawData) of samples |
---|
9 |
(modeled by the class Sample) in chronological order. A RawData object is |
---|
10 |
initialized with a string representing the daily file data: |
---|
11 |
|
---|
12 |
rawDataHandle = open('20070601.dat') |
---|
13 |
rawDataString = rawDataHandle.read() |
---|
14 |
rawDataObject = RawData(rawDataString) |
---|
15 |
|
---|
16 |
Each Sample object has attributes for a Header and Body object. The Samples |
---|
17 |
within a RawData object may also be accessed by time using a string of the |
---|
18 |
format YYYY-MM-DD-HH-MM as in index on the RawData object to return the first |
---|
19 |
matching Sample in the RawData object: |
---|
20 |
|
---|
21 |
rawDataObject[0] # the first Sample object of the day |
---|
22 |
rawDataObject['2007-06-01-09-15'] # the Sample object for 9:15am |
---|
23 |
rawDataObject[15].header # the Header object of the 16th Sample |
---|
24 |
rawDataObject['2007-06-01-09-15'].body # the Body object for 9:15am |
---|
25 |
|
---|
26 |
Header objects act as dictionaries. Access each sample-wide parameter of |
---|
27 |
interest using the header parameter name as a keyword on the Header object: |
---|
28 |
|
---|
29 |
rawDataObject[15].header['VAL2'] # the number of validations for beam 2 |
---|
30 |
rawDataObject['2007-06-01-09-15'].header['SPU3'] # normalized false signal |
---|
31 |
# probability on beam 3 |
---|
32 |
rawDataObject[0].header['SNR1'] # signal to noise on beam 1 |
---|
33 |
|
---|
34 |
Consult your Sodar documentation for a complete list of header parameters. |
---|
35 |
Different sodar models have different sets of header parameters. This model |
---|
36 |
seeks to be model agnostic, and parses the header parameter names from the |
---|
37 |
raw data itself. |
---|
38 |
|
---|
39 |
Body objects act as lists of dictionaries. The dictionaries access |
---|
40 |
altitude-specific parameters by name as keywords. The dictionaries are in |
---|
41 |
altitude-ascending order. Each dictionary may also by accessed by indexing with |
---|
42 |
an altitude string: |
---|
43 |
|
---|
44 |
rawDataObject[15].body[0] # the data for the lowest altitude, 16th sample |
---|
45 |
rawDataObject['2007-06-01-09-15'].body['70'] # the data for 70 meters |
---|
46 |
rawDataObject[15].body[0]['SPEED'] # wind speed at lowest altitude |
---|
47 |
rawDataObject['2007-06-01-09-15'].body['70']['DIR'] # wind direction |
---|
48 |
# at 70 meters |
---|
49 |
|
---|
50 |
The body attribute of a Sample object may also be indexed directly on a Sample |
---|
51 |
object for the most convenient semantics: |
---|
52 |
|
---|
53 |
rawDataObject[15][0]['SPEED'] # wind speed at lowest altitude, 16th sample |
---|
54 |
rawDataObject['2007-06-01-09-15']['70']['DIR'] # wind direction, |
---|
55 |
# 70 meters, 9:15am |
---|
56 |
""" |
---|
57 |
|
---|
58 |
__author__ = 'Chris Calloway' |
---|
59 |
__email__ = 'cbc@unc.edu' |
---|
60 |
__copyright__ = 'Copyright 2007 UNC-CH Department of Marine Science' |
---|
61 |
__license__ = 'GPL2' |
---|
62 |
|
---|
63 |
import re |
---|
64 |
|
---|
65 |
class RawData(list): |
---|
66 |
|
---|
67 |
"""Class to handle raw daily sodar file data. |
---|
68 |
|
---|
69 |
A chronologically ordered list of samples. |
---|
70 |
""" |
---|
71 |
|
---|
72 |
def __init__(self, data): |
---|
73 |
"""Parse raw daily sodar file data.""" |
---|
74 |
super(RawData, self).__init__() |
---|
75 |
# Divide daily string into list of Samples separated by $. |
---|
76 |
self.extend([Sample(sample) |
---|
77 |
for sample in |
---|
78 |
[sample.strip() for sample in data.split('$')] |
---|
79 |
if sample.strip()]) |
---|
80 |
|
---|
81 |
def __getitem__(self, index): |
---|
82 |
"""Allow sample retrieval by Sample time in header.""" |
---|
83 |
try: |
---|
84 |
return super(RawData,self).__getitem__(index) |
---|
85 |
except TypeError: |
---|
86 |
return self._find(index) |
---|
87 |
|
---|
88 |
def _find(self, index): |
---|
89 |
"""Find Sample in RawData. |
---|
90 |
|
---|
91 |
Where sample time of form YYYY-MM-DD-HH-MM. |
---|
92 |
""" |
---|
93 |
|
---|
94 |
try: |
---|
95 |
year,month,day,hour,minute = index.split('-') |
---|
96 |
except ValueError: |
---|
97 |
raise ValueError('RawData index by date must be '\ |
---|
98 |
'"YYYY-MM-DD-HH-MM"') |
---|
99 |
except AttributeError: |
---|
100 |
raise AttributeError('RawData index by date must be '\ |
---|
101 |
'"YYYY-MM-DD-HH-MM"') |
---|
102 |
for sample in self: |
---|
103 |
try: |
---|
104 |
if sample.header['YEAR'].rjust(4,'0') != year: continue |
---|
105 |
if sample.header['MONTH'].rjust(2,'0') != month: continue |
---|
106 |
if sample.header['DAY'].rjust(2,'0') != day: continue |
---|
107 |
if sample.header['HOUR'].rjust(2,'0') != hour: continue |
---|
108 |
if sample.header['MIN'].rjust(2,'0') != minute: continue |
---|
109 |
return sample |
---|
110 |
except TypeError: # sample.header may not exist |
---|
111 |
continue |
---|
112 |
raise IndexError('RawData index out of range') |
---|
113 |
|
---|
114 |
|
---|
115 |
class Sample(object): |
---|
116 |
|
---|
117 |
"""A single sample from raw daily sodar file data. |
---|
118 |
|
---|
119 |
(A header and a body attribute.) |
---|
120 |
""" |
---|
121 |
|
---|
122 |
def __init__(self,sample): |
---|
123 |
"""Separate Sample into Header and Body objects.""" |
---|
124 |
super(Sample, self).__init__() |
---|
125 |
# first three groups of lines are the header; rest is body |
---|
126 |
samplePattern = re.compile(r'''(?P<header>.*?\n\n.*?\n\n.*?\n\n) |
---|
127 |
(?P<body>.*$) |
---|
128 |
''',re.DOTALL | re.VERBOSE) |
---|
129 |
self.__dict__.update(samplePattern.match(sample.strip()).groupdict()) |
---|
130 |
# getattr with default covers parsing invalid Samples |
---|
131 |
self.header = getattr(self, 'header', None) |
---|
132 |
if self.header is not None: |
---|
133 |
self.header = Header(self.header) |
---|
134 |
self.body = getattr(self, 'body', None) |
---|
135 |
if self.body is not None: |
---|
136 |
self.body = Body(self.body) |
---|
137 |
|
---|
138 |
def __getitem__(self, index): |
---|
139 |
"""Index Sample by body attribute.""" |
---|
140 |
try: |
---|
141 |
return self.body[index] |
---|
142 |
except TypeError: # sample.body may not exist |
---|
143 |
raise IndexError('Sample index out of range') |
---|
144 |
|
---|
145 |
def data(self): |
---|
146 |
"""Create a deep copy as a dictionary of header and body data.""" |
---|
147 |
return {'header':self.header.data(), |
---|
148 |
'body':self.body.data()} |
---|
149 |
|
---|
150 |
|
---|
151 |
class Header(dict): |
---|
152 |
|
---|
153 |
"""A raw sodar data sample header. |
---|
154 |
|
---|
155 |
(A dictionary of sample-wide parameters.) |
---|
156 |
""" |
---|
157 |
|
---|
158 |
def __init__(self, header): |
---|
159 |
|
---|
160 |
"""Identify discreet header parameter names and values. |
---|
161 |
|
---|
162 |
Every other line contains parameter keys; |
---|
163 |
every other line contains parameter values. |
---|
164 |
""" |
---|
165 |
|
---|
166 |
super(Header, self).__init__() |
---|
167 |
headerLines = [headerLine.strip() |
---|
168 |
for headerLine in header.split('\n') |
---|
169 |
if headerLine.strip()] |
---|
170 |
#fix for bad match between names and values |
---|
171 |
self.update(dict(zip(" ".join(headerLines[::2]).split(), |
---|
172 |
" ".join(headerLines[1::2]).split()))) |
---|
173 |
|
---|
174 |
def data(self): |
---|
175 |
"""Create a deep/shallow copy of the data as a dictionary.""" |
---|
176 |
return self.copy() |
---|
177 |
|
---|
178 |
|
---|
179 |
class Body(list): |
---|
180 |
|
---|
181 |
"""A raw sodar data sample body. |
---|
182 |
|
---|
183 |
(A list of dictionariess at each altitude.) |
---|
184 |
""" |
---|
185 |
|
---|
186 |
def __init__(self, body): |
---|
187 |
|
---|
188 |
"""Identify discreet body parameter names and values. |
---|
189 |
|
---|
190 |
The first line contains parameter keys; |
---|
191 |
the remaining lines contains parameter values, |
---|
192 |
one set of parameters for a single altitude per line. |
---|
193 |
""" |
---|
194 |
|
---|
195 |
super(Body, self).__init__() |
---|
196 |
bodyLines = [bodyLine.strip() |
---|
197 |
for bodyLine in body.split('\n') |
---|
198 |
if bodyLine.strip()] |
---|
199 |
bodyKeys = bodyLines[0].split() |
---|
200 |
#fix for bad match between names and values |
---|
201 |
self.extend([dict(zip(bodyKeys, bodyLine.split())) |
---|
202 |
for bodyLine in bodyLines[1:]]) |
---|
203 |
self.reverse() |
---|
204 |
|
---|
205 |
def __getitem__(self, index): |
---|
206 |
"""Return raw altitude data by altitude string.""" |
---|
207 |
try: |
---|
208 |
return super(Body, self).__getitem__(index) |
---|
209 |
except TypeError: |
---|
210 |
return self._find(index) |
---|
211 |
|
---|
212 |
def _find(self, index): |
---|
213 |
"""Find raw altitude data in Body.""" |
---|
214 |
for altitude in self: |
---|
215 |
try: |
---|
216 |
if altitude['ALT'] != index: continue |
---|
217 |
except KeyError: |
---|
218 |
continue |
---|
219 |
return altitude |
---|
220 |
raise IndexError('Body index, out of range') |
---|
221 |
|
---|
222 |
def data(self): |
---|
223 |
"""Create a deep copy of the data as a list of dictionaries.""" |
---|
224 |
return [altitude.copy() for altitude in self] |
---|
225 |
|
---|
226 |
|
---|
227 |
def _main(): |
---|
228 |
"""Process as script from command line.""" |
---|
229 |
import urllib2 |
---|
230 |
try: |
---|
231 |
rawDataHandle = urllib2.urlopen('http://nemo.isis.unc.edu/'\ |
---|
232 |
'data/nccoos/level0/ims/sodar/'\ |
---|
233 |
'2008_01/20080101.dat') |
---|
234 |
rawDataString = rawDataHandle.read() |
---|
235 |
rawDataHandle.close() |
---|
236 |
except: |
---|
237 |
raise IOError("Failure to read raw test data") |
---|
238 |
rawDataObject = RawData(rawDataString) |
---|
239 |
print rawDataObject['2008-01-01-09-15']['70']['SPEED'] |
---|
240 |
|
---|
241 |
if __name__ == "__main__": |
---|
242 |
_main() |
---|