# From Official Airline Guide - Worldwide Edition (1987). Countries not
# listed here do not observe DST, according to the OAG. Time zone names
# are pure inventions, and none are supplied for countries not observing
# DST; updates from natives would be appreciated. The times that DST
# starts and ends are based on the assumption that they switch a 2AM just
# as everybody else does.
###############################################################################
# The OAG lists October 25, 1987 and February 12, 1988 as the starting and
# ending dates, giving them as "estimated date(s) based on previous year". We
# infer a rule here from one example, always a dangerous practice.... Yes,
# they really do switch on Saturday, according to the OAG.
# "Brazil/Acre" is for the Territory of Acre; "Brazil/DeNoronha" is for
# From Bob Devine (January 28, 1988):
# The only information I found is that there was no DST up to 1985.
# But there was some before 1952!
# From U. S. Naval Observatory (January 16, 1989):
# BRAZIL WEST 5 H BEHIND UTC TERRITORY OF ACRE
# BRAZIL WEST 4 H BEHIND UTC ACRE OCT 23, '88-FEB 11,
# BRAZIL CENTRAL 4 H BEHIND UTC MANAUS
# BRAZIL CENTRAL 3 H BEHIND UTC MANAUS OCT 23, '88-FEB 11,
# BRAZIL CENTRAL '89 (ESTIMATED)
# BRAZIL EAST 3 H BEHIND UTC COASTAL STATES, RIO, SAO
# BRAZIL EAST PAULO, BRASILIA
# BRAZIL EAST 2 H BEHIND UTC COASTAL STATES, RIO, SAO
# BRAZIL PAULO, BRASILIA OCT 23,
# BRAZIL 2 H BEHIND UTC ATLANTIC ISLANDS, FERNANDO
# BRAZIL 1 H BEHIND UTC OCT 23, '88-FEB 11, '89
# BRAZIL 3 H BEHIND UTC FOR MOST MAJOR AIRPORTS.
# Rule NAME FROM TO TYPE IN ON AT SAVE LETTER/S
Rule Brazil 1987 max - Feb Sat<=14 2:00 0 S
Rule Brazil 1987 max - Oct Sat<=28 2:00 1:00 D
# Zone NAME GMTOFF RULES/SAVE FORMAT [UNTIL]
Zone Brazil/East -3:00 Brazil E%sT
Zone Brazil/West -4:00 Brazil W%sT
Zone Brazil/Acre -5:00 Brazil A%sT
Zone Brazil/DeNoronha -2:00 Brazil F%sT
###############################################################################
# The OAG lists October 11, 1987 and March 12, 1988 as the starting and
# ending dates, giving them as "estimated date(s) based on previous year."
# From Bob Devine (January 28, 1988):
# Chile has had 2nd Sunday in October to 2nd Sunday in March DST since 1977.
# Switch is at midnight. OAG is right.
# Rule NAME FROM TO TYPE IN ON AT SAVE LETTER/S
Rule Chile 1977 max - Oct Sun>=8 0:00 1:00 D
Rule Chile 1978 max - Mar Sun>=8 0:00 0 S
# Zone NAME GMTOFF RULES/SAVE FORMAT [UNTIL]
Zone Chile/Continental -4:00 Chile C%sT
Zone Chile/EasterIsland -6:00 Chile E%sT
###############################################################################
# From Bob Devine (January 28, 1988):
# Paraguay: First day in October to last in March. Midnight switch??
# From U. S. Naval Observatory (January 19, 1989):
# PARAGUAY 4 H BEHIND UTC
# PARAGUAY 3 H BEHIND UTC OCT 1, '88-MAR 31, '89
# From Arthur David Olson (January 21, 1989):
# Still don't know time of day when switch occurs. Punt.
###############################################################################
# From Bob Devine (January 28, 1988):
# Argentina: first Sunday in October to first Sunday in April since 1976.
# Double Summer time from 1969 to 1974. Switches at midnight.
# From U. S. Naval Observatory (January 19, 19889):
# ARGENTINA 3 H BEHIND UTC
# From Arthur David Olson (January 21, 1989):
# OAG, USNO and Bob have different data. Punt.
projects / unix-history / blob