file-cmtime-summary

#!/usr/bin/python3
#
# TODO: improve speed. Don't be stupid, mostly - right now things get statted multiple times.
# 
import os,datetime,sys
from datetime import datetime

#take filenames, and/or use os.walk?

datetimes = []
year_count    = { }
month_count   = { 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0, 10:0, 11:0, 12:0 }
month_names   = { 1:'jan', 2:'feb', 3:'mar', 4:'apr', 5:'may', 6:'jun', 7:'jul', 8:'aug', 9:'sep', 10:'oct', 11:'nov', 12:'dec' }
weekday_count = { 0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0 }
weekday_names = { 0:'mon', 1:'tue', 2:'wed', 3:'thu', 4:'fri', 5:'sat', 6:'sun' }
ym_count      = { }


def sameline(s,stream=sys.stderr):
    """ Used for newlineless tally: Prints the given string on the given stream (defaults to stderr),
        then clears the rest of the line and goes to the start of the same line.
        (so to go on with normal printing you'ld want a sameline('') first)
        Requries ANSI capability (TODO: test for that?)

        Tries to prevent new lines caused by wrapping of long strings (filename)
        by printing only first so-many characters of a filename.
    """
    stream.write(s)
    stream.write('\x1b[K\r') #clear rest of line, send cursor to start of line
    stream.flush()


used=0

try:
    args = sys.argv[1:]
    if len(args)==0:
        print( "No arguments, using current directory")
        args=['.']
    #print args

    for bdir in args:
        i=0
        for r,ds,fs in os.walk(bdir):
            for fn in fs:
                i+=1

                #if not '.py' in fn:
                #    continue
                ffn = os.path.join(r,fn)

                try:
                    used+=1
                    # perhaps count ctime as well?
                    for timestamp in (os.stat(ffn).st_mtime, os.stat(ffn).st_ctime):

                        mtime = os.stat(ffn).st_mtime
                        dt = datetime.fromtimestamp(mtime)

                        weekday_count[dt.weekday()] += 1
                        month_count[dt.month]       += 1
                        if dt.year not in year_count:
                            year_count[dt.year]=1
                        else:
                            year_count[dt.year]+=1

                        ym='%04d_%02d'%(dt.year,dt.month)
                        if ym not in ym_count:
                            ym_count[ym]=1
                        else:
                            ym_count[ym]+=1

                except OSError: # probably unresovable symlink?
                    pass

                if i%33==0:
                    msg = 'scanned %d files'%i
                    if used<i:
                        msg+=', used %d'%used
                    sameline(msg)

except KeyboardInterrupt:
    print( "Ctrl-C, showing partial result.")
            
print()
print( "Summarizing %d files"%used)
print()

# TODO: show it histogram-stlye?

histwidth = 80

if 0:
    print( "Years:")
    maxcount = max(year_count.values())
    for year in sorted(year_count):
        print( '  %9s: %6d'%(year, year_count[year]/2), end=' ')
        print( '*'*int(histwidth*year_count[year]/maxcount) )
    print

print( "Months-in-years:")
maxcount = max(ym_count.values())
for ym in sorted(ym_count):
    y,m = ym.split('_')
    print( '  %s %04d: %6d'%(month_names[int(m,10)], int(y),  ym_count[ym]/2), end=' ')
    print( '*'*int(histwidth*ym_count[ym]/maxcount))
print()

if 1:
    print( "Months:")
    for month in sorted(month_count):
        print( '  %8s: %6d'%(month_names[month], month_count[month]/2), end=' ')
        print( '*'*int(histwidth*month_count[month]/maxcount))
    print()

if 0:
    print( "Weekdays:")
    maxcount = max(weekday_count.values())
    for weekday in sorted(weekday_count):
        print( '  %8s: %6d'%(weekday_names[weekday], weekday_count[weekday]/2), end=' ')
        print( '*'*int(histwidth*weekday_count[weekday]/maxcount))
    print()