extractor.py

# opensourced
import re
import string

# Predefined strings.
numbers = "(^a(?=\s)|one|two|three|four|five|six|seven|\
            eight|nine|ten|eleven|twelve|thirteen|fourteen|\
            fifteen|sixteen|seventeen|eighteen|nineteen|twenty|\
            thirty|forty|fifty|sixty|seventy|eighty|ninety|hundred|thousand)"

day = "(monday|tuesday|wednesday|thursday|friday|saturday|sunday)"
week_day = "(monday|tuesday|wednesday|thursday|friday|saturday|sunday)"
month = "(january|february|march|april|may|june|july|august|september|\
          october|november|december)"

dmy = "(year|day|week|month)"
rel_day = "(today|yesterday|tomorrow|tonight|tonite)"
exp1 = "(before|after|earlier|later|ago)"
exp2 = "(this|next|last)"
iso = "\d+[/-]\d+[/-]\d+ \d+:\d+:\d+\.\d+"
year = "((?<=\s)\d{4}|^\d{4})"
date = "([012]?[0-9]|3[01])"

regxp1 = "((\d+|(" + numbers + "[-\s]?)+) " + dmy + "s? " + exp1 + ")"
regxp2 = "(" + exp2 + " (" + dmy + "|" + week_day + "|" + month + "))"
regxp3 = "(" + date + " " + month + " " + year + ")"
regxp4 = "(" + month + " " + date + "[th|st|rd]?[,]? " + year + ")"

reg1 = re.compile(regxp1, re.IGNORECASE)
reg2 = re.compile(regxp2, re.IGNORECASE)
reg3 = re.compile(rel_day, re.IGNORECASE)
reg4 = re.compile(iso)
reg5 = re.compile(year)
reg6 = re.compile(regxp3, re.IGNORECASE)
reg7 = re.compile(regxp4, re.IGNORECASE)

def extractDate(text):

    # Initialization
    timex_found = []

    # re.findall() finds all the substring matches, keep only the full
    # matching string. Captures expressions such as 'number of days' ago, etc.
    found = reg1.findall(text)
    found = [a[0] for a in found if len(a) > 1]
    for timex in found:
        timex_found.append(timex)

    # Variations of this thursday, next year, etc
    found = reg2.findall(text)
    found = [a[0] for a in found if len(a) > 1]
    for timex in found:
        timex_found.append(timex)

    # today, tomorrow, etc
    found = reg3.findall(text)
    for timex in found:
        timex_found.append(timex)

    # ISO
    found = reg4.findall(text)
    for timex in found:
        timex_found.append(timex)

    # Dates
    found = reg6.findall(text)
    found = [a[0] for a in found if len(a) > 1]
    for timex in found:
        timex_found.append(timex)

    found = reg7.findall(text)
    found = [a[0] for a in found if len(a) > 1]
    for timex in found:
        timex_found.append(timex)

    # Year
    found = reg5.findall(text)
    for timex in found:
        timex_found.append(timex)
    # Tag only temporal expressions which haven't been tagged.
    #for timex in timex_found:
    #    text = re.sub(timex + '(?!</TIMEX2>)', '<TIMEX2>' + timex + '</TIMEX2>', text)

    return timex_found