Line data Source code
1 : % Copyright (C) 2020-2021, 2022 John E. Davis
2 : %
3 : % This file is part of the S-Lang Library and may be distributed under the
4 : % terms of the GNU General Public License. See the file COPYING for
5 : % more information.
6 : %---------------------------------------------------------------------------
7 : % The code here attemps to convert a human readable representation of a
8 : % timestamp, such as Wed May 13 02:38:34 2020 to a Unix time (number
9 : % of secs since the Unix epoch)
10 : %
11 : % Public funtions:
12 : %
13 : % timestamp_parse:
14 : % Parses a timestamp and Returns the number of seconds since
15 : % the Unix EPOCH (1970-01-01T00:00:00Z)
16 : %
17 : private variable Months
18 1 : = ["jan", "feb", "mar", "apr", "may", "jun",
19 : "jul", "aug", "sep", "oct", "nov", "dec"];
20 :
21 : % There is an extensive list of timezone abbreviations at
22 : % <https://www.timeanddate.com/time/zones/>. The problem with
23 : % abbreviations is that they are not unique. For example, CST
24 : % could refer to Austrailia, North America, or China.
25 : % The ones here are used by slrn.
26 1 : private variable TZMap = Assoc_Type[Int_Type, 0];
27 1 : TZMap["EDT"] = -400; % US Eastern Daylight Time
28 1 : TZMap["EST"] = -500; % US Eastern Standard Time
29 1 : TZMap["CDT"] = -500; % US Central Daylight Time
30 1 : TZMap["CST"] = -600; % US Central
31 1 : TZMap["MDT"] = -600; % US Mountain Daylight Time
32 1 : TZMap["MST"] = -700; % US Mountain
33 1 : TZMap["PDT"] = -700; % US Pacific Daylight Time
34 1 : TZMap["PST"] = -800; % US Pacific
35 1 : TZMap["GMT"] = 0;
36 1 : TZMap["UTC"] = 0;
37 1 : TZMap["Z"] = 0;
38 1 : TZMap["CET"] = 100; % Central European
39 1 : TZMap["MET"] = 100; % Middle European
40 1 : TZMap["MEZ"] = 100; % Middle European
41 1 : TZMap["EET"] = 200; % Eastern European
42 1 : TZMap["MSK"] = 300; % Moscow
43 1 : TZMap["HKT"] = 800; % Hong Kong
44 1 : TZMap["JST"] = 900; % Japan Standard
45 1 : TZMap["KST"] = 900; % Korean Standard
46 1 : TZMap["CAST"] = 930; % Central Autsralian
47 1 : TZMap["EAST"] = 1000; % Eastern Autsralian
48 1 : TZMap["NZST"] = 1200; % New Zealand Autsralian
49 :
50 : private define map_tzstr_string (tzstr)
51 : {
52 20 : return TZMap[strtrim (strup(tzstr))];
53 : }
54 :
55 : private variable Cumulative_Days_Per_Month =
56 1 : int (cumsum ([0, 31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31]));
57 :
58 1 : private variable TS_Formats = {};
59 : private define add_ts_format (re, indices, month_is_int)
60 : {
61 7 : variable s = struct
62 : {
63 7 : re = re,
64 7 : month_is_int = month_is_int,
65 7 : indices = indices,
66 : };
67 :
68 7 : list_append (TS_Formats, s);
69 : }
70 : % Tue 12 May 2020 04:37:54 [PM] EDT
71 1 : add_ts_format (`^[a-zA-Z,]*` % weekday
72 : + ` *\(\d\d?\)[ -]\([a-zA-Z]+\)[ -]\(\d\d\d*\),?` % day month year
73 : + ` *\(\d\d?\)[:.]\(\d\d\)[:.]?\(\d*\)` % hh:mm:ss
74 : + ` *\(.*\)`, % AM/PM + tz
75 : [1,2,3,4,5,6,7], 0);
76 :
77 : % Sun, Dec 04, 1994 11:05:52 GMT
78 1 : add_ts_format (`^[a-zA-Z,]+`
79 : + ` +\([a-zA-Z]\{3,\}\) \(\d+\),? *\(\d\d\d*\)`% month, day, year
80 : + ` +\(\d\d?\):\(\d\d\):?\(\d*\)`% hh:mm:ss
81 : + ` *\(.*\)`, % tz
82 : [2,1,3,4,5,6,7], 0);
83 :
84 : % Dec 04, 1994 11:05:52 GMT
85 1 : add_ts_format (`^\([a-zA-Z]\{3,\}\) \(\d+\),? *\(\d\d\d*\)`% month, day, year
86 : + ` +\(\d\d?\):\(\d\d\):?\(\d*\)`% hh:mm:ss
87 : + ` *\(.*\)`, % tz
88 : [2,1,3,4,5,6,7], 0);
89 :
90 : % 2020-09-12T21:17:30 <tz-offset>
91 1 : add_ts_format (`^\(\d\d\d\d\)-?\(\d\d\)-?\(\d\d\)`
92 : + `T\(\d\d\):?\(\d\d\):?\(\d\d\)`
93 : + ` *\(.*\)`,
94 : [3,2,1,4,5,6,7], 1);
95 :
96 : % 5/12/2020, 5:31:57 PM
97 1 : add_ts_format (`\(\d\d?\)/\(\d\d?\)/\(\d\d\d*\),?`
98 : + ` *\(\d\d?\):\(\d\d\):?\(\d*\)`
99 : + ` *\(.*\)`,
100 : [2,1,3,4,5,6,7], 1);
101 :
102 : % Dec 4 11:05:52 2020
103 1 : add_ts_format (`^\([a-zA-Z]\{3,\}\) +\(\d+\),?` % month, day
104 : + ` +\(\d\d?\):\(\d\d\):?\(\d*\),?`% hh:mm:ss
105 : + ` +\(\d\d\d*\)` % year
106 : + ` *\(.*\)`, % tz
107 : [2,1,6,3,4,5,7], 0);
108 :
109 : % Tue Dec 4 11:05:52 2020
110 1 : add_ts_format (`^[A-Za-z,]+`
111 : + ` +\([a-zA-Z]\{3,\}\) +\(\d+\),?` % month, day
112 : + ` +\(\d\d?\):\(\d\d\):?\(\d*\),?`% hh:mm:ss
113 : + ` +\(\d\d\d*\)` % year
114 : + ` *\(.*\)`, % tz
115 : [2,1,6,3,4,5,7], 0);
116 :
117 1 : private variable Last_TS_Index = 0;
118 : private define guess_local_timezone_offset ()
119 : {
120 2 : variable now = _time(), tm = gmtime(now);
121 1 : tm.tm_isdst = -1; % Force a lookup to see if DST is in effect
122 1 : variable secs = now - mktime (tm);
123 1 : variable hours = secs/3600;
124 1 : return 100*hours + (secs - 3600*hours)/60;
125 : }
126 :
127 : define timestamp_parse (timestamp)
128 : {
129 29 : timestamp = strtrim (timestamp);
130 : variable day, month, year, hours, minutes, secs, tz, tzstr;
131 29 : variable num = length (TS_Formats);
132 29 : loop (num)
133 : {
134 80 : variable fmt = TS_Formats[Last_TS_Index];
135 :
136 80 : variable matches = string_matches (timestamp, fmt.re);
137 80 : if (matches == NULL)
138 : {
139 51 : Last_TS_Index = (Last_TS_Index + 1) mod num;
140 51 : continue;
141 : }
142 29 : variable ind = fmt.indices;
143 29 : day = atoi (matches[ind[0]]);
144 29 : month = matches[ind[1]];
145 29 : year = atoi (matches[ind[2]]);
146 29 : hours = atoi (matches[ind[3]]);
147 29 : minutes = atoi (matches[ind[4]]);
148 29 : secs = atoi (matches[ind[5]]);
149 29 : tzstr = matches[ind[6]];
150 :
151 29 : if (fmt.month_is_int)
152 10 : month = atoi (month) - 1; % 0 to 11
153 : else
154 : {
155 21 : if (strbytelen (month) > 3) month = month[[0,1,2]];
156 19 : month = wherefirst (Months == strlow(month));
157 19 : if (month == NULL) return NULL;
158 : }
159 29 : break;
160 : }
161 0 : then return NULL;
162 :
163 29 : if (year < 100)
164 : {
165 : % No century
166 4 : year += 1900;
167 8 : if (year < 1950) year += 100;
168 : }
169 29 : tzstr = strtrim (tzstr);
170 :
171 : % Take care of AM/PM
172 29 : if (((tzstr[0] == 'A') || (tzstr[0] == 'P'))
173 : && (tzstr[1] == 'M')
174 : && ((tzstr[2] == 0) || (tzstr[2] == ' ')))
175 : {
176 9 : if (tzstr[0] == 'P')
177 9 : hours += 12;
178 9 : tzstr = strtrim (tzstr[[2:]]);
179 : }
180 :
181 29 : tzstr = strreplace (tzstr, ":", "");
182 29 : if (tzstr == "")
183 1 : tz = guess_local_timezone_offset ();
184 : else
185 : {
186 28 : tz = atoi (tzstr);
187 28 : if ((tz == 0) && (tzstr[0] != '+') && (tzstr[0] != '-'))
188 20 : tz = map_tzstr_string (tzstr);
189 : }
190 :
191 29 : day--; % offset from 0
192 : % Compute the cumulative number of days, accounting for a leap year
193 29 : day += Cumulative_Days_Per_Month[month];
194 29 : if ((month > 2)
195 : && (0 == (year mod 4))
196 : && ((year mod 100) || (0 == (year mod 400))))
197 26 : day++;
198 :
199 29 : year -= 1970; % Unix EPOCH
200 29 : day += 365*year + (year+1)/4; % leap year, every 4 years from 72
201 :
202 : % The TZ is hhmm form, so 600 is 6 hours, and 0 minutes
203 29 : hours -= tz/100;
204 29 : minutes -= tz mod 100;
205 :
206 29 : return secs + 60L*(minutes + 60L*(hours + 24L*day));
207 : }
|
