|
1 ;---------------------------------------------------------------------- |
|
2 ; add year to output |
|
3 ; |
|
4 ; This example reads an ASCII file that is formatted a specific way, and |
|
5 ; writes out the results to a netCDF file. |
|
6 ; |
|
7 ; The first line in the ASCII file must be a header, with each field |
|
8 ; separated by a single character delimiter (like a ","). The rest of |
|
9 ; the file must be such that each row contains all fields, each |
|
10 ; separated by the designated delimiter. |
|
11 ; |
|
12 ; The fields can be integer, float, double, character, or string. |
|
13 ; String fields cannot be written to a netCDF file. They have to |
|
14 ; be read in as character arrays and written out that way. |
|
15 ;---------------------------------------------------------------------- |
|
16 |
|
17 ;---------------------------------------------------------------------- |
|
18 ; This function returns the index locations of the given delimiter |
|
19 ; in a row or several rows of strings. |
|
20 ;---------------------------------------------------------------------- |
|
21 function delim_indices(strings,nfields,delimiter) |
|
22 local cstrings, cdelim |
|
23 begin |
|
24 nrows = dimsizes(strings) |
|
25 ; |
|
26 ; Handle special case if we only have one string. Make sure it |
|
27 ; is put into a 2D array. |
|
28 ; |
|
29 if(nrows.eq.1) then |
|
30 cstrings = new((/1,strlen(strings)+1/),character) |
|
31 end if |
|
32 |
|
33 cstrings = stringtochar(strings) ; Convert to characters. |
|
34 cdelim = stringtochar(delimiter) ; Convert delimiter to character. |
|
35 ; |
|
36 ; Som error checking here. Make sure delimiter is one character. |
|
37 ; |
|
38 nc = dimsizes(cdelim) |
|
39 rank = dimsizes(nc) |
|
40 if(rank.ne.1.or.(rank.eq.1.and.nc.ne.2)) then |
|
41 print("delim_indices: fatal: the delimiter you've selected") |
|
42 print("must be a single character. Can't continue.") |
|
43 exit |
|
44 end if |
|
45 |
|
46 ; |
|
47 ; Create array to hold indices of delimiter locations, and then loop |
|
48 ; through each row and find all the delimiters. Make sure each row has |
|
49 ; the correct number of delimiters. |
|
50 ; |
|
51 ndelims = nfields-1 |
|
52 cindices = new((/nrows,ndelims/),integer) |
|
53 do i = 0, nrows-1 |
|
54 ii = ind(cstrings(i,:).eq.cdelim(0)) |
|
55 ; |
|
56 ; Make sure there were delimiters on this row. If not, we just quit. |
|
57 ; This could probably be modified to do this more gracefully. |
|
58 ; |
|
59 if(any(ismissing(ii))) then |
|
60 print("delim_indices: fatal: I didn't find any delimiters") |
|
61 print("('" + delimiter + "') on row " + i + ". Can't continue.") |
|
62 exit |
|
63 end if |
|
64 if(dimsizes(ii).ne.ndelims) then |
|
65 print("delim_indices: fatal: I expected to find " + ndelims) |
|
66 print("delimiters on row " + i + ". Instead, I found " + dimsizes(ii) + ".") |
|
67 print("Can't continue.") |
|
68 exit |
|
69 end if |
|
70 |
|
71 cindices(i,:) = ii |
|
72 |
|
73 delete(ii) ; For next time through loop |
|
74 end do |
|
75 |
|
76 return(cindices) |
|
77 end |
|
78 |
|
79 ;---------------------------------------------------------------------- |
|
80 ; This function reads in a particular field from a string array, |
|
81 ; given the field number to read (fields start at #1 and go to #nfield), |
|
82 ; and the indices of the delimiters. |
|
83 ; |
|
84 ; It returns either an integer, float, double, character, or a string, |
|
85 ; depending on the input flag "return_type". |
|
86 ;---------------------------------------------------------------------- |
|
87 function read_field(strings,ifield,indices,return_type) |
|
88 local nstring, cstrings, nf, tmp_str |
|
89 begin |
|
90 nrows = dimsizes(strings) |
|
91 ; |
|
92 ; Handle special case if we only have one string. Make sure it |
|
93 ; is put into a 2D array. |
|
94 ; |
|
95 if(nrows.eq.1) then |
|
96 cstrings = new((/1,strlen(strings)+1/),character) |
|
97 end if |
|
98 |
|
99 cstrings = stringtochar(strings) |
|
100 nf = dimsizes(indices(0,:))+1 ; indices is nrows x (nfields-1) |
|
101 |
|
102 ; |
|
103 ; Error checking. Make sure user has entered a valid field. |
|
104 ; |
|
105 if(ifield.le.0.or.ifield.gt.nf) then |
|
106 print("read_field: fatal: you've selected a field that is") |
|
107 print("out-of-range of the number of fields that you have (" + nf + ").") |
|
108 exit |
|
109 end if |
|
110 |
|
111 ; |
|
112 ; Set up array to return. For string, int, float, or double arrays, |
|
113 ; we don't have to do anything special. For character arrays, |
|
114 ; however, we do. |
|
115 ; |
|
116 if(return_type.ne."character") then |
|
117 return_array = new(nrows,return_type) |
|
118 else |
|
119 ; |
|
120 ; We don't know what the biggest character array is at this point, so |
|
121 ; make it bigger than necessary, and then resize later as necessary. |
|
122 ; |
|
123 tmp_return_array = new((/nrows,dimsizes(cstrings(0,:))/),"character") |
|
124 |
|
125 max_len = 0 ; Use to keep track of max lengths of strings. |
|
126 end if |
|
127 |
|
128 do i = 0,nrows-1 |
|
129 ; |
|
130 ; Special case of first field in row. |
|
131 ; |
|
132 if(ifield.eq.1) then |
|
133 ibeg = 0 |
|
134 iend = indices(i,ifield-1)-1 |
|
135 else |
|
136 ; |
|
137 ; Special case of first field in row. |
|
138 ; |
|
139 if(ifield.eq.nf) then |
|
140 ibeg = indices(i,ifield-2)+1 |
|
141 iend = dimsizes(cstrings(i,:))-1 |
|
142 ; |
|
143 ; Any field between first and last field. |
|
144 ; |
|
145 else |
|
146 ibeg = indices(i,ifield-2)+1 |
|
147 iend = indices(i,ifield-1)-1 |
|
148 end if |
|
149 end if |
|
150 ; |
|
151 ; Here's the code that pulls off the correct string, and converts it |
|
152 ; to float if desired. |
|
153 ; |
|
154 if(return_type.eq."integer") then |
|
155 return_array(i) = stringtointeger(chartostring(cstrings(i,ibeg:iend))) |
|
156 end if |
|
157 if(return_type.eq."float") then |
|
158 return_array(i) = stringtofloat(chartostring(cstrings(i,ibeg:iend))) |
|
159 end if |
|
160 if(return_type.eq."double") then |
|
161 return_array(i) = stringtodouble(chartostring(cstrings(i,ibeg:iend))) |
|
162 end if |
|
163 if(return_type.eq."string") then |
|
164 return_array(i) = chartostring(cstrings(i,ibeg:iend)) |
|
165 end if |
|
166 if(return_type.eq."character") then |
|
167 if( (iend-ibeg+1) .gt. max_len) then |
|
168 max_len = iend-ibeg+1 |
|
169 end if |
|
170 tmp_return_array(i,0:iend-ibeg) = cstrings(i,ibeg:iend) |
|
171 end if |
|
172 end do |
|
173 |
|
174 if(return_type.eq."character") then |
|
175 return_array = new((/nrows,max_len/),"character") |
|
176 return_array = tmp_return_array(:,0:max_len-1) |
|
177 end if |
|
178 |
|
179 return(return_array) |
|
180 end |
|
181 |
|
182 |
|
183 ;---------------------------------------------------------------------- |
|
184 ; This function reads in string fields only to get the maximum string |
|
185 ; length. |
|
186 ;---------------------------------------------------------------------- |
|
187 function get_maxlen(strings,ifield,indices) |
|
188 local nstring, cstrings, nf, tmp_str |
|
189 begin |
|
190 nrows = dimsizes(strings) |
|
191 ; |
|
192 ; Handle special case if we only have one string. Make sure it |
|
193 ; is put into a 2D array. |
|
194 ; |
|
195 if(nrows.eq.1) then |
|
196 cstrings = new((/1,strlen(strings)+1/),character) |
|
197 end if |
|
198 |
|
199 cstrings = stringtochar(strings) |
|
200 nf = dimsizes(indices(0,:))+1 ; indices is nrows x (nfields-1) |
|
201 |
|
202 ; |
|
203 ; Error checking. Make sure user has entered a valid field. |
|
204 ; |
|
205 if(ifield.le.0.or.ifield.gt.nf) then |
|
206 print("read_field: fatal: you've selected a field that is") |
|
207 print("out-of-range of the number of fields that you have (" + nf + ").") |
|
208 exit |
|
209 end if |
|
210 ; |
|
211 ; We don't know what the biggest character array is at this point, so |
|
212 ; make it bigger than necessary, and then resize later as necessary. |
|
213 ; |
|
214 tmp_return_array = new((/nrows,dimsizes(cstrings(0,:))/),"character") |
|
215 |
|
216 max_len = 0 ; Use to keep track of max lengths of strings. |
|
217 |
|
218 do i = 0,nrows-1 |
|
219 ; |
|
220 ; Special case of first field in row. |
|
221 ; |
|
222 if(ifield.eq.1) then |
|
223 ibeg = 0 |
|
224 iend = indices(i,ifield-1)-1 |
|
225 else |
|
226 ; |
|
227 ; Special case of first field in row. |
|
228 ; |
|
229 if(ifield.eq.nf) then |
|
230 ibeg = indices(i,ifield-2)+1 |
|
231 iend = dimsizes(cstrings(i,:))-1 |
|
232 ; |
|
233 ; Any field between first and last field. |
|
234 ; |
|
235 else |
|
236 ibeg = indices(i,ifield-2)+1 |
|
237 iend = indices(i,ifield-1)-1 |
|
238 end if |
|
239 end if |
|
240 if( (iend-ibeg+1) .gt. max_len) then |
|
241 max_len = iend-ibeg+1 |
|
242 end if |
|
243 end do |
|
244 |
|
245 return(max_len) |
|
246 end |
|
247 |
|
248 ;---------------------------------------------------------------------- |
|
249 ; Main code. |
|
250 ;---------------------------------------------------------------------- |
|
251 begin |
|
252 |
|
253 ; Set up defaults here. We are hard-coding the field types here. |
|
254 ; You can set up this script to try to determine the field types |
|
255 ; automatically, but this is a bit tedious. Maybe later. |
|
256 |
|
257 station = "USHa1" |
|
258 year = 1991 |
|
259 nfields = 30 ; # of fields |
|
260 delimiter = "," ; field delimiter |
|
261 |
|
262 filename = station+year+"_L4_m.txt" ; ASCII" file to read. |
|
263 cdf_file = station+year+"_L4_m.nc" ; netCDF file to write. |
|
264 |
|
265 ; In this case, fields #2-#2 are integers, |
|
266 ; and the rest of the fields are floats. |
|
267 |
|
268 var_types = new(nfields,string) |
|
269 var_strlens = new(nfields,integer) ; var to hold strlens, just in case. |
|
270 |
|
271 var_types = "float" ; Most are floats. |
|
272 var_types(0:1) = "integer" |
|
273 |
|
274 if(isfilepresent(cdf_file)) |
|
275 print("Warning: '" + cdf_file + "' exists. Will remove it.") |
|
276 system("/bin/rm " + cdf_file) |
|
277 end if |
|
278 |
|
279 ; Read in data as strings. This will create a string array that has the |
|
280 ; same number of strings as there are rows in the file. We will then need |
|
281 ; to parse each string later. |
|
282 |
|
283 read_data = asciiread(filename,-1,"string") |
|
284 header = read_data(0) ; Header. Use for variable names. |
|
285 data = read_data(1:) ; Get rid of first line which is a header. |
|
286 nrows = dimsizes(data) ; Number of rows. |
|
287 |
|
288 ; Read in locations of delimiters in each string row. |
|
289 |
|
290 hindices = delim_indices(header,nfields,delimiter) ; header row |
|
291 dindices = delim_indices(data,nfields,delimiter) ; rest of file |
|
292 |
|
293 print (hindices) |
|
294 print (dindices) |
|
295 |
|
296 ; Read in the field names which will become variable names on |
|
297 ; the netCDF file. |
|
298 |
|
299 var_names = new(nfields,string) |
|
300 |
|
301 do i=0,nfields-1 |
|
302 var_names(i) = read_field(header,i+1,hindices,"string") |
|
303 end do |
|
304 |
|
305 ; Write out this netCDF file efficiently so it will be faster. |
|
306 ; Try to predefine everything before you write to it. |
|
307 |
|
308 f = addfile(cdf_file,"c") |
|
309 setfileoption(f,"DefineMode",True) ; Enter predefine phase. |
|
310 |
|
311 ; Write global attributes to file. It's okay to do this before |
|
312 ; predefining the file's variables. We are still in "define" mode. |
|
313 |
|
314 fAtt = True |
|
315 fAtt@description = "Data read in from " + filename + " ASCII file." |
|
316 fAtt@creation_date = systemfunc ("date") |
|
317 fileattdef( f, fAtt ) |
|
318 |
|
319 ; Write dimension names to file. If there are no character variables, |
|
320 ; then there's only one dimension name ("nvalues"). |
|
321 |
|
322 nyear = -1 |
|
323 nmonth = 12 |
|
324 |
|
325 dim_names = (/ "year", "month" /) |
|
326 dim_sizes = (/ nyear , nmonth /) |
|
327 dimUnlim = (/ True , False /) |
|
328 |
|
329 filedimdef( f, dim_names, dim_sizes, dimUnlim ) |
|
330 |
|
331 ; Define each variable on the file. |
|
332 |
|
333 filevardef( f, "year", "integer", "year" ) |
|
334 |
|
335 ; Don't deal with variable Month (i=0). |
|
336 |
|
337 do i=1,nfields-1 |
|
338 filevardef(f, var_names(i), var_types(i), dim_names) |
|
339 end do |
|
340 |
|
341 ; Loop through each field, read the values for that field, print |
|
342 ; information about the variable, and then write it to the netCDF |
|
343 ; file. |
|
344 |
|
345 do i=1,nfields-1 |
|
346 ifield = i+1 ; Fields start at #1, not #0. |
|
347 |
|
348 ; Note: you can't write strings to a netCDF file, so these have |
|
349 ; to be written out as character arrays. |
|
350 |
|
351 tmp_data = new((/1,nmonth/),var_types(i)) |
|
352 |
|
353 out_data = read_field(data,ifield,dindices,var_types(i)) |
|
354 |
|
355 tmp_data(0,:) = out_data(:) |
|
356 |
|
357 ; Print some info about the variable. |
|
358 |
|
359 print("") |
|
360 print("Writing variable '" + var_names(i) + "' (field #" + ifield + ").") |
|
361 print("Type is " + var_types(i) + ".") |
|
362 print("min/max = " + min(tmp_data) + "/" + max(tmp_data)) |
|
363 |
|
364 if(any(ismissing(tmp_data))) then |
|
365 print("This variable does contain missing values.") |
|
366 else |
|
367 print("This variable doesn't contain missing values.") |
|
368 end if |
|
369 |
|
370 f->$var_names(i)$ = tmp_data ; Write to netCDF file. |
|
371 |
|
372 delete(tmp_data) ; Delete for next round. |
|
373 delete(out_data) |
|
374 end do |
|
375 |
|
376 f->year = year |
|
377 end |