Created
July 5, 2012 14:49
-
-
Save bertspaan/3054114 to your computer and use it in GitHub Desktop.
Parser for Dutch natural language time strings
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/local/Cellar/python3/3.2.3/bin/python3.2 | |
from pypeg2 import * | |
import datetime | |
numbers = [ | |
"nul", | |
"een", | |
"twee", | |
"drie", | |
"vier", | |
"vijf", | |
"zes", | |
"zeven", | |
"acht", | |
"negen", | |
"tien", | |
"elf", | |
"twaalf", | |
"dertien", | |
"veertien", | |
"vijftien", | |
"zestien", | |
"zeventien", | |
"achttien", | |
"negentien", | |
"twintig", | |
"eenentwintig", | |
"tweeëntwintig", | |
"drieëntwintig", | |
"vierentwintig", | |
"vijfentwintig", | |
"zesentwintig", | |
"zevenentwintig", | |
"achtentwintig", | |
"negenentwintig", | |
"dertig", | |
"eenendertig", | |
"tweeëndertig", | |
"drieëndertig", | |
"vierendertig", | |
"vijfendertig", | |
"zesendertig", | |
"zevenendertig", | |
"achtendertig", | |
"negenendertig", | |
"veertig", | |
"eenenveertig", | |
"tweeënveertig", | |
"drieënveertig", | |
"vierenveertig", | |
"vijfenveertig", | |
"zesenveertig", | |
"zevenenveertig", | |
"achtenveertig", | |
"negenenveertig", | |
"vijftig", | |
"eenenvijftig", | |
"tweeënvijftig", | |
"drieënvijftig", | |
"vierenvijftig", | |
"vijfenvijftig", | |
"zesenvijftig", | |
"zevenenvijftig", | |
"achtenvijftig", | |
"negenenvijftig", | |
"zestig" | |
] | |
times = [ | |
"12 56", | |
"12:56", | |
"twaalf dertig", | |
"negen vijftien", | |
"kwart voor twee", | |
"tien voor twee", | |
"drie over half tien", | |
"13 uur 52", | |
"half drie", | |
"zevenenveertig over tien", | |
"zeven voor half vier", | |
"achttien vijftien", | |
"zeven", | |
"negen uur", | |
"8 uur", | |
"achttien uur vijftien" | |
] | |
class Number(str): | |
grammar = word | |
class Hours(object): | |
grammar = flag("half", K("half")), attr("value", Number), optional("uur") | |
class SignValue(Keyword): | |
grammar = Enum( K("voor"), K("over") ) | |
class Sign(object): | |
grammar = attr("value", SignValue) | |
class Minutes(object): | |
grammar = attr("value", Number) | |
class Time(List): | |
grammar = [ | |
(Minutes, Sign, Hours), | |
(Minutes, Sign, Hours), | |
(Hours, optional(":"), Minutes), | |
Hours | |
] | |
def string_to_int(str): | |
if str == "kwart": | |
return 15 | |
for i in range(0, 60): | |
if str == numbers[i]: | |
return i | |
return int(str) | |
def to_time(thing): | |
minutes_str = "" | |
hours_str = "" | |
half = False | |
sign = 1 | |
for element in thing: | |
if isinstance(element, Hours): | |
hours_str = element.value | |
half = element.half | |
elif isinstance(element, Minutes): | |
minutes_str = element.value | |
elif isinstance(element, Sign): | |
if element.value == "voor": | |
sign = -1 | |
minutes = 0 | |
if len(hours_str) > 0: | |
minutes = string_to_int(hours_str) * 60 | |
if half: | |
minutes -= 30 | |
if len(minutes_str) > 0: | |
minutes += sign * string_to_int(minutes_str) | |
hours = minutes // 60 | |
minutes = minutes - (60 * hours) | |
today = datetime.date.today() #+ datetime.timedelta(days=1) | |
return datetime.datetime.combine(today, datetime.time(hours, minutes)) | |
for time_str in times: | |
f = parse(time_str, Time) | |
time = to_time(f) | |
print(time_str, " => ", time) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment