source: nappy/trunk/nappy/nappy_api.py @ 5125

Subversion URL: http://proj.badc.rl.ac.uk/svn/ndg/nappy/trunk/nappy/nappy_api.py@5125
Revision 5125, 17.5 KB checked in by astephen, 12 years ago (diff)

Added some fixes after command-line testing.

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 list of tuples/lists.
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 list and write
89# them to a CSV file.
90nappy.convertCDMSObjectsToCSV(cdms_vars, global_atttributes, 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 cdms2 as cdms
129except:
130    try:
131        import cdms
132    except:
133        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."
134        cdms = False
135
136# Import local modules
137import nappy.utils.common_utils
138import nappy.utils.compare_na
139
140# Bring some utils into the API
141compareNA = nappy.utils.compare_na.compareNA
142readFFI = nappy.utils.common_utils.readFFI
143chooseFFI = nappy.utils.common_utils.chooseFFI
144getNAFileClass = nappy.utils.common_utils.getNAFileClass
145__version__ = nappy.utils.common_utils.getVersion()
146default_delimiter = nappy.utils.common_utils.getDefault("default_delimiter")
147default_float_format = nappy.utils.common_utils.getDefault("default_float_format")
148
149
150def openNAFile(filename, mode="r", na_dict=None):
151    """
152    Function wrapper around the NASA Ames File classes. Any NASA Ames
153    file can be opened through this function and the appropriate read or
154    write NASA Ames File class instance is returned.
155    """
156    if mode == "r":
157        ffi = readFFI(filename)
158        return apply(getNAFileClass(ffi), (filename, mode))
159
160    elif mode == "w":
161        if na_dict.has_key('FFI') and type(na_dict['FFI']) == type(3):
162            ffi = na_dict['FFI']
163        else:
164            ffi = chooseFFI(na_dict)
165            na_dict['FFI'] = ffi
166            print "\nFormat identified as:", ffi   
167        return apply(getNAFileClass(ffi), (filename,), {"mode":mode, "na_dict":na_dict})
168    else:
169        raise Exception("File mode not recognised '" + mode + "'.")
170
171
172def convertNAToNC(na_file, nc_file=None, mode="w", variables=None, aux_variables=None,
173                 global_attributes=[("Conventions", "CF-1.0")],
174                 time_units=None, time_warning=True,
175                 rename_variables={}):
176    """
177    Takes a NASA Ames file and converts to a NetCDF file. Options are:
178
179    na_file - the input NASA Ames file.
180    nc_file - name for the output NetCDF file (default is to replace ".na" from NASA Ames
181              file with ".nc").
182    mode - is the file mode, either "w" for write or "a" for append
183    variables - is a list of variable names that you wish to be converted. If not set then
184              nappy will attempt to convert all files.
185    aux_var_list - is a list of auxiliary variables names that you wish to be converted.
186              If not set then nappy will use any compatible variables it finds as
187              auxiliary variables.
188    global_attributes - is a list of global attributes to add to the output file.
189    rename_variables - is a dictionary of {old_name: new_name} variable ID pairs that nappy
190              should use to rename variables before it writes them to file. 
191    time_units - is a valid time units string such as "hours since 2003-04-30 10:00:00" to
192              use for time units if there is a valid time axis.
193    time_warning - suppresses the time units warning for invalid time units if set to False.
194    """
195    arg_dict = vars()
196    for arg_out in ("nc_file", "mode"):
197        del arg_dict[arg_out]
198
199    import nappy.nc_interface.na_to_nc
200    convertor = apply(nappy.nc_interface.na_to_nc.NAToNC, [], arg_dict)
201    convertor.convert()
202    if nc_file == None:
203        nc_file = nappy.utils.getFileNameWithNewExtension(na_file, "nc")
204    convertor.writeNCFile(nc_file, mode)
205    return nc_file   
206 
207
208def convertNAToCSV(na_file, csv_file=None, annotation=False, no_header=False):
209    """
210    Reads in a NASA Ames file and writes it out a new CSV file which is identical to the
211    input file except that commas are used as the delimiter. Arguments are:
212
213    na_file - NASA Ames file path
214    csv_file - CSV file path (default is to replace ".na" from NASA Ames file with ".csv").
215    annotation - if set to True write the output file with an additional left-hand column
216                 describing the contents of each header line.
217    no_header - if set to True then only the data blocks are written to file.
218    """
219    fin = openNAFile(na_file)
220    fin.readData()
221    na_dict = fin.getNADict()
222    fin.close()
223
224    if csv_file == None:
225        csv_file = nappy.utils.getFileNameWithNewExtension(nc_file, "csv")
226    fout = openNAFile(csv_file, "w", na_dict=na_dict)
227    fout.write(delimiter=",", annotation=annotation)
228    fout.close()
229    return True
230
231
232def convertNCToNA(nc_file, na_file=None, var_ids=None, na_items_to_override={},
233            only_return_file_names=False, exclude_vars=[],
234            requested_ffi=None, delimiter=default_delimiter, float_format=default_float_format, 
235            size_limit=None, annotation=False, no_header=False,
236            ):
237    """
238    Takes a NetCDF file and converts the contents to one or more NASA Ames files.
239    Arguments are:
240
241    nc_file - is the name of input file (NetCDF).
242    na_file - is the name of output file (default is to replace ".nc" from NASA Ames
243              file with ".na"). If multiple files produced then this name will be used
244              as the base name.
245    var_ids - is a list of variables (as ids) to include in the output file(s).
246    na_items_to_override - is a dictionary of {key: value} pairs to overwrite in output
247              files. Typically the keys are in: 
248              ("DATE", "RDATE", "ANAME", "MNAME","ONAME", "ORG", "SNAME", "VNAME".)
249    only_return_file_names - if set to True then only return a list of file names that
250              would be written (i.e. don't convert actual file).
251    exclude_vars - is a list of variables (as ids) to exclude in the output file(s).
252    requested_ffi - is the NASA Ames File Format Index (FFI) you wish to write to. Note
253              that there are only limited options available depending on the data
254              structures found.
255    delimiter - the delimiter you wish to use between data items in the output file such
256              as "   ", "\t" or ",".
257    float_format - a python formatting string such as "%s", "%g" or "%5.2f" used for
258              formatting floats when written to file.
259    size_limit - if format FFI is 1001 then chop files up into size_limit rows of data.
260    annotation - if set to True write the output file with an additional left-hand column
261              describing the contents of each header line.
262    no_header - if set to True then only the data blocks are written to file.
263    """
264    arg_dict = vars()
265    for arg_out in ("na_file", "only_return_file_names", "delimiter", "float_format", 
266                    "size_limit", "annotation", "no_header"):
267        del arg_dict[arg_out]
268
269    if na_file == None:
270        na_file =  nappy.utils.getFileNameWithNewExtension(nc_file, "na")
271
272    import nappy.nc_interface.nc_to_na
273    convertor = apply(nappy.nc_interface.nc_to_na.NCToNA, [], arg_dict)
274    convertor.convert()
275
276    # If user only wants files then only give them that
277    if only_return_file_names == True:
278        return convertor.constructNAFileNames(na_file)
279    else:
280        convertor.writeNAFiles(na_file, delimiter=delimiter, float_format=float_format, 
281                               size_limit=size_limit, annotation=annotation, no_header=no_header)
282        print convertor.output_message
283        output_files_written = convertor.output_files_written
284        print output_files_written
285        return output_files_written
286
287   
288def convertNCToCSV(nc_file, csv_file=None, **arg_dict):
289    """
290    Reads in a NetCDF file and writes the data out to a CSV file following the
291    NASA Ames standard.
292    """
293    if csv_file == None:
294        csv_file = nappy.utils.getFileNameWithNewExtension(nc_file, "csv")
295        arg_dict["na_file"] = csv_file
296        arg_dict["delimiter"] = ","
297 
298    return apply(convertNCToNA, [nc_file], arg_dict)
299   
300
301def convertCDMSObjectsToNA(cdms_vars, global_attributes, na_file, 
302              na_items_to_override={}, requested_ffi=None, delimiter=default_delimiter, 
303              float_format=default_float_format, size_limit=None, annotation=False, no_header=False,
304              ):
305    """
306    Takes a list of cdms variables and a list of global attributes and
307    writes them to one or more NASA Ames files. Arguments are:
308 
309    cdms_vars - is a list of CDMS variables
310    global_attributes - is a list of (key, value) pairs for header
311    na_items_to_override - is a dictionary of {key: value} pairs to overwrite in
312                output files. Typically the keys are in: 
313                ("DATE", "RDATE", "ANAME", "MNAME","ONAME", "ORG", "SNAME", "VNAME", "SCOM", "NCOM".)
314    requested_ffi - is the NASA Ames File Format Index (FFI) you wish to write to.
315                Note that there are only limited options available depending on the data
316                structures found.
317    delimiter - the delimiter you wish to use between data items in the output file
318                such as "   ", "\t" or ",".
319    float_format - a python formatting string such as "%s", "%g" or "%5.2f" used for
320                formatting floats when written to file.
321    size_limit - if format FFI is 1001 then chop files up into size_limit rows of data.
322    annotation - if set to True write the output file with an additional left-hand
323                column describing the contents of each header line.
324    no_header - if set to True then only the data blocks are written to file.
325    """
326    import nappy.nc_interface.cdms_objs_to_na_file
327    convertor = nappy.nc_interface.cdms_objs_to_na_file.CDMSObjectsToNAFile(cdms_vars, global_attributes=global_attributes, 
328                        na_items_to_override=na_items_to_override, requested_ffi=requested_ffi,
329                        )
330    convertor.convert()
331    na_files = convertor.writeNAFiles(na_file, delimiter=delimiter, float_format=float_format, 
332                                      annotation=annotation, no_header=no_header) 
333    return convertor.output_files_written
334
335
336def convertCDMSObjectsToCSV(cdms_vars, global_attributes, csv_file, **arg_dict):
337    """
338    Takes a list of cdms variables and a global attributes list and
339    writes them to one or more CSV files.
340    """
341    arg_dict["delimiter"] = ","
342    return apply(convertCDMSObjectsToNA, [cdms_vars, global_attributes, csv_file], arg_dict)
343
344
345def writeNADictToNC(na_dict, nc_file, mode="w"):
346    """
347    Writes an NASA Ames dictionary object called na_dict to a NetCDF file called nc_file.
348    Can set mode="a" or mode="w" to either append to existing nc_file or write new one.
349    Note that mode="a" might not always work.
350    """
351    # Note, this needs to pretend that the na_dict exists, do this by instantiating NACore and cheating...
352    import nappy.na_file.na_core
353    na_file_obj = nappy.na_file.na_core.NACore()
354    na_file_obj.setNADict(na_dict)
355    # Fake up some required methods
356    def fakeCaller():pass
357    na_file_obj.readData = fakeCaller
358    import nappy.nc_interface.na_to_cdms
359    convertor = nappy.nc_interface.na_to_cdms.NAToCDMS(na_file_obj)
360    (cdms_primary_vars, cdms_aux_vars, global_attributes) = convertor.convert()
361
362    # Now write them out
363    fout = cdms.open(nc_file, mode=mode)
364    for var in (cdms_primary_vars + cdms_aux_vars):
365        fout.write(var)
366
367    # Write global attributes
368    for (att, value) in global_attributes:
369        setattr(fout, att, value)
370       
371    fout.close()
372    print "NetCDF file '%s' written successfully." % file_name
373    return True
374
375
376def readCDMSObjectsFromNA(na_file):
377    """
378    Reads the NASA Ames file and converts to CDMS objects.
379    Returns a tuple containing:
380      * a list of primary NASA Ames variables as CDMS variables
381      * a list of auxiliary NASA Ames variables as CDMS variables,
382      * a list of global attributes
383    """
384    cdms_var_list = []
385    global_attributes = {}
386
387    # Open the NA file
388    na_file_obj = openNAFile(na_file)
389    import nappy.nc_interface.na_to_cdms
390    convertor = nappy.nc_interface.na_to_cdms.NADictToCdmsObjects(na_file_obj)
391    (cdms_vars_primary, cdms_vars_aux, global_attributes) = convertor.convert()
392    return (cdms_vars_primary, cdms_vars_aux, global_attributes)
393
394
395def getCDMSVariableFromNA(na_file, var):
396    """
397    Returns a CDMS variable object (TransientVariable) identified by the var argument which
398    can either be an integer index in the list of variables or the name of the variable.
399    The variable is created from the variables found in the NASA Ames file na_file.
400    """
401    na_file_obj = openNAFile(na_file)
402    import nappy.nc_interface.na_to_cdms
403    convertor = nappy.nc_interface.na_to_cdms.NADictToCdmsObjects(na_file_obj, variables=[var])
404    (cdms_primary_vars, cdms_aux_vars, global_attributes) = convertor.convert()
405    # Must now be a primary var
406    return cdms_primary_vars[0]
407
408
Note: See TracBrowser for help on using the repository browser.