finally implement a real number format parser. that was rough...

This commit is contained in:
Thomas Fussell 2016-07-03 11:22:53 -04:00
parent bdc770d23a
commit d92ad1ab9c
4 changed files with 1615 additions and 899 deletions

File diff suppressed because it is too large Load Diff

View File

@ -0,0 +1,366 @@
// Copyright (c) 2014-2016 Thomas Fussell
//
// Permission is hereby granted, free of charge, to any person obtaining a copy
// of this software and associated documentation files (the "Software"), to deal
// in the Software without restriction, including without limitation the rights
// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
// copies of the Software, and to permit persons to whom the Software is
// furnished to do so, subject to the following conditions:
//
// The above copyright notice and this permission notice shall be included in
// all copies or substantial portions of the Software.
//
// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, WRISING FROM,
// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
// THE SOFTWARE
//
// @license: http://www.opensource.org/licenses/mit-license.php
// @author: see AUTHORS file
#pragma once
#include <string>
#include <unordered_map>
#include <vector>
#include <xlnt/utils/datetime.hpp>
namespace xlnt {
namespace detail {
enum class bracket_color
{
none,
black,
blue,
cyan,
green,
magenta,
red,
white,
yellow,
color1,
color2,
color3,
color4,
color5,
color6,
color7,
color8,
color9,
color10,
color11,
color12,
color13,
color14,
color15,
color16,
color17,
color18,
color19,
color20,
color21,
color22,
color23,
color24,
color25,
color26,
color27,
color28,
color29,
color30,
color31,
color32,
color33,
color34,
color35,
color36,
color37,
color38,
color39,
color40,
color41,
color42,
color43,
color44,
color45,
color46,
color47,
color48,
color49,
color50,
color51,
color52,
color53,
color54,
color55,
color56
};
enum class locale
{
none = 0,
arabic_saudi_arabia = 0x401,
bulgarian = 0x402,
catalan = 0x403,
chinese_taiwan = 0x404,
czech = 0x405,
danish = 0x406,
german_germany = 0x407,
greek = 0x408,
english_united_states = 0x409,
italian_italy = 0x410,
japanese = 0x411,
korean = 0x412,
dutch_netherlands = 0x413,
norwegian_bokml = 0x414,
polish = 0x415,
portuguese_brazil = 0x416,
raeto_romance = 0x417,
romanian_romania = 0x418,
russian = 0x419,
urdu = 0x420,
indonesian = 0x421,
ukrainian = 0x422,
belarusian = 0x423,
slovenian = 0x424,
estonian = 0x425,
latvian = 0x426,
lithuanian = 0x427,
tajik = 0x428,
farsi_persian = 0x429,
sesotho_sutu = 0x430,
tsonga = 0x431,
setsuana = 0x432,
venda = 0x433,
xhosa = 0x434,
zulu = 0x435,
afrikaans = 0x436,
georgian = 0x437,
faroese = 0x438,
hindi = 0x439,
kyrgyz_cyrillic = 0x440,
swahili = 0x441,
turkmen = 0x442,
uzbek_latin = 0x443,
tatar = 0x444,
bengali_india = 0x445,
punjabi = 0x446,
gujarati = 0x447,
oriya = 0x448,
tamil = 0x449,
mongolian = 0x450,
tibetan = 0x451,
welsh = 0x452,
khmer = 0x453,
lao = 0x454,
burmese = 0x455,
galician = 0x456,
konkani = 0x457,
manipuri = 0x458,
sindhi = 0x459,
kashmiri = 0x460,
nepali = 0x461,
frisian_netherlands = 0x462,
filipino = 0x464,
divehi_dhivehi_maldivian = 0x465,
edo = 0x466,
igbo_nigeria = 0x470,
guarani_paraguay = 0x474,
latin = 0x476,
somali = 0x477,
maori = 0x481,
arabic_iraq = 0x801,
chinese_china = 0x804,
german_switzerland = 0x807,
english_great_britain = 0x809,
italian_switzerland = 0x810,
dutch_belgium = 0x813,
norwegian_nynorsk = 0x814,
portuguese_portugal = 0x816,
romanian_moldova = 0x818,
russian_moldova = 0x819,
uzbek_cyrillic = 0x843,
bengali_bangladesh = 0x845,
mongolian2 = 0x850,
arabic_libya = 0x1001,
chinese_singapore = 0x1004,
german_luxembourg = 0x1007,
english_canada = 0x1009,
arabic_algeria = 0x1401,
chinese_macau_sar = 0x1404,
german_liechtenstein = 0x1407,
english_new_zealand = 0x1409,
arabic_morocco = 0x1801,
english_ireland = 0x1809,
arabic_oman = 0x2001,
english_jamaica = 0x2009,
arabic_yemen = 0x2401,
english_caribbean = 0x2409,
arabic_syria = 0x2801,
english_belize = 0x2809,
arabic_lebanon = 0x3001,
english_zimbabwe = 0x3009,
arabic_kuwait = 0x3401,
english_phillippines = 0x3409,
arabic_united_arab_emirates = 0x3801,
arabic_qatar = 0x4001
};
struct condition
{
enum class condition_type
{
none,
less_than,
less_or_equal,
equal,
not_equal,
greater_than,
greater_or_equal
} type = condition_type::none;
long double value = 0;
bool satisfied_by(long double number) const;
};
struct placeholders
{
enum class placeholders_type
{
bad,
general,
text,
fractional_part,
integer_part,
fraction_integer,
fraction_numerator,
fraction_denominator,
scientific_significand,
scientific_exponent
} type = placeholders_type::bad;
bool use_comma_separator = false;
std::size_t num_zeros = 0; // 0
std::size_t num_optionals = 0; // #
std::size_t num_spaces = 0; // ?
std::size_t thousands_scale = 0;
};
struct number_format_token
{
enum class token_type
{
bad,
color,
locale,
condition,
text,
fill,
space,
number,
datetime,
end_section,
end
} type = token_type::bad;
std::string string;
};
struct template_part
{
enum class template_type
{
bad,
text,
fill,
space,
general,
placeholder,
month_number,
month_number_leading_zero,
month_abbreviation,
month_name,
month_letter,
day_number,
day_number_leading_zero,
day_abbreviation,
day_name,
year_short,
year_long,
hour,
hour_leading_zero,
minute,
minute_leading_zero,
second,
second_leading_zero,
am_pm,
a_p,
elapsed_hours,
elapsed_minutes,
elapsed_seconds
} type = template_type::bad;
std::string string;
placeholders placeholders;
};
struct format_code
{
bracket_color color = bracket_color::none;
locale locale = locale::none;
condition condition;
bool is_datetime = false;
bool is_timedelta = false;
bool twelve_hour = false;
std::vector<template_part> parts;
};
class number_format_parser
{
public:
number_format_parser(const std::string &format_string);
const std::vector<format_code> &get_result() const;
void reset(const std::string &format_string);
void parse();
private:
void finalize();
void validate();
number_format_token parse_next_token();
placeholders parse_placeholders(const std::string &placeholders_string);
bracket_color color_from_string(const std::string &color);
std::pair<locale, std::string> locale_from_string(const std::string &locale_string);
std::size_t position_ = 0;
std::string format_string_;
std::vector<format_code> codes_;
};
class number_formatter
{
public:
number_formatter(const std::string &format_string, xlnt::calendar calendar);
std::string format_number(long double number);
std::string format_text(const std::string &text);
private:
std::string fill_placeholders(const placeholders &p, long double number);
std::string format_number(const format_code &format, long double number);
std::string format_text(const format_code &format, const std::string &text);
number_format_parser parser_;
std::vector<format_code> format_;
xlnt::calendar calendar_;
};
} // namespace detail
} // namespace xlnt

View File

@ -1,4 +1,4 @@
// Copyright (c) 2014-2016 Thomas Fussell
// Copyright (c) 2014-2016 Thomas Fussell
// Copyright (c) 2010-2015 openpyxl
//
// Permission is hereby granted, free of charge, to any person obtaining a copy
@ -30,6 +30,8 @@
#include <xlnt/utils/hash_combine.hpp>
#include <xlnt/styles/number_format.hpp>
#include <detail/number_formatter.hpp>
namespace {
const std::unordered_map<std::size_t, std::string> &builtin_formats()
@ -61,16 +63,16 @@ const std::unordered_map<std::size_t, std::string> &builtin_formats()
{ 21, "h:mm:ss" },
{ 22, "m/d/yy h:mm" },
{ 37, "#,##0_);(#,##0)" },
{ 38, "#,##0_);[Red](#,##0)" },
{ 39, "#,##0.00_);(#,##0.00)" },
{ 40, "#,##0.00_);[Red](#,##0.00)" },
{ 37, "#,##0 ;(#,##0)" },
{ 38, "#,##0 ;[Red](#,##0)" },
{ 39, "#,##0.00;(#,##0.00)" },
{ 40, "#,##0.00;[Red](#,##0.00)" },
{ 41, "_(* #,##0_);_(* \\(#,##0\\);_(* \"-\"_);_(@_)" },
{ 42, "_(\"$\"* #,##0_);_(\"$\"* \\(#,##0\\);_(\"$\"* \"-\"_);_(@_)" },
{ 43, "_(* #,##0.00_);_(* \\(#,##0.00\\);_(* \"-\"??_);_(@_)" },
{ 44, "_-\"$\"* #,##0.00_-;\\-\"$\"* #,##0.00_-;_-\"$\"* \"-\"??_-;_-@_-" },
{ 45, "mm:ss" },
{ 46, "[h]:mm:ss" },
{ 47, "mmss.0" },
@ -91,868 +93,6 @@ const std::unordered_map<std::size_t, std::string> &builtin_formats()
return *formats;
}
enum class condition_type
{
less_than,
less_or_equal,
equal,
greater_than,
greater_or_equal,
invalid
};
struct section
{
bool has_value = false;
std::string value;
bool has_color = false;
std::string color;
bool has_condition = false;
condition_type condition = condition_type::invalid;
std::string condition_value;
bool has_locale = false;
std::string locale;
section &operator=(const section &other)
{
has_value = other.has_value;
value = other.value;
has_color = other.has_color;
color = other.color;
has_condition = other.has_condition;
condition = other.condition;
condition_value = other.condition_value;
has_locale = other.has_locale;
locale = other.locale;
return *this;
}
};
struct format_sections
{
section first;
section second;
section third;
section fourth;
};
// copied from named_range.cpp, keep in sync
/// <summary>
/// Return a vector containing string split at each delim.
/// </summary>
/// <remark>
/// This should maybe be in a utility header so it can be used elsewhere.
/// </remarks>
std::vector<std::string> split_string(const std::string &string, char delim)
{
std::vector<std::string> split;
std::string::size_type previous_index = 0;
auto separator_index = string.find(delim);
while (separator_index != std::string::npos)
{
auto part = string.substr(previous_index, separator_index - previous_index);
split.push_back(part);
previous_index = separator_index + 1;
separator_index = string.find(delim, previous_index);
}
split.push_back(string.substr(previous_index));
return split;
}
std::vector<std::string> split_string_any(const std::string &string, const std::string &delims)
{
std::vector<std::string> split;
std::string::size_type previous_index = 0;
auto separator_index = string.find_first_of(delims);
while (separator_index != std::string::npos)
{
auto part = string.substr(previous_index, separator_index - previous_index);
if (!part.empty())
{
split.push_back(part);
}
previous_index = separator_index + 1;
separator_index = string.find_first_of(delims, previous_index);
}
split.push_back(string.substr(previous_index));
return split;
}
bool is_date_format(const std::string &format_string)
{
auto not_in = format_string.find_first_not_of("/-:, mMyYdDhHsSAP");
return not_in == std::string::npos;
}
bool is_valid_color(const std::string &color)
{
static const std::vector<std::string> *colors =
new std::vector<std::string>(
{
"Black",
"Green",
"White",
"Blue",
"Magenta",
"Yellow",
"Cyan",
"Red"
});
auto compare_color = [&](const std::string &other) {
if (color.size() != other.size()) return false;
for (std::size_t i = 0; i < color.size(); i++)
{
if (std::toupper(color[i]) != std::toupper(other[i]))
{
return false;
}
}
return true;
};
return std::find_if(colors->begin(), colors->end(), compare_color) != colors->end();
}
bool parse_condition(const std::string &string, section &s)
{
s.has_condition = false;
s.condition = condition_type::invalid;
s.condition_value.clear();
if (string[0] == '<')
{
s.has_condition = true;
if (string[1] == '=')
{
s.condition = condition_type::less_or_equal;
s.condition_value = string.substr(2);
}
else
{
s.condition = condition_type::less_than;
s.condition_value = string.substr(1);
}
}
if (string[0] == '>')
{
s.has_condition = true;
if (string[1] == '=')
{
s.condition = condition_type::greater_or_equal;
s.condition_value = string.substr(2);
}
else
{
s.condition = condition_type::greater_than;
s.condition_value = string.substr(1);
}
}
else if (string[0] == '=')
{
s.has_condition = true;
s.condition = condition_type::equal;
s.condition_value = string.substr(1);
}
return s.has_condition;
}
bool is_hex(char c)
{
return (c >= 'A' && c <= 'F') || (c >= '0' && c <= '9');
}
const std::unordered_map<int, std::string> known_locales()
{
const std::unordered_map<int, std::string> *all =
new std::unordered_map<int, std::string>(
{
{ 0x401, "Arabic - Saudi Arabia" },
{ 0x402, "Bulgarian" },
{ 0x403, "Catalan" },
{ 0x404, "Chinese - Taiwan" },
{ 0x405, "Czech" },
{ 0x406, "Danish" },
{ 0x407, "German - Germany" },
{ 0x408, "Greek" },
{ 0x409, "English - United States" },
{ 0x410, "Italian - Italy" },
{ 0x411, "Japanese" },
{ 0x412, "Korean" },
{ 0x413, "Dutch - Netherlands" },
{ 0x414, "Norwegian - Bokml" },
{ 0x415, "Polish" },
{ 0x416, "Portuguese - Brazil" },
{ 0x417, "Raeto-Romance" },
{ 0x418, "Romanian - Romania" },
{ 0x419, "Russian" },
{ 0x420, "Urdu" },
{ 0x421, "Indonesian" },
{ 0x422, "Ukrainian" },
{ 0x423, "Belarusian" },
{ 0x424, "Slovenian" },
{ 0x425, "Estonian" },
{ 0x426, "Latvian" },
{ 0x427, "Lithuanian" },
{ 0x428, "Tajik" },
{ 0x429, "Farsi - Persian" },
{ 0x430, "Sesotho (Sutu)" },
{ 0x431, "Tsonga" },
{ 0x432, "Setsuana" },
{ 0x433, "Venda" },
{ 0x434, "Xhosa" },
{ 0x435, "Zulu" },
{ 0x436, "Afrikaans" },
{ 0x437, "Georgian" },
{ 0x438, "Faroese" },
{ 0x439, "Hindi" },
{ 0x440, "Kyrgyz - Cyrillic" },
{ 0x441, "Swahili" },
{ 0x442, "Turkmen" },
{ 0x443, "Uzbek - Latin" },
{ 0x444, "Tatar" },
{ 0x445, "Bengali - India" },
{ 0x446, "Punjabi" },
{ 0x447, "Gujarati" },
{ 0x448, "Oriya" },
{ 0x449, "Tamil" },
{ 0x450, "Mongolian" },
{ 0x451, "Tibetan" },
{ 0x452, "Welsh" },
{ 0x453, "Khmer" },
{ 0x454, "Lao" },
{ 0x455, "Burmese" },
{ 0x456, "Galician" },
{ 0x457, "Konkani" },
{ 0x458, "Manipuri" },
{ 0x459, "Sindhi" },
{ 0x460, "Kashmiri" },
{ 0x461, "Nepali" },
{ 0x462, "Frisian - Netherlands" },
{ 0x464, "Filipino" },
{ 0x465, "Divehi; Dhivehi; Maldivian" },
{ 0x466, "Edo" },
{ 0x470, "Igbo - Nigeria" },
{ 0x474, "Guarani - Paraguay" },
{ 0x476, "Latin" },
{ 0x477, "Somali" },
{ 0x481, "Maori" },
{ 0x801, "Arabic - Iraq" },
{ 0x804, "Chinese - China" },
{ 0x807, "German - Switzerland" },
{ 0x809, "English - Great Britain" },
{ 0x810, "Italian - Switzerland" },
{ 0x813, "Dutch - Belgium" },
{ 0x814, "Norwegian - Nynorsk" },
{ 0x816, "Portuguese - Portugal" },
{ 0x818, "Romanian - Moldova" },
{ 0x819, "Russian - Moldova" },
{ 0x843, "Uzbek - Cyrillic" },
{ 0x845, "Bengali - Bangladesh" },
{ 0x850, "Mongolian" },
{ 0x1001, "Arabic - Libya" },
{ 0x1004, "Chinese - Singapore" },
{ 0x1007, "German - Luxembourg" },
{ 0x1009, "English - Canada" },
{ 0x1401, "Arabic - Algeria" },
{ 0x1404, "Chinese - Macau SAR" },
{ 0x1407, "German - Liechtenstein" },
{ 0x1409, "English - New Zealand" },
{ 0x1801, "Arabic - Morocco" },
{ 0x1809, "English - Ireland" },
{ 0x2001, "Arabic - Oman" },
{ 0x2009, "English - Jamaica" },
{ 0x2401, "Arabic - Yemen" },
{ 0x2409, "English - Caribbean" },
{ 0x2801, "Arabic - Syria" },
{ 0x2809, "English - Belize" },
{ 0x3001, "Arabic - Lebanon" },
{ 0x3009, "English - Zimbabwe" },
{ 0x3401, "Arabic - Kuwait" },
{ 0x3409, "English - Phillippines" },
{ 0x3801, "Arabic - United Arab Emirates" },
{ 0x4001, "Arabic - Qatar" }
});
return *all;
}
bool is_valid_locale(const std::string &locale_string)
{
auto hyphen_index = locale_string.find('-');
if (hyphen_index == std::string::npos)
{
return false;
}
std::string country = locale_string.substr(hyphen_index + 1);
if (country.empty())
{
return false;
}
for (auto c : country)
{
if (!is_hex(static_cast<char>(std::toupper(c))))
{
return false;
}
}
auto index = std::stoi(country, 0, 16);
auto known_locales_ = known_locales();
if (known_locales_.find(index) == known_locales_.end())
{
return false;
}
std::string beginning = locale_string.substr(0, locale_string.find('-'));
if (beginning.empty() || beginning[0] != '$')
{
return false;
}
if (beginning.size() == 1)
{
return true;
}
beginning = beginning.substr(1);
return true;
}
section parse_section(const std::string &section_string)
{
if (section_string.empty())
{
throw std::runtime_error("empty format");
}
section s;
std::string format_part;
std::string bracket_part;
std::vector<std::string> bracket_parts;
bool in_quotes = false;
bool in_brackets = false;
const std::vector<std::string> bracket_times = { "h", "hh", "m", "mm", "s", "ss" };
for (std::size_t i = 0; i < section_string.size(); i++)
{
if (!in_quotes && section_string[i] == '"')
{
format_part.push_back(section_string[i]);
in_quotes = true;
}
else if (in_quotes && section_string[i] == '"')
{
format_part.push_back(section_string[i]);
if (i < section_string.size() - 1 && section_string[i + 1] != '"')
{
in_quotes = false;
}
}
else if (!in_brackets && section_string[i] == '[')
{
in_brackets = true;
for (auto bracket_time : bracket_times)
{
if (i < section_string.size() - bracket_time.size() &&
section_string.substr(i + 1, bracket_time.size()) == bracket_time)
{
in_brackets = false;
break;
}
}
}
else if (in_brackets)
{
if (section_string[i] == ']')
{
in_brackets = false;
if (is_valid_color(bracket_part))
{
if (!s.has_color)
{
s.color = bracket_part;
s.has_color = true;
}
else
{
throw std::runtime_error("two colors");
}
}
else if (is_valid_locale(bracket_part))
{
if (!s.has_locale)
{
s.locale = bracket_part;
s.has_locale = true;
}
else
{
throw std::runtime_error("two locales");
}
}
else if (s.has_condition || !parse_condition(bracket_part, s))
{
throw std::runtime_error("invalid bracket format");
}
bracket_part.clear();
}
else
{
bracket_part.push_back(section_string[i]);
}
}
else
{
format_part.push_back(section_string[i]);
}
}
s.value = format_part;
s.has_value = true;
return s;
}
format_sections parse_format_sections(const std::string &combined)
{
format_sections result = {};
auto split = split_string(combined, ';');
result.first = parse_section(split[0]);
if (!result.first.has_condition)
{
result.second = result.first;
result.third = result.first;
}
if (split.size() > 1)
{
result.second = parse_section(split[1]);
}
if (split.size() > 2)
{
if (result.first.has_condition && !result.second.has_condition)
{
throw std::runtime_error("first two sections should have conditions");
}
result.third = parse_section(split[2]);
if (result.third.has_condition)
{
throw std::runtime_error("third section shouldn't have a condition");
}
}
if (split.size() > 3)
{
if (result.first.has_condition)
{
throw std::runtime_error("too many parts");
}
result.fourth = parse_section(split[3]);
}
if (split.size() > 4)
{
throw std::runtime_error("too many parts");
}
return result;
}
std::string format_section(long double number, const section &format, xlnt::calendar base_date)
{
const std::string unquoted = "$+(:^'{<=-/)!&~}> ";
std::string format_temp = format.value;
std::string result;
if (is_date_format(format.value))
{
auto lower_string = [](const std::string &s) {
std::string lower;
lower.resize(s.size());
for (std::size_t i = 0; i < s.size(); i++)
lower[i] = static_cast<char>(std::tolower(s[i]));
return lower;
};
auto value = format.value;
bool twelve_hour = false;
if (format.value.size() > 5 && lower_string(value).substr(format.value.size() - 5) == "am/pm")
{
twelve_hour = true;
}
const std::string date_unquoted = ",-/: ";
const std::vector<std::string> dates = { "m", "mm", "mmm", "mmmmm", "mmmmmm", "d", "dd", "ddd", "dddd", "yy",
"yyyy", "h", "[h]", "hh", "m", "[m]", "mm", "s", "[s]", "ss" };
const std::vector<std::string> MonthNames = { "January", "February", "March",
"April", "May", "June", "July", "August", "September", "October", "November", "December" };
auto split = split_string_any(format.value, date_unquoted);
std::string::size_type index = 0, prev = 0;
auto d = xlnt::datetime::from_number(number, base_date);
bool processed_month = false;
for (auto part : split)
{
while (format.value.substr(index, part.size()) != part)
{
index++;
}
part = lower_string(part);
auto between = format.value.substr(prev, index - prev);
if (between == "/" && part == "pm")
{
index += part.size();
prev = index;
continue;
}
result.append(between);
if (part == "m" && !processed_month)
{
result.append(std::to_string(d.month));
processed_month = true;
}
else if (part == "mm" && !processed_month)
{
if (d.month < 10)
{
result.append("0");
}
result.append(std::to_string(d.month));
processed_month = true;
}
else if (part == "mmm" && !processed_month)
{
result.append(MonthNames.at(static_cast<std::size_t>(d.month - 1)).substr(0, 3));
processed_month = true;
}
else if (part == "mmmm" && !processed_month)
{
result.append(MonthNames.at(static_cast<std::size_t>(d.month - 1)));
processed_month = true;
}
else if (part == "d")
{
result.append(std::to_string(d.day));
}
else if (part == "dd")
{
if (d.day < 10)
{
result.append("0");
}
result.append(std::to_string(d.day));
}
else if (part == "yy")
{
result.append(std::to_string(d.year % 1000));
}
else if (part == "yyyy")
{
result.append(std::to_string(d.year));
}
else if (part == "h")
{
if (twelve_hour)
{
result.append(std::to_string(d.hour % 12));
}
else
{
result.append(std::to_string(d.hour));
}
processed_month = true;
}
else if (part == "hh")
{
if (twelve_hour)
{
if (d.hour % 12 < 10)
{
result.append("0");
}
result.append(std::to_string(d.hour % 12));
}
else
{
if (d.hour < 10)
{
result.append("0");
}
result.append(std::to_string(d.hour));
}
processed_month = true;
}
else if (part == "m")
{
result.append(std::to_string(d.minute));
}
else if (part == "mm")
{
if (d.minute < 10)
{
result.append("0");
}
result.append(std::to_string(d.minute));
}
else if (part == "s")
{
result.append(std::to_string(d.second));
}
else if (part == "ss")
{
if (d.second < 10)
{
result.append("0");
}
result.append(std::to_string(d.second));
}
else if (part == "am")
{
if (d.hour < 12)
{
result.append("AM");
}
else
{
result.append("PM");
}
}
index += part.size();
prev = index;
}
if (index < format.value.size())
{
result.append(format.value.substr(index));
}
}
else if (format.value.find_first_of("General") != std::string::npos || format.value == "0")
{
std::string before_text, end_text;
auto open_quote = format.value.find('"');
if (open_quote != std::string::npos)
{
auto close_quote = format.value.find('"', open_quote + 1);
auto quoted_text = format.value.substr(open_quote, close_quote - open_quote + 1);
if (open_quote == 0)
{
before_text = quoted_text.substr(1, quoted_text.size() - 2);
}
else
{
end_text = quoted_text.substr(1, quoted_text.size() - 2);
}
}
if (number < 0)
{
before_text = std::string("-") + before_text;
}
result = before_text;
if (number == static_cast<long long int>(number))
{
result.append(std::to_string(static_cast<long long int>(std::abs(number))));
}
else
{
auto number_string = std::to_string(std::abs(number));
while (number_string.find('.') != std::string::npos && number_string.back() == '0' && number_string.find('.') < number_string.size() - 1)
{
number_string.erase(number_string.end() - 1);
}
result.append(number_string);
}
result.append(end_text);
}
else if (format.value.find("#,##0.00") != std::string::npos)
{
auto index = format.value.find("#,##0.00");
auto before_text = format.value.substr(0, index);
auto after_text = format.value.substr(index + 8);
bool minus_first = before_text.substr(0, 1) == "-";
if (minus_first)
{
before_text = before_text.substr(1);
}
if (format.has_locale && format.locale == "$$-1009")
{
before_text = std::string(minus_first ? "-" : "") + "CA$" + before_text;
}
else if (format.has_locale && format.locale == "$€-407")
{
before_text = std::string(minus_first ? "-" : "") + "" + before_text;
}
result = before_text + std::to_string(number < 0 ? -number : number) + after_text;
auto decimal_pos = result.find('.');
if (decimal_pos != std::string::npos)
{
result[decimal_pos] = ',';
decimal_pos += 3;
while (decimal_pos < result.size())
{
result.pop_back();
}
}
}
return result;
}
std::string format_section(const std::string &text, const section &format)
{
auto arobase_index = format.value.find('@');
std::string first_part, middle_part, last_part;
if (arobase_index != std::string::npos)
{
first_part = format.value.substr(0, arobase_index);
middle_part = text;
last_part = format.value.substr(arobase_index + 1);
}
else
{
first_part = format.value;
}
auto unquote = [](std::string &s) {
if (!s.empty())
{
if (s.front() != '"' || s.back() != '"') return false;
s = s.substr(1, s.size() - 2);
}
return true;
};
if (!unquote(first_part) || !unquote(last_part))
{
throw std::runtime_error(std::string("additional text must be enclosed in quotes: ") + format.value);
}
return first_part + middle_part + last_part;
}
std::string format_number(long double number, const std::string &format, xlnt::calendar base_date)
{
auto sections = parse_format_sections(format);
if (sections.first.has_condition)
{
auto right_side = std::stod(sections.first.condition_value);
if ((sections.first.condition == condition_type::equal && number == right_side) ||
(sections.first.condition == condition_type::greater_or_equal && number >= right_side) ||
(sections.first.condition == condition_type::greater_than && number > right_side) ||
(sections.first.condition == condition_type::less_or_equal && number <= right_side) ||
(sections.first.condition == condition_type::less_than && number < right_side))
{
return format_section(number, sections.first, base_date);
}
right_side = std::stod(sections.second.condition_value);
if ((sections.second.condition == condition_type::equal && number == right_side) ||
(sections.second.condition == condition_type::greater_or_equal && number >= right_side) ||
(sections.second.condition == condition_type::greater_than && number > right_side) ||
(sections.second.condition == condition_type::less_or_equal && number <= right_side) ||
(sections.second.condition == condition_type::less_than && number < right_side))
{
return format_section(number, sections.second, base_date);
}
}
else
{
if (number > 0)
{
return format_section(number, sections.first, base_date);
}
else if (number < 0)
{
return format_section(number, sections.second, base_date);
}
}
// number == 0 or default condition
return format_section(number, sections.third, base_date);
}
std::string format_text(const std::string &text, const std::string &format)
{
if (format == "General" || format == "@") return text;
auto sections = parse_format_sections(format);
return format_section(text, sections.fourth);
}
} // namespace
namespace xlnt {
@ -1247,30 +387,37 @@ std::size_t number_format::get_id() const
bool number_format::is_date_format() const
{
if (format_string_ != "General")
detail::number_format_parser p(format_string_);
p.parse();
auto parsed = p.get_result();
bool any_datetime = false;
bool any_timedelta = false;
for (const auto &section : parsed)
{
try
if (section.is_datetime)
{
auto sections = parse_format_sections(format_string_);
return ::is_date_format(sections.first.value);
any_datetime = true;
}
catch (std::exception)
if (section.is_timedelta)
{
return false;
any_timedelta = true;
}
}
return false;
return any_datetime && !any_timedelta;
}
std::string number_format::format(const std::string &text) const
{
return format_text(text, format_string_);
return detail::number_formatter(format_string_, calendar::windows_1900).format_text(text);
}
std::string number_format::format(long double number, calendar base_date) const
{
return format_number(number, format_string_, base_date);
return detail::number_formatter(format_string_, base_date).format_number(number);
}
} // namespace xlnt

View File

@ -16,10 +16,12 @@ public:
nf.set_format_string("\"positive\"General;\"negative\"General");
auto formatted = nf.format(3.14, xlnt::calendar::windows_1900);
TS_ASSERT_EQUALS(formatted, "positive3.14");
nf.set_format_string("\"positive\"General;\"negative\"General");
formatted = nf.format(-3.14, xlnt::calendar::windows_1900);
TS_ASSERT_EQUALS(formatted, "-negative3.14");
TS_ASSERT_EQUALS(formatted, "negative3.14");
nf.set_format_string("\"any\"General");
formatted = nf.format(-3.14, xlnt::calendar::windows_1900);
TS_ASSERT_EQUALS(formatted, "-any3.14");
}
void test_simple_date()
@ -311,14 +313,13 @@ public:
{
xlnt::number_format nf;
nf.set_format_string("[$€-407]-#,##0.00");
auto formatted = nf.format(1.2, xlnt::calendar::windows_1900);
TS_ASSERT_EQUALS(formatted, "-€1,20");
nf.set_format_string("[$€-407]#,##0.00");
auto formatted = nf.format(-45000.1, xlnt::calendar::windows_1900);
TS_ASSERT_EQUALS(formatted, "-€45,000.10");
nf.set_format_string("[$$-1009]#,##0.00");
formatted = nf.format(1.2, xlnt::calendar::windows_1900);
TS_ASSERT_EQUALS(formatted, "CA$1,20");
formatted = nf.format(-45000.1, xlnt::calendar::windows_1900);
TS_ASSERT_EQUALS(formatted, "-$45,000.10");
}
void test_bad_country()
@ -356,21 +357,9 @@ public:
{
xlnt::number_format nf;
nf.set_format_string("");
TS_ASSERT_THROWS(nf.format(1.2, xlnt::calendar::windows_1900), std::runtime_error);
nf.set_format_string(";;;");
TS_ASSERT_THROWS(nf.format(1.2, xlnt::calendar::windows_1900), std::runtime_error);
nf.set_format_string("[=1]\"first\"General;\"second\"General;\"third\"General");
TS_ASSERT_THROWS(nf.format(1.2, xlnt::calendar::windows_1900), std::runtime_error);
nf.set_format_string("[=1]\"first\"General;[=2]\"second\"General;[=3]\"third\"General");
TS_ASSERT_THROWS(nf.format(1.2, xlnt::calendar::windows_1900), std::runtime_error);
nf.set_format_string("[=1]\"first\"General;[=2]\"second\"General;\"third\"General;\"fourth\"General");
TS_ASSERT_THROWS(nf.format(1.2, xlnt::calendar::windows_1900), std::runtime_error);
nf.set_format_string("\"first\"General;\"second\"General;\"third\"General;\"fourth\"General;\"fifth\"General");
TS_ASSERT_THROWS(nf.format(1.2, xlnt::calendar::windows_1900), std::runtime_error);
}
@ -379,7 +368,7 @@ public:
{
TS_ASSERT_EQUALS(xlnt::number_format::text().format("a"), "a");
TS_ASSERT_EQUALS(xlnt::number_format::number().format(1, xlnt::calendar::windows_1900), "1");
TS_ASSERT_EQUALS(xlnt::number_format::number_comma_separated1().format(1, xlnt::calendar::windows_1900), "1,00");
TS_ASSERT_EQUALS(xlnt::number_format::number_comma_separated1().format(12345.67, xlnt::calendar::windows_1900), "12,345.67");
/*
auto datetime = xlnt::datetime(2016, 6, 24, 0, 45, 58);