diff --git a/.gitmodules b/.gitmodules
deleted file mode 100644
index e69de29..0000000
diff --git a/loveframes/third-party/utf8/LICENSE b/loveframes/third-party/utf8/LICENSE
new file mode 100644
index 0000000..fd3b301
--- /dev/null
+++ b/loveframes/third-party/utf8/LICENSE
@@ -0,0 +1,21 @@
+The MIT License (MIT)
+
+Copyright (c) 2016 Stepets
+
+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, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
diff --git a/loveframes/third-party/utf8/README.md b/loveframes/third-party/utf8/README.md
new file mode 100644
index 0000000..1b2d803
--- /dev/null
+++ b/loveframes/third-party/utf8/README.md
@@ -0,0 +1,62 @@
+# utf8.lua
+pure-lua 5.3 regex library for Lua 5.3, Lua 5.1, LuaJIT
+
+This library provides simple way to add UTF-8 support into your application.
+
+#### Example:
+```Lua
+local utf8 = require('.utf8'):init()
+for k,v in pairs(utf8) do
+  string[k] = v
+end
+
+local str = "пыщпыщ ололоо я водитель нло"
+print(str:find("(.л.+)н"))
+-- 8	26	ололоо я водитель
+
+print(str:gsub("ло+", "보라"))
+-- пыщпыщ о보라보라 я водитель н보라	3
+
+print(str:match("^п[лопыщ ]*я"))
+-- пыщпыщ ололоо я
+```
+
+#### Usage:
+
+This library can be used as drop-in replacement for vanilla string library. It exports all vanilla functions under `raw` sub-object.
+
+```Lua
+local utf8 = require('.utf8'):init()
+local str = "пыщпыщ ололоо я водитель нло"
+utf8.gsub(str, "ло+", "보라")
+-- пыщпыщ о보라보라 я водитель н보라	3
+utf8.raw.gsub(str, "ло+", "보라")
+-- пыщпыщ о보라보라о я водитель н보라	3
+```
+
+It also provides all functions from Lua 5.3 UTF-8 [module](https://www.lua.org/manual/5.3/manual.html#6.5) except `utf8.len (s [, i [, j]])`. If you need to validate your strings use `utf8.validate(str, byte_pos)` or iterate over with `utf8.validator`.
+
+#### Installation:
+
+Download repository to your project folder. (no rockspecs yet)
+
+As of Lua 5.3 default `utf8` module has precedence over user-provided. In this case you can specify full module path (`.utf8`).
+
+#### Configuration:
+
+Library is highly modular. You can provide your implementation for almost any function used. Library already has several back-ends:
+- [Runtime character class processing](charclass/runtime/init.lua) using hardcoded codepoint ranges or using native functions through `ffi`.
+- [Basic functions](primitives/init.lua) for working with UTF-8 characters have specializations for `ffi`-enabled runtime and for tarantool.
+
+Probably most interesting [customizations](init.lua) are `utf8.config.loadstring` and `utf8.config.cache` if you want to precompile your regexes.
+
+```Lua
+local utf8 = require('.utf8')
+utf8.config = {
+  cache = my_smart_cache,
+}
+utf8:init()
+```
+Customization is done before initialization. If you want, you can change configuration after `init`, it might work for everything but modules. All of them should be reloaded.
+
+#### [Documentation:](test/test.lua)
diff --git a/loveframes/third-party/utf8/begins/compiletime/parser.lua b/loveframes/third-party/utf8/begins/compiletime/parser.lua
new file mode 100644
index 0000000..c5b0dbc
--- /dev/null
+++ b/loveframes/third-party/utf8/begins/compiletime/parser.lua
@@ -0,0 +1,17 @@
+return function(utf8)
+
+utf8.config.begins = utf8.config.begins or {
+  utf8:require "begins.compiletime.vanilla"
+}
+
+function utf8.regex.compiletime.begins.parse(regex, c, bs, ctx)
+  for _, m in ipairs(utf8.config.begins) do
+    local functions, move = m.parse(regex, c, bs, ctx)
+    utf8.debug("begins", _, c, bs, nbs, move, functions)
+    if functions then
+      return functions, move
+    end
+  end
+end
+
+end
diff --git a/loveframes/third-party/utf8/begins/compiletime/vanilla.lua b/loveframes/third-party/utf8/begins/compiletime/vanilla.lua
new file mode 100644
index 0000000..69da4ca
--- /dev/null
+++ b/loveframes/third-party/utf8/begins/compiletime/vanilla.lua
@@ -0,0 +1,61 @@
+return function(utf8)
+
+local matchers = {
+  sliding = function()
+    return [[
+    add(function(ctx) -- sliding
+        local saved = ctx:clone()
+        local start_pos = ctx.pos
+        while ctx.pos <= 1 + utf8len(ctx.str) do
+            debug('starting from', ctx, "start_pos", start_pos)
+            ctx.result.start = ctx.pos
+            ctx:next_function()
+            ctx:get_function()(ctx)
+
+            ctx = saved:clone()
+            start_pos = start_pos + 1
+            ctx.pos = start_pos
+        end
+        ctx:terminate()
+    end)
+]]
+  end,
+  fromstart = function(ctx)
+    return [[
+    add(function(ctx) -- fromstart
+        local saved = ctx:clone()
+        ctx.result.start = ctx.pos
+        ctx:next_function()
+        ctx:get_function()(ctx)
+        ctx:terminate()
+    end)
+]]
+  end,
+}
+
+local function default()
+  return matchers.sliding()
+end
+
+local function parse(regex, c, bs, ctx)
+  if bs ~= 1 then return end
+
+  local functions
+  local skip = 0
+
+  if c == '^' then
+    functions = matchers.fromstart()
+    skip = 1
+  else
+    functions = matchers.sliding()
+  end
+
+  return functions, skip
+end
+
+return {
+  parse = parse,
+  default = default,
+}
+
+end
diff --git a/loveframes/third-party/utf8/charclass/compiletime/builder.lua b/loveframes/third-party/utf8/charclass/compiletime/builder.lua
new file mode 100644
index 0000000..b1d9e47
--- /dev/null
+++ b/loveframes/third-party/utf8/charclass/compiletime/builder.lua
@@ -0,0 +1,124 @@
+return function(utf8)
+
+local byte = utf8.byte
+local unpack = utf8.config.unpack
+
+local builder = {}
+local mt = {__index = builder}
+
+utf8.regex.compiletime.charclass.builder = builder
+
+function builder.new()
+  return setmetatable({}, mt)
+end
+
+function builder:invert()
+  self.inverted = true
+  return self
+end
+
+function builder:internal() -- is it enclosed in []
+  self.internal = true
+  return self
+end
+
+function builder:with_codes(...)
+  local codes = {...}
+  self.codes = self.codes or {}
+
+  for _, v in ipairs(codes) do
+    table.insert(self.codes, type(v) == "number" and v or byte(v))
+  end
+
+  table.sort(self.codes)
+  return self
+end
+
+function builder:with_ranges(...)
+  local ranges = {...}
+  self.ranges = self.ranges or {}
+
+  for _, v in ipairs(ranges) do
+    table.insert(self.ranges, v)
+  end
+
+  return self
+end
+
+function builder:with_classes(...)
+  local classes = {...}
+  self.classes = self.classes or {}
+
+  for _, v in ipairs(classes) do
+    table.insert(self.classes, v)
+  end
+
+  return self
+end
+
+function builder:without_classes(...)
+  local not_classes = {...}
+  self.not_classes = self.not_classes or {}
+
+  for _, v in ipairs(not_classes) do
+    table.insert(self.not_classes, v)
+  end
+
+  return self
+end
+
+function builder:include(b)
+  if not b.inverted then
+    if b.codes then
+      self:with_codes(unpack(b.codes))
+    end
+    if b.ranges then
+      self:with_ranges(unpack(b.ranges))
+    end
+    if b.classes then
+      self:with_classes(unpack(b.classes))
+    end
+    if b.not_classes then
+      self:without_classes(unpack(b.not_classes))
+    end
+  else
+    self.includes = self.includes or {}
+    self.includes[#self.includes + 1] = b
+  end
+  return self
+end
+
+function builder:build()
+  local codes_list = table.concat(self.codes or {}, ', ')
+  local ranges_list = ''
+  for i, r in ipairs(self.ranges or {}) do ranges_list = ranges_list .. (i > 1 and ', {' or '{') .. tostring(r[1]) .. ', ' .. tostring(r[2]) .. '}' end
+  local classes_list = ''
+  if self.classes then classes_list = "'" .. table.concat(self.classes, "', '") .. "'" end
+  local not_classes_list = ''
+  if self.not_classes then not_classes_list = "'" .. table.concat(self.not_classes, "', '") .. "'" end
+
+  local subs_list = ''
+  for i, r in ipairs(self.includes or {}) do subs_list = subs_list .. (i > 1 and ', ' or '') .. r:build() .. '' end
+
+  local src = [[cl.new():with_codes(
+      ]] .. codes_list .. [[
+    ):with_ranges(
+      ]] .. ranges_list .. [[
+    ):with_classes(
+      ]] .. classes_list .. [[
+    ):without_classes(
+      ]] .. not_classes_list .. [[
+    ):with_subs(
+      ]] .. subs_list .. [[
+    )]]
+
+  if self.inverted then
+    src = src .. ':invert()'
+  end
+
+  return src
+end
+
+return builder
+
+end
diff --git a/loveframes/third-party/utf8/charclass/compiletime/parser.lua b/loveframes/third-party/utf8/charclass/compiletime/parser.lua
new file mode 100644
index 0000000..4f1d4a9
--- /dev/null
+++ b/loveframes/third-party/utf8/charclass/compiletime/parser.lua
@@ -0,0 +1,21 @@
+return function(utf8)
+
+utf8.config.compiletime_charclasses = utf8.config.compiletime_charclasses or {
+  utf8:require "charclass.compiletime.vanilla",
+  utf8:require "charclass.compiletime.range",
+  utf8:require "charclass.compiletime.stub",
+}
+
+function utf8.regex.compiletime.charclass.parse(regex, c, bs, ctx)
+  utf8.debug("parse charclass():", regex, c, bs, regex[bs])
+  for _, p in ipairs(utf8.config.compiletime_charclasses) do
+    local charclass, nbs = p(regex, c, bs, ctx)
+    if charclass then
+      ctx.prev_class = charclass:build()
+      utf8.debug("cc", ctx.prev_class, _, c, bs, nbs)
+      return charclass, nbs
+    end
+  end
+end
+
+end
diff --git a/loveframes/third-party/utf8/charclass/compiletime/range.lua b/loveframes/third-party/utf8/charclass/compiletime/range.lua
new file mode 100644
index 0000000..a76cfd1
--- /dev/null
+++ b/loveframes/third-party/utf8/charclass/compiletime/range.lua
@@ -0,0 +1,44 @@
+return function(utf8)
+
+local cl = utf8.regex.compiletime.charclass.builder
+
+local next = utf8.util.next
+
+return function(str, c, bs, ctx)
+  if not ctx.internal then return end
+
+  local nbs = bs
+
+  local r1, r2
+
+  local c, nbs = c, bs
+  if c == '%' then
+    c, nbs = next(str, nbs)
+    r1 = c
+  else
+    r1 = c
+  end
+
+  utf8.debug("range r1", r1, nbs)
+
+  c, nbs = next(str, nbs)
+  if c ~= '-' then return end
+
+  c, nbs = next(str, nbs)
+  if c == '%' then
+    c, nbs = next(str, nbs)
+    r2 = c
+  elseif c ~= '' then
+    r2 = c
+  end
+
+  utf8.debug("range r2", r2, nbs)
+
+  if r1 and r2 then
+    return cl.new():with_ranges{utf8.byte(r1), utf8.byte(r2)}, utf8.next(str, nbs) - bs
+  else
+    return
+  end
+end
+
+end
diff --git a/loveframes/third-party/utf8/charclass/compiletime/stub.lua b/loveframes/third-party/utf8/charclass/compiletime/stub.lua
new file mode 100644
index 0000000..395d05c
--- /dev/null
+++ b/loveframes/third-party/utf8/charclass/compiletime/stub.lua
@@ -0,0 +1,9 @@
+return function(utf8)
+
+local cl = utf8.regex.compiletime.charclass.builder
+
+return function(str, c, bs, ctx)
+  return cl.new():with_codes(c), utf8.next(str, bs) - bs
+end
+
+end
diff --git a/loveframes/third-party/utf8/charclass/compiletime/vanilla.lua b/loveframes/third-party/utf8/charclass/compiletime/vanilla.lua
new file mode 100644
index 0000000..3307519
--- /dev/null
+++ b/loveframes/third-party/utf8/charclass/compiletime/vanilla.lua
@@ -0,0 +1,112 @@
+return function(utf8)
+
+local cl = utf8:require "charclass.compiletime.builder"
+
+local next = utf8.util.next
+
+local token = 1
+
+local function parse(str, c, bs, ctx)
+  local tttt = token
+  token = token + 1
+
+  local class
+  local nbs = bs
+  utf8.debug("cc_parse", tttt, str, c, nbs, next(str, nbs))
+
+  if c == '%' then
+    c, nbs = next(str, bs)
+    local _c = utf8.raw.lower(c)
+    local matched
+    if _c == 'a' then
+      matched = ('alpha')
+    elseif _c == 'c' then
+      matched = ('cntrl')
+    elseif _c == 'd' then
+      matched = ('digit')
+    elseif _c == 'g' then
+      matched = ('graph')
+    elseif _c == 'l' then
+      matched = ('lower')
+    elseif _c == 'p' then
+      matched = ('punct')
+    elseif _c == 's' then
+      matched = ('space')
+    elseif _c == 'u' then
+      matched = ('upper')
+    elseif _c == 'w' then
+      matched = ('alnum')
+    elseif _c == 'x' then
+      matched = ('xdigit')
+    end
+
+    if matched then
+      if _c ~= c then
+        class = cl.new():without_classes(matched)
+      else
+        class = cl.new():with_classes(matched)
+      end
+    end
+  elseif c == '[' then
+    local old_internal = ctx.internal
+    ctx.internal = true
+    class = cl.new()
+    local firstletter = true
+    while true do
+      local prev_nbs = nbs
+      c, nbs = next(str, nbs)
+      utf8.debug("next", tttt, c, nbs)
+      if c == '^' and firstletter then
+        class:invert()
+      elseif c == ']' then
+        utf8.debug('] on pos', tttt, nbs)
+        break
+      elseif c == '' then
+        error "malformed pattern (missing ']')"
+      else
+        local sub_class, skip = utf8.regex.compiletime.charclass.parse(str, c, nbs, ctx)
+        nbs = prev_nbs + skip
+        utf8.debug("include", tttt, bs, prev_nbs, nbs, skip)
+        class:include(sub_class)
+      end
+      firstletter = false
+    end
+    ctx.internal = old_internal
+  elseif c == '.' then
+    if not ctx.internal then
+      class = cl.new():invert()
+    else
+      class = cl.new():with_codes(c)
+    end
+  end
+
+  return class, utf8.next(str, nbs) - bs
+end
+
+return parse
+
+end
+
+--[[
+    x: (where x is not one of the magic characters ^$()%.[]*+-?) represents the character x itself.
+    .: (a dot) represents all characters.
+    %a: represents all letters.
+    %c: represents all control characters.
+    %d: represents all digits.
+    %g: represents all printable characters except space.
+    %l: represents all lowercase letters.
+    %p: represents all punctuation characters.
+    %s: represents all space characters.
+    %u: represents all uppercase letters.
+    %w: represents all alphanumeric characters.
+    %x: represents all hexadecimal digits.
+    %x: (where x is any non-alphanumeric character) represents the character x. This is the standard way to escape the magic characters. Any non-alphanumeric character (including all punctuation characters, even the non-magical) can be preceded by a '%' when used to represent itself in a pattern.
+    [set]: represents the class which is the union of all characters in set. A range of characters can be specified by separating the end characters of the range, in ascending order, with a '-'. All classes %x described above can also be used as components in set. All other characters in set represent themselves. For example, [%w_] (or [_%w]) represents all alphanumeric characters plus the underscore, [0-7] represents the octal digits, and [0-7%l%-] represents the octal digits plus the lowercase letters plus the '-' character.
+
+    You can put a closing square bracket in a set by positioning it as the first character in the set. You can put a hyphen in a set by positioning it as the first or the last character in the set. (You can also use an escape for both cases.)
+
+    The interaction between ranges and classes is not defined. Therefore, patterns like [%a-z] or [a-%%] have no meaning.
+    [^set]: represents the complement of set, where set is interpreted as above.
+
+For all classes represented by single letters (%a, %c, etc.), the corresponding uppercase letter represents the complement of the class. For instance, %S represents all non-space characters.
+]]
diff --git a/loveframes/third-party/utf8/charclass/runtime/base.lua b/loveframes/third-party/utf8/charclass/runtime/base.lua
new file mode 100644
index 0000000..70b3c88
--- /dev/null
+++ b/loveframes/third-party/utf8/charclass/runtime/base.lua
@@ -0,0 +1,164 @@
+return function(utf8)
+
+local class = {}
+local mt = {__index = class}
+
+local utf8gensub = utf8.gensub
+
+function class.new()
+  return setmetatable({}, mt)
+end
+
+function class:invert()
+  self.inverted = true
+  return self
+end
+
+function class:with_codes(...)
+  local codes = {...}
+  self.codes = self.codes or {}
+
+  for _, v in ipairs(codes) do
+    table.insert(self.codes, v)
+  end
+
+  table.sort(self.codes)
+  return self
+end
+
+function class:with_ranges(...)
+  local ranges = {...}
+  self.ranges = self.ranges or {}
+
+  for _, v in ipairs(ranges) do
+    table.insert(self.ranges, v)
+  end
+
+  return self
+end
+
+function class:with_classes(...)
+  local classes = {...}
+  self.classes = self.classes or {}
+
+  for _, v in ipairs(classes) do
+    table.insert(self.classes, v)
+  end
+
+  return self
+end
+
+function class:without_classes(...)
+  local not_classes = {...}
+  self.not_classes = self.not_classes or {}
+
+  for _, v in ipairs(not_classes) do
+    table.insert(self.not_classes, v)
+  end
+
+  return self
+end
+
+function class:with_subs(...)
+  local subs = {...}
+  self.subs = self.subs or {}
+
+  for _, v in ipairs(subs) do
+    table.insert(self.subs, v)
+  end
+
+  return self
+end
+
+function class:in_codes(item)
+  if not self.codes then return false end
+
+  local head, tail = 1, #self.codes
+  local mid = math.floor((head + tail)/2)
+  while (tail - head) > 1 do
+    if self.codes[mid] > item then
+      tail = mid
+    else
+      head = mid
+    end
+    mid = math.floor((head + tail)/2)
+  end
+  if self.codes[head] == item then
+    return true, head
+  elseif self.codes[tail] == item then
+    return true, tail
+  else
+    return false
+  end
+end
+
+function class:in_ranges(char_code)
+  if not self.ranges then return false end
+
+  for _,r in ipairs(self.ranges) do
+    if r[1] <= char_code and char_code <= r[2] then
+      return true
+    end
+  end
+  return false
+end
+
+function class:in_classes(char_code)
+  if not self.classes then return false end
+
+  for _, class in ipairs(self.classes) do
+    if self:is(class, char_code) then
+      return true
+    end
+  end
+  return false
+end
+
+function class:in_not_classes(char_code)
+  if not self.not_classes then return false end
+
+  for _, class in ipairs(self.not_classes) do
+    if self:is(class, char_code) then
+      return true
+    end
+  end
+  return false
+end
+
+function class:is(class, char_code)
+  error("not implemented")
+end
+
+function class:in_subs(char_code)
+  if not self.subs or #self.subs == 0 then return false end
+
+  for _, c in ipairs(self.subs) do
+    if not c:test(char_code) then
+      return false
+    end
+  end
+  return true
+end
+
+function class:test(char_code)
+  local result = self:do_test(char_code)
+  utf8.debug('class:test', result, "'" .. (char_code and utf8.char(char_code) or 'nil') .. "'", char_code)
+  return result
+end
+
+function class:do_test(char_code)
+  if not char_code then return false end
+  local found = (self:in_codes(char_code) or self:in_ranges(char_code) or self:in_classes(char_code) or self:in_subs(char_code)) and not self:in_not_classes(char_code)
+  utf8.debug('class:do_test', 'found', found, 'inverted', self.inverted, 'result', self.inverted and not found or found)
+  -- utf8.debug(self:in_codes(char_code), self:in_ranges(char_code), self:in_classes(char_code), self:in_subs(char_code), not self:in_not_classes(char_code))
+  -- ternary if ideom (self.inverted and not found or found) doesn't work with booleans >_<
+  if self.inverted then
+    return not found
+  else
+    return found
+  end
+end
+
+return class
+
+end
diff --git a/loveframes/third-party/utf8/charclass/runtime/dummy.lua b/loveframes/third-party/utf8/charclass/runtime/dummy.lua
new file mode 100644
index 0000000..1faddc1
--- /dev/null
+++ b/loveframes/third-party/utf8/charclass/runtime/dummy.lua
@@ -0,0 +1,41 @@
+return function(utf8)
+
+local base = utf8:require "charclass.runtime.base"
+
+local dummy = setmetatable({}, {__index = base})
+local mt = {__index = dummy}
+
+function dummy.new()
+  return setmetatable({}, mt)
+end
+
+function dummy:with_classes(...)
+  local classes = {...}
+  for _, c in ipairs(classes) do
+    if c == 'alpha' then self:with_ranges({65, 90}, {97, 122})
+    elseif c == 'cntrl' then self:with_ranges({0, 31}):with_codes(127)
+    elseif c == 'digit' then self:with_ranges({48, 57})
+    elseif c == 'graph' then self:with_ranges({1, 8}, {14, 31}, {33, 132}, {134, 159}, {161, 5759}, {5761, 8191}, {8203, 8231}, {8234, 8238}, {8240, 8286}, {8288, 12287})
+    elseif c == 'lower' then self:with_ranges({97, 122})
+    elseif c == 'punct' then self:with_ranges({33, 47}, {58, 64}, {91, 96}, {123, 126})
+    elseif c == 'space' then self:with_ranges({9, 13}):with_codes(32, 133, 160, 5760):with_ranges({8192, 8202}):with_codes(8232, 8233, 8239, 8287, 12288)
+    elseif c == 'upper' then self:with_ranges({65, 90})
+    elseif c == 'alnum' then self:with_ranges({48, 57}, {65, 90}, {97, 122})
+    elseif c == 'xdigit' then self:with_ranges({48, 57}, {65, 70}, {97, 102})
+    end
+  end
+  return self
+end
+
+function dummy:without_classes(...)
+  local classes = {...}
+  if #classes > 0 then
+    return self:with_subs(dummy.new():with_classes(...):invert())
+  else
+    return self
+  end
+end
+
+return dummy
+
+end
diff --git a/loveframes/third-party/utf8/charclass/runtime/init.lua b/loveframes/third-party/utf8/charclass/runtime/init.lua
new file mode 100644
index 0000000..e71d037
--- /dev/null
+++ b/loveframes/third-party/utf8/charclass/runtime/init.lua
@@ -0,0 +1,22 @@
+return function(utf8)
+
+local provided = utf8.config.runtime_charclasses
+
+if provided then
+  if type(provided) == "table" then
+    return provided
+  elseif type(provided) == "function" then
+    return provided(utf8)
+  else
+    return utf8:require(provided)
+  end
+end
+
+local ffi = pcall(require, "ffi")
+if not ffi then
+  return utf8:require "charclass.runtime.dummy"
+else
+  return utf8:require "charclass.runtime.native"
+end
+
+end
diff --git a/loveframes/third-party/utf8/charclass/runtime/native.lua b/loveframes/third-party/utf8/charclass/runtime/native.lua
new file mode 100644
index 0000000..f7b7890
--- /dev/null
+++ b/loveframes/third-party/utf8/charclass/runtime/native.lua
@@ -0,0 +1,47 @@
+return function(utf8)
+
+os.setlocale(utf8.config.locale, "ctype")
+
+local ffi = require("ffi")
+ffi.cdef[[
+  int iswalnum(int c);
+  int iswalpha(int c);
+  int iswascii(int c);
+  int iswblank(int c);
+  int iswcntrl(int c);
+  int iswdigit(int c);
+  int iswgraph(int c);
+  int iswlower(int c);
+  int iswprint(int c);
+  int iswpunct(int c);
+  int iswspace(int c);
+  int iswupper(int c);
+  int iswxdigit(int c);
+]]
+
+local base = utf8:require "charclass.runtime.base"
+
+local native = setmetatable({}, {__index = base})
+local mt = {__index = native}
+
+function native.new()
+  return setmetatable({}, mt)
+end
+
+function native:is(class, char_code)
+  if class == 'alpha' then return ffi.C.iswalpha(char_code) ~= 0
+  elseif class == 'cntrl' then return ffi.C.iswcntrl(char_code) ~= 0
+  elseif class == 'digit' then return ffi.C.iswdigit(char_code) ~= 0
+  elseif class == 'graph' then return ffi.C.iswgraph(char_code) ~= 0
+  elseif class == 'lower' then return ffi.C.iswlower(char_code) ~= 0
+  elseif class == 'punct' then return ffi.C.iswpunct(char_code) ~= 0
+  elseif class == 'space' then return ffi.C.iswspace(char_code) ~= 0
+  elseif class == 'upper' then return ffi.C.iswupper(char_code) ~= 0
+  elseif class == 'alnum' then return ffi.C.iswalnum(char_code) ~= 0
+  elseif class == 'xdigit' then return ffi.C.iswxdigit(char_code) ~= 0
+  end
+end
+
+return native
+
+end
diff --git a/loveframes/third-party/utf8/context/compiletime.lua b/loveframes/third-party/utf8/context/compiletime.lua
new file mode 100644
index 0000000..621204d
--- /dev/null
+++ b/loveframes/third-party/utf8/context/compiletime.lua
@@ -0,0 +1,18 @@
+return function(utf8)
+
+local begins = utf8.config.begins
+local ends = utf8.config.ends
+
+return {
+  new = function()
+    return {
+      prev_class = nil,
+      begins = begins[1].default(),
+      ends = ends[1].default(),
+      funcs = {},
+      internal = false, -- hack for ranges, flags if parser is in []
+    }
+  end
+}
+
+end
diff --git a/loveframes/third-party/utf8/context/runtime.lua b/loveframes/third-party/utf8/context/runtime.lua
new file mode 100644
index 0000000..a333749
--- /dev/null
+++ b/loveframes/third-party/utf8/context/runtime.lua
@@ -0,0 +1,74 @@
+return function(utf8)
+
+local utf8unicode = utf8.byte
+local utf8sub = utf8.sub
+local utf8len = utf8.len
+local rawgsub = utf8.raw.gsub
+
+local util = utf8.util
+
+local ctx = {}
+local mt = {
+  __index = ctx,
+  __tostring = function(self)
+    return rawgsub([[str: '${str}', char: ${pos} '${char}', func: ${func_pos}]], "${(.-)}", {
+      str = self.str,
+      pos = self.pos,
+      char = self:get_char(),
+      func_pos = self.func_pos,
+    })
+  end
+}
+
+function ctx.new(obj)
+  obj = obj or {}
+  return setmetatable({
+    pos = obj.pos or 1,
+    str = obj.str or nil,
+    starts = obj.starts or nil,
+    functions = obj.functions or {},
+    func_pos = obj.func_pos or 1,
+    ends = obj.ends or nil,
+    result = obj.result and util.copy(obj.result) or {},
+    captures = obj.captures and util.copy(obj.captures, true) or {active = {}},
+  }, mt)
+end
+
+function ctx:clone()
+  return self:new()
+end
+
+function ctx:next_char()
+  self.pos = self.pos + 1
+end
+
+function ctx:get_char()
+  return utf8sub(self.str, self.pos, self.pos)
+end
+
+function ctx:get_charcode()
+  if utf8len(self.str) < self.pos then return nil end
+  return utf8unicode(self:get_char())
+end
+
+function ctx:next_function()
+  self.func_pos = self.func_pos + 1
+end
+
+function ctx:get_function()
+  return self.functions[self.func_pos]
+end
+
+function ctx:done()
+  utf8.debug('done', self)
+  coroutine.yield(self, self.result, self.captures)
+end
+
+function ctx:terminate()
+  utf8.debug('terminate', self)
+  coroutine.yield(nil)
+end
+
+return ctx
+
+end
diff --git a/loveframes/third-party/utf8/ends/compiletime/parser.lua b/loveframes/third-party/utf8/ends/compiletime/parser.lua
new file mode 100644
index 0000000..25f93ca
--- /dev/null
+++ b/loveframes/third-party/utf8/ends/compiletime/parser.lua
@@ -0,0 +1,17 @@
+return function(utf8)
+
+utf8.config.ends = utf8.config.ends or {
+  utf8:require "ends.compiletime.vanilla"
+}
+
+function utf8.regex.compiletime.ends.parse(regex, c, bs, ctx)
+  for _, m in ipairs(utf8.config.ends) do
+    local functions, move = m.parse(regex, c, bs, ctx)
+    utf8.debug("ends", _, c, bs, nbs, move, functions)
+    if functions then
+      return functions, move
+    end
+  end
+end
+
+end
diff --git a/loveframes/third-party/utf8/ends/compiletime/vanilla.lua b/loveframes/third-party/utf8/ends/compiletime/vanilla.lua
new file mode 100644
index 0000000..520c7a9
--- /dev/null
+++ b/loveframes/third-party/utf8/ends/compiletime/vanilla.lua
@@ -0,0 +1,45 @@
+return function(utf8)
+
+local matchers = {
+  any = function()
+    return [[
+  add(function(ctx) -- any
+    ctx.result.finish = ctx.pos - 1
+    ctx:done()
+  end)
+]]
+  end,
+  toend = function(ctx)
+    return [[
+  add(function(ctx) -- toend
+    ctx.result.finish = ctx.pos - 1
+    if ctx.pos == #ctx.str + 1 then ctx:done() end
+  end)
+]]
+  end,
+}
+
+local len = utf8.raw.len
+
+local function default()
+  return matchers.any()
+end
+
+local function parse(regex, c, bs, ctx)
+  local functions
+  local skip = 0
+
+  if bs == len(regex) and c == '$' then
+    functions = matchers.toend()
+    skip = 1
+  end
+
+  return functions, skip
+end
+
+return {
+  parse = parse,
+  default = default,
+}
+
+end
diff --git a/loveframes/third-party/utf8/functions/lua53.lua b/loveframes/third-party/utf8/functions/lua53.lua
new file mode 100644
index 0000000..ee05271
--- /dev/null
+++ b/loveframes/third-party/utf8/functions/lua53.lua
@@ -0,0 +1,134 @@
+return function(utf8)
+
+local utf8sub = utf8.sub
+local utf8gensub = utf8.gensub
+local unpack = utf8.config.unpack
+local get_matcher_function = utf8:require 'regex_parser'
+
+local function utf8find(str, regex, init, plain)
+  local func = get_matcher_function(regex, plain)
+  init = ((init or 1) < 0) and (utf8.len(str) + init + 1) or init
+  local ctx, result, captures = func(str, init, utf8)
+  if not ctx then return nil end
+
+  utf8.debug('ctx:', ctx)
+  utf8.debug('result:', result)
+  utf8.debug('captures:', captures)
+
+  return result.start, result.finish, unpack(captures)
+end
+
+local function utf8match(str, regex, init)
+  local func = get_matcher_function(regex, plain, utf8)
+  local ctx, result, captures = func(str, init, utf8)
+  if not ctx then return nil end
+
+  utf8.debug('ctx:', ctx)
+  utf8.debug('result:', result)
+  utf8.debug('captures:', captures)
+
+  if #captures > 0 then return unpack(captures) end
+
+  return utf8sub(str, result.start, result.finish)
+end
+
+local function utf8gmatch(str, regex)
+  regex = (utf8sub(regex,1,1) ~= '^') and regex or '%' .. regex
+  local func = get_matcher_function(regex, plain, utf8)
+  local ctx, result, captures
+  local continue_pos = 1
+
+  return function()
+    ctx, result, captures = func(str, continue_pos, utf8)
+
+    if not ctx then return nil end
+
+    utf8.debug('ctx:', ctx)
+    utf8.debug('result:', result)
+    utf8.debug('captures:', captures)
+
+    continue_pos = math.max(result.finish + 1, result.start + 1)
+    if #captures > 0 then
+      return unpack(captures)
+    else
+      return utf8sub(str, result.start, result.finish)
+    end
+  end
+end
+
+local function replace(repl, args)
+  local ret = ''
+  if type(repl) == 'string' then
+    local ignore = false
+    local num
+    for _, c in utf8gensub(repl) do
+      if not ignore then
+        if c == '%' then
+          ignore = true
+        else
+          ret = ret .. c
+        end
+      else
+        num = tonumber(c)
+        if num then
+          ret = ret .. args[num]
+        else
+          ret = ret .. c
+        end
+        ignore = false
+      end
+    end
+  elseif type(repl) == 'table' then
+    ret = repl[args[1] or args[0]] or ''
+  elseif type(repl) == 'function' then
+    if #args > 0 then
+      ret = repl(unpack(args, 1)) or ''
+    else
+      ret = repl(args[0]) or ''
+    end
+  end
+  return ret
+end
+
+local function utf8gsub(str, regex, repl, limit)
+  limit = limit or -1
+  local subbed = ''
+  local prev_sub_finish = 1
+
+  regex = (utf8sub(regex,1,1) ~= '^') and regex or '%' .. regex
+  local func = get_matcher_function(regex, plain, utf8)
+  local ctx, result, captures
+  local continue_pos = 1
+
+  local n = 0
+  while limit ~= n do
+    ctx, result, captures = func(str, continue_pos, utf8)
+    if not ctx then break end
+
+    utf8.debug('ctx:', ctx)
+    utf8.debug('result:', result)
+    utf8.debug('result:', utf8sub(str, result.start, result.finish))
+    utf8.debug('captures:', captures)
+
+    continue_pos = math.max(result.finish + 1, result.start + 1)
+    local args = {[0] = utf8sub(str, result.start, result.finish), unpack(captures)}
+
+    subbed = subbed .. utf8sub(str, prev_sub_finish, result.start - 1)
+    subbed = subbed .. replace(repl, args)
+    prev_sub_finish = result.finish + 1
+    n = n + 1
+
+  end
+
+  return subbed .. utf8sub(str, prev_sub_finish), n
+end
+
+-- attaching high-level functions
+utf8.find    = utf8find
+utf8.match   = utf8match
+utf8.gmatch  = utf8gmatch
+utf8.gsub    = utf8gsub
+
+return utf8
+
+end
diff --git a/loveframes/third-party/utf8/init.lua b/loveframes/third-party/utf8/init.lua
new file mode 100644
index 0000000..13cf69a
--- /dev/null
+++ b/loveframes/third-party/utf8/init.lua
@@ -0,0 +1,57 @@
+local module_path = ...
+module_path = module_path:match("^(.-)init$") or (module_path .. '.')
+local utf8 = {
+  config = {},
+  default = {
+    debug = nil,
+    logger = io.write,
+    loadstring = (loadstring or load),
+    unpack = (unpack or table.unpack),
+    cache = {
+      regex = setmetatable({},{
+        __mode = 'kv'
+      }),
+      plain = setmetatable({},{
+        __mode = 'kv'
+      }),
+    },
+    locale = "C.UTF-8",
+  },
+  regex = {
+    compiletime = {
+      charclass = {},
+      begins = {},
+      ends = {},
+      modifier = {},
+    }
+  },
+  util = {},
+}
+
+function utf8:require(name)
+  local full_module_path = module_path .. name
+  if package.loaded[full_module_path] then
+    return package.loaded[full_module_path]
+  end
+
+  local mod = require(full_module_path)
+  if type(mod) == 'function' then
+    mod = mod(self)
+    package.loaded[full_module_path] = mod
+  end
+  return mod
+end
+
+function utf8:init()
+  for k, v in pairs(self.default) do
+    self.config[k] = self.config[k] or v
+  end
+
+  self:require "util"
+  self:require "primitives.init"
+  self:require "functions.lua53"
+
+  return self
+end
+
+return utf8
diff --git a/loveframes/third-party/utf8/modifier/compiletime/frontier.lua b/loveframes/third-party/utf8/modifier/compiletime/frontier.lua
new file mode 100644
index 0000000..0c3a912
--- /dev/null
+++ b/loveframes/third-party/utf8/modifier/compiletime/frontier.lua
@@ -0,0 +1,49 @@
+return function(utf8)
+
+local matchers = {
+  frontier = function(class, name)
+    local class_name = 'class' .. name
+    return [[
+  local ]] .. class_name .. [[ = ]] .. class .. [[
+
+  add(function(ctx) -- frontier
+    ctx.pos = ctx.pos - 1
+    local prev_charcode = ctx:get_charcode()
+    ctx:next_char()
+    debug("frontier pos", ctx.pos, "prev_charcode", prev_charcode, "charcode", ctx:get_charcode())
+    if ]] .. class_name .. [[:test(prev_charcode) then return end
+    if ]] .. class_name .. [[:test(ctx:get_charcode()) then
+      ctx:next_function()
+      return ctx:get_function()(ctx)
+    end
+  end)
+]]
+  end,
+  simple = utf8:require("modifier.compiletime.simple").simple,
+}
+
+local function parse(regex, c, bs, ctx)
+  local functions, nbs, class
+
+  if c == '%' then
+    if utf8.raw.sub(regex, bs + 1, bs + 1) ~= 'f' then return end
+    if utf8.raw.sub(regex, bs + 2, bs + 2) ~= '[' then error("missing '[' after '%f' in pattern") end
+
+    functions = {}
+    if ctx.prev_class then
+      table.insert(functions, matchers.simple(ctx.prev_class, tostring(bs)))
+      ctx.prev_class = nil
+    end
+    class, nbs = utf8.regex.compiletime.charclass.parse(regex, '[', bs + 2, ctx)
+    nbs = nbs + 2
+    table.insert(functions, matchers.frontier(class:build(), tostring(bs)))
+  end
+
+  return functions, nbs
+end
+
+return {
+  parse = parse,
+}
+
+end
diff --git a/loveframes/third-party/utf8/modifier/compiletime/parser.lua b/loveframes/third-party/utf8/modifier/compiletime/parser.lua
new file mode 100644
index 0000000..0fb2e53
--- /dev/null
+++ b/loveframes/third-party/utf8/modifier/compiletime/parser.lua
@@ -0,0 +1,20 @@
+return function(utf8)
+
+utf8.config.modifier = utf8.config.modifier or {
+  utf8:require "modifier.compiletime.vanilla",
+  utf8:require "modifier.compiletime.frontier",
+  utf8:require "modifier.compiletime.stub",
+}
+
+function utf8.regex.compiletime.modifier.parse(regex, c, bs, ctx)
+  for _, m in ipairs(utf8.config.modifier) do
+    local functions, move = m.parse(regex, c, bs, ctx)
+    utf8.debug("mod", _, c, bs, nbs, move, functions and utf8.config.unpack(functions))
+    if functions then
+      ctx.prev_class = nil
+      return functions, move
+    end
+  end
+end
+
+end
diff --git a/loveframes/third-party/utf8/modifier/compiletime/simple.lua b/loveframes/third-party/utf8/modifier/compiletime/simple.lua
new file mode 100644
index 0000000..d1e3943
--- /dev/null
+++ b/loveframes/third-party/utf8/modifier/compiletime/simple.lua
@@ -0,0 +1,23 @@
+return function(utf8)
+
+local matchers = {
+  simple = function(class, name)
+    local class_name = 'class' .. name
+    return [[
+  local ]] .. class_name .. [[ = ]] .. class .. [[
+
+  add(function(ctx) -- simple
+    debug(ctx, 'simple', ']] .. class_name .. [[')
+    if ]] .. class_name .. [[:test(ctx:get_charcode()) then
+      ctx:next_char()
+      ctx:next_function()
+      return ctx:get_function()(ctx)
+    end
+  end)
+]]
+  end,
+}
+
+return matchers
+
+end
diff --git a/loveframes/third-party/utf8/modifier/compiletime/stub.lua b/loveframes/third-party/utf8/modifier/compiletime/stub.lua
new file mode 100644
index 0000000..16458b3
--- /dev/null
+++ b/loveframes/third-party/utf8/modifier/compiletime/stub.lua
@@ -0,0 +1,28 @@
+return function(utf8)
+
+local matchers = utf8:require("modifier.compiletime.simple")
+
+local function parse(regex, c, bs, ctx)
+  local functions
+
+  if ctx.prev_class then
+    functions = { matchers.simple(ctx.prev_class, tostring(bs)) }
+    ctx.prev_class = nil
+  end
+
+  return functions, 0
+end
+
+local function check(ctx)
+  if ctx.prev_class then
+    table.insert(ctx.funcs, matchers.simple(ctx.prev_class, tostring(bs)))
+    ctx.prev_class = nil
+  end
+end
+
+return {
+  parse = parse,
+  check = check,
+}
+
+end
diff --git a/loveframes/third-party/utf8/modifier/compiletime/vanilla.lua b/loveframes/third-party/utf8/modifier/compiletime/vanilla.lua
new file mode 100644
index 0000000..0723583
--- /dev/null
+++ b/loveframes/third-party/utf8/modifier/compiletime/vanilla.lua
@@ -0,0 +1,226 @@
+return function(utf8)
+
+local utf8unicode = utf8.byte
+local sub = utf8.raw.sub
+
+local matchers = {
+  star = function(class, name)
+    local class_name = 'class' .. name
+    return [[
+  local ]] .. class_name .. [[ = ]] .. class .. [[
+
+  add(function(ctx) -- star
+    debug(ctx, 'star', ']] .. class_name .. [[')
+    local saved = {ctx:clone()}
+    while ]] .. class_name .. [[:test(ctx:get_charcode()) do
+      ctx:next_char()
+      table.insert(saved, ctx:clone())
+      debug('#saved <<', #saved)
+    end
+    while #saved > 0 do
+      ctx = table.remove(saved)
+      ctx:next_function()
+      ctx:get_function()(ctx)
+      debug('#saved >>', #saved)
+    end
+  end)
+]]
+  end,
+  minus = function(class, name)
+    local class_name = 'class' .. name
+    return [[
+  local ]] .. class_name .. [[ = ]] .. class .. [[
+
+  add(function(ctx) -- minus
+    debug(ctx, 'minus', ']] .. class_name .. [[')
+
+    repeat
+      local saved = ctx:clone()
+      ctx:next_function()
+      ctx:get_function()(ctx)
+      ctx = saved
+      local match = ]] .. class_name .. [[:test(ctx:get_charcode())
+      ctx:next_char()
+    until not match
+  end)
+]]
+  end,
+  question = function(class, name)
+    local class_name = 'class' .. name
+    return [[
+  local ]] .. class_name .. [[ = ]] .. class .. [[
+
+  add(function(ctx) -- question
+    debug(ctx, 'question', ']] .. class_name .. [[')
+    local saved = ctx:clone()
+    if ]] .. class_name .. [[:test(ctx:get_charcode()) then
+      ctx:next_char()
+      ctx:next_function()
+      ctx:get_function()(ctx)
+    end
+    ctx = saved
+    ctx:next_function()
+    return ctx:get_function()(ctx)
+  end)
+]]
+  end,
+  capture_start = function(number)
+    return [[
+  add(function(ctx)
+    debug(ctx, 'capture_start', ']] .. tostring(number) .. [[')
+    table.insert(ctx.captures.active, { id = ]] .. tostring(number) .. [[, start_byte = byte_pos, start = ctx.pos })
+    ctx:next_function()
+    return ctx:get_function()(ctx)
+  end)
+]]
+  end,
+  capture_finish = function(number)
+    return [[
+  add(function(ctx)
+    debug(ctx, 'capture_finish', ']] .. tostring(number) .. [[')
+    local cap = table.remove(ctx.captures.active)
+    cap.finish_byte = byte_pos
+    cap.finish = ctx.pos
+    ctx.captures[cap.id] = utf8sub(ctx.str, cap.start, cap.finish - 1)
+    debug('capture#' .. tostring(cap.id), '[' .. tostring(cap.start).. ',' .. tostring(cap.finish) .. ']' , 'is', ctx.captures[cap.id])
+    ctx:next_function()
+    return ctx:get_function()(ctx)
+  end)
+]]
+  end,
+  capture = function(number)
+    return [[
+  add(function(ctx)
+    debug(ctx, 'capture', ']] .. tostring(number) .. [[')
+    local cap = ctx.captures[ ]] .. tostring(number) .. [[ ]
+    local len = utf8len(cap)
+		local check = utf8sub(ctx.str, ctx.pos, ctx.pos + len - 1)
+    debug("capture check:", cap, check)
+		if cap == check then
+			ctx.pos = ctx.pos + len
+			ctx:next_function()
+      return ctx:get_function()(ctx)
+		end
+  end)
+]]
+  end,
+  balancer = function(pair, name)
+    local class_name = 'class' .. name
+    return [[
+
+  add(function(ctx) -- balancer
+    local d, b = ]] .. tostring(utf8unicode(pair[1])) .. [[, ]] .. tostring(utf8unicode(pair[2])) .. [[
+    if ctx:get_charcode() ~= d then return end
+    local balance = 0
+    repeat
+      local c = ctx:get_charcode()
+      if c == nil then return end
+
+      if c == d then
+        balance = balance + 1
+      elseif c == b then
+        balance = balance - 1
+      end
+      debug("balancer: balance=", balance, ", d=", d, ", b=", b, ", charcode=", ctx:get_charcode())
+      ctx:next_char()
+    until balance == 0
+    ctx:next_function()
+    return ctx:get_function()(ctx)
+  end)
+]]
+  end,
+  simple = utf8:require("modifier.compiletime.simple").simple,
+}
+
+local next = utf8.util.next
+
+local function parse(regex, c, bs, ctx)
+  local functions, nbs = nil, bs
+  if c == '%' then
+    c, nbs = next(regex, bs)
+    utf8.debug("next", c, bs)
+    if utf8.raw.find('123456789', c, 1, true) then
+      functions = { matchers.capture(tonumber(c)) }
+      nbs = utf8.next(regex, nbs)
+    elseif c == 'b' then
+      local d, b
+      d, nbs = next(regex, nbs)
+      b, nbs = next(regex, nbs)
+      functions = { matchers.balancer({d, b}, tostring(bs)) }
+      nbs = utf8.next(regex, nbs)
+    end
+
+    if functions and ctx.prev_class then
+      table.insert(functions, 1, matchers.simple(ctx.prev_class, tostring(bs)))
+    end
+  elseif c == '*' and ctx.prev_class then
+    functions = {
+      matchers.star(
+        ctx.prev_class,
+        tostring(bs)
+      )
+    }
+    nbs = bs + 1
+  elseif c == '+' and ctx.prev_class then
+    functions = {
+      matchers.simple(
+        ctx.prev_class,
+        tostring(bs)
+      ),
+      matchers.star(
+        ctx.prev_class,
+        tostring(bs)
+      )
+    }
+    nbs = bs + 1
+  elseif c == '-' and ctx.prev_class then
+    functions = {
+      matchers.minus(
+        ctx.prev_class,
+        tostring(bs)
+      )
+    }
+    nbs = bs + 1
+  elseif c == '?' and ctx.prev_class then
+    functions = {
+      matchers.question(
+        ctx.prev_class,
+        tostring(bs)
+      )
+    }
+    nbs = bs + 1
+  elseif c == '(' then
+    ctx.capture = ctx.capture or {balance = 0, id = 0}
+    ctx.capture.balance = ctx.capture.balance + 1
+    ctx.capture.id = ctx.capture.id + 1
+    functions = { matchers.capture_start(ctx.capture.id) }
+    if ctx.prev_class then
+      table.insert(functions, 1, matchers.simple(ctx.prev_class, tostring(bs)))
+    end
+    nbs = bs + 1
+  elseif c == ')' then
+    ctx.capture = ctx.capture or {balance = 0, id = 0}
+    functions = { matchers.capture_finish(ctx.capture.id) }
+
+    ctx.capture.balance = ctx.capture.balance - 1
+    assert(ctx.capture.balance >= 0, 'invalid capture: "(" missing')
+
+    if ctx.prev_class then
+      table.insert(functions, 1, matchers.simple(ctx.prev_class, tostring(bs)))
+    end
+    nbs = bs + 1
+  end
+
+  return functions, nbs - bs
+end
+
+local function check(ctx)
+  if ctx.capture then assert(ctx.capture.balance == 0, 'invalid capture: ")" missing') end
+end
+
+return {
+  parse = parse,
+  check = check,
+}
+
+end
diff --git a/loveframes/third-party/utf8/primitives/dummy.lua b/loveframes/third-party/utf8/primitives/dummy.lua
new file mode 100644
index 0000000..90a5d7e
--- /dev/null
+++ b/loveframes/third-party/utf8/primitives/dummy.lua
@@ -0,0 +1,522 @@
+-- $Id: utf8.lua 179 2009-04-03 18:10:03Z pasta $
+--
+-- Provides UTF-8 aware string functions implemented in pure lua:
+-- * utf8len(s)
+-- * utf8sub(s, i, j)
+-- * utf8reverse(s)
+-- * utf8char(unicode)
+-- * utf8unicode(s, i, j)
+-- * utf8gensub(s, sub_len)
+-- * utf8find(str, regex, init, plain)
+-- * utf8match(str, regex, init)
+-- * utf8gmatch(str, regex, all)
+-- * utf8gsub(str, regex, repl, limit)
+--
+-- If utf8data.lua (containing the lower<->upper case mappings) is loaded, these
+-- additional functions are available:
+-- * utf8upper(s)
+-- * utf8lower(s)
+--
+-- All functions behave as their non UTF-8 aware counterparts with the exception
+-- that UTF-8 characters are used instead of bytes for all units.
+
+--[[
+Copyright (c) 2006-2007, Kyle Smith
+All rights reserved.
+
+Contributors:
+	Alimov Stepan
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    * Redistributions of source code must retain the above copyright notice,
+      this list of conditions and the following disclaimer.
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+    * Neither the name of the author nor the names of its contributors may be
+      used to endorse or promote products derived from this software without
+      specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+--]]
+
+-- ABNF from RFC 3629
+--
+-- UTF8-octets = *( UTF8-char )
+-- UTF8-char   = UTF8-1 / UTF8-2 / UTF8-3 / UTF8-4
+-- UTF8-1      = %x00-7F
+-- UTF8-2      = %xC2-DF UTF8-tail
+-- UTF8-3      = %xE0 %xA0-BF UTF8-tail / %xE1-EC 2( UTF8-tail ) /
+--               %xED %x80-9F UTF8-tail / %xEE-EF 2( UTF8-tail )
+-- UTF8-4      = %xF0 %x90-BF 2( UTF8-tail ) / %xF1-F3 3( UTF8-tail ) /
+--               %xF4 %x80-8F 2( UTF8-tail )
+-- UTF8-tail   = %x80-BF
+--
+return function(utf8)
+
+local byte    = string.byte
+local char    = string.char
+local dump    = string.dump
+local find    = string.find
+local format  = string.format
+local len     = string.len
+local lower   = string.lower
+local rep     = string.rep
+local sub     = string.sub
+local upper   = string.upper
+
+local function utf8symbollen(byte)
+  return not byte and 0 or (byte < 0x80 and 1) or (byte >= 0xF0 and 4) or (byte >= 0xE0 and 3) or (byte >= 0xC0 and 2) or 1
+end
+
+local function utf8charbytes(str, bs)
+  return utf8symbollen(byte(str, bs))
+end
+
+local function utf8next(str, bs)
+  return bs + utf8charbytes(str, bs)
+end
+
+-- returns the number of characters in a UTF-8 string
+local function utf8len (str)
+  local bs = 1
+  local bytes = len(str)
+  local length = 0
+
+  while bs <= bytes do
+    length = length + 1
+    bs = utf8next(str, bs)
+  end
+
+  return length
+end
+
+-- functions identically to string.sub except that i and j are UTF-8 characters
+-- instead of bytes
+local function utf8sub (s, i, j)
+  -- argument defaults
+  j = j or -1
+
+  local bs = 1
+  local bytes = len(s)
+  local length = 0
+
+  local l = (i >= 0 and j >= 0) or utf8len(s)
+  i = (i >= 0) and i or l + i + 1
+  j = (j >= 0) and j or l + j + 1
+
+  if i > j then
+    return ""
+  end
+
+  local start, finish = 1, bytes
+
+  while bs <= bytes do
+    length = length + 1
+
+    if length == i then
+      start = bs
+    end
+
+    bs = utf8next(s, bs)
+
+    if length == j then
+      finish = bs - 1
+      break
+    end
+  end
+
+  if i > length then start = bytes + 1 end
+  if j < 1 then finish = 0 end
+
+  return sub(s, start, finish)
+end
+
+-- http://en.wikipedia.org/wiki/Utf8
+-- http://developer.coronalabs.com/code/utf-8-conversion-utility
+local function utf8char(...)
+  local codes = {...}
+  local result = {}
+
+  for _, unicode in ipairs(codes) do
+
+    if unicode <= 0x7F then
+      result[#result + 1] = unicode
+    elseif unicode <= 0x7FF then
+      local b0 = 0xC0 + math.floor(unicode / 0x40);
+      local b1 = 0x80 + (unicode % 0x40);
+      result[#result + 1] = b0
+      result[#result + 1] = b1
+    elseif unicode <= 0xFFFF then
+      local b0 = 0xE0 +  math.floor(unicode / 0x1000);
+      local b1 = 0x80 + (math.floor(unicode / 0x40) % 0x40);
+      local b2 = 0x80 + (unicode % 0x40);
+      result[#result + 1] = b0
+      result[#result + 1] = b1
+      result[#result + 1] = b2
+    elseif unicode <= 0x10FFFF then
+      local code = unicode
+      local b3= 0x80 + (code % 0x40);
+      code       = math.floor(code / 0x40)
+      local b2= 0x80 + (code % 0x40);
+      code       = math.floor(code / 0x40)
+      local b1= 0x80 + (code % 0x40);
+      code       = math.floor(code / 0x40)
+      local b0= 0xF0 + code;
+
+      result[#result + 1] = b0
+      result[#result + 1] = b1
+      result[#result + 1] = b2
+      result[#result + 1] = b3
+    else
+      error 'Unicode cannot be greater than U+10FFFF!'
+    end
+
+  end
+
+  return char(utf8.config.unpack(result))
+end
+
+
+local shift_6  = 2^6
+local shift_12 = 2^12
+local shift_18 = 2^18
+
+local utf8unicode
+utf8unicode = function(str, ibs, jbs)
+  if ibs > jbs then return end
+
+  local ch,bytes
+
+  bytes = utf8charbytes(str, ibs)
+  if bytes == 0 then return end
+  ch  = sub(str,ibs,ibs-1+bytes)
+
+  local unicode
+
+  if bytes == 1 then unicode = byte(ch) end
+  if bytes == 2 then
+    local byte0,byte1 = byte(ch,1,2)
+    if byte0 and byte1 then
+      local code0,code1 = byte0-0xC0,byte1-0x80
+      unicode = code0*shift_6 + code1
+    else
+      unicode = byte0
+    end
+  end
+  if bytes == 3 then
+    local byte0,byte1,byte2 = byte(ch,1,3)
+    if byte0 and byte1 and byte2 then
+      local code0,code1,code2 = byte0-0xE0,byte1-0x80,byte2-0x80
+      unicode = code0*shift_12 + code1*shift_6 + code2
+    else
+      unicode = byte0
+    end
+  end
+  if bytes == 4 then
+    local byte0,byte1,byte2,byte3 = byte(ch,1,4)
+    if byte0 and byte1 and byte2 and byte3 then
+      local code0,code1,code2,code3 = byte0-0xF0,byte1-0x80,byte2-0x80,byte3-0x80
+      unicode = code0*shift_18 + code1*shift_12 + code2*shift_6 + code3
+    else
+      unicode = byte0
+    end
+  end
+
+  return unicode,utf8unicode(str, ibs+bytes, jbs)
+end
+
+local function utf8byte(str, i, j)
+  if #str == 0 then return end
+
+  local ibs, jbs
+
+  if i or j then
+    i = i or 1
+    j = j or i
+
+    local str_len = utf8len(str)
+    i = i < 0 and str_len + i + 1 or i
+    j = j < 0 and str_len + j + 1 or j
+    j = j > str_len and str_len or j
+
+    if i > j then return end
+
+    for p = 1, i - 1 do
+      ibs = utf8next(str, ibs or 1)
+    end
+
+    if i == j then
+      jbs = ibs
+    else
+      for p = 1, j - 1 do
+        jbs = utf8next(str, jbs or 1)
+      end
+    end
+
+    if not ibs or not jbs then
+      return nil
+    end
+  else
+    ibs, jbs = 1, 1
+  end
+
+  return utf8unicode(str, ibs, jbs)
+end
+
+local function utf8gensub(str, sub_len)
+  sub_len = sub_len or 1
+  local max_len = #str
+  return function(skip_ptr, bs)
+    bs = (bs and bs or 1) + (skip_ptr and (skip_ptr[1] or 0) or 0)
+
+    nbs = bs
+    if bs > max_len then return nil end
+    for i = 1, sub_len do
+      nbs = utf8next(str, nbs)
+    end
+
+    return nbs, sub(str, bs, nbs - 1), bs
+  end
+end
+
+local function utf8reverse (s)
+  local result = ''
+  for _, w in utf8gensub(s) do result = w .. result end
+  return result
+end
+
+local function utf8validator(str, bs)
+  bs = bs or 1
+
+  if type(str) ~= "string" then
+    error("bad argument #1 to 'utf8charbytes' (string expected, got ".. type(str).. ")")
+  end
+  if type(bs) ~= "number" then
+    error("bad argument #2 to 'utf8charbytes' (number expected, got ".. type(bs).. ")")
+  end
+
+  local c = byte(str, bs)
+  if not c then return end
+
+  -- determine bytes needed for character, based on RFC 3629
+
+  -- UTF8-1
+  if c >= 0 and c <= 127 then
+    return bs + 1
+  elseif c >= 128 and c <= 193 then
+    return bs + 1, bs, 1, c
+      -- UTF8-2
+  elseif c >= 194 and c <= 223 then
+    local c2 = byte(str, bs + 1)
+    if not c2 or c2 < 128 or c2 > 191 then
+      return bs + 2, bs, 2, c2
+    end
+
+    return bs + 2
+      -- UTF8-3
+  elseif c >= 224 and c <= 239 then
+    local c2 = byte(str, bs + 1)
+
+    if not c2 then
+      return bs + 2, bs, 2, c2
+    end
+
+    -- validate byte 2
+    if c == 224 and (c2 < 160 or c2 > 191) then
+      return bs + 2, bs, 2, c2
+    elseif c == 237 and (c2 < 128 or c2 > 159) then
+      return bs + 2, bs, 2, c2
+    elseif c2 < 128 or c2 > 191 then
+      return bs + 2, bs, 2, c2
+    end
+
+    local c3 = byte(str, bs + 2)
+    if not c3 or c3 < 128 or c3 > 191 then
+      return bs + 3, bs, 3, c3
+    end
+
+    return bs + 3
+      -- UTF8-4
+  elseif c >= 240 and c <= 244 then
+    local c2 = byte(str, bs + 1)
+
+    if not c2 then
+      return bs + 2, bs, 2, c2
+    end
+
+    -- validate byte 2
+    if c == 240 and (c2 < 144 or c2 > 191) then
+      return bs + 2, bs, 2, c2
+    elseif c == 244 and (c2 < 128 or c2 > 143) then
+      return bs + 2, bs, 2, c2
+    elseif c2 < 128 or c2 > 191 then
+      return bs + 2, bs, 2, c2
+    end
+
+    local c3 = byte(str, bs + 2)
+    if not c3 or c3 < 128 or c3 > 191 then
+      return bs + 3, bs, 3, c3
+    end
+
+    local c4 = byte(str, bs + 3)
+    if not c4 or c4 < 128 or c4 > 191 then
+      return bs + 4, bs, 4, c4
+    end
+
+    return bs + 4
+  else -- c > 245
+    return bs + 1, bs, 1, c
+  end
+end
+
+local function utf8validate(str, byte_pos)
+  local result = {}
+  for nbs, bs, part, code in utf8validator, str, byte_pos do
+    if bs then
+      result[#result + 1] = { pos = bs, part = part, code = code }
+    end
+  end
+  return #result == 0, result
+end
+
+local function utf8codes(str)
+  local max_len = #str
+  local bs = 1
+  return function(skip_ptr)
+    if bs > max_len then return nil end
+    local pbs = bs
+    bs = utf8next(str, pbs)
+
+    return pbs, utf8unicode(str, pbs, pbs), pbs
+  end
+end
+
+
+--[[--
+differs from Lua 5.3 utf8.offset in accepting any byte positions (not only head byte) for all n values
+
+h - head, c - continuation, t - tail
+hhhccthccthccthcthhh
+        ^ start byte pos
+searching current charracter head by moving backwards
+hhhccthccthccthcthhh
+      ^ head
+
+n == 0: current position
+n > 0: n jumps forward
+n < 0: n more scans backwards
+--]]--
+local function utf8offset(str, n, bs)
+  local l = #str
+  if not bs then
+    if n < 0 then
+      bs = l + 1
+    else
+      bs = 1
+    end
+  end
+  if bs < 0 or bs > l + 1 then
+    error("bad argument #3 to 'offset' (position out of range)")
+  end
+
+  if n == 0 then
+    if bs == l + 1 then
+      return bs
+    end
+    while true do
+      local b = byte(str, bs)
+      if 0 < b and b < 127
+      or 194 < b and b < 244 then
+        return bs
+      end
+      bs = bs - 1
+      if bs < 1 then
+        return
+      end
+    end
+  elseif n < 0 then
+    bs = bs - 1
+    repeat
+      if bs < 1 then
+        return
+      end
+
+      local b = byte(str, bs)
+      if 0 < b and b < 127
+      or 194 < b and b < 244 then
+        n = n + 1
+      end
+      bs = bs - 1
+    until n == 0
+    return bs + 1
+  else
+    while true do
+      if bs > l then
+        return
+      end
+
+      local b = byte(str, bs)
+      if 0 < b and b < 127
+      or 194 < b and b < 244 then
+        n = n - 1
+        for i = 1, n do
+          if bs > l then
+            return
+          end
+          bs = utf8next(str, bs)
+        end
+        return bs
+      end
+      bs = bs - 1
+    end
+  end
+
+end
+
+utf8.len       = utf8len
+utf8.sub       = utf8sub
+utf8.reverse   = utf8reverse
+utf8.char      = utf8char
+utf8.unicode   = utf8unicode
+utf8.byte      = utf8byte
+utf8.next      = utf8next
+utf8.gensub    = utf8gensub
+utf8.validator = utf8validator
+utf8.validate  = utf8validate
+utf8.dump      = dump
+utf8.format    = format
+utf8.lower     = lower
+utf8.upper     = upper
+utf8.rep       = rep
+utf8.raw = {}
+for k,v in pairs(string) do
+  utf8.raw[k] = v
+end
+
+utf8.charpattern = '[\0-\127\194-\244][\128-\191]*'
+utf8.offset = utf8offset
+local ok, utf8_53 = pcall(require, "utf8")
+if ok then
+  utf8.codes = utf8_53.codes
+  utf8.codepoint = utf8_53.codepoint
+  utf8.len53 = utf8_53.len
+else
+  utf8.codes = utf8codes
+  utf8.codepoint = utf8unicode
+end
+
+return utf8
+
+end
diff --git a/loveframes/third-party/utf8/primitives/init.lua b/loveframes/third-party/utf8/primitives/init.lua
new file mode 100644
index 0000000..df28ef3
--- /dev/null
+++ b/loveframes/third-party/utf8/primitives/init.lua
@@ -0,0 +1,23 @@
+return function(utf8)
+
+local provided = utf8.config.primitives
+
+if provided then
+  if type(provided) == "table" then
+    return provided
+  elseif type(provided) == "function" then
+    return provided(utf8)
+  else
+    return utf8:require(provided)
+  end
+end
+
+if pcall(require, "tarantool") then
+  return utf8:require "primitives.tarantool"
+elseif pcall(require, "ffi") then
+  return utf8:require "primitives.native"
+else
+  return utf8:require "primitives.dummy"
+end
+
+end
diff --git a/loveframes/third-party/utf8/primitives/native.lua b/loveframes/third-party/utf8/primitives/native.lua
new file mode 100644
index 0000000..3653839
--- /dev/null
+++ b/loveframes/third-party/utf8/primitives/native.lua
@@ -0,0 +1,46 @@
+return function(utf8)
+
+os.setlocale(utf8.config.locale, "ctype")
+
+local ffi = require("ffi")
+ffi.cdef[[
+  int towupper(int c);
+  int towlower(int c);
+]]
+
+utf8:require "primitives.dummy"
+
+function utf8.lower(str)
+  local bs = 1
+  local nbs
+  local bytes = utf8.raw.len(str)
+  local res = {}
+
+  while bs <= bytes do
+    nbs = utf8.next(str, bs)
+    local cp = utf8.unicode(str, bs, nbs)
+    res[#res + 1] = ffi.C.towlower(cp)
+    bs = nbs
+  end
+
+  return utf8.char(utf8.config.unpack(res))
+end
+
+function utf8.upper(str)
+  local bs = 1
+  local nbs
+  local bytes = utf8.raw.len(str)
+  local res = {}
+
+  while bs <= bytes do
+    nbs = utf8.next(str, bs)
+    local cp = utf8.unicode(str, bs, nbs)
+    res[#res + 1] = ffi.C.towupper(cp)
+    bs = nbs
+  end
+
+  return utf8.char(utf8.config.unpack(res))
+end
+
+return utf8
+end
diff --git a/loveframes/third-party/utf8/primitives/tarantool.lua b/loveframes/third-party/utf8/primitives/tarantool.lua
new file mode 100644
index 0000000..c38acf6
--- /dev/null
+++ b/loveframes/third-party/utf8/primitives/tarantool.lua
@@ -0,0 +1,13 @@
+return function(utf8)
+
+utf8:require "primitives.dummy"
+
+local tnt_utf8 = utf8.config.tarantool_utf8 or require("utf8")
+
+utf8.lower = tnt_utf8.lower
+utf8.upper = tnt_utf8.upper
+utf8.len = tnt_utf8.len
+utf8.char = tnt_utf8.char
+
+return utf8
+end
diff --git a/loveframes/third-party/utf8/regex_parser.lua b/loveframes/third-party/utf8/regex_parser.lua
new file mode 100644
index 0000000..0d3b4fe
--- /dev/null
+++ b/loveframes/third-party/utf8/regex_parser.lua
@@ -0,0 +1,78 @@
+return function(utf8)
+
+utf8:require "modifier.compiletime.parser"
+utf8:require "charclass.compiletime.parser"
+utf8:require "begins.compiletime.parser"
+utf8:require "ends.compiletime.parser"
+
+local gensub = utf8.gensub
+local sub = utf8.sub
+
+local parser_context = utf8:require "context.compiletime"
+
+return function(regex, plain)
+  utf8.debug("regex", regex)
+  local ctx = parser_context:new()
+
+  local skip = {0}
+  for nbs, c, bs in gensub(regex, 0), skip do
+    repeat -- continue
+      skip[1] = 0
+
+      c = utf8.raw.sub(regex, bs, utf8.next(regex, bs) - 1)
+
+      local functions, move = utf8.regex.compiletime.begins.parse(regex, c, bs, ctx)
+      if functions then
+        ctx.begins = functions
+        skip[1] = move
+      end
+      if skip[1] ~= 0 then break end
+
+      local functions, move = utf8.regex.compiletime.ends.parse(regex, c, bs, ctx)
+      if functions then
+        ctx.ends = functions
+        skip[1] = move
+      end
+      if skip[1] ~= 0 then break end
+
+      local functions, move = utf8.regex.compiletime.modifier.parse(regex, c, bs, ctx)
+      if functions then
+        for _, f in ipairs(functions) do
+          ctx.funcs[#ctx.funcs + 1] = f
+        end
+        skip[1] = move
+      end
+      if skip[1] ~= 0 then break end
+
+      local charclass, move = utf8.regex.compiletime.charclass.parse(regex, c, bs, ctx)
+      if charclass then skip[1] = move end
+    until true -- continue
+  end
+
+  for _, m in ipairs(utf8.config.modifier) do
+    if m.check then m.check(ctx) end
+  end
+
+  local src = [[
+  return function(str, init, utf8)
+      local ctx = utf8:require("context.runtime").new({str = str, pos = init or 1})
+      local cl = utf8:require("charclass.runtime.init")
+      local utf8sub = utf8.sub
+      local utf8len = utf8.len
+      local debug = utf8.debug
+      local function add(fun)
+          ctx.functions[#ctx.functions + 1] = fun
+      end
+  ]] .. ctx.begins
+  for _, v in ipairs(ctx.funcs) do src = src .. v end
+  src = src .. ctx.ends .. [[
+      return coroutine.wrap(ctx:get_function())(ctx)
+  end
+  ]]
+
+  utf8.debug(regex, src)
+
+  return assert(utf8.config.loadstring(src, (plain and "plain " or "") .. regex))()
+end
+
+end
diff --git a/loveframes/third-party/utf8/util.lua b/loveframes/third-party/utf8/util.lua
new file mode 100644
index 0000000..7723626
--- /dev/null
+++ b/loveframes/third-party/utf8/util.lua
@@ -0,0 +1,64 @@
+return function(utf8)
+
+function utf8.util.copy(obj, deep)
+  if type(obj) == 'table' then
+    local result = {}
+    if deep then
+      for k,v in pairs(obj) do
+        result[k] = utf8.util.copy(v, true)
+      end
+    else
+      for k,v in pairs(obj) do
+        result[k] = v
+      end
+    end
+    return result
+  else
+    return obj
+  end
+end
+
+local function dump(val, tab)
+  tab = tab or ''
+
+  if type(val) == 'table' then
+    utf8.config.logger('{\n')
+    for k,v in pairs(val) do
+      utf8.config.logger(tab .. tostring(k) .. " = ")
+      dump(v, tab .. '\t')
+      utf8.config.logger("\n")
+    end
+    utf8.config.logger(tab .. '}\n')
+  else
+    utf8.config.logger(tostring(val))
+  end
+end
+
+function utf8.util.debug(...)
+  local t = {...}
+  for _, v in ipairs(t) do
+    if type(v) == "table" and not (getmetatable(v) or {}).__tostring then
+      dump(v, '\t')
+    else
+      utf8.config.logger(tostring(v), " ")
+    end
+  end
+
+  utf8.config.logger('\n')
+end
+
+function utf8.debug(...)
+  if utf8.config.debug then
+    utf8.config.debug(...)
+  end
+end
+
+function utf8.util.next(str, bs)
+  local nbs1 = utf8.next(str, bs)
+  local nbs2 = utf8.next(str, nbs1)
+  return utf8.raw.sub(str, nbs1, nbs2 - 1), nbs1
+end
+
+return utf8.util
+
+end