Mercurial > getan
view scripts/getan-daily-report @ 540:91aaf455c44e
Add new template zeiterfassung-hierarchy1
* Add template for getan-report for the use case where hierarchical
project keys are used to select the workpackages within a project.
The grouping within the new template is done the way it is, because
a second attribute lookup is not possible in Jinja2, when having
just an entry it is not possible to get the `projects[entry.project_id].key`.
author | Bernhard Reiter <bernhard@intevation.de> |
---|---|
date | Thu, 30 Jan 2020 10:47:20 +0100 |
parents | a4044956e08c |
children |
line wrap: on
line source
#!/usr/bin/env python3 """ write a daily report accessing a database from http://hg.intevation.de/getan 1.1dev3. """ # hastily done, written to learn the getan database format and its manipulation # Free Software under GNU GPL v>=3 # 20130109 bernhard@intevation.de # 20140103 bernhard@intevation.de: # started from 2013/getan-writeout-timesorted.py # ported to python3. Removed the dependency for functions from worklog.py. # the timesorted variant can be uncommented in the code for now # 20140109 bernhard@intevation.de: # Total time output format improved. # 20140120 bernhard@intevation.de: # added command line options, requires argparse module now (e.g. Python>=3.2) # 20141104 bernhard@intevation.de: # migration to argparse complete, added -t option # TODO: # * use python v>=3.2 variants in the code where noted. import argparse import datetime import locale import logging import sqlite3 factor = {'privat': 0} # sets the locale for all categories to the user’s default setting #(typically specified in the LANG environment variable). locale.setlocale(locale.LC_ALL, '') l = logging l.basicConfig(level=logging.INFO, # l.basicConfig(level=logging.DEBUG, format='%(message)s') # format='%(asctime)s %(levelname)s %(message)s') def hhhmm_from_timedelta(td): """Return a string '-HHH:MM' from the timedelta parameter. Accounts for way the integer division works for negative numbers: -2 // 60 == -1 -2 % 60 == 58 by first working on the positive number and then adding the minus to the string. For python >=3.1. Successor of hhmm_from_timedelta() from http://intevation.de/cgi-bin/viewcvs-misc.cgi/worklog.py/ . """ total_minutes = abs(round(td.days * 24 * 60 + td.seconds / 60)) # variant for Python v>3.2: # total_minutes = abs(round(td/datetime.timedelta(minutes=1))) hours = total_minutes // 60 minutes = total_minutes % 60 h_string = "{}".format(hours) if(td.days < 0): h_string = "-" + h_string return "{:>3s}:{:02d}".format(h_string, minutes) def self_test(): """Run some simple tests on hhhmm_from_timedelta(). e.g. run like python3 -c 'from getan_report_20140103 import *; self_test()' """ l.info(hhhmm_from_timedelta(datetime.timedelta(minutes=1))) l.info(hhhmm_from_timedelta(datetime.timedelta(minutes=-2))) def main(): parser = argparse.ArgumentParser(description=__doc__) parser.add_argument("-t", action='store_true', help="timesorted output and default reportday to today") parser.add_argument("dbfilename") parser.add_argument("reportday", nargs='?', help="day to report yyyy-mm-dd") args = parser.parse_args() l.debug(args) if args.reportday: try: report_range_start = \ datetime.datetime.strptime(args.reportday, "%Y-%m-%d") except ValueError: report_range_start = \ datetime.datetime.strptime(args.reportday, "%Y%m%d") elif args.t: # start with today 00:00 report_range_start = datetime.datetime.combine( datetime.date.today(), datetime.time()) else: # start with yesterday 00:00 report_range_start = datetime.datetime.combine( datetime.date.today() - datetime.timedelta(days=1), datetime.time()) report_range_end = report_range_start + datetime.timedelta(days=1) l.info("Opening sqlite3 database '%s'" % args.dbfilename) conn = sqlite3.connect( args.dbfilename, detect_types=sqlite3.PARSE_DECLTYPES | sqlite3.PARSE_COLNAMES) c = conn.cursor() tasks = {} task_total = {} c.execute('select * from projects') for t in c: l.debug(t) tasks[t[0]] = t[2] task_total[t[0]] = datetime.timedelta() # from getan 1.0 20130103 # CREATE TABLE entries ( # id INTEGER PRIMARY KEY AUTOINCREMENT, # project_id INTEGER REFERENCES projects(id), # start_time TIMESTAMP NOT NULL, # stop_time TIMESTAMP NOT NULL, # description VARCHAR(256), # # CHECK (strftime('%s', start_time) <= strftime('%s', stop_time)) # ); total_time = datetime.timedelta() if args.t: c.execute('select * from entries order by start_time') else: c.execute('select * from entries order by project_id') for e in c: l.debug(e) # let us ignore microseconds start_time = e[2].replace(microsecond = 0) stop_time = e[3].replace(microsecond = 0) length = stop_time - start_time desc = tasks[e[1]] if start_time >= report_range_start and start_time < report_range_end: if args.t: print("{0:%Y-%m-%d %H:%M}-\n" "{1:%Y-%m-%d %H:%M} {4} {2}: {3}\n". format(start_time, stop_time, desc, e[4], hhhmm_from_timedelta(length))) else: print("{0} {2}: {3} {4}". format(start_time, stop_time, desc, e[4], hhhmm_from_timedelta(length))) if desc in factor: # python3.1 does not allow timedelta division. # TODO: Make python3.1 save or update to python3.2. # l.info("applying factor %f to entry %s" % (factor[desc], e)) # length = (length * int(factor[desc]*1000))/1000 # Until python3.2 we only honor a factor of zero: if factor[desc] == 0: length = datetime.timedelta(0) l.info("not counting {}".format(e)) else: l.info("ignoring factor {}".factor[desc]) total_time += length task_total[e[1]] += length print(report_range_start.strftime("%Y%m%d %a"), end=" ") print("(" + hhhmm_from_timedelta(total_time).strip() + ")") for t in tasks: if task_total[t] != datetime.timedelta(0): print("\t" + tasks[t], hhhmm_from_timedelta(task_total[t])) c.close() if __name__ == "__main__": main()