timelib.py 文件源码

def fn_getdatetime_list(fn):
    """Extract all datetime strings from input filename
    """
    #Want to split last component
    fn = os.path.split(os.path.splitext(fn)[0])[-1]
    import re
    #WV01_12JUN152223255-P1BS_R1C1-102001001B3B9800__WV01_12JUN152224050-P1BS_R1C1-102001001C555C00-DEM_4x.tif
    #Need to parse above with month name 
    #Note: made this more restrictive to avoid false matches:
    #'20130304_1510_1030010020770600_1030010020CEAB00-DEM_4x'
    #This is a problem, b/c 2015/17/00:
    #WV02_20130315_10300100207D5600_1030010020151700
    #This code should be obsolete before 2019 
    #Assume new filenames
    #fn = fn[0:13]
    #Use cascading re find to pull out timestamps
    #Note: Want to be less restrictive here - could have a mix of YYYYMMDD_HHMM, YYYYMMDD and YYYY in filename
    #Should probably search for all possibilities, then prune  
    #NOTE: these don't include seconds in the time
    #NOTE: could have 20130304_1510__20130304__whatever in filename
    #The current approach will only catch the first datetime 
    dstr = None
    dstr = re.findall(r'(?:^|_|-)(?:19|20)[0-9][0-9](?:0[1-9]|1[012])(?:0[1-9]|[12][0-9]|3[01])[_T](?:0[0-9]|1[0-9]|2[0-3])[0-5][0-9]', fn)
    if not dstr:
        dstr = re.findall(r'(?:^|_|-)(?:19|20)[0-9][0-9](?:0[1-9]|1[012])(?:0[1-9]|[12][0-9]|3[01])(?:0[0-9]|1[0-9]|2[0-3])[0-5][0-9]', fn)
    if not dstr:
        dstr = re.findall(r'(?:^|_|-)(?:19|20)[0-9][0-9](?:0[1-9]|1[012])(?:0[1-9]|[12][0-9]|3[01])(?:_|-)', fn)
        #This should pick up dates separated by a dash
        #dstr = re.findall(r'(?:^|_|-)(?:19|20)[0-9][0-9](?:0[1-9]|1[012])(?:0[1-9]|[12][0-9]|3[01])', fn)
    if not dstr:
        dstr = re.findall(r'(?:^|_|-)(?:19|20)[0-9][0-9](?:_|-)', fn)
    #This is for USGS archive filenames
    if not dstr:
        dstr = re.findall(r'[0-3][0-9][a-z][a-z][a-z][0-9][0-9]', fn)
    #if not dstr:
    #    dstr = re.findall(r'(?:^|_)(?:19|20)[0-9][0-9]', fn)
    #This is a hack to remove peripheral underscores and dashes
    dstr = [d.lstrip('_').rstrip('_') for d in dstr]
    dstr = [d.lstrip('-').rstrip('-') for d in dstr]
    #This returns an empty list of nothing is found
    out = [strptime_fuzzy(s) for s in dstr]
    #This is USGS archive format
    #out = [datetime.strptime(s, '%d%b%y') for s in dstr][0]
    return out