Allow storage of date in ISO (RFC 3339) format

diff --git a/twitter/archiver.py b/twitter/archiver.py
index 587b63e6e4abbc4a34f22f9ff157b0e2213e441c..91a8aa7641bcd420803a08cc44f2a196a7c53e52 100644 (file)
--- a/twitter/archiver.py
+++ b/twitter/archiver.py
@@ -21,6 +21,7 @@ OPTIONS
  -r --redirect-sites   follow redirects for this comma separated list of hosts
  -d --dms <file>       archive own direct messages (both received and
                        sent) into given file name.
  -r --redirect-sites   follow redirects for this comma separated list of hosts
  -d --dms <file>       archive own direct messages (both received and
                        sent) into given file name.

+ -i --isoformat        store dates in ISO format (specifically RFC 3339)

 
 AUTHENTICATION
     Authenticate to Twitter using OAuth to archive tweets of private profiles
 
 AUTHENTICATION
     Authenticate to Twitter using OAuth to archive tweets of private profiles


@@ -30,7 +31,8 @@ AUTHENTICATION
 
 from __future__ import print_function
 
 
 from __future__ import print_function
 

-import os, sys, time, calendar, functools
+import os, sys, time as _time, calendar, functools
+from datetime import time, date, datetime

 from getopt import gnu_getopt as getopt, GetoptError
 
 try:
 from getopt import gnu_getopt as getopt, GetoptError
 
 try:


@@ -51,11 +53,12 @@ from .oauth_dance import oauth_dance
 from .auth import NoAuth
 from .util import Fail, err, expand_line, parse_host_list
 from .follow import lookup
 from .auth import NoAuth
 from .util import Fail, err, expand_line, parse_host_list
 from .follow import lookup

+from .timezones import utc as UTC, Local

 
 def parse_args(args, options):
     """Parse arguments from command-line to set options."""
 
 def parse_args(args, options):
     """Parse arguments from command-line to set options."""

-    long_opts = ['help', 'oauth', 'save-dir=', 'api-rate', 'timeline=', 'mentions=', 'favorites', 'follow-redirects',"redirect-sites=", 'dms=']
-    short_opts = "hos:at:m:vfr:d:"
+    long_opts = ['help', 'oauth', 'save-dir=', 'api-rate', 'timeline=', 'mentions=', 'favorites', 'follow-redirects',"redirect-sites=", 'dms=', 'isoformat']
+    short_opts = "hos:at:m:vfr:d:i"

     opts, extra_args = getopt(args, short_opts, long_opts)
 
     for opt, arg in opts:
     opts, extra_args = getopt(args, short_opts, long_opts)
 
     for opt, arg in opts:


@@ -80,6 +83,8 @@ def parse_args(args, options):
             options['redirect-sites'] = arg
         elif opt in ('-d', '--dms'):
             options['dms'] = arg
             options['redirect-sites'] = arg
         elif opt in ('-d', '--dms'):
             options['dms'] = arg

+        elif opt in ('-i', '--isoformat'):
+            options['isoformat'] = True

 
     options['extra_args'] = extra_args
 
 
     options['extra_args'] = extra_args
 


@@ -124,14 +129,18 @@ def save_tweets(filename, tweets):
 
     archive.close()
 
 
     archive.close()
 

-def format_date(utc, to_localtime=True):
+def format_date(utc, to_localtime=True, isoformat=False):

     """Parse Twitter's UTC date into UTC or local time."""
     """Parse Twitter's UTC date into UTC or local time."""

-    u = time.strptime(utc.replace('+0000','UTC'), '%a %b %d %H:%M:%S %Z %Y')
-    if to_localtime and time.timezone != 0:
-        t = time.localtime(calendar.timegm(u))
-        return time.strftime("%Y-%m-%d %H:%M:%S", t) + " " + time.tzname[1]
+    u = datetime.strptime(utc.replace('+0000','UTC'), '%a %b %d %H:%M:%S %Z %Y')
+    unew = datetime.combine(u.date(), time(u.time().hour,
+        u.time().minute, u.time().second, tzinfo=UTC))
+
+    if to_localtime and _time.timezone != 0:
+        unew = unew.astimezone(Local)
+    if isoformat:
+        return unew.isoformat()

     else:
     else:

-        return time.strftime("%Y-%m-%d %H:%M:%S UTC", u)
+        return unew.strftime('%Y-%m-%d %H:%M:%S %Z')

 
 def expand_format_text(hosts, text):
     """Following redirects in links."""
 
 def expand_format_text(hosts, text):
     """Following redirects in links."""


@@ -169,7 +178,7 @@ def statuses_resolve_uids(twitter, tl):
 
     return new_tl
 
 
     return new_tl
 

-def statuses_portion(twitter, screen_name, max_id=None, mentions=False, favorites=False, received_dms=None):
+def statuses_portion(twitter, screen_name, max_id=None, mentions=False, favorites=False, received_dms=None, isoformat=False):

     """Get a portion of the statuses of a screen name."""
     kwargs = dict(count=200, include_rts=1, screen_name=screen_name)
     if max_id:
     """Get a portion of the statuses of a screen name."""
     kwargs = dict(count=200, include_rts=1, screen_name=screen_name)
     if max_id:


@@ -206,24 +215,24 @@ def statuses_portion(twitter, screen_name, max_id=None, mentions=False, favorite
         # the recipient was, we synthesise a mention. If we're not
         # operating on DMs, behave as normal
         if received_dms == None:
         # the recipient was, we synthesise a mention. If we're not
         # operating on DMs, behave as normal
         if received_dms == None:

-          tweets[t['id']] = "%s <%s> %s" % (format_date(t['created_at']),
+          tweets[t['id']] = "%s <%s> %s" % (format_date(t['created_at'], isoformat=isoformat),

                                             t['user']['screen_name'],
                                             format_text(text))
         else:
                                             t['user']['screen_name'],
                                             format_text(text))
         else:

-          tweets[t['id']] = "%s <%s> @%s %s" % (format_date(t['created_at']),
+          tweets[t['id']] = "%s <%s> @%s %s" % (format_date(t['created_at'], isoformat=isoformat),

                                             t['sender_screen_name'],
                                             t['recipient']['screen_name'],
                                             format_text(text))
     return tweets
 
                                             t['sender_screen_name'],
                                             t['recipient']['screen_name'],
                                             format_text(text))
     return tweets
 

-def statuses(twitter, screen_name, tweets, mentions=False, favorites=False, received_dms=None):
+def statuses(twitter, screen_name, tweets, mentions=False, favorites=False, received_dms=None, isoformat=False):

     """Get all the statuses for a screen name."""
     max_id = None
     fail = Fail()
     # get portions of statuses, incrementing max id until no new tweets appear
     while True:
         try:
     """Get all the statuses for a screen name."""
     max_id = None
     fail = Fail()
     # get portions of statuses, incrementing max id until no new tweets appear
     while True:
         try:

-            portion = statuses_portion(twitter, screen_name, max_id, mentions, favorites, received_dms)
+            portion = statuses_portion(twitter, screen_name, max_id, mentions, favorites, received_dms, isoformat)

         except TwitterError as e:
             if e.e.code == 401:
                 err("Fail: %i Unauthorized (tweets of that user are protected)"
         except TwitterError as e:
             if e.e.code == 401:
                 err("Fail: %i Unauthorized (tweets of that user are protected)"


@@ -290,6 +299,7 @@ def main(args=sys.argv[1:]):
         'favorites': False,
         'follow-redirects': False,
         'redirect-sites': None,
         'favorites': False,
         'follow-redirects': False,
         'redirect-sites': None,

+        'isoformat': False,

     }
     try:
         parse_args(args, options)
     }
     try:
         parse_args(args, options)


@@ -356,7 +366,7 @@ def main(args=sys.argv[1:]):
                 % str(e))
 
         try:
                 % str(e))
 
         try:

-            statuses(twitter, "", tweets, options['mentions'], options['favorites'])
+            statuses(twitter, "", tweets, options['mentions'], options['favorites'], isoformat=options['isoformat'])

         except KeyboardInterrupt:
             err()
             err("Interrupted")
         except KeyboardInterrupt:
             err()
             err("Interrupted")


@@ -384,8 +394,8 @@ def main(args=sys.argv[1:]):
                 % str(e))
 
         try:
                 % str(e))
 
         try:

-            statuses(twitter, "", dms, received_dms=True)
-            statuses(twitter, "", dms, received_dms=False)
+            statuses(twitter, "", dms, received_dms=True, isoformat=options['isoformat'])
+            statuses(twitter, "", dms, received_dms=False, isoformat=options['isoformat'])

         except KeyboardInterrupt:
             err()
             err("Interrupted")
         except KeyboardInterrupt:
             err()
             err("Interrupted")


@@ -418,7 +428,7 @@ def main(args=sys.argv[1:]):
         new = 0
         before = len(tweets)
         try:
         new = 0
         before = len(tweets)
         try:

-            statuses(twitter, user, tweets, options['mentions'], options['favorites'])
+            statuses(twitter, user, tweets, options['mentions'], options['favorites'], isoformat=options['isoformat'])

         except KeyboardInterrupt:
             err()
             err("Interrupted")
         except KeyboardInterrupt:
             err()
             err("Interrupted")

diff --git a/twitter/timezones.py b/twitter/timezones.py
new file mode 100644 (file)
index 0000000..b513e0e
--- /dev/null
+++ b/twitter/timezones.py
@@ -0,0 +1,81 @@
+# Retrieved from http://docs.python.org/2/library/datetime.html on 2013-05-24
+from datetime import tzinfo, timedelta, datetime
+
+ZERO = timedelta(0)
+HOUR = timedelta(hours=1)
+
+# A UTC class.
+
+class UTC(tzinfo):
+    """UTC"""
+
+    def utcoffset(self, dt):
+        return ZERO
+
+    def tzname(self, dt):
+        return "UTC"
+
+    def dst(self, dt):
+        return ZERO
+
+utc = UTC()
+
+# A class building tzinfo objects for fixed-offset time zones.
+# Note that FixedOffset(0, "UTC") is a different way to build a
+# UTC tzinfo object.
+
+class FixedOffset(tzinfo):
+    """Fixed offset in minutes east from UTC."""
+
+    def __init__(self, offset, name):
+        self.__offset = timedelta(minutes = offset)
+        self.__name = name
+
+    def utcoffset(self, dt):
+        return self.__offset
+
+    def tzname(self, dt):
+        return self.__name
+
+    def dst(self, dt):
+        return ZERO
+
+# A class capturing the platform's idea of local time.
+
+import time as _time
+
+STDOFFSET = timedelta(seconds = -_time.timezone)
+if _time.daylight:
+    DSTOFFSET = timedelta(seconds = -_time.altzone)
+else:
+    DSTOFFSET = STDOFFSET
+
+DSTDIFF = DSTOFFSET - STDOFFSET
+
+class LocalTimezone(tzinfo):
+
+    def utcoffset(self, dt):
+        if self._isdst(dt):
+            return DSTOFFSET
+        else:
+            return STDOFFSET
+
+    def dst(self, dt):
+        if self._isdst(dt):
+            return DSTDIFF
+        else:
+            return ZERO
+
+    def tzname(self, dt):
+        return _time.tzname[self._isdst(dt)]
+
+    def _isdst(self, dt):
+        tt = (dt.year, dt.month, dt.day,
+              dt.hour, dt.minute, dt.second,
+              dt.weekday(), 0, 0)
+        stamp = _time.mktime(tt)
+        tt = _time.localtime(stamp)
+        return tt.tm_isdst > 0
+
+Local = LocalTimezone()
+