--- /dev/null
+Return-Path: <jani@nikula.org>\r
+X-Original-To: notmuch@notmuchmail.org\r
+Delivered-To: notmuch@notmuchmail.org\r
+Received: from localhost (localhost [127.0.0.1])\r
+ by olra.theworths.org (Postfix) with ESMTP id 8E9FB431FB6\r
+ for <notmuch@notmuchmail.org>; Sun, 21 Oct 2012 14:22:58 -0700 (PDT)\r
+X-Virus-Scanned: Debian amavisd-new at olra.theworths.org\r
+X-Amavis-Alert: BAD HEADER SECTION, Duplicate header field: "References"\r
+X-Spam-Flag: NO\r
+X-Spam-Score: -0.7\r
+X-Spam-Level: \r
+X-Spam-Status: No, score=-0.7 tagged_above=-999 required=5\r
+ tests=[RCVD_IN_DNSWL_LOW=-0.7] autolearn=disabled\r
+Received: from olra.theworths.org ([127.0.0.1])\r
+ by localhost (olra.theworths.org [127.0.0.1]) (amavisd-new, port 10024)\r
+ with ESMTP id eZSbcm6hwDsw for <notmuch@notmuchmail.org>;\r
+ Sun, 21 Oct 2012 14:22:46 -0700 (PDT)\r
+Received: from mail-lb0-f181.google.com (mail-lb0-f181.google.com\r
+ [209.85.217.181]) (using TLSv1 with cipher RC4-SHA (128/128 bits))\r
+ (No client certificate requested)\r
+ by olra.theworths.org (Postfix) with ESMTPS id 17CA2431FC7\r
+ for <notmuch@notmuchmail.org>; Sun, 21 Oct 2012 14:22:44 -0700 (PDT)\r
+Received: by mail-lb0-f181.google.com with SMTP id gg6so1515820lbb.26\r
+ for <notmuch@notmuchmail.org>; Sun, 21 Oct 2012 14:22:43 -0700 (PDT)\r
+X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;\r
+ d=google.com; s=20120113;\r
+ h=from:to:cc:subject:date:message-id:x-mailer:in-reply-to:references\r
+ :in-reply-to:references:mime-version:content-type\r
+ :content-transfer-encoding:x-gm-message-state;\r
+ bh=gV4daA1OiBG2ORrUkBPTJk0g/NYGu2VSSSUGZVqwUCk=;\r
+ b=JmCFToACqzJ6h7s/eyoQ0SMx8PYbyllCuyNqwLVQOcdizMWLhQhm5KmFcNlUoGzFEg\r
+ yoLxvyamE38oJDL2ukfhMtbQzvWZZCFYX0Pi8T+6RnihYsQ9TTLDC7exfsXHhWD8uy9j\r
+ CtLW+TAdD8AH/9l6TDvU0Y+/1ghgIW1C9JMl9skR4grovXZQAgLoR1lkPq8vckfdnNCG\r
+ AOuiQUbzdSCE7nCOqIKZxRoBAn4R/1BjwkJek8Rd/4i6kI3sF6ofzsWHgrTjiXTCrBiU\r
+ yHKCGtc7HxC7FyIp691m0HFEmx5L0THBxXgCdtMC3xWG6pyqD6gUI31IMpHopvKnnjXe\r
+ Bp3A==\r
+Received: by 10.152.48.111 with SMTP id k15mr6507155lan.17.1350854563259;\r
+ Sun, 21 Oct 2012 14:22:43 -0700 (PDT)\r
+Received: from localhost (dsl-hkibrasgw4-fe51df00-27.dhcp.inet.fi.\r
+ [80.223.81.27])\r
+ by mx.google.com with ESMTPS id n2sm2495589lbg.2.2012.10.21.14.22.40\r
+ (version=SSLv3 cipher=OTHER); Sun, 21 Oct 2012 14:22:42 -0700 (PDT)\r
+From: Jani Nikula <jani@nikula.org>\r
+To: notmuch@notmuchmail.org\r
+Subject: [PATCH v5 2/9] parse-time-string: add a date/time parser to notmuch\r
+Date: Mon, 22 Oct 2012 00:22:25 +0300\r
+Message-Id:\r
+ <a90d3b687895a26f765539d6c0420038a74ee42f.1350854171.git.jani@nikula.org>\r
+X-Mailer: git-send-email 1.7.10.4\r
+In-Reply-To: <cover.1350854171.git.jani@nikula.org>\r
+References: <cover.1350854171.git.jani@nikula.org>\r
+In-Reply-To: <cover.1350854171.git.jani@nikula.org>\r
+References: <cover.1350854171.git.jani@nikula.org>\r
+MIME-Version: 1.0\r
+Content-Type: text/plain; charset=UTF-8\r
+Content-Transfer-Encoding: 8bit\r
+X-Gm-Message-State:\r
+ ALoCoQmO33wjAyAhoCTiXbRgPmQt36TUNYcC37ZLH1fEwVKhFowtalV3qXkZtMeUg8OWzi5FAZPM\r
+X-BeenThere: notmuch@notmuchmail.org\r
+X-Mailman-Version: 2.1.13\r
+Precedence: list\r
+List-Id: "Use and development of the notmuch mail system."\r
+ <notmuch.notmuchmail.org>\r
+List-Unsubscribe: <http://notmuchmail.org/mailman/options/notmuch>,\r
+ <mailto:notmuch-request@notmuchmail.org?subject=unsubscribe>\r
+List-Archive: <http://notmuchmail.org/pipermail/notmuch>\r
+List-Post: <mailto:notmuch@notmuchmail.org>\r
+List-Help: <mailto:notmuch-request@notmuchmail.org?subject=help>\r
+List-Subscribe: <http://notmuchmail.org/mailman/listinfo/notmuch>,\r
+ <mailto:notmuch-request@notmuchmail.org?subject=subscribe>\r
+X-List-Received-Date: Sun, 21 Oct 2012 21:22:58 -0000\r
+\r
+Add a date/time parser to notmuch, to be used for adding date range\r
+query support for notmuch lib later on. Add the parser to a directory\r
+of its own to make it independent of the rest of the notmuch code\r
+base.\r
+\r
+Signed-off-by: Jani Nikula <jani@nikula.org>\r
+---\r
+ Makefile | 2 +-\r
+ parse-time-string/Makefile | 5 +\r
+ parse-time-string/Makefile.local | 12 +\r
+ parse-time-string/README | 9 +\r
+ parse-time-string/parse-time-string.c | 1477 +++++++++++++++++++++++++++++++++\r
+ parse-time-string/parse-time-string.h | 102 +++\r
+ 6 files changed, 1606 insertions(+), 1 deletion(-)\r
+ create mode 100644 parse-time-string/Makefile\r
+ create mode 100644 parse-time-string/Makefile.local\r
+ create mode 100644 parse-time-string/README\r
+ create mode 100644 parse-time-string/parse-time-string.c\r
+ create mode 100644 parse-time-string/parse-time-string.h\r
+\r
+diff --git a/Makefile b/Makefile\r
+index e5e2e3a..bb9c316 100644\r
+--- a/Makefile\r
++++ b/Makefile\r
+@@ -3,7 +3,7 @@\r
+ all:\r
+ \r
+ # List all subdirectories here. Each contains its own Makefile.local\r
+-subdirs = compat completion emacs lib man util test\r
++subdirs = compat completion emacs lib man parse-time-string util test\r
+ \r
+ # We make all targets depend on the Makefiles themselves.\r
+ global_deps = Makefile Makefile.config Makefile.local \\r
+diff --git a/parse-time-string/Makefile b/parse-time-string/Makefile\r
+new file mode 100644\r
+index 0000000..fa25832\r
+--- /dev/null\r
++++ b/parse-time-string/Makefile\r
+@@ -0,0 +1,5 @@\r
++all:\r
++ $(MAKE) -C .. all\r
++\r
++.DEFAULT:\r
++ $(MAKE) -C .. $@\r
+diff --git a/parse-time-string/Makefile.local b/parse-time-string/Makefile.local\r
+new file mode 100644\r
+index 0000000..53534f3\r
+--- /dev/null\r
++++ b/parse-time-string/Makefile.local\r
+@@ -0,0 +1,12 @@\r
++dir := parse-time-string\r
++extra_cflags += -I$(srcdir)/$(dir)\r
++\r
++libparse-time-string_c_srcs := $(dir)/parse-time-string.c\r
++\r
++libparse-time-string_modules := $(libparse-time-string_c_srcs:.c=.o)\r
++\r
++$(dir)/libparse-time-string.a: $(libparse-time-string_modules)\r
++ $(call quiet,AR) rcs $@ $^\r
++\r
++SRCS := $(SRCS) $(libparse-time-string_c_srcs)\r
++CLEAN := $(CLEAN) $(libparse-time-string_modules) $(dir)/libparse-time-string.a\r
+diff --git a/parse-time-string/README b/parse-time-string/README\r
+new file mode 100644\r
+index 0000000..300ff1f\r
+--- /dev/null\r
++++ b/parse-time-string/README\r
+@@ -0,0 +1,9 @@\r
++PARSE TIME STRING\r
++=================\r
++\r
++parse_time_string() is a date/time parser originally written for\r
++notmuch by Jani Nikula <jani@nikula.org>. However, there is nothing\r
++notmuch specific in it, and it should be kept reusable for other\r
++projects, and ready to be packaged on its own as needed. Please do not\r
++add dependencies on or references to anything notmuch specific. The\r
++parser should only depend on the C library.\r
+diff --git a/parse-time-string/parse-time-string.c b/parse-time-string/parse-time-string.c\r
+new file mode 100644\r
+index 0000000..942041a\r
+--- /dev/null\r
++++ b/parse-time-string/parse-time-string.c\r
+@@ -0,0 +1,1477 @@\r
++/*\r
++ * parse time string - user friendly date and time parser\r
++ * Copyright © 2012 Jani Nikula\r
++ *\r
++ * This program is free software: you can redistribute it and/or modify\r
++ * it under the terms of the GNU General Public License as published by\r
++ * the Free Software Foundation, either version 2 of the License, or\r
++ * (at your option) any later version.\r
++ *\r
++ * This program is distributed in the hope that it will be useful,\r
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of\r
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the\r
++ * GNU General Public License for more details.\r
++ *\r
++ * You should have received a copy of the GNU General Public License\r
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.\r
++ *\r
++ * Author: Jani Nikula <jani@nikula.org>\r
++ */\r
++\r
++#include <assert.h>\r
++#include <ctype.h>\r
++#include <errno.h>\r
++#include <limits.h>\r
++#include <stdio.h>\r
++#include <stdarg.h>\r
++#include <stdbool.h>\r
++#include <stdlib.h>\r
++#include <string.h>\r
++#include <strings.h>\r
++#include <time.h>\r
++#include <sys/time.h>\r
++#include <sys/types.h>\r
++\r
++#include "parse-time-string.h"\r
++\r
++/*\r
++ * IMPLEMENTATION DETAILS\r
++ *\r
++ * At a high level, the parsing is done in two phases: 1) actual\r
++ * parsing of the input string and storing the parsed data into\r
++ * 'struct state', and 2) processing of the data in 'struct state'\r
++ * according to current time (or provided reference time) and\r
++ * rounding. This is evident in the main entry point function\r
++ * parse_time_string().\r
++ *\r
++ * 1) The parsing phase - parse_input()\r
++ *\r
++ * Parsing is greedy and happens from left to right. The parsing is as\r
++ * unambiguous as possible; only unambiguous date/time formats are\r
++ * accepted. Redundant or contradictory absolute date/time in the\r
++ * input (e.g. date specified multiple times/ways) is not\r
++ * accepted. Relative date/time on the other hand just accumulates if\r
++ * present multiple times (e.g. "5 days 5 days" just turns into 10\r
++ * days).\r
++ *\r
++ * Parsing decisions are made on the input format, not value. For\r
++ * example, "20/5/2005" fails because the recognized format here is\r
++ * MM/D/YYYY, even though the values would suggest DD/M/YYYY.\r
++ *\r
++ * Parsing is mostly stateless in the sense that parsing decisions are\r
++ * not made based on the values of previously parsed data, or whether\r
++ * certain data is present in the first place. (There are a few\r
++ * exceptions to the latter part, though, such as parsing of time zone\r
++ * that would otherwise look like plain time.)\r
++ *\r
++ * When the parser encounters a number that is not greedily parsed as\r
++ * part of a format, the interpretation is postponed until the next\r
++ * token is parsed. The parser for the next token may consume the\r
++ * previously postponed number. For example, when parsing "20 May" the\r
++ * meaning of "20" is not known until "May" is parsed. If the parser\r
++ * for the next token does not consume the postponed number, the\r
++ * number is handled as a "lone" number before parser for the next\r
++ * token finishes.\r
++ *\r
++ * 2) The processing phase - create_output()\r
++ *\r
++ * Once the parser in phase 1 has finished, 'struct state' contains\r
++ * all the information from the input string, and it's no longer\r
++ * needed. Since the parser does not even handle the concept of "now",\r
++ * the processing initializes the fields referring to the current\r
++ * date/time.\r
++ *\r
++ * If requested, the result is rounded towards past or future. The\r
++ * idea behind rounding is to support parsing date/time ranges in an\r
++ * obvious way. For example, for a range defined as two dates (without\r
++ * time), one would typically want to have an inclusive range from the\r
++ * beginning of start date to the end of the end date. The caller\r
++ * would use rounding towards past in the start date, and towards\r
++ * future in the end date.\r
++ *\r
++ * The absolute date and time is shifted by the relative date and\r
++ * time, and time zone adjustments are made. Daylight saving time\r
++ * (DST) is specifically *not* handled at all.\r
++ *\r
++ * Finally, the result is stored to time_t.\r
++ */\r
++\r
++#define unused(x) x __attribute__ ((unused))\r
++\r
++/* XXX: Redefine these to add i18n support. The keyword table uses\r
++ * N_() to mark strings to be translated; they are accessed\r
++ * dynamically using _(). */\r
++#define _(s) (s) /* i18n: define as gettext (s) */\r
++#define N_(s) (s) /* i18n: define as gettext_noop (s) */\r
++\r
++#define ARRAY_SIZE(a) (sizeof (a) / sizeof (a[0]))\r
++\r
++/*\r
++ * Field indices in the tm and set arrays of struct state.\r
++ *\r
++ * NOTE: There's some code that depends on the ordering of this enum.\r
++ */\r
++enum field {\r
++ /* Keep SEC...YEAR in this order. */\r
++ TM_ABS_SEC, /* seconds */\r
++ TM_ABS_MIN, /* minutes */\r
++ TM_ABS_HOUR, /* hours */\r
++ TM_ABS_MDAY, /* day of the month */\r
++ TM_ABS_MON, /* month */\r
++ TM_ABS_YEAR, /* year */\r
++\r
++ TM_ABS_WDAY, /* day of the week. special: may be relative */\r
++ TM_ABS_ISDST, /* daylight saving time */\r
++\r
++ TM_AMPM, /* am vs. pm */\r
++ TM_TZ, /* timezone in minutes */\r
++\r
++ /* Keep SEC...YEAR in this order. */\r
++ TM_REL_SEC, /* seconds relative to absolute or reference time */\r
++ TM_REL_MIN, /* minutes ... */\r
++ TM_REL_HOUR, /* hours ... */\r
++ TM_REL_DAY, /* days ... */\r
++ TM_REL_MON, /* months ... */\r
++ TM_REL_YEAR, /* years ... */\r
++ TM_REL_WEEK, /* weeks ... */\r
++\r
++ TM_NONE, /* not a field */\r
++\r
++ TM_SIZE = TM_NONE,\r
++ TM_FIRST_ABS = TM_ABS_SEC,\r
++ TM_FIRST_REL = TM_REL_SEC,\r
++};\r
++\r
++/* Values for the set array of struct state. */\r
++enum field_set {\r
++ FIELD_UNSET, /* The field has not been touched by parser. */\r
++ FIELD_SET, /* The field has been set by parser. */\r
++ FIELD_NOW, /* The field will be set to reference time. */\r
++};\r
++\r
++static enum field\r
++next_abs_field (enum field field)\r
++{\r
++ /* NOTE: Depends on the enum ordering. */\r
++ return field < TM_ABS_YEAR ? field + 1 : TM_NONE;\r
++}\r
++\r
++static enum field\r
++abs_to_rel_field (enum field field)\r
++{\r
++ assert (field <= TM_ABS_YEAR);\r
++\r
++ /* NOTE: Depends on the enum ordering. */\r
++ return field + (TM_FIRST_REL - TM_FIRST_ABS);\r
++}\r
++\r
++/* Get epoch value for field. */\r
++static int\r
++field_epoch (enum field field)\r
++{\r
++ if (field == TM_ABS_MDAY || field == TM_ABS_MON)\r
++ return 1;\r
++ else if (field == TM_ABS_YEAR)\r
++ return 1970;\r
++ else\r
++ return 0;\r
++}\r
++\r
++/* The parsing state. */\r
++struct state {\r
++ int tm[TM_SIZE]; /* parsed date and time */\r
++ enum field_set set[TM_SIZE]; /* set status of tm */\r
++\r
++ enum field last_field; /* Previously set field. */\r
++ char delim;\r
++\r
++ int postponed_length; /* Number of digits in postponed value. */\r
++ int postponed_value;\r
++ char postponed_delim; /* The delimiter preceding postponed number. */\r
++};\r
++\r
++/*\r
++ * Helpers for postponed numbers.\r
++ *\r
++ * postponed_length is the number of digits in postponed value. 0\r
++ * means there is no postponed number. -1 means there is a postponed\r
++ * number, but it comes from a keyword, and it doesn't have digits.\r
++ */\r
++static int\r
++get_postponed_length (struct state *state)\r
++{\r
++ return state->postponed_length;\r
++}\r
++\r
++/*\r
++ * Consume a previously postponed number. Return true if a number was\r
++ * in fact postponed, false otherwise. Store the postponed number's\r
++ * value in *v, length in the input string in *n (or -1 if the number\r
++ * was written out and parsed as a keyword), and the preceding\r
++ * delimiter to *d.\r
++ */\r
++static bool\r
++get_postponed_number (struct state *state, int *v, int *n, char *d)\r
++{\r
++ if (!state->postponed_length)\r
++ return false;\r
++\r
++ if (n)\r
++ *n = state->postponed_length;\r
++\r
++ if (v)\r
++ *v = state->postponed_value;\r
++\r
++ if (d)\r
++ *d = state->postponed_delim;\r
++\r
++ state->postponed_length = 0;\r
++ state->postponed_value = 0;\r
++ state->postponed_delim = 0;\r
++\r
++ return true;\r
++}\r
++\r
++static int parse_postponed_number (struct state *state, enum field next_field);\r
++\r
++/*\r
++ * Postpone a number to be handled later. If one exists already,\r
++ * handle it first. n may be -1 to indicate a keyword that has no\r
++ * number length.\r
++ */\r
++static int\r
++set_postponed_number (struct state *state, int v, int n)\r
++{\r
++ int r;\r
++ char d = state->delim;\r
++\r
++ /* Parse a previously postponed number, if any. */\r
++ r = parse_postponed_number (state, TM_NONE);\r
++ if (r)\r
++ return r;\r
++\r
++ state->postponed_length = n;\r
++ state->postponed_value = v;\r
++ state->postponed_delim = d;\r
++\r
++ return 0;\r
++}\r
++\r
++static void\r
++set_delim (struct state *state, char delim)\r
++{\r
++ state->delim = delim;\r
++}\r
++\r
++static void\r
++unset_delim (struct state *state)\r
++{\r
++ state->delim = 0;\r
++}\r
++\r
++/*\r
++ * Field set/get/mod helpers.\r
++ */\r
++\r
++/* Return true if field has been set. */\r
++static bool\r
++is_field_set (struct state *state, enum field field)\r
++{\r
++ assert (field < ARRAY_SIZE (state->tm));\r
++\r
++ return field < ARRAY_SIZE (state->set) &&\r
++ state->set[field] != FIELD_UNSET;\r
++}\r
++\r
++static void\r
++unset_field (struct state *state, enum field field)\r
++{\r
++ assert (field < ARRAY_SIZE (state->tm));\r
++\r
++ state->set[field] = FIELD_UNSET;\r
++ state->tm[field] = 0;\r
++}\r
++\r
++/*\r
++ * Set field to value. A field can only be set once to ensure the\r
++ * input does not contain redundant and potentially conflicting data.\r
++ */\r
++static int\r
++set_field (struct state *state, enum field field, int value)\r
++{\r
++ int r;\r
++\r
++ assert (field < ARRAY_SIZE (state->tm));\r
++\r
++ /* Fields can only be set once. */\r
++ if (field < ARRAY_SIZE (state->set) && state->set[field] != FIELD_UNSET)\r
++ return -PARSE_TIME_ERR_ALREADYSET;\r
++\r
++ state->set[field] = FIELD_SET;\r
++\r
++ /* Parse a previously postponed number, if any. */\r
++ r = parse_postponed_number (state, field);\r
++ if (r)\r
++ return r;\r
++\r
++ unset_delim (state);\r
++\r
++ state->tm[field] = value;\r
++ state->last_field = field;\r
++\r
++ return 0;\r
++}\r
++\r
++/*\r
++ * Mark n fields in fields to be set to the reference date/time in the\r
++ * specified time zone, or local timezone if not specified. The fields\r
++ * will be initialized after parsing is complete and timezone is\r
++ * known.\r
++ */\r
++static int\r
++set_fields_to_now (struct state *state, enum field *fields, size_t n)\r
++{\r
++ size_t i;\r
++ int r;\r
++\r
++ for (i = 0; i < n; i++) {\r
++ r = set_field (state, fields[i], 0);\r
++ if (r)\r
++ return r;\r
++ state->set[fields[i]] = FIELD_NOW;\r
++ }\r
++\r
++ return 0;\r
++}\r
++\r
++/* Modify field by adding value to it. To be used on relative fields,\r
++ * which can be modified multiple times (to accumulate). */\r
++static int\r
++mod_field (struct state *state, enum field field, int value)\r
++{\r
++ int r;\r
++\r
++ assert (field < ARRAY_SIZE (state->tm)); /* assert relative??? */\r
++\r
++ if (field < ARRAY_SIZE (state->set))\r
++ state->set[field] = FIELD_SET;\r
++\r
++ /* Parse a previously postponed number, if any. */\r
++ r = parse_postponed_number (state, field);\r
++ if (r)\r
++ return r;\r
++\r
++ unset_delim (state);\r
++\r
++ state->tm[field] += value;\r
++ state->last_field = field;\r
++\r
++ return 0;\r
++}\r
++\r
++/*\r
++ * Get field value. Make sure the field is set before query. It's most\r
++ * likely an error to call this while parsing (for example fields set\r
++ * as FIELD_NOW will only be set to some value after parsing).\r
++ */\r
++static int\r
++get_field (struct state *state, enum field field)\r
++{\r
++ assert (field < ARRAY_SIZE (state->tm));\r
++\r
++ return state->tm[field];\r
++}\r
++\r
++/*\r
++ * Validity checkers.\r
++ */\r
++static bool is_valid_12hour (int h)\r
++{\r
++ return h >= 0 && h <= 12;\r
++}\r
++\r
++static bool is_valid_time (int h, int m, int s)\r
++{\r
++ /* Allow 24:00:00 to denote end of day. */\r
++ if (h == 24 && m == 0 && s == 0)\r
++ return true;\r
++\r
++ return h >= 0 && h <= 23 && m >= 0 && m <= 59 && s >= 0 && s <= 59;\r
++}\r
++\r
++static bool is_valid_mday (int mday)\r
++{\r
++ return mday >= 1 && mday <= 31;\r
++}\r
++\r
++static bool is_valid_mon (int mon)\r
++{\r
++ return mon >= 1 && mon <= 12;\r
++}\r
++\r
++static bool is_valid_year (int year)\r
++{\r
++ return year >= 1970;\r
++}\r
++\r
++static bool is_valid_date (int year, int mon, int mday)\r
++{\r
++ return is_valid_year (year) && is_valid_mon (mon) && is_valid_mday (mday);\r
++}\r
++\r
++/* Unset indicator for time and date set helpers. */\r
++#define UNSET -1\r
++\r
++/* Time set helper. No input checking. Use UNSET (-1) to leave unset. */\r
++static int\r
++set_abs_time (struct state *state, int hour, int min, int sec)\r
++{\r
++ int r;\r
++\r
++ if (hour != UNSET) {\r
++ if ((r = set_field (state, TM_ABS_HOUR, hour)))\r
++ return r;\r
++ }\r
++\r
++ if (min != UNSET) {\r
++ if ((r = set_field (state, TM_ABS_MIN, min)))\r
++ return r;\r
++ }\r
++\r
++ if (sec != UNSET) {\r
++ if ((r = set_field (state, TM_ABS_SEC, sec)))\r
++ return r;\r
++ }\r
++\r
++ return 0;\r
++}\r
++\r
++/* Date set helper. No input checking. Use UNSET (-1) to leave unset. */\r
++static int\r
++set_abs_date (struct state *state, int year, int mon, int mday)\r
++{\r
++ int r;\r
++\r
++ if (year != UNSET) {\r
++ if ((r = set_field (state, TM_ABS_YEAR, year)))\r
++ return r;\r
++ }\r
++\r
++ if (mon != UNSET) {\r
++ if ((r = set_field (state, TM_ABS_MON, mon)))\r
++ return r;\r
++ }\r
++\r
++ if (mday != UNSET) {\r
++ if ((r = set_field (state, TM_ABS_MDAY, mday)))\r
++ return r;\r
++ }\r
++\r
++ return 0;\r
++}\r
++\r
++/*\r
++ * Keyword parsing and handling.\r
++ */\r
++struct keyword;\r
++typedef int (*setter_t)(struct state *state, struct keyword *kw);\r
++\r
++struct keyword {\r
++ const char *name; /* keyword */\r
++ enum field field; /* field to set, or FIELD_NONE if N/A */\r
++ int value; /* value to set, or 0 if N/A */\r
++ setter_t set; /* function to use for setting, if non-NULL */\r
++};\r
++\r
++/*\r
++ * Setter callback functions for keywords.\r
++ */\r
++static int\r
++kw_set_default (struct state *state, struct keyword *kw)\r
++{\r
++ return set_field (state, kw->field, kw->value);\r
++}\r
++\r
++static int\r
++kw_set_rel (struct state *state, struct keyword *kw)\r
++{\r
++ int multiplier = 1;\r
++\r
++ /* Get a previously set multiplier, if any. */\r
++ get_postponed_number (state, &multiplier, NULL, NULL);\r
++\r
++ /* Accumulate relative field values. */\r
++ return mod_field (state, kw->field, multiplier * kw->value);\r
++}\r
++\r
++static int\r
++kw_set_number (struct state *state, struct keyword *kw)\r
++{\r
++ /* -1 = no length, from keyword. */\r
++ return set_postponed_number (state, kw->value, -1);\r
++}\r
++\r
++static int\r
++kw_set_month (struct state *state, struct keyword *kw)\r
++{\r
++ int n = get_postponed_length (state);\r
++\r
++ /* Consume postponed number if it could be mday. This handles "20\r
++ * January". */\r
++ if (n == 1 || n == 2) {\r
++ int r, v;\r
++\r
++ get_postponed_number (state, &v, NULL, NULL);\r
++\r
++ if (!is_valid_mday (v))\r
++ return -PARSE_TIME_ERR_INVALIDDATE;\r
++\r
++ r = set_field (state, TM_ABS_MDAY, v);\r
++ if (r)\r
++ return r;\r
++ }\r
++\r
++ return set_field (state, kw->field, kw->value);\r
++}\r
++\r
++static int\r
++kw_set_ampm (struct state *state, struct keyword *kw)\r
++{\r
++ int n = get_postponed_length (state);\r
++\r
++ /* Consume postponed number if it could be hour. This handles\r
++ * "5pm". */\r
++ if (n == 1 || n == 2) {\r
++ int r, v;\r
++\r
++ get_postponed_number (state, &v, NULL, NULL);\r
++\r
++ if (!is_valid_12hour (v))\r
++ return -PARSE_TIME_ERR_INVALIDTIME;\r
++\r
++ r = set_abs_time (state, v, 0, 0);\r
++ if (r)\r
++ return r;\r
++ }\r
++\r
++ return set_field (state, kw->field, kw->value);\r
++}\r
++\r
++static int\r
++kw_set_timeofday (struct state *state, struct keyword *kw)\r
++{\r
++ return set_abs_time (state, kw->value, 0, 0);\r
++}\r
++\r
++static int\r
++kw_set_today (struct state *state, unused (struct keyword *kw))\r
++{\r
++ enum field fields[] = { TM_ABS_YEAR, TM_ABS_MON, TM_ABS_MDAY };\r
++\r
++ return set_fields_to_now (state, fields, ARRAY_SIZE (fields));\r
++}\r
++\r
++static int\r
++kw_set_now (struct state *state, unused (struct keyword *kw))\r
++{\r
++ enum field fields[] = { TM_ABS_HOUR, TM_ABS_MIN, TM_ABS_SEC };\r
++\r
++ return set_fields_to_now (state, fields, ARRAY_SIZE (fields));\r
++}\r
++\r
++static int\r
++kw_set_ordinal (struct state *state, struct keyword *kw)\r
++{\r
++ int n, v;\r
++\r
++ /* Require a postponed number. */\r
++ if (!get_postponed_number (state, &v, &n, NULL))\r
++ return -PARSE_TIME_ERR_DATEFORMAT;\r
++\r
++ /* Ordinals are mday. */\r
++ if (n != 1 && n != 2)\r
++ return -PARSE_TIME_ERR_DATEFORMAT;\r
++\r
++ /* Be strict about st, nd, rd, and lax about th. */\r
++ if (strcasecmp (kw->name, "st") == 0 && v != 1 && v != 21 && v != 31)\r
++ return -PARSE_TIME_ERR_INVALIDDATE;\r
++ else if (strcasecmp (kw->name, "nd") == 0 && v != 2 && v != 22)\r
++ return -PARSE_TIME_ERR_INVALIDDATE;\r
++ else if (strcasecmp (kw->name, "rd") == 0 && v != 3 && v != 23)\r
++ return -PARSE_TIME_ERR_INVALIDDATE;\r
++ else if (strcasecmp (kw->name, "th") == 0 && !is_valid_mday (v))\r
++ return -PARSE_TIME_ERR_INVALIDDATE;\r
++\r
++ return set_field (state, TM_ABS_MDAY, v);\r
++}\r
++\r
++/*\r
++ * Accepted keywords.\r
++ *\r
++ * A keyword may optionally contain a '|' to indicate the minimum\r
++ * match length. Without one, full match is required. It's advisable\r
++ * to keep the minimum match parts unique across all keywords.\r
++ *\r
++ * If keyword begins with upper case letter, then the matching will be\r
++ * case sensitive. Otherwise the matching is case insensitive.\r
++ *\r
++ * If setter is NULL, set_default will be used.\r
++ *\r
++ * Note: Order matters. Matching is greedy, longest match is used, but\r
++ * of equal length matches the first one is used, unless there's an\r
++ * equal length case sensitive match which trumps case insensitive\r
++ * matches.\r
++ */\r
++static struct keyword keywords[] = {\r
++ /* Weekdays. */\r
++ { N_("sun|day"), TM_ABS_WDAY, 0, NULL },\r
++ { N_("mon|day"), TM_ABS_WDAY, 1, NULL },\r
++ { N_("tue|sday"), TM_ABS_WDAY, 2, NULL },\r
++ { N_("wed|nesday"), TM_ABS_WDAY, 3, NULL },\r
++ { N_("thu|rsday"), TM_ABS_WDAY, 4, NULL },\r
++ { N_("fri|day"), TM_ABS_WDAY, 5, NULL },\r
++ { N_("sat|urday"), TM_ABS_WDAY, 6, NULL },\r
++\r
++ /* Months. */\r
++ { N_("jan|uary"), TM_ABS_MON, 1, kw_set_month },\r
++ { N_("feb|ruary"), TM_ABS_MON, 2, kw_set_month },\r
++ { N_("mar|ch"), TM_ABS_MON, 3, kw_set_month },\r
++ { N_("apr|il"), TM_ABS_MON, 4, kw_set_month },\r
++ { N_("may"), TM_ABS_MON, 5, kw_set_month },\r
++ { N_("jun|e"), TM_ABS_MON, 6, kw_set_month },\r
++ { N_("jul|y"), TM_ABS_MON, 7, kw_set_month },\r
++ { N_("aug|ust"), TM_ABS_MON, 8, kw_set_month },\r
++ { N_("sep|tember"), TM_ABS_MON, 9, kw_set_month },\r
++ { N_("oct|ober"), TM_ABS_MON, 10, kw_set_month },\r
++ { N_("nov|ember"), TM_ABS_MON, 11, kw_set_month },\r
++ { N_("dec|ember"), TM_ABS_MON, 12, kw_set_month },\r
++\r
++ /* Durations. */\r
++ { N_("y|ears"), TM_REL_YEAR, 1, kw_set_rel },\r
++ { N_("w|eeks"), TM_REL_WEEK, 1, kw_set_rel },\r
++ { N_("d|ays"), TM_REL_DAY, 1, kw_set_rel },\r
++ { N_("h|ours"), TM_REL_HOUR, 1, kw_set_rel },\r
++ { N_("hr|s"), TM_REL_HOUR, 1, kw_set_rel },\r
++ { N_("m|inutes"), TM_REL_MIN, 1, kw_set_rel },\r
++ /* M=months, m=minutes */\r
++ { N_("M"), TM_REL_MON, 1, kw_set_rel },\r
++ { N_("mins"), TM_REL_MIN, 1, kw_set_rel },\r
++ { N_("mo|nths"), TM_REL_MON, 1, kw_set_rel },\r
++ { N_("s|econds"), TM_REL_SEC, 1, kw_set_rel },\r
++ { N_("secs"), TM_REL_SEC, 1, kw_set_rel },\r
++\r
++ /* Numbers. */\r
++ { N_("one"), TM_NONE, 1, kw_set_number },\r
++ { N_("two"), TM_NONE, 2, kw_set_number },\r
++ { N_("three"), TM_NONE, 3, kw_set_number },\r
++ { N_("four"), TM_NONE, 4, kw_set_number },\r
++ { N_("five"), TM_NONE, 5, kw_set_number },\r
++ { N_("six"), TM_NONE, 6, kw_set_number },\r
++ { N_("seven"), TM_NONE, 7, kw_set_number },\r
++ { N_("eight"), TM_NONE, 8, kw_set_number },\r
++ { N_("nine"), TM_NONE, 9, kw_set_number },\r
++ { N_("ten"), TM_NONE, 10, kw_set_number },\r
++ { N_("dozen"), TM_NONE, 12, kw_set_number },\r
++ { N_("hundred"), TM_NONE, 100, kw_set_number },\r
++\r
++ /* Special number forms. */\r
++ { N_("this"), TM_NONE, 0, kw_set_number },\r
++ { N_("last"), TM_NONE, 1, kw_set_number },\r
++\r
++ /* Other special keywords. */\r
++ { N_("yesterday"), TM_REL_DAY, 1, kw_set_rel },\r
++ { N_("today"), TM_NONE, 0, kw_set_today },\r
++ { N_("now"), TM_NONE, 0, kw_set_now },\r
++ { N_("noon"), TM_NONE, 12, kw_set_timeofday },\r
++ { N_("midnight"), TM_NONE, 0, kw_set_timeofday },\r
++ { N_("am"), TM_AMPM, 0, kw_set_ampm },\r
++ { N_("a.m."), TM_AMPM, 0, kw_set_ampm },\r
++ { N_("pm"), TM_AMPM, 1, kw_set_ampm },\r
++ { N_("p.m."), TM_AMPM, 1, kw_set_ampm },\r
++ { N_("st"), TM_NONE, 0, kw_set_ordinal },\r
++ { N_("nd"), TM_NONE, 0, kw_set_ordinal },\r
++ { N_("rd"), TM_NONE, 0, kw_set_ordinal },\r
++ { N_("th"), TM_NONE, 0, kw_set_ordinal },\r
++\r
++ /* Timezone codes: offset in minutes. XXX: Add more codes. */\r
++ { N_("pst"), TM_TZ, -8*60, NULL },\r
++ { N_("mst"), TM_TZ, -7*60, NULL },\r
++ { N_("cst"), TM_TZ, -6*60, NULL },\r
++ { N_("est"), TM_TZ, -5*60, NULL },\r
++ { N_("ast"), TM_TZ, -4*60, NULL },\r
++ { N_("nst"), TM_TZ, -(3*60+30), NULL },\r
++\r
++ { N_("gmt"), TM_TZ, 0, NULL },\r
++ { N_("utc"), TM_TZ, 0, NULL },\r
++\r
++ { N_("wet"), TM_TZ, 0, NULL },\r
++ { N_("cet"), TM_TZ, 1*60, NULL },\r
++ { N_("eet"), TM_TZ, 2*60, NULL },\r
++ { N_("fet"), TM_TZ, 3*60, NULL },\r
++\r
++ { N_("wat"), TM_TZ, 1*60, NULL },\r
++ { N_("cat"), TM_TZ, 2*60, NULL },\r
++ { N_("eat"), TM_TZ, 3*60, NULL },\r
++};\r
++\r
++/*\r
++ * Compare strings s and keyword. Return number of matching chars on\r
++ * match, 0 for no match. Match must be at least n chars, or all of\r
++ * keyword if n < 0, otherwise it's not a match. Use match_case for\r
++ * case sensitive matching.\r
++ */\r
++static size_t\r
++match_keyword (const char *s, const char *keyword, ssize_t n, bool match_case)\r
++{\r
++ ssize_t i;\r
++\r
++ if (!n)\r
++ return 0;\r
++\r
++ for (i = 0; *s && *keyword; i++, s++, keyword++) {\r
++ if (match_case) {\r
++ if (*s != *keyword)\r
++ break;\r
++ } else {\r
++ if (tolower ((unsigned char) *s) !=\r
++ tolower ((unsigned char) *keyword))\r
++ break;\r
++ }\r
++ }\r
++\r
++ if (n > 0)\r
++ return i < n ? 0 : i;\r
++ else\r
++ return *keyword ? 0 : i;\r
++}\r
++\r
++/*\r
++ * Parse a keyword. Return < 0 on error, number of parsed chars on\r
++ * success.\r
++ */\r
++static ssize_t\r
++parse_keyword (struct state *state, const char *s)\r
++{\r
++ unsigned int i;\r
++ size_t n, max_n = 0;\r
++ struct keyword *kw = NULL;\r
++ int r;\r
++\r
++ /* Match longest keyword */\r
++ for (i = 0; i < ARRAY_SIZE (keywords); i++) {\r
++ /* Match case if keyword begins with upper case letter. */\r
++ bool mcase = isupper ((unsigned char) keywords[i].name[0]);\r
++ ssize_t minlen = -1;\r
++ char keyword[128];\r
++ char *p;\r
++\r
++ strncpy (keyword, _(keywords[i].name), sizeof (keyword));\r
++\r
++ /* Truncate too long keywords. XXX: Make this dynamic? */\r
++ keyword[sizeof (keyword) - 1] = '\0';\r
++\r
++ /* Minimum match length. */\r
++ p = strchr (keyword, '|');\r
++ if (p) {\r
++ minlen = p - keyword;\r
++\r
++ /* Remove the minimum match length separator. */\r
++ memmove (p, p + 1, strlen (p + 1) + 1);\r
++ }\r
++\r
++ n = match_keyword (s, keyword, minlen, mcase);\r
++ if (n > max_n || (n == max_n && mcase)) {\r
++ max_n = n;\r
++ kw = &keywords[i];\r
++ }\r
++ }\r
++\r
++ if (!kw)\r
++ return -PARSE_TIME_ERR_KEYWORD;\r
++\r
++ if (kw->set)\r
++ r = kw->set (state, kw);\r
++ else\r
++ r = kw_set_default (state, kw);\r
++\r
++ if (r < 0)\r
++ return r;\r
++\r
++ return max_n;\r
++}\r
++\r
++/*\r
++ * Non-keyword parsers and their helpers.\r
++ */\r
++\r
++static int\r
++set_user_tz (struct state *state, char sign, int hour, int min)\r
++{\r
++ int tz = hour * 60 + min;\r
++\r
++ assert (sign == '+' || sign == '-');\r
++\r
++ if (hour < 0 || hour > 14 || min < 0 || min > 59 || min % 15)\r
++ return -PARSE_TIME_ERR_INVALIDTIME;\r
++\r
++ if (sign == '-')\r
++ tz = -tz;\r
++\r
++ return set_field (state, TM_TZ, tz);\r
++}\r
++\r
++/*\r
++ * Parse a previously postponed number if one exists. Independent\r
++ * parsing of a postponed number when it wasn't consumed during\r
++ * parsing of the following token.\r
++ */\r
++static int\r
++parse_postponed_number (struct state *state, unused (enum field next_field))\r
++{\r
++ int v, n;\r
++ char d;\r
++\r
++ /* Bail out if there's no postponed number. */\r
++ if (!get_postponed_number (state, &v, &n, &d))\r
++ return 0;\r
++\r
++ if (n == 1 || n == 2) {\r
++ /* Notable exception: Previous field affects parsing. This\r
++ * handles "January 20". */\r
++ if (state->last_field == TM_ABS_MON) {\r
++ /* D[D] */\r
++ if (!is_valid_mday (v))\r
++ return -PARSE_TIME_ERR_INVALIDDATE;\r
++\r
++ return set_field (state, TM_ABS_MDAY, v);\r
++ } else if (n == 2) {\r
++ /* XXX: Only allow if last field is hour, min, or sec? */\r
++ if (d == '+' || d == '-') {\r
++ /* +/-HH */\r
++ return set_user_tz (state, d, v, 0);\r
++ }\r
++ }\r
++ } else if (n == 4) {\r
++ /* Notable exception: Value affects parsing. Time zones are\r
++ * always at most 1400 and we don't understand years before\r
++ * 1970. */\r
++ if (!is_valid_year (v)) {\r
++ if (d == '+' || d == '-') {\r
++ /* +/-HHMM */\r
++ return set_user_tz (state, d, v / 100, v % 100);\r
++ }\r
++ } else {\r
++ /* YYYY */\r
++ return set_field (state, TM_ABS_YEAR, v);\r
++ }\r
++ } else if (n == 6) {\r
++ /* HHMMSS */\r
++ int hour = v / 10000;\r
++ int min = (v / 100) % 100;\r
++ int sec = v % 100;\r
++\r
++ if (!is_valid_time (hour, min, sec))\r
++ return -PARSE_TIME_ERR_INVALIDTIME;\r
++\r
++ return set_abs_time (state, hour, min, sec);\r
++ } else if (n == 8) {\r
++ /* YYYYMMDD */\r
++ int year = v / 10000;\r
++ int mon = (v / 100) % 100;\r
++ int mday = v % 100;\r
++\r
++ if (!is_valid_date (year, mon, mday))\r
++ return -PARSE_TIME_ERR_INVALIDDATE;\r
++\r
++ return set_abs_date (state, year, mon, mday);\r
++ } else {\r
++ return -PARSE_TIME_ERR_FORMAT;\r
++ }\r
++\r
++ return -PARSE_TIME_ERR_FORMAT;\r
++}\r
++\r
++static int tm_get_field (const struct tm *tm, enum field field);\r
++\r
++static int\r
++set_timestamp (struct state *state, time_t t)\r
++{\r
++ struct tm tm;\r
++ enum field f;\r
++ int r;\r
++\r
++ if (gmtime_r (&t, &tm) == NULL)\r
++ return -PARSE_TIME_ERR_LIB;\r
++\r
++ for (f = TM_ABS_SEC; f != TM_NONE; f = next_abs_field (f)) {\r
++ r = set_field (state, f, tm_get_field (&tm, f));\r
++ if (r)\r
++ return r;\r
++ }\r
++\r
++ r = set_field (state, TM_TZ, 0);\r
++ if (r)\r
++ return r;\r
++\r
++ /* XXX: Prevent TM_AMPM with timestamp, e.g. "@123456 pm" */\r
++\r
++ return 0;\r
++}\r
++\r
++/* Parse a single number. Typically postpone parsing until later. */\r
++static int\r
++parse_single_number (struct state *state, unsigned long v,\r
++ unsigned long n)\r
++{\r
++ assert (n);\r
++\r
++ if (state->delim == '@')\r
++ return set_timestamp (state, (time_t) v);\r
++\r
++ if (v > INT_MAX)\r
++ return -PARSE_TIME_ERR_FORMAT;\r
++\r
++ return set_postponed_number (state, v, n);\r
++}\r
++\r
++static bool\r
++is_time_sep (char c)\r
++{\r
++ return c == ':';\r
++}\r
++\r
++static bool\r
++is_date_sep (char c)\r
++{\r
++ return c == '/' || c == '-' || c == '.';\r
++}\r
++\r
++static bool\r
++is_sep (char c)\r
++{\r
++ return is_time_sep (c) || is_date_sep (c);\r
++}\r
++\r
++/* Two-digit year: 00...69 is 2000s, 70...99 1900s, if n == 0 keep\r
++ * unset. */\r
++static int\r
++expand_year (unsigned long year, size_t n)\r
++{\r
++ if (n == 2) {\r
++ return (year < 70 ? 2000 : 1900) + year;\r
++ } else if (n == 4) {\r
++ return year;\r
++ } else {\r
++ return UNSET;\r
++ }\r
++}\r
++\r
++/* Parse a date number triplet. */\r
++static int\r
++parse_date (struct state *state, char sep,\r
++ unsigned long v1, unsigned long v2, unsigned long v3,\r
++ size_t n1, size_t n2, size_t n3)\r
++{\r
++ int year = UNSET, mon = UNSET, mday = UNSET;\r
++\r
++ assert (is_date_sep (sep));\r
++\r
++ switch (sep) {\r
++ case '/': /* Date: M[M]/D[D][/YY[YY]] or M[M]/YYYY */\r
++ if (n1 != 1 && n1 != 2)\r
++ return -PARSE_TIME_ERR_DATEFORMAT;\r
++\r
++ if ((n2 == 1 || n2 == 2) && (n3 == 0 || n3 == 2 || n3 == 4)) {\r
++ /* M[M]/D[D][/YY[YY]] */\r
++ year = expand_year (v3, n3);\r
++ mon = v1;\r
++ mday = v2;\r
++ } else if (n2 == 4 && n3 == 0) {\r
++ /* M[M]/YYYY */\r
++ year = v2;\r
++ mon = v1;\r
++ } else {\r
++ return -PARSE_TIME_ERR_DATEFORMAT;\r
++ }\r
++ break;\r
++\r
++ case '-': /* Date: YYYY-MM[-DD] or DD-MM[-YY[YY]] or MM-YYYY */\r
++ if (n1 == 4 && n2 == 2 && (n3 == 0 || n3 == 2)) {\r
++ /* YYYY-MM[-DD] */\r
++ year = v1;\r
++ mon = v2;\r
++ if (n3)\r
++ mday = v3;\r
++ } else if (n1 == 2 && n2 == 2 && (n3 == 0 || n3 == 2 || n3 == 4)) {\r
++ /* DD-MM[-YY[YY]] */\r
++ year = expand_year (v3, n3);\r
++ mon = v2;\r
++ mday = v1;\r
++ } else if (n1 == 2 && n2 == 4 && n3 == 0) {\r
++ /* MM-YYYY */\r
++ year = v2;\r
++ mon = v1;\r
++ } else {\r
++ return -PARSE_TIME_ERR_DATEFORMAT;\r
++ }\r
++ break;\r
++\r
++ case '.': /* Date: D[D].M[M][.[YY[YY]]] */\r
++ if ((n1 != 1 && n1 != 2) || (n2 != 1 && n2 != 2) ||\r
++ (n3 != 0 && n3 != 2 && n3 != 4))\r
++ return -PARSE_TIME_ERR_DATEFORMAT;\r
++\r
++ year = expand_year (v3, n3);\r
++ mon = v2;\r
++ mday = v1;\r
++ break;\r
++ }\r
++\r
++ if (year != UNSET && !is_valid_year (year))\r
++ return -PARSE_TIME_ERR_INVALIDDATE;\r
++\r
++ if (mon != UNSET && !is_valid_mon (mon))\r
++ return -PARSE_TIME_ERR_INVALIDDATE;\r
++\r
++ if (mday != UNSET && !is_valid_mday (mday))\r
++ return -PARSE_TIME_ERR_INVALIDDATE;\r
++\r
++ return set_abs_date (state, year, mon, mday);\r
++}\r
++\r
++/* Parse a time number triplet. */\r
++static int\r
++parse_time (struct state *state, char sep,\r
++ unsigned long v1, unsigned long v2, unsigned long v3,\r
++ size_t n1, size_t n2, size_t n3)\r
++{\r
++ assert (is_time_sep (sep));\r
++\r
++ if ((n1 != 1 && n1 != 2) || n2 != 2 || (n3 != 0 && n3 != 2))\r
++ return -PARSE_TIME_ERR_TIMEFORMAT;\r
++\r
++ /*\r
++ * Notable exception: Previously set fields affect\r
++ * parsing. Interpret (+|-)HH:MM as time zone only if hour and\r
++ * minute have been set.\r
++ *\r
++ * XXX: This could be fixed by restricting the delimiters\r
++ * preceding time. For '+' it would be justified, but for '-' it\r
++ * might be inconvenient. However prefer to allow '-' as an\r
++ * insignificant delimiter preceding time for convenience, and\r
++ * handle '+' the same way for consistency between positive and\r
++ * negative time zones.\r
++ */\r
++ if (is_field_set (state, TM_ABS_HOUR) &&\r
++ is_field_set (state, TM_ABS_MIN) &&\r
++ n1 == 2 && n2 == 2 && n3 == 0 &&\r
++ (state->delim == '+' || state->delim == '-')) {\r
++ return set_user_tz (state, state->delim, v1, v2);\r
++ }\r
++\r
++ if (!is_valid_time (v1, v2, v3))\r
++ return -PARSE_TIME_ERR_INVALIDTIME;\r
++\r
++ return set_abs_time (state, v1, v2, n3 ? v3 : 0);\r
++}\r
++\r
++/* strtoul helper that assigns length. */\r
++static unsigned long\r
++strtoul_len (const char *s, const char **endp, size_t *len)\r
++{\r
++ unsigned long val = strtoul (s, (char **) endp, 10);\r
++\r
++ *len = *endp - s;\r
++ return val;\r
++}\r
++\r
++/*\r
++ * Parse a (group of) number(s). Return < 0 on error, number of parsed\r
++ * chars on success.\r
++ */\r
++static ssize_t\r
++parse_number (struct state *state, const char *s)\r
++{\r
++ int r;\r
++ unsigned long v1, v2, v3 = 0;\r
++ size_t n1, n2, n3 = 0;\r
++ const char *p = s;\r
++ char sep;\r
++\r
++ v1 = strtoul_len (p, &p, &n1);\r
++\r
++ if (is_sep (*p) && isdigit ((unsigned char) *(p + 1))) {\r
++ sep = *p;\r
++ v2 = strtoul_len (p + 1, &p, &n2);\r
++ } else {\r
++ /* A single number. */\r
++ r = parse_single_number (state, v1, n1);\r
++ if (r)\r
++ return r;\r
++\r
++ return p - s;\r
++ }\r
++\r
++ /* A group of two or three numbers? */\r
++ if (*p == sep && isdigit ((unsigned char) *(p + 1)))\r
++ v3 = strtoul_len (p + 1, &p, &n3);\r
++\r
++ if (is_time_sep (sep))\r
++ r = parse_time (state, sep, v1, v2, v3, n1, n2, n3);\r
++ else\r
++ r = parse_date (state, sep, v1, v2, v3, n1, n2, n3);\r
++\r
++ if (r)\r
++ return r;\r
++\r
++ return p - s;\r
++}\r
++\r
++/*\r
++ * Parse delimiter(s). Throw away all except the last one, which is\r
++ * stored for parsing the next non-delimiter. Return < 0 on error,\r
++ * number of parsed chars on success.\r
++ *\r
++ * XXX: We might want to be more strict here.\r
++ */\r
++static ssize_t\r
++parse_delim (struct state *state, const char *s)\r
++{\r
++ const char *p = s;\r
++\r
++ /*\r
++ * Skip non-alpha and non-digit, and store the last for further\r
++ * processing.\r
++ */\r
++ while (*p && !isalnum ((unsigned char) *p)) {\r
++ set_delim (state, *p);\r
++ p++;\r
++ }\r
++\r
++ return p - s;\r
++}\r
++\r
++/*\r
++ * Parse a date/time string. Return < 0 on error, number of parsed\r
++ * chars on success.\r
++ */\r
++static ssize_t\r
++parse_input (struct state *state, const char *s)\r
++{\r
++ const char *p = s;\r
++ ssize_t n;\r
++ int r;\r
++\r
++ while (*p) {\r
++ if (isalpha ((unsigned char) *p)) {\r
++ n = parse_keyword (state, p);\r
++ } else if (isdigit ((unsigned char) *p)) {\r
++ n = parse_number (state, p);\r
++ } else {\r
++ n = parse_delim (state, p);\r
++ }\r
++\r
++ if (n <= 0) {\r
++ if (n == 0)\r
++ n = -PARSE_TIME_ERR;\r
++\r
++ return n;\r
++ }\r
++\r
++ p += n;\r
++ }\r
++\r
++ /* Parse a previously postponed number, if any. */\r
++ r = parse_postponed_number (state, TM_NONE);\r
++ if (r < 0)\r
++ return r;\r
++\r
++ return p - s;\r
++}\r
++\r
++/*\r
++ * Processing the parsed input.\r
++ */\r
++\r
++/*\r
++ * Initialize reference time to tm. Use time zone in state if\r
++ * specified, otherwise local time. Use now for reference time if\r
++ * non-NULL, otherwise current time.\r
++ */\r
++static int\r
++initialize_now (struct state *state, struct tm *tm, const time_t *now)\r
++{\r
++ time_t t;\r
++\r
++ if (now) {\r
++ t = *now;\r
++ } else {\r
++ if (time (&t) == (time_t) -1)\r
++ return -PARSE_TIME_ERR_LIB;\r
++ }\r
++\r
++ if (is_field_set (state, TM_TZ)) {\r
++ /* Some other time zone. */\r
++\r
++ /* Adjust now according to the TZ. */\r
++ t += get_field (state, TM_TZ) * 60;\r
++\r
++ /* It's not gm, but this doesn't mess with the TZ. */\r
++ if (gmtime_r (&t, tm) == NULL)\r
++ return -PARSE_TIME_ERR_LIB;\r
++ } else {\r
++ /* Local time. */\r
++ if (localtime_r (&t, tm) == NULL)\r
++ return -PARSE_TIME_ERR_LIB;\r
++ }\r
++\r
++ return 0;\r
++}\r
++\r
++/*\r
++ * Normalize tm according to mktime(3). Both mktime(3) and\r
++ * localtime_r(3) use local time, but they cancel each other out here,\r
++ * making this function agnostic to time zone.\r
++ */\r
++static int\r
++normalize_tm (struct tm *tm)\r
++{\r
++ time_t t = mktime (tm);\r
++\r
++ if (t == (time_t) -1)\r
++ return -PARSE_TIME_ERR_LIB;\r
++\r
++ if (!localtime_r (&t, tm))\r
++ return -PARSE_TIME_ERR_LIB;\r
++\r
++ return 0;\r
++}\r
++\r
++/* Get field out of a struct tm. */\r
++static int\r
++tm_get_field (const struct tm *tm, enum field field)\r
++{\r
++ switch (field) {\r
++ case TM_ABS_SEC: return tm->tm_sec;\r
++ case TM_ABS_MIN: return tm->tm_min;\r
++ case TM_ABS_HOUR: return tm->tm_hour;\r
++ case TM_ABS_MDAY: return tm->tm_mday;\r
++ case TM_ABS_MON: return tm->tm_mon + 1; /* 0- to 1-based */\r
++ case TM_ABS_YEAR: return 1900 + tm->tm_year;\r
++ case TM_ABS_WDAY: return tm->tm_wday;\r
++ case TM_ABS_ISDST: return tm->tm_isdst;\r
++ default:\r
++ assert (false);\r
++ break;\r
++ }\r
++\r
++ return 0;\r
++}\r
++\r
++/* Modify hour according to am/pm setting. */\r
++static int\r
++fixup_ampm (struct state *state)\r
++{\r
++ int hour, hdiff = 0;\r
++\r
++ if (!is_field_set (state, TM_AMPM))\r
++ return 0;\r
++\r
++ if (!is_field_set (state, TM_ABS_HOUR))\r
++ return -PARSE_TIME_ERR_TIMEFORMAT;\r
++\r
++ hour = get_field (state, TM_ABS_HOUR);\r
++ if (!is_valid_12hour (hour))\r
++ return -PARSE_TIME_ERR_INVALIDTIME;\r
++\r
++ if (get_field (state, TM_AMPM)) {\r
++ /* 12pm is noon. */\r
++ if (hour != 12)\r
++ hdiff = 12;\r
++ } else {\r
++ /* 12am is midnight, beginning of day. */\r
++ if (hour == 12)\r
++ hdiff = -12;\r
++ }\r
++\r
++ mod_field (state, TM_REL_HOUR, -hdiff);\r
++\r
++ return 0;\r
++}\r
++\r
++/* Combine absolute and relative fields, and round. */\r
++static int\r
++create_output (struct state *state, time_t *t_out, const time_t *ref,\r
++ int round)\r
++{\r
++ struct tm tm = { .tm_isdst = -1 };\r
++ struct tm now;\r
++ time_t t;\r
++ enum field f;\r
++ int r;\r
++ int week_round = PARSE_TIME_NO_ROUND;\r
++\r
++ r = initialize_now (state, &now, ref);\r
++ if (r)\r
++ return r;\r
++\r
++ /* Initialize fields flagged as "now" to reference time. */\r
++ for (f = TM_ABS_SEC; f != TM_NONE; f = next_abs_field (f)) {\r
++ if (state->set[f] == FIELD_NOW) {\r
++ state->tm[f] = tm_get_field (&now, f);\r
++ state->set[f] = FIELD_SET;\r
++ }\r
++ }\r
++\r
++ /*\r
++ * If WDAY is set but MDAY is not, we consider WDAY relative\r
++ *\r
++ * XXX: This fails on stuff like "two months monday" because two\r
++ * months ago wasn't the same day as today. Postpone until we know\r
++ * date?\r
++ */\r
++ if (is_field_set (state, TM_ABS_WDAY) &&\r
++ !is_field_set (state, TM_ABS_MDAY)) {\r
++ int wday = get_field (state, TM_ABS_WDAY);\r
++ int today = tm_get_field (&now, TM_ABS_WDAY);\r
++ int rel_days;\r
++\r
++ if (today > wday)\r
++ rel_days = today - wday;\r
++ else\r
++ rel_days = today + 7 - wday;\r
++\r
++ /* This also prevents special week rounding from happening. */\r
++ mod_field (state, TM_REL_DAY, rel_days);\r
++\r
++ unset_field (state, TM_ABS_WDAY);\r
++ }\r
++\r
++ r = fixup_ampm (state);\r
++ if (r)\r
++ return r;\r
++\r
++ /*\r
++ * Iterate fields from most accurate to least accurate, and set\r
++ * unset fields according to requested rounding.\r
++ */\r
++ for (f = TM_ABS_SEC; f != TM_NONE; f = next_abs_field (f)) {\r
++ if (round != PARSE_TIME_NO_ROUND) {\r
++ enum field r = abs_to_rel_field (f);\r
++\r
++ if (is_field_set (state, f) || is_field_set (state, r)) {\r
++ if (round >= PARSE_TIME_ROUND_UP && f != TM_ABS_SEC) {\r
++ mod_field (state, r, -1);\r
++ if (round == PARSE_TIME_ROUND_UP_INCLUSIVE)\r
++ mod_field (state, TM_REL_SEC, 1);\r
++ }\r
++ round = PARSE_TIME_NO_ROUND; /* No more rounding. */\r
++ } else {\r
++ if (f == TM_ABS_MDAY &&\r
++ is_field_set (state, TM_REL_WEEK)) {\r
++ /* Week is most accurate. */\r
++ week_round = round;\r
++ round = PARSE_TIME_NO_ROUND;\r
++ } else {\r
++ set_field (state, f, field_epoch (f));\r
++ }\r
++ }\r
++ }\r
++\r
++ if (!is_field_set (state, f))\r
++ set_field (state, f, tm_get_field (&now, f));\r
++ }\r
++\r
++ /* Special case: rounding with week accuracy. */\r
++ if (week_round != PARSE_TIME_NO_ROUND) {\r
++ /* Temporarily set more accurate fields to now. */\r
++ set_field (state, TM_ABS_SEC, tm_get_field (&now, TM_ABS_SEC));\r
++ set_field (state, TM_ABS_MIN, tm_get_field (&now, TM_ABS_MIN));\r
++ set_field (state, TM_ABS_HOUR, tm_get_field (&now, TM_ABS_HOUR));\r
++ set_field (state, TM_ABS_MDAY, tm_get_field (&now, TM_ABS_MDAY));\r
++ }\r
++\r
++ /*\r
++ * Set all fields. They may contain out of range values before\r
++ * normalization by mktime(3).\r
++ */\r
++ tm.tm_sec = get_field (state, TM_ABS_SEC) - get_field (state, TM_REL_SEC);\r
++ tm.tm_min = get_field (state, TM_ABS_MIN) - get_field (state, TM_REL_MIN);\r
++ tm.tm_hour = get_field (state, TM_ABS_HOUR) - get_field (state, TM_REL_HOUR);\r
++ tm.tm_mday = get_field (state, TM_ABS_MDAY) -\r
++ get_field (state, TM_REL_DAY) - 7 * get_field (state, TM_REL_WEEK);\r
++ tm.tm_mon = get_field (state, TM_ABS_MON) - get_field (state, TM_REL_MON);\r
++ tm.tm_mon--; /* 1- to 0-based */\r
++ tm.tm_year = get_field (state, TM_ABS_YEAR) - get_field (state, TM_REL_YEAR) - 1900;\r
++\r
++ /*\r
++ * It's always normal time.\r
++ *\r
++ * XXX: This is probably not a solution that universally\r
++ * works. Just make sure DST is not taken into account. We don't\r
++ * want rounding to be affected by DST.\r
++ */\r
++ tm.tm_isdst = -1;\r
++\r
++ /* Special case: rounding with week accuracy. */\r
++ if (week_round != PARSE_TIME_NO_ROUND) {\r
++ /* Normalize to get proper tm.wday. */\r
++ r = normalize_tm (&tm);\r
++ if (r < 0)\r
++ return r;\r
++\r
++ /* Set more accurate fields back to zero. */\r
++ tm.tm_sec = 0;\r
++ tm.tm_min = 0;\r
++ tm.tm_hour = 0;\r
++ tm.tm_isdst = -1;\r
++\r
++ /* Monday is the true 1st day of week, but this is easier. */\r
++ if (week_round >= PARSE_TIME_ROUND_UP) {\r
++ tm.tm_mday += 7 - tm.tm_wday;\r
++ if (week_round == PARSE_TIME_ROUND_UP_INCLUSIVE)\r
++ tm.tm_sec--;\r
++ } else {\r
++ tm.tm_mday -= tm.tm_wday;\r
++ }\r
++ }\r
++\r
++ if (is_field_set (state, TM_TZ)) {\r
++ /* tm is in specified TZ, convert to UTC for timegm(3). */\r
++ tm.tm_min -= get_field (state, TM_TZ);\r
++ t = timegm (&tm);\r
++ } else {\r
++ /* tm is in local time. */\r
++ t = mktime (&tm);\r
++ }\r
++\r
++ if (t == (time_t) -1)\r
++ return -PARSE_TIME_ERR_LIB;\r
++\r
++ *t_out = t;\r
++\r
++ return 0;\r
++}\r
++\r
++/* Internally, all errors are < 0. parse_time_string() returns errors > 0. */\r
++#define EXTERNAL_ERR(r) (-r)\r
++\r
++int\r
++parse_time_string (const char *s, time_t *t, const time_t *ref, int round)\r
++{\r
++ struct state state = { .last_field = TM_NONE };\r
++ int r;\r
++\r
++ if (!s || !t)\r
++ return EXTERNAL_ERR (-PARSE_TIME_ERR);\r
++\r
++ r = parse_input (&state, s);\r
++ if (r < 0)\r
++ return EXTERNAL_ERR (r);\r
++\r
++ r = create_output (&state, t, ref, round);\r
++ if (r < 0)\r
++ return EXTERNAL_ERR (r);\r
++\r
++ return 0;\r
++}\r
+diff --git a/parse-time-string/parse-time-string.h b/parse-time-string/parse-time-string.h\r
+new file mode 100644\r
+index 0000000..bfa4ee3\r
+--- /dev/null\r
++++ b/parse-time-string/parse-time-string.h\r
+@@ -0,0 +1,102 @@\r
++/*\r
++ * parse time string - user friendly date and time parser\r
++ * Copyright © 2012 Jani Nikula\r
++ *\r
++ * This program is free software: you can redistribute it and/or modify\r
++ * it under the terms of the GNU General Public License as published by\r
++ * the Free Software Foundation, either version 2 of the License, or\r
++ * (at your option) any later version.\r
++ *\r
++ * This program is distributed in the hope that it will be useful,\r
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of\r
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the\r
++ * GNU General Public License for more details.\r
++ *\r
++ * You should have received a copy of the GNU General Public License\r
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.\r
++ *\r
++ * Author: Jani Nikula <jani@nikula.org>\r
++ */\r
++\r
++#ifndef PARSE_TIME_STRING_H\r
++#define PARSE_TIME_STRING_H\r
++\r
++#ifdef __cplusplus\r
++extern "C" {\r
++#endif\r
++\r
++#include <time.h>\r
++\r
++/* return values for parse_time_string() */\r
++enum {\r
++ PARSE_TIME_OK = 0,\r
++ PARSE_TIME_ERR, /* unspecified error */\r
++ PARSE_TIME_ERR_LIB, /* library call failed */\r
++ PARSE_TIME_ERR_ALREADYSET, /* attempt to set unit twice */\r
++ PARSE_TIME_ERR_FORMAT, /* generic date/time format error */\r
++ PARSE_TIME_ERR_DATEFORMAT, /* date format error */\r
++ PARSE_TIME_ERR_TIMEFORMAT, /* time format error */\r
++ PARSE_TIME_ERR_INVALIDDATE, /* date value error */\r
++ PARSE_TIME_ERR_INVALIDTIME, /* time value error */\r
++ PARSE_TIME_ERR_KEYWORD, /* unknown keyword */\r
++};\r
++\r
++/* round values for parse_time_string() */\r
++enum {\r
++ PARSE_TIME_ROUND_DOWN = -1,\r
++ PARSE_TIME_NO_ROUND = 0,\r
++ PARSE_TIME_ROUND_UP = 1,\r
++ PARSE_TIME_ROUND_UP_INCLUSIVE = 2,\r
++};\r
++\r
++/**\r
++ * parse_time_string() - user friendly date and time parser\r
++ * @s: string to parse\r
++ * @t: pointer to time_t to store parsed time in\r
++ * @ref: pointer to time_t containing reference date/time, or NULL\r
++ * @round: PARSE_TIME_NO_ROUND, PARSE_TIME_ROUND_DOWN, or\r
++ * PARSE_TIME_ROUND_UP\r
++ *\r
++ * Parse a date/time string 's' and store the parsed date/time result\r
++ * in 't'.\r
++ *\r
++ * A reference date/time is used for determining the "date/time units"\r
++ * (roughly equivalent to struct tm members) not specified by 's'. If\r
++ * 'ref' is non-NULL, it must contain a pointer to a time_t to be used\r
++ * as reference date/time. Otherwise, the current time is used.\r
++ *\r
++ * If 's' does not specify a full date/time, the 'round' parameter\r
++ * specifies if and how the result should be rounded as follows:\r
++ *\r
++ * PARSE_TIME_NO_ROUND: All date/time units that are not specified\r
++ * by 's' are set to the corresponding unit derived from the\r
++ * reference date/time.\r
++ *\r
++ * PARSE_TIME_ROUND_DOWN: All date/time units that are more accurate\r
++ * than the most accurate unit specified by 's' are set to the\r
++ * smallest valid value for that unit. Rest of the unspecified units\r
++ * are set as in PARSE_TIME_NO_ROUND.\r
++ *\r
++ * PARSE_TIME_ROUND_UP: All date/time units that are more accurate\r
++ * than the most accurate unit specified by 's' are set to the\r
++ * smallest valid value for that unit. The most accurate unit\r
++ * specified by 's' is incremented by one (and this is rolled over\r
++ * to the less accurate units as necessary), unless the most\r
++ * accurate unit is seconds. Rest of the unspecified units are set\r
++ * as in PARSE_TIME_NO_ROUND.\r
++ *\r
++ * PARSE_TIME_ROUND_UP_INCLUSIVE: Same as PARSE_TIME_ROUND_UP, minus\r
++ * one second, unless the most accurate unit specified by 's' is\r
++ * seconds. This is useful for callers that require a value for\r
++ * inclusive comparison of the result.\r
++ *\r
++ * Return 0 (PARSE_TIME_OK) for succesfully parsed date/time, or one\r
++ * of PARSE_TIME_ERR_* on error. 't' is not modified on error.\r
++ */\r
++int parse_time_string (const char *s, time_t *t, const time_t *ref, int round);\r
++\r
++#ifdef __cplusplus\r
++}\r
++#endif\r
++\r
++#endif /* PARSE_TIME_STRING_H */\r
+-- \r
+1.7.10.4\r
+\r
projects / notmuch-archives.git / commitdiff