source: nappy/trunk/nappy/nappy_api.py @ 3628

Subversion URL: http://proj.badc.rl.ac.uk/svn/ndg/nappy/trunk/nappy/nappy_api.py@3628
Revision 3628, 16.9 KB checked in by astephen, 12 years ago (diff)

Fixing defaults for delimiter and float format throughout.

Line 
1"""
2nappy_api.py
3============
4
5Top-level API module that allows user to access most of the useful stuff in
6nappy. API examples:
7
8 1. Working with NASA Ames file objects
9 2. Converting between formats (NASA Ames, NetCDF and CSV)
10 3. Comparing NASA Ames files (and/or CSV files)
11 4. General NASA Ames utilities
12
13 1. Working with NASA Ames file objects
14
15# Start python interactive shell
16$ python
17
18# Import the nappy package
19import nappy
20
21# Let's open a NASA Ames file and examine its contents
22f = nappy.openNAFile("data_files/2010.na")
23
24# Get number of header lines
25n_lines = f.getNumHeaderLines()
26
27# Get Organisation from header
28org = f.getOrg()
29# Get the Normal Comments (SCOM) lines.
30norm_comms = f.getNormalComments()
31
32# Get the Special Comments (SCOM) lines.
33spec_comms = f.getSpecialComments()
34
35# Get a list of metadata for all main (non-auxiliary or independent) variables
36var_list = getVariables()
37
38# Get Auxiliary variable metadata for auxiliary variable number 2
39(variable, units, miss, scale) = f.getAuxVariable(2)
40
41# Get scale factor for primary variable number 3
42scale_factor = f.getScaleFactor(3)
43
44# Get missing value for primary variable number 1
45missing = f.getMissingValue(1)
46
47# Let's get the contents dictionary of the whole file
48na_dict = f.getNADict()
49
50# Let's write the na_dict object to a new NASA Ames file
51fout = openNAFile("test_outputs/mytest.na", mode="w", na_dict=na_dict)
52fout.write()
53fout.close()
54
55 2. Converting between formats (NASA Ames, NetCDF and CSV)
56
57# Let's convert a NASA Ames file into a NetCDF file, and add some of our own global attributes
58glob_atts = {"Project": "Really important scientific project involving worms",
59             "Errata": "I meant worm holes!"}
60na_file = "data_files/1020.na"
61nc_file = "test_outputs/try_1020.nc"
62nappy.convertNAToNC(na_file, nc_file, global_attributes=glob_atts)
63
64# Let's convert a NASA Ames file to a CSV and add an annotation column to explain the header
65nappy.convertNAToCSV(na_file, annotation=True)
66
67# Let's read a NetCDF and write one (or more) output NASA Ames files,
68# but only including and variables "temp" and "ozone". Also let's write
69# the output using tabs as the delimiters and a float format of "%6.3f".
70nappy.convertNCToNA("data_files/test1.nc", "test_outputs/test1nc.na",
71              var_ids=("temp", "ozone"), delimiter="\t", float_format="%6.3f")
72 
73# Let's convert a NetCDF file to one (or more) CSV files and don't write the header at all
74nappy.convertNCToCSV("data_files/test1.nc", "test_outputs/test1nc_no_header.csv",
75                     no_header=True)
76
77# Let's take some in-memory CDMS objects and write them to one, or more, NASA Ames file(s).
78# We need to give it a list of cdms variables and a global attributes dictionary.
79# We also want to instruct nappy to overwrite the content of its
80# MNAME (Mission Name) header line with our specific mission name.
81# Also, tell nappy to write the output to NASA Ames file format index (FFI) 2310
82# because we know it is compatible.
83nappy.convertCDMSObjectsToNA([cdms_var_1, cdms_var_2], {"Institute": "British Atmospheric Data Centre"},
84              na_file="test_outputs/cdms_to_na.na",
85              na_items_to_override={"MNAME": "Atlantic Divergence Mission 2009"},
86              requested_ffi=2310)
87
88# Let's take a list of cdms variables and a global attributes dictionary and write
89# them to a CSV file.
90nappy.convertCDMSObjectsToCSV(cdms_vars, global_atts_dict, csv_file)
91
92# Let's take a NASA Ames dictionary object, and write it to a NetCDF file
93nappy.writeNADictToNC(na_dict, nc_file, mode="w")
94
95# Let's try and write a second na_dict object to the same NetCDF file using mode="a".
96nappy.writeNADictToNC(na_dict_2, nc_file, mode="a")
97
98# Now let's read in a NASA Ames file and convert the contents in-memory into
99# CDMS objects so that we can manipulate them with NetCDF-compatible tools
100(cdms_vars_primary, cdms_vars_aux, global_attributes) = nappy.readCDMSObjectsFromNA(na_file)
101
102# Actually, I only want to get a single variable from that file, so I'll try
103temp_var = getCDMSVariableFromNA(na_file, "temperature")
104
105 3. Comparing NASA Ames files (and/or CSV files)
106
107# I'd like to compare a NASA Ames and CSV file to check they are the same.
108# It will allow for different formatting of numbers as long as the values
109# are the same. Compare both header and body by setting as True (default).
110result = nappy.compareNA(na_file, csv_file, header=True, body=True,
111            number_clever=True, delimiter_1="    ", delimiter_2=",")
112
113 4. General NASA Ames utilities
114
115# Get the FFI from a NASA Ames file
116ffi = nappy.readFFI(na_file)
117
118# Given a NASA Ames dictionary (na_dict) get an appropriate FFI.
119ffi = nappy.chooseFFI(na_dict)
120
121
122"""
123
124# Import standard library modules
125
126# Import third-party software
127try:
128    import cdms
129except:
130    print "WARNING: You cannot use NAPpy's NetCDF conversion tools as your system does not have CDMS installed, or it is not in your sys.path."
131    cdms = False
132
133# Import local modules
134import nappy.utils.common_utils
135import nappy.utils.compare_na
136
137# Bring some utils into the API
138compareNA = nappy.utils.compare_na.compareNA
139readFFI = nappy.utils.common_utils.readFFI
140chooseFFI = nappy.utils.common_utils.chooseFFI
141getNAFileClass = nappy.utils.common_utils.getNAFileClass
142__version__ = nappy.utils.common_utils.getVersion()
143
144
145def openNAFile(filename, mode="r", na_dict=None):
146    """
147    Function wrapper around the NASA Ames File classes. Any NASA Ames
148    file can be opened through this function and the appropriate read or
149    write NASA Ames File class instance is returned.
150    """
151    if mode == "r":
152        ffi = readFFI(filename)
153        return apply(getNAFileClass(ffi), (filename, mode))
154
155    elif mode == "w":
156        if na_dict.has_key('FFI') and type(na_dict['FFI']) == type(3):
157            ffi = na_dict['FFI']
158        else:
159            ffi = chooseFFI(na_dict)
160            na_dict['FFI'] = ffi
161            print "\nFormat identified as:", ffi   
162        return apply(getNAFileClass(ffi), (filename,), {"mode":mode, "na_dict":na_dict})
163    else:
164        raise Exception("File mode not recognised '" + mode + "'.")
165
166
167def convertNAToNC(na_file, nc_file=None, mode="w", variables=None, aux_variables=None,
168                 global_attributes={"Conventions":"CF-1.0"},
169                 time_units=None, time_warning=True,
170                 rename_variables={}):
171    """
172    Takes a NASA Ames file and converts to a NetCDF file. Options are:
173
174    na_file - the input NASA Ames file.
175    nc_file - name for the output NetCDF file (default is to replace ".na" from NASA Ames
176              file with ".nc").
177    mode - is the file mode, either "w" for write or "a" for append
178    variables - is a list of variable names that you wish to be converted. If not set then
179              nappy will attempt to convert all files.
180    aux_var_list - is a list of auxiliary variables names that you wish to be converted.
181              If not set then nappy will use any compatible variables it finds as
182              auxiliary variables.
183    global_attributes - is a dictionary of global attributes to add to the output file.
184    rename_variables - is a dictionary of {old_name: new_name} variable ID pairs that nappy
185              should use to rename variables before it writes them to file. 
186    time_units - is a valid time units string such as "hours since 2003-04-30 10:00:00" to
187              use for time units if there is a valid time axis.
188    time_warning - suppresses the time units warning for invalid time units if set to False.
189    """
190    arg_dict = vars()
191    for arg_out in ("nc_file", "mode"):
192        del arg_dict[arg_out]
193
194    convertor = apply(nappy.nc_convertor.na_to_nc.NAToNC, [], arg_dict)
195    convertor.convert()
196    if nc_file == None:
197        nc_file = nappy.utils.getFileNameWithNewExtension(na_file, "nc")
198    convertor.writeNCFile(nc_file, mode)
199    print "SHould this return nc file path?"
200    return True   
201 
202
203def convertNAToCSV(na_file, csv_file=None, annotation=False, no_header=False):
204    """
205    Reads in a NASA Ames file and writes it out a new CSV file which is identical to the
206    input file except that commas are used as the delimiter. Arguments are:
207
208    na_file - NASA Ames file path
209    csv_file - CSV file path (default is to replace ".na" from NASA Ames file with ".csv").
210    annotation - if set to True write the output file with an additional left-hand column
211                 describing the contents of each header line.
212    no_header - if set to True then only the data blocks are written to file.
213    """
214    fin = openNAFile(na_file)
215    fin.readData()
216    na_dict = fin.getNADict()
217    fin.close()
218
219    if csv_file == None:
220        csv_file = nappy.utils.getFileNameWithNewExtension(nc_file, "csv")
221    fout = openNAFile(csv_file, "w", na_dict=na_dict)
222    fout.write(delimiter=",", annotation=annotation)
223    fout.close()
224    return True
225
226
227def convertNCToNA(nc_file, na_file=None, var_ids=None, na_items_to_override={},
228            only_return_file_names=False, exclude_vars=[],
229            requested_ffi=None, delimiter=default_delimiter, float_format=default_float_format, 
230            size_limit=None, annotation=False, no_header=False):
231    """
232    Takes a NetCDF file and converts the contents to one or more NASA Ames files.
233    Arguments are:
234
235    nc_file - is the name of input file (NetCDF).
236    na_file - is the name of output file (default is to replace ".nc" from NASA Ames
237              file with ".na"). If multiple files produced then this name will be used
238              as the base name.
239    var_ids - is a list of variables (as ids) to include in the output file(s).
240    na_items_to_override - is a dictionary of {key: value} pairs to overwrite in output
241              files. Typically the keys are in: 
242              ("DATE", "RDATE", "ANAME", "MNAME","ONAME", "ORG", "SNAME", "VNAME".)
243    only_return_file_names - if set to True then only return a list of file names that
244              would be written (i.e. don't convert actual file).
245    exclude_vars - is a list of variables (as ids) to exclude in the output file(s).
246    requested_ffi - is the NASA Ames File Format Index (FFI) you wish to write to. Note
247              that there are only limited options available depending on the data
248              structures found.
249    delimiter - the delimiter you wish to use between data items in the output file such
250              as "   ", "\t" or ",".
251    float_format - a python formatting string such as "%s", "%g" or "%5.2f" used for
252              formatting floats when written to file.
253    size_limit - if format FFI is 1001 then chop files up into size_limit rows of data.
254    annotation - if set to True write the output file with an additional left-hand column
255              describing the contents of each header line.
256    no_header - if set to True then only the data blocks are written to file.
257    """
258    arg_dict = vars()
259    for arg_out in ("na_file", "only_return_file_names", "delimiter", "float_format", 
260                    "size_limit", "annotation", "no_header"):
261        del arg_dict[arg_out]
262
263    if na_file == None:
264        na_file =  nappy.utils.getFileNameWithNewExtension(nc_file, "na")
265
266    convertor = apply(nappy.nc_convertor.nc_to_na.NCToNA, [], arg_dict)
267    convertor.convert()
268
269    # If user only wants files then only give them that
270    if only_return_file_names == True:
271        return convertor.constructNAFileNames(na_file)
272    else:
273        convertor.writeNAFiles(na_file, delimiter=delimiter, float_format=float_format, 
274                               size_limit=size_limit, annotation=annotation, no_header=no_header)
275        print convertor.output_message
276        print "Should NCToNA return a list of the na file names produced?"
277        return True   
278
279   
280def convertNCToCSV(nc_file, csv_file=None, **arg_dict):
281    """
282    Reads in a NetCDF file and writes the data out to a CSV file following the
283    NASA Ames standard.
284    """
285    if csv_file == None:
286        csv_file = nappy.utils.getFileNameWithNewExtension(nc_file, "csv")
287        arg_dict["na_file"] = csv_file
288        arg_dict["delimiter"] = ","
289 
290    return apply(convertNCToNA, [nc_file], arg_dict)
291   
292
293def convertCDMSObjectsToNA(cdms_vars, global_atts_dict, na_file, 
294              na_items_to_override={}, requested_ffi=None, delimiter=default_delimiter, 
295              float_format=default_float_format, size_limit=None, annotation=False, no_header=False):
296    """
297    Takes a list of cdms variables and a global attributes dictionary and
298    writes them to one or more NASA Ames files. Arguments are:
299 
300    cdms_vars - is a list of CDMS variables
301    global_atts_dict - is a dictionary of {key: value} pairs for header
302    na_items_to_override - is a dictionary of {key: value} pairs to overwrite in
303                output files. Typically the keys are in: 
304                ("DATE", "RDATE", "ANAME", "MNAME","ONAME", "ORG", "SNAME", "VNAME".)
305    requested_ffi - is the NASA Ames File Format Index (FFI) you wish to write to.
306                Note that there are only limited options available depending on the data
307                structures found.
308    delimiter - the delimiter you wish to use between data items in the output file
309                such as "   ", "\t" or ",".
310    float_format - a python formatting string such as "%s", "%g" or "%5.2f" used for
311                formatting floats when written to file.
312    size_limit - if format FFI is 1001 then chop files up into size_limit rows of data.
313    annotation - if set to True write the output file with an additional left-hand
314                column describing the contents of each header line.
315    no_header - if set to True then only the data blocks are written to file.
316    """
317    convertor = nappy.nc_interface.cdms_to_na.CDMSToNA(cdms_vars, global_atts=global_atts_dict, 
318                       na_items_to_override=na_items_to_override, requested_ffi=requested_ffi)
319    convertor.convert()
320    na_files = convertor.writeNAFiles(na_file, delimiter=delimiter, float_format=float_format, 
321                                      annotation=annotation, no_header=no_header) 
322    print "SHOULD WE RETURN FILES WRITTEM????"
323    return True
324
325
326def convertCDMSObjectsToCSV(cdms_vars, global_atts_dict, csv_file, **arg_dict):
327    """
328    Takes a list of cdms variables and a global attributes dictionary and
329    writes them to one or more CSV files.
330    """
331    arg_dict["delimiter"] = ","
332    return apply(convertCDMSObjectsToNA, [cdms_vars, global_atts_dict, csv_file], arg_dict)
333
334
335def writeNADictToNC(na_dict, nc_file, mode="w"):
336    """
337    Writes an NASA Ames dictionary object called na_dict to a NetCDF file called nc_file.
338    Can set mode="a" or mode="w" to either append to existing nc_file or write new one.
339    Note that mode="a" might not always work.
340    """
341    # Note, this needs to pretend that the na_dict exists, do this by instantiating NACore and cheating...
342    na_file_obj = nappy.na_file.na_core.NACore()
343    na_file_obj.setNADict(na_dict)
344    # Fake up some required methods
345    def fakeCaller():pass
346    na_file_obj.readData = fakeCaller
347    convertor = nappy.na_to_cdms.NAToCDMS(na_file_obj)
348    (cdms_primary_vars, cdms_aux_vars, global_attributes) = convertor.convert()
349
350    # Now write them out
351    fout = cdms.open(nc_file, mode=mode)
352    for var in (cdms_primary_vars + cdms_aux_vars):
353        fout.write(var)
354
355    # Write global attributes
356    for (att, value) in global_attributes.items():
357        setattr(fout, att, value)
358       
359    fout.close()
360    print "NetCDF file '%s' written successfully." % file_name
361    return True
362
363
364def readCDMSObjectsFromNA(na_file):
365    """
366    Reads the NASA Ames file and converts to CDMS objects.
367    Returns a tuple containing:
368      * a list of primary NASA Ames variables as CDMS variables
369      * a list of auxiliary NASA Ames variables as CDMS variables,
370      * a dictionary of global attributes
371    """
372    cdms_var_list = []
373    global_attributes = {}
374
375    # Open the NA file
376    na_file_obj = openNAFile(na_file)
377    convertor = nappy.nc_interface.na_to_cdms.NADictToCdmsObjects(na_file_obj)
378    (cdms_vars_primary, cdms_vars_aux, global_attributes) = convertor.convert()
379    return (cdms_vars_primary, cdms_vars_aux, global_attributes)
380
381
382def getCDMSVariableFromNA(na_file, var):
383    """
384    Returns a CDMS variable object (TransientVariable) identified by the var argument which
385    can either be an integer index in the list of variables or the name of the variable.
386    The variable is created from the variables found in the NASA Ames file na_file.
387    """
388    na_file_obj = openNAFile(na_file)
389    convertor = nappy.nc_interface.na_to_cdms.NADictToCdmsObjects(na_file_obj, variables=[var])
390    (cdms_primary_vars, cdms_aux_vars, global_attributes) = convertor.convert()
391    # Must now be a primary var
392    return cdms_primary_vars[0]
393
394
Note: See TracBrowser for help on using the repository browser.