str_get_cols
Returns an array of substrings, given a start and end index into the given string.
Available in version 5.1.1 and later.
Prototype
function str_get_cols ( string_val : string, start_col [1] : integer, end_col [1] : integer ) return_val [dimsizes(string_val)] : string
Arguments
string_valAn array of strings of any dimensionality.
start_colend_col
Scalar integers representing the start column and end column to be grabbed from a string. Numbering is 0-based, so the first column is 0.
Description
This function returns an array of substrings with the same dimensionality as the input strings, but with the strings subscripted from start_col to end_col.
You can use negative index values to indicate a count from the end of the string. For example, -1 represents the last character in a string, -2 the second-to-the-last character, etc.
If end_col is greater than start_col, then the substring will be reversed (see example below).
If just one of start_col and/or end_col are out-of-bounds, then the string will be subscripted only to the end or beginning as appropriate. If both are out-of-bounds, then an empty string ("") will be returned.If input strings are missing, then missing values will be returned.
See Also
str_fields_count, str_get_field
Examples
Example 1
Assume you have an ASCII file, "asc3.txt":
200306130209 0.38 25.28 10088 233.95 6 92 9.99 9.99 99999.0 0.0 -9.99 167.9 p p p 1782 BOS_1 ATL 3 200306130209 0.38 25.28 10088 233.95 6 92 9.99 9.99 99999.0 0.0 -9.99 167.9 p p p 1782 ORD_2 ATL 3 200306122341 -45.10 168.70 914 279.35 4 272 9.99 9.99 1.1 0.0 0.01 -99.9 p p p 4552 DTW_3 MSP 3 200306122341 -45.10 168.70 914 279.35 4 272 9.99 9.99 1.1 0.0 0.01 -99.9 p p p 4552 SDF_4 MHR 3Use str_fields_count to count the number of fields separated by one or more spaces, str_get_field to retrieve the fields, and then str_get_cols to retrieve subsets of these fields:
flnm = "asc3.txt" strs = asciiread(flnm,-1,"string") delim = " " nfields = str_fields_count(strs(0), delim) ; Count the fields separated ; by one or more spaces. print(nfields) ; nfields = 20 field = 1 subStrings = str_get_field(strs, field, delim) print(subStrings) ; (/"200306130209","200306130209",...,/) year = str_get_cols(subStrings, 0, 3) print(year) ; (/"2003","2003","2003","2003"/) month = str_get_cols(subStrings, 4, 5) print(month) ; (/"06","06","06","06"/) day = stringtoint(str_get_cols(subStrings, 6, 7)) print(day) ; (/13,13,12,12/) hour = stringtoint(str_get_cols(subStrings, 8, 9)) print(hour) ; (/2,2,23,23) minute = stringtoint(str_get_cols(subStrings, 10, 11)) print(minute) ; (/9,9,41,41/)Example 2
Assume you have the same ASCII file and the same strs arrays as in example 1, but that you want to read in the first 12 characters and reverse them:
date = str_get_cols(strs, 11, 0) print(date) ; (/"902031603002","902031603002",.../)To read in the second-to-the-last field, you can do this one of two ways:
field = nfields-1 subStrings = str_get_field(strs, field, delim) print(subStrings) ; (/"ATL","ATL","MSP","MHR"/)Or:
subStrings = str_get_cols(strs,-5,-3) print(subStrings) ; (/"ATL","ATL","MSP","MHR"/)This second method is only good if you don't have extra whitespace at the end of your lines in the file.
Example 3
How can the following file (sample_sounding.txt) be read?
PRES HGHT TEMP DWPT RELH MIXR DRCT SKNT THTA THTE
hPa m C C % g/kg deg knot K K
-------------------------------------------------------------------------
1013.0 17 22.8 16.8 69 12.02 20 10 294.9 329.5
1012.0 26 22.8 16.9 70 12.15 350 14 294.9 330.0
1000.0 136 22.4 18.8 80 345 16 295.6 335.5
984.0 275 21.3 83 345 20 295.8 335.5
930.0 763 17.6 97 350 14 296.8
Use a combination of str_get_cols and tofloat.
When tofloat encounters a non-numeric type (here, spaces)
it issues a warning message. It uses the default float _FillValue (= 9.96921e+36).
The simple function below results in a cleaner code and changes the return _FillValue.
To suppress these warning messages use the answer to How can I turn off NCL warning messages?
undef("strflt") function strflt(dstr[*]:string, colStrt[1]:integer, colLast[1]:integer) begin dat = tofloat( str_get_cols(dstr, colStrt, colLast) ) dat@_FillValue = -999.0 ; change from 9.96921e+36 return(dat) end err = NhlGetErrorObjectId() setvalues err "errLevel" : "Fatal" ; only report Fatal errors end setvalues diri = "./" fili = "sample_sounding.txt" dstr = asciiread(diri+fili, -1, "string") nrow = dimsizes(dstr) nhead = 3 nlvl = nrow-nhead p = strflt(dstr(nhead:), 0, 7) h = strflt(dstr(nhead:),10,14) ts = strflt(dstr(nhead:),16,21) td = strflt(dstr(nhead:),22,28) rh = strflt(dstr(nhead:),30,35) mr = strflt(dstr(nhead:),37,42) wd = strflt(dstr(nhead:),44,49) ws = strflt(dstr(nhead:),50,56) ta = strflt(dstr(nhead:),57,63) te = strflt(dstr(nhead:),64,70) print(p) print(td)The output looks like:
Variable: pstr
Type: float
Total Size: 20 bytes
5 values
Number of Dimensions: 1
Dimensions and sizes: [5]
Coordinates:
Number Of Attributes: 1
_FillValue : -999
(0) 1013
(1) 1012
(2) 1000
(3) 984
(4) 930
=============
Variable: tdstr
Type: float
Total Size: 20 bytes
5 values
Number of Dimensions: 1
Dimensions and sizes: [5]
Coordinates:
Number Of Attributes: 1
_FillValue : -999
(0) 16.8
(1) 16.9
(2) 18.8
(3) -999
(4) -999