Question

我有一份水文模型文件输出(出口)。

Units CFS
Type INST-VAL
1 01 Jan 1997, 02:00 1933.0
2 01 Jan 1997, 04:00 1918.0
3 01 Jan 1997, 06:00 1918.0
4 01 Jan 1997, 08:00 1904.0
5 01 Jan 1997, 10:00 1904.0
...

并且已经将 Python(2.6)编码为优化过程的投入格式:

import re
o=open("C:documents and settingscmjawdydesktopPyOut.txt","w")
data=open("C:documents and settingscmjawdydesktopexport.txt").read()
Step1=re.sub(":00",":00:00",data)
Step2=re.sub(" Jan ","/01/",Step1)
Step3=re.sub(",","",Step2)
FindIDs=re.compile("^[0-9]*s",re.M)
Step4=re.sub(FindIDs,"SiteXXX ",Step3)
o.write(Step4)
o.close()

屈服:

Units CFS
Type INST-VAL
SiteXXX 01/01/1997 02:00:00 1933.0
SiteXXX 01/01/1997 04:00:00 1918.0
SiteXXX 01/01/1997 06:00:00 1918.0
SiteXXX 01/01/1997 08:00:00 1904.0
SiteXXX 01/01/1997 10:00:00 1904.0
...

问题在于,我的优化软件的小时数为24小时,而第二天的小时数为00。因此,我需要将X日24:00改为X+1日00:00,同时保持同样的格式。它看上去的是,有24小时/直径。这些是我绝对的第一行,即任何电脑语言,我找不到改变这一文本的leg。

Answer 1

import datetime
s =    1 01 Jan 1997, 02:00 1933.0
2 01 Jan 1997, 04:00 1918.0
3 01 Jan 1997, 06:00 1918.0
4 01 Jan 1997, 08:00 1904.0
5 01 Jan 1997, 10:00 1904.0
6 01 Jan 1997, 24:00 1000.0   
for row in s.split( 
 ):
    prefix = row[:2]
    sdate = row[2:-7]
    suffix = row[-7:]
    if sdate[13:15] ==  24 :
        offset = datetime.timedelta(1)
        sdate = sdate[:13] +  00  + sdate[15:]
    else:
        offset = datetime.timedelta(0)
    dt = datetime.datetime.strptime(sdate,  %d %b %Y, %H:%M ) + offset
    print prefix + dt.strftime( %d/%m/%Y %H:%M:%S ) + suffix

结果:

1 01/01/1997 02:00:00 1933.0
2 01/01/1997 04:00:00 1918.0
3 01/01/1997 06:00:00 1918.0
4 01/01/1997 08:00:00 1904.0
5 01/01/1997 10:00:00 1904.0
6 02/01/1997 00:00:00 1000.0

Answer 2

下面的法典解决了几天之后的第二天的问题,以便找到一条线上的244天:1月31日、2月28日、2月29日、6月30日、12月31日等。

import re


ss1 =    Units CFS 
Type INST-VAL
1 01 Jan 1997, 02:00 1933.0
2 12 Feb 1997, 04:00 1918.0
3 26 May 1997, 06:00 1918.0
4 15 Aug 1997, 08:00 1904.0
5 09 Dec 1997, 10:00 1904.0   

ss2 =    Units CFS 
Type INST-VAL
1 31 Jan 1997, 11:00 1933.0
2 28 Feb 1997, 11:00 1918.0
2 29 Feb 1997, 11:00 1918.0
3 31 Mar 1997, 11:00 1918.0
4 30 Sep 1997, 11:00 1904.0
5 31 Dec 1997, 11:00 1904.0   

ss3 =    Units CFS 
Type INST-VAL
1 31 Jan 1997, 24:00 1933.0
2 28 Feb 2011, 24:00 1700.2
2 29 Feb 2011, 24:00 1700.0
2 28 Feb 2012, 24:00 1801.8
2 29 Feb 2012, 24:00 1801.0
3 31 Mar 1997, 24:00 1918.0
4 30 Sep 1997, 24:00 1904.0
5 31 Dec 1997, 24:00 1904.0   


bis = ( 1904 ,  1908 ,  1912 ,  1916 ,  1920 ,  1924 ,  1928 ,  1932 ,  1936 ,  1940 ,
        1944 ,  1948 ,  1952 ,  1956 ,  1960 ,  1964 ,  1968 ,  1972 ,  1976 ,  1980 ,
        1984 ,  1988 ,  1992 ,  1996 ,  2000 ,  2004 ,  2008 ,  2012 ,  2016 ,  2020 ,
        2024 ,  2028 ,  2032 ,  2036 ,  2040 ,  2044 ,  2048 ,  2052 ,  2056 ,  2060 ,
        2064 ,  2068 ,  2072 ,  2076 ,  2080 ,  2084 ,  2088 ,  2092 ,  2096 ,  2104 )

months = dict(zip( Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec .split(),xrange(1,13)))

firstday_nextmonth = {( 31 , Jan ): 01/02/ ,                        ( 31 , Mar ): 01/04/ ,
                      ( 30 , Apr ): 01/05/ , ( 31 , May ): 01/06/ , ( 30 , Jun ): 01/07/ ,
                      ( 31 , Jul ): 01/08/ , ( 31 , Aug ): 01/09/ , ( 30 , Sep ): 01/10/ ,
                      ( 31 , Oct ): 01/11/ , ( 30 , Nov ): 01/12/ , ( 31 , Dec ): 01/01/ }


di = { 1 : MADRID  , 2 : HUAHINE , 3 : MOSCOW  , 4 : OSAKA   , 5 : VALPAR. }


def repl(mat, fdnm = firstday_nextmonth, months = months, sites = di, bisextiles = bis):
    d,m,y = mat.group(2,3,4)

    if mat.group(5)== 24:00 :
        if (d,m)==( 31 , Dec ):
            dmy =  01/01/%d  % (int(y)+1)
        elif (d,m) == ( 29 , Feb ):
            if y in bisextiles:
                dmy =  01/03/  + y
            else:
                dmy =  !!!!!!!!!! 
        elif (d,m)==( 28 , Feb ):
            if y in bisextiles:
                dmy =  29/02/  + y
            else:
                dmy =  01/03/  + y
        elif (d,m) in fdnm:
            dmy = fdnm[(d,m)] + y
        else:
            dmy =  %02d/%02d/%s  % (int(mat.group(2))+1,months[m],y)

    elif (d,m) == ( 29 , Feb ) and y not in bisextiles:
        dmy =  !!!!!!!!!! 

    else:
        dmy =  %s/%02d/%s  % (d,months[m],y)

    return  %s %s %s:00  % (sites[mat.group(1)],
                            dmy,
                            mat.group(5).replace( 24:00 , 00:00 ))  




reg = re.compile( ^(d+) ([012]d|30|31) ([a-z]+) (d{4}), (dd:dd)(?= d+.d+) ,
                 re.IGNORECASE|re.MULTILINE)

for ss in (ss1,ss2,ss3):
    print ss
    print
    print reg.sub(repl,ss)
    print  
=========================================================

结果

Units CFS 
Type INST-VAL
1 01 Jan 1997, 02:00 1933.0
2 12 Feb 1997, 04:00 1918.0
3 26 May 1997, 06:00 1918.0
4 15 Aug 1997, 08:00 1904.0
5 09 Dec 1997, 10:00 1904.0

Units CFS 
Type INST-VAL
MADRID  01/01/1997 02:00:00 1933.0
HUAHINE 12/02/1997 04:00:00 1918.0
MOSCOW  26/05/1997 06:00:00 1918.0
OSAKA   15/08/1997 08:00:00 1904.0
VALPAR. 09/12/1997 10:00:00 1904.0

=========================================================

Units CFS 
Type INST-VAL
1 31 Jan 1997, 11:00 1933.0
2 28 Feb 1997, 11:00 1918.0
2 29 Feb 1997, 11:00 1918.0
3 31 Mar 1997, 11:00 1918.0
4 30 Sep 1997, 11:00 1904.0
5 31 Dec 1997, 11:00 1904.0

Units CFS 
Type INST-VAL
MADRID  31/01/1997 11:00:00 1933.0
HUAHINE 28/02/1997 11:00:00 1918.0
HUAHINE !!!!!!!!!! 11:00:00 1918.0
MOSCOW  31/03/1997 11:00:00 1918.0
OSAKA   30/09/1997 11:00:00 1904.0
VALPAR. 31/12/1997 11:00:00 1904.0

=========================================================

Units CFS 
Type INST-VAL
1 31 Jan 1997, 24:00 1933.0
2 28 Feb 2011, 24:00 1700.2
2 29 Feb 2011, 24:00 1700.0
2 28 Feb 2012, 24:00 1801.8
2 29 Feb 2012, 24:00 1801.0
3 31 Mar 1997, 24:00 1918.0
4 30 Sep 1997, 24:00 1904.0
5 31 Dec 1997, 24:00 1904.0

Units CFS 
Type INST-VAL
MADRID  01/02/1997 00:00:00 1933.0
HUAHINE 01/03/2011 00:00:00 1700.2
HUAHINE !!!!!!!!!! 00:00:00 1700.0
HUAHINE 29/02/2012 00:00:00 1801.8
HUAHINE 01/03/2012 00:00:00 1801.0
MOSCOW  01/04/1997 00:00:00 1918.0
OSAKA   01/10/1997 00:00:00 1904.0
VALPAR. 01/01/1998 00:00:00 1904.0

=========================================================

友情链接