[ncl-talk] Problem with str_match
Wei Huang
huangwei at ucar.edu
Mon Mar 9 13:45:28 MDT 2015
Stephane,
Mary just pointed to me that we recorded the in-compatible changes of
str_match at:
http://www.ncl.ucar.edu/prev_releases.shtml#ChangesTostr_matchFunctions6.2.0
As I said in my last email, we will discuss on this again within NCL group.
Thanks,
Wei
================================================
1850 Table Mesa Dr.
Boulder, CO 80307
Phone: 303-497-8924
On Mon, Mar 9, 2015 at 11:32 AM, Wei Huang <huangwei at ucar.edu> wrote:
> Stephane,
>
> For your first question, sorry we do not have a different version (using
> regex) around.
> The second one is a hard question. Yes, we try pretty hard to maintain
> backward compatible,
> and as you can see, most of NCL functions are, but that is not always the
> case.
> Like here str_match is one of them.
>
> As I said, str_match was developed using regex, then there were quite few
> issues with it,
> and we switched to plain character string match. it solved some issues,
> then caused some
> new one, like the ones Sophie (and probably you) got, and certainly the
> backward compatible
> issues as those two schemes functioning pretty different.
>
> Let say this, we will have a NCL group meeting tomorrow. Let me rise this
> issue at our group
> meeting, and discuss what we should handle this issue and then come back
> to you.
>
> At this point, if you have any comments/suggestions/requirements, please
> let us know, so we can
> discuss it. Certainly we welcome other people chip-in their ideas as well.
>
> Thanks,
>
> Wei
>
> ================================================
> 1850 Table Mesa Dr.
> Boulder, CO 80307
> Phone: 303-497-8924
>
> On Mon, Mar 9, 2015 at 10:28 AM, SENESI Stéphane <stephane.senesi at meteo.fr
> > wrote:
>
>> Hi Wei Huang
>>
>> Is there an new function that would match the old behaviour of str_match
>> ?
>> Is there any commitment from NCL on backward compatibility on basic
>> functions?
>>
>> Best regards
>>
>> S
>>
>> ------------------------------
>>
>> *De: *"TYTECA Sophie" <sophie.tyteca at meteo.fr>
>> *À: *"SENESI Stéphane" <stephane.senesi at meteo.fr>
>> *Envoyé: *Lundi 9 Mars 2015 17:01:19
>> *Objet: *Fwd: [ncl-talk] Problem with str_match
>>
>> Voici l'explication, mais cela ne nous arrange pas!
>>
>> ----- Météo-France -----
>> TYTECA SOPHIE
>> CNRM/GMGEC/VDR
>> sophie.tyteca at meteo.fr
>> Fixe : +33 561079379
>> ------------------------------
>> *De: *"Wei Huang" <huangwei at ucar.edu>
>> *À: *"TYTECA Sophie" <sophie.tyteca at meteo.fr>
>> *Cc: *ncl-talk at ucar.edu
>> *Envoyé: *Lundi 9 Mars 2015 16:40:46
>> *Objet: *Re: [ncl-talk] Problem with str_match
>>
>>
>> Sophie,
>>
>> In the earlier version, NCL used regex for string match, which is why it
>> worked there.
>> The current version, NCL has switched to character match, which is why it
>> failed there.
>>
>> Here is a short code (based on your case) on how to use NCL str_match.
>>
>> ;NCL does not support regex string match/compare, the following code will
>> return missing.
>> varname="rntcre"
>> a=str_match(varname,"r[lsn][ats]cre")
>> print (a)
>>
>> ;NCL supports plain character string match, the following code will
>> return the match.
>> b=str_match(varname,"ntcre")
>> print (b)
>> ------------------------
>> Here is the print out from the code above.
>> -------------------------------
>> Variable: a
>> Type: string
>> Total Size: 8 bytes
>> 1 values
>> Number of Dimensions: 1
>> Dimensions and sizes: [1]
>> Coordinates:
>> Number Of Attributes: 1
>> _FillValue : missing
>> (0) missing
>>
>>
>> Variable: b
>> Type: string
>> Total Size: 8 bytes
>> 1 values
>> Number of Dimensions: 1
>> Dimensions and sizes: [1]
>> Coordinates:
>> (0) rntcre
>>
>>
>> Please let ncl-talk know if there is anything can help.
>>
>> Wei
>>
>>
>> ================================================
>> 1850 Table Mesa Dr.
>> Boulder, CO 80307
>> Phone: 303-497-8924
>>
>> On Mon, Mar 9, 2015 at 9:27 AM, TYTECA Sophie <sophie.tyteca at meteo.fr>
>> wrote:
>>
>>> Hello,
>>>
>>> I have a problem with str_match function and the NCL last version .
>>> the result gives "missing" in my string
>>>
>>> Here my machine: Linux x86_64 GNU/Linux
>>>
>>> *ncl*
>>> Copyright (C) 1995-2014 - All Rights Reserved
>>> University Corporation for Atmospheric Research
>>> NCAR Command Language *Version 6.2.1*
>>> The use of this software is governed by a License Agreement.
>>> See http://www.ncl.ucar.edu/ for more details.
>>> *ncl 0> varname="rntcre"*
>>> *ncl 1> a=str_match(varname,"r[lsn][ats]cre")*
>>> *ncl 2> print (a)*
>>>
>>> Variable: a
>>> Type: string
>>> Total Size: 8 bytes
>>> 1 values
>>> Number of Dimensions: 1
>>> Dimensions and sizes: [1]
>>> Coordinates:
>>> Number Of Attributes: 1
>>> _FillValue : missing
>>> (0) missing
>>>
>>> this function works with NCL previous version:
>>>
>>> *ncl*
>>> Copyright (C) 1995-2012 - All Rights Reserved
>>> University Corporation for Atmospheric Research
>>> NCAR Command Language *Version 6.1.0*
>>> The use of this software is governed by a License Agreement.
>>> See http://www.ncl.ucar.edu/ for more details.
>>> *ncl 0> varname="rntcre"*
>>> *ncl 1> a=str_match(varname,"r[lsn][ats]cre")*
>>> *ncl 2> print (a)*
>>>
>>> Variable: a
>>> Type: string
>>> Total Size: 8 bytes
>>> 1 values
>>> Number of Dimensions: 1
>>> Dimensions and sizes: [1]
>>> Coordinates:
>>> (0) rntcre
>>>
>>> Could you help me, please?
>>> Thank you very much
>>>
>>> ----- Météo-France -----
>>> TYTECA SOPHIE
>>> CNRM/GMGEC/VDR
>>> sophie.tyteca at meteo.fr
>>> Fixe : +33 561079379
>>>
>>>
>>>
>>> _______________________________________________
>>> ncl-talk mailing list
>>> List instructions, subscriber options, unsubscribe:
>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>
>>>
>>
>>
>>
>>
>> --
>> ----- Météo-France -----
>> SENESI STEPHANE
>> CNRM/GMGEC/ASTER
>> stephane.senesi at meteo.fr
>> Fixe : +33 561079931
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20150309/261dcce7/attachment.html
More information about the ncl-talk
mailing list