From c90f36e3dd179d2de96f4f5fe38d8dc9a9de6dfe Mon Sep 17 00:00:00 2001 From: Emile Date: Fri, 25 Oct 2024 15:55:50 +0200 Subject: vendor --- vendor/golang.org/x/net/html/atom/atom.go | 78 + vendor/golang.org/x/net/html/atom/table.go | 783 +++++++++ vendor/golang.org/x/net/html/const.go | 111 ++ vendor/golang.org/x/net/html/doc.go | 127 ++ vendor/golang.org/x/net/html/doctype.go | 156 ++ vendor/golang.org/x/net/html/entity.go | 2253 +++++++++++++++++++++++++ vendor/golang.org/x/net/html/escape.go | 339 ++++ vendor/golang.org/x/net/html/foreign.go | 222 +++ vendor/golang.org/x/net/html/node.go | 225 +++ vendor/golang.org/x/net/html/parse.go | 2460 ++++++++++++++++++++++++++++ vendor/golang.org/x/net/html/render.go | 293 ++++ vendor/golang.org/x/net/html/token.go | 1272 ++++++++++++++ 12 files changed, 8319 insertions(+) create mode 100644 vendor/golang.org/x/net/html/atom/atom.go create mode 100644 vendor/golang.org/x/net/html/atom/table.go create mode 100644 vendor/golang.org/x/net/html/const.go create mode 100644 vendor/golang.org/x/net/html/doc.go create mode 100644 vendor/golang.org/x/net/html/doctype.go create mode 100644 vendor/golang.org/x/net/html/entity.go create mode 100644 vendor/golang.org/x/net/html/escape.go create mode 100644 vendor/golang.org/x/net/html/foreign.go create mode 100644 vendor/golang.org/x/net/html/node.go create mode 100644 vendor/golang.org/x/net/html/parse.go create mode 100644 vendor/golang.org/x/net/html/render.go create mode 100644 vendor/golang.org/x/net/html/token.go (limited to 'vendor/golang.org/x/net/html') diff --git a/vendor/golang.org/x/net/html/atom/atom.go b/vendor/golang.org/x/net/html/atom/atom.go new file mode 100644 index 0000000..cd0a8ac --- /dev/null +++ b/vendor/golang.org/x/net/html/atom/atom.go @@ -0,0 +1,78 @@ +// Copyright 2012 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Package atom provides integer codes (also known as atoms) for a fixed set of +// frequently occurring HTML strings: tag names and attribute keys such as "p" +// and "id". +// +// Sharing an atom's name between all elements with the same tag can result in +// fewer string allocations when tokenizing and parsing HTML. Integer +// comparisons are also generally faster than string comparisons. +// +// The value of an atom's particular code is not guaranteed to stay the same +// between versions of this package. Neither is any ordering guaranteed: +// whether atom.H1 < atom.H2 may also change. The codes are not guaranteed to +// be dense. The only guarantees are that e.g. looking up "div" will yield +// atom.Div, calling atom.Div.String will return "div", and atom.Div != 0. +package atom // import "golang.org/x/net/html/atom" + +// Atom is an integer code for a string. The zero value maps to "". +type Atom uint32 + +// String returns the atom's name. +func (a Atom) String() string { + start := uint32(a >> 8) + n := uint32(a & 0xff) + if start+n > uint32(len(atomText)) { + return "" + } + return atomText[start : start+n] +} + +func (a Atom) string() string { + return atomText[a>>8 : a>>8+a&0xff] +} + +// fnv computes the FNV hash with an arbitrary starting value h. +func fnv(h uint32, s []byte) uint32 { + for i := range s { + h ^= uint32(s[i]) + h *= 16777619 + } + return h +} + +func match(s string, t []byte) bool { + for i, c := range t { + if s[i] != c { + return false + } + } + return true +} + +// Lookup returns the atom whose name is s. It returns zero if there is no +// such atom. The lookup is case sensitive. +func Lookup(s []byte) Atom { + if len(s) == 0 || len(s) > maxAtomLen { + return 0 + } + h := fnv(hash0, s) + if a := table[h&uint32(len(table)-1)]; int(a&0xff) == len(s) && match(a.string(), s) { + return a + } + if a := table[(h>>16)&uint32(len(table)-1)]; int(a&0xff) == len(s) && match(a.string(), s) { + return a + } + return 0 +} + +// String returns a string whose contents are equal to s. In that sense, it is +// equivalent to string(s) but may be more efficient. +func String(s []byte) string { + if a := Lookup(s); a != 0 { + return a.String() + } + return string(s) +} diff --git a/vendor/golang.org/x/net/html/atom/table.go b/vendor/golang.org/x/net/html/atom/table.go new file mode 100644 index 0000000..2a93886 --- /dev/null +++ b/vendor/golang.org/x/net/html/atom/table.go @@ -0,0 +1,783 @@ +// Code generated by go generate gen.go; DO NOT EDIT. + +//go:generate go run gen.go + +package atom + +const ( + A Atom = 0x1 + Abbr Atom = 0x4 + Accept Atom = 0x1a06 + AcceptCharset Atom = 0x1a0e + Accesskey Atom = 0x2c09 + Acronym Atom = 0xaa07 + Action Atom = 0x27206 + Address Atom = 0x6f307 + Align Atom = 0xb105 + Allowfullscreen Atom = 0x2080f + Allowpaymentrequest Atom = 0xc113 + Allowusermedia Atom = 0xdd0e + Alt Atom = 0xf303 + Annotation Atom = 0x1c90a + AnnotationXml Atom = 0x1c90e + Applet Atom = 0x31906 + Area Atom = 0x35604 + Article Atom = 0x3fc07 + As Atom = 0x3c02 + Aside Atom = 0x10705 + Async Atom = 0xff05 + Audio Atom = 0x11505 + Autocomplete Atom = 0x2780c + Autofocus Atom = 0x12109 + Autoplay Atom = 0x13c08 + B Atom = 0x101 + Base Atom = 0x3b04 + Basefont Atom = 0x3b08 + Bdi Atom = 0xba03 + Bdo Atom = 0x14b03 + Bgsound Atom = 0x15e07 + Big Atom = 0x17003 + Blink Atom = 0x17305 + Blockquote Atom = 0x1870a + Body Atom = 0x2804 + Br Atom = 0x202 + Button Atom = 0x19106 + Canvas Atom = 0x10306 + Caption Atom = 0x23107 + Center Atom = 0x22006 + Challenge Atom = 0x29b09 + Charset Atom = 0x2107 + Checked Atom = 0x47907 + Cite Atom = 0x19c04 + Class Atom = 0x56405 + Code Atom = 0x5c504 + Col Atom = 0x1ab03 + Colgroup Atom = 0x1ab08 + Color Atom = 0x1bf05 + Cols Atom = 0x1c404 + Colspan Atom = 0x1c407 + Command Atom = 0x1d707 + Content Atom = 0x58b07 + Contenteditable Atom = 0x58b0f + Contextmenu Atom = 0x3800b + Controls Atom = 0x1de08 + Coords Atom = 0x1ea06 + Crossorigin Atom = 0x1fb0b + Data Atom = 0x4a504 + Datalist Atom = 0x4a508 + Datetime Atom = 0x2b808 + Dd Atom = 0x2d702 + Default Atom = 0x10a07 + Defer Atom = 0x5c705 + Del Atom = 0x45203 + Desc Atom = 0x56104 + Details Atom = 0x7207 + Dfn Atom = 0x8703 + Dialog Atom = 0xbb06 + Dir Atom = 0x9303 + Dirname Atom = 0x9307 + Disabled Atom = 0x16408 + Div Atom = 0x16b03 + Dl Atom = 0x5e602 + Download Atom = 0x46308 + Draggable Atom = 0x17a09 + Dropzone Atom = 0x40508 + Dt Atom = 0x64b02 + Em Atom = 0x6e02 + Embed Atom = 0x6e05 + Enctype Atom = 0x28d07 + Face Atom = 0x21e04 + Fieldset Atom = 0x22608 + Figcaption Atom = 0x22e0a + Figure Atom = 0x24806 + Font Atom = 0x3f04 + Footer Atom = 0xf606 + For Atom = 0x25403 + ForeignObject Atom = 0x2540d + Foreignobject Atom = 0x2610d + Form Atom = 0x26e04 + Formaction Atom = 0x26e0a + Formenctype Atom = 0x2890b + Formmethod Atom = 0x2a40a + Formnovalidate Atom = 0x2ae0e + Formtarget Atom = 0x2c00a + Frame Atom = 0x8b05 + Frameset Atom = 0x8b08 + H1 Atom = 0x15c02 + H2 Atom = 0x2de02 + H3 Atom = 0x30d02 + H4 Atom = 0x34502 + H5 Atom = 0x34f02 + H6 Atom = 0x64d02 + Head Atom = 0x33104 + Header Atom = 0x33106 + Headers Atom = 0x33107 + Height Atom = 0x5206 + Hgroup Atom = 0x2ca06 + Hidden Atom = 0x2d506 + High Atom = 0x2db04 + Hr Atom = 0x15702 + Href Atom = 0x2e004 + Hreflang Atom = 0x2e008 + Html Atom = 0x5604 + HttpEquiv Atom = 0x2e80a + I Atom = 0x601 + Icon Atom = 0x58a04 + Id Atom = 0x10902 + Iframe Atom = 0x2fc06 + Image Atom = 0x30205 + Img Atom = 0x30703 + Input Atom = 0x44b05 + Inputmode Atom = 0x44b09 + Ins Atom = 0x20403 + Integrity Atom = 0x23f09 + Is Atom = 0x16502 + Isindex Atom = 0x30f07 + Ismap Atom = 0x31605 + Itemid Atom = 0x38b06 + Itemprop Atom = 0x19d08 + Itemref Atom = 0x3cd07 + Itemscope Atom = 0x67109 + Itemtype Atom = 0x31f08 + Kbd Atom = 0xb903 + Keygen Atom = 0x3206 + Keytype Atom = 0xd607 + Kind Atom = 0x17704 + Label Atom = 0x5905 + Lang Atom = 0x2e404 + Legend Atom = 0x18106 + Li Atom = 0xb202 + Link Atom = 0x17404 + List Atom = 0x4a904 + Listing Atom = 0x4a907 + Loop Atom = 0x5d04 + Low Atom = 0xc303 + Main Atom = 0x1004 + Malignmark Atom = 0xb00a + Manifest Atom = 0x6d708 + Map Atom = 0x31803 + Mark Atom = 0xb604 + Marquee Atom = 0x32707 + Math Atom = 0x32e04 + Max Atom = 0x33d03 + Maxlength Atom = 0x33d09 + Media Atom = 0xe605 + Mediagroup Atom = 0xe60a + Menu Atom = 0x38704 + Menuitem Atom = 0x38708 + Meta Atom = 0x4b804 + Meter Atom = 0x9805 + Method Atom = 0x2a806 + Mglyph Atom = 0x30806 + Mi Atom = 0x34702 + Min Atom = 0x34703 + Minlength Atom = 0x34709 + Mn Atom = 0x2b102 + Mo Atom = 0xa402 + Ms Atom = 0x67402 + Mtext Atom = 0x35105 + Multiple Atom = 0x35f08 + Muted Atom = 0x36705 + Name Atom = 0x9604 + Nav Atom = 0x1303 + Nobr Atom = 0x3704 + Noembed Atom = 0x6c07 + Noframes Atom = 0x8908 + Nomodule Atom = 0xa208 + Nonce Atom = 0x1a605 + Noscript Atom = 0x21608 + Novalidate Atom = 0x2b20a + Object Atom = 0x26806 + Ol Atom = 0x13702 + Onabort Atom = 0x19507 + Onafterprint Atom = 0x2360c + Onautocomplete Atom = 0x2760e + Onautocompleteerror Atom = 0x27613 + Onauxclick Atom = 0x61f0a + Onbeforeprint Atom = 0x69e0d + Onbeforeunload Atom = 0x6e70e + Onblur Atom = 0x56d06 + Oncancel Atom = 0x11908 + Oncanplay Atom = 0x14d09 + Oncanplaythrough Atom = 0x14d10 + Onchange Atom = 0x41b08 + Onclick Atom = 0x2f507 + Onclose Atom = 0x36c07 + Oncontextmenu Atom = 0x37e0d + Oncopy Atom = 0x39106 + Oncuechange Atom = 0x3970b + Oncut Atom = 0x3a205 + Ondblclick Atom = 0x3a70a + Ondrag Atom = 0x3b106 + Ondragend Atom = 0x3b109 + Ondragenter Atom = 0x3ba0b + Ondragexit Atom = 0x3c50a + Ondragleave Atom = 0x3df0b + Ondragover Atom = 0x3ea0a + Ondragstart Atom = 0x3f40b + Ondrop Atom = 0x40306 + Ondurationchange Atom = 0x41310 + Onemptied Atom = 0x40a09 + Onended Atom = 0x42307 + Onerror Atom = 0x42a07 + Onfocus Atom = 0x43107 + Onhashchange Atom = 0x43d0c + Oninput Atom = 0x44907 + Oninvalid Atom = 0x45509 + Onkeydown Atom = 0x45e09 + Onkeypress Atom = 0x46b0a + Onkeyup Atom = 0x48007 + Onlanguagechange Atom = 0x48d10 + Onload Atom = 0x49d06 + Onloadeddata Atom = 0x49d0c + Onloadedmetadata Atom = 0x4b010 + Onloadend Atom = 0x4c609 + Onloadstart Atom = 0x4cf0b + Onmessage Atom = 0x4da09 + Onmessageerror Atom = 0x4da0e + Onmousedown Atom = 0x4e80b + Onmouseenter Atom = 0x4f30c + Onmouseleave Atom = 0x4ff0c + Onmousemove Atom = 0x50b0b + Onmouseout Atom = 0x5160a + Onmouseover Atom = 0x5230b + Onmouseup Atom = 0x52e09 + Onmousewheel Atom = 0x53c0c + Onoffline Atom = 0x54809 + Ononline Atom = 0x55108 + Onpagehide Atom = 0x5590a + Onpageshow Atom = 0x5730a + Onpaste Atom = 0x57f07 + Onpause Atom = 0x59a07 + Onplay Atom = 0x5a406 + Onplaying Atom = 0x5a409 + Onpopstate Atom = 0x5ad0a + Onprogress Atom = 0x5b70a + Onratechange Atom = 0x5cc0c + Onrejectionhandled Atom = 0x5d812 + Onreset Atom = 0x5ea07 + Onresize Atom = 0x5f108 + Onscroll Atom = 0x60008 + Onsecuritypolicyviolation Atom = 0x60819 + Onseeked Atom = 0x62908 + Onseeking Atom = 0x63109 + Onselect Atom = 0x63a08 + Onshow Atom = 0x64406 + Onsort Atom = 0x64f06 + Onstalled Atom = 0x65909 + Onstorage Atom = 0x66209 + Onsubmit Atom = 0x66b08 + Onsuspend Atom = 0x67b09 + Ontimeupdate Atom = 0x400c + Ontoggle Atom = 0x68408 + Onunhandledrejection Atom = 0x68c14 + Onunload Atom = 0x6ab08 + Onvolumechange Atom = 0x6b30e + Onwaiting Atom = 0x6c109 + Onwheel Atom = 0x6ca07 + Open Atom = 0x1a304 + Optgroup Atom = 0x5f08 + Optimum Atom = 0x6d107 + Option Atom = 0x6e306 + Output Atom = 0x51d06 + P Atom = 0xc01 + Param Atom = 0xc05 + Pattern Atom = 0x6607 + Picture Atom = 0x7b07 + Ping Atom = 0xef04 + Placeholder Atom = 0x1310b + Plaintext Atom = 0x1b209 + Playsinline Atom = 0x1400b + Poster Atom = 0x2cf06 + Pre Atom = 0x47003 + Preload Atom = 0x48607 + Progress Atom = 0x5b908 + Prompt Atom = 0x53606 + Public Atom = 0x58606 + Q Atom = 0xcf01 + Radiogroup Atom = 0x30a + Rb Atom = 0x3a02 + Readonly Atom = 0x35708 + Referrerpolicy Atom = 0x3d10e + Rel Atom = 0x48703 + Required Atom = 0x24c08 + Reversed Atom = 0x8008 + Rows Atom = 0x9c04 + Rowspan Atom = 0x9c07 + Rp Atom = 0x23c02 + Rt Atom = 0x19a02 + Rtc Atom = 0x19a03 + Ruby Atom = 0xfb04 + S Atom = 0x2501 + Samp Atom = 0x7804 + Sandbox Atom = 0x12907 + Scope Atom = 0x67505 + Scoped Atom = 0x67506 + Script Atom = 0x21806 + Seamless Atom = 0x37108 + Section Atom = 0x56807 + Select Atom = 0x63c06 + Selected Atom = 0x63c08 + Shape Atom = 0x1e505 + Size Atom = 0x5f504 + Sizes Atom = 0x5f505 + Slot Atom = 0x1ef04 + Small Atom = 0x20605 + Sortable Atom = 0x65108 + Sorted Atom = 0x33706 + Source Atom = 0x37806 + Spacer Atom = 0x43706 + Span Atom = 0x9f04 + Spellcheck Atom = 0x4740a + Src Atom = 0x5c003 + Srcdoc Atom = 0x5c006 + Srclang Atom = 0x5f907 + Srcset Atom = 0x6f906 + Start Atom = 0x3fa05 + Step Atom = 0x58304 + Strike Atom = 0xd206 + Strong Atom = 0x6dd06 + Style Atom = 0x6ff05 + Sub Atom = 0x66d03 + Summary Atom = 0x70407 + Sup Atom = 0x70b03 + Svg Atom = 0x70e03 + System Atom = 0x71106 + Tabindex Atom = 0x4be08 + Table Atom = 0x59505 + Target Atom = 0x2c406 + Tbody Atom = 0x2705 + Td Atom = 0x9202 + Template Atom = 0x71408 + Textarea Atom = 0x35208 + Tfoot Atom = 0xf505 + Th Atom = 0x15602 + Thead Atom = 0x33005 + Time Atom = 0x4204 + Title Atom = 0x11005 + Tr Atom = 0xcc02 + Track Atom = 0x1ba05 + Translate Atom = 0x1f209 + Tt Atom = 0x6802 + Type Atom = 0xd904 + Typemustmatch Atom = 0x2900d + U Atom = 0xb01 + Ul Atom = 0xa702 + Updateviacache Atom = 0x460e + Usemap Atom = 0x59e06 + Value Atom = 0x1505 + Var Atom = 0x16d03 + Video Atom = 0x2f105 + Wbr Atom = 0x57c03 + Width Atom = 0x64905 + Workertype Atom = 0x71c0a + Wrap Atom = 0x72604 + Xmp Atom = 0x12f03 +) + +const hash0 = 0x81cdf10e + +const maxAtomLen = 25 + +var table = [1 << 9]Atom{ + 0x1: 0xe60a, // mediagroup + 0x2: 0x2e404, // lang + 0x4: 0x2c09, // accesskey + 0x5: 0x8b08, // frameset + 0x7: 0x63a08, // onselect + 0x8: 0x71106, // system + 0xa: 0x64905, // width + 0xc: 0x2890b, // formenctype + 0xd: 0x13702, // ol + 0xe: 0x3970b, // oncuechange + 0x10: 0x14b03, // bdo + 0x11: 0x11505, // audio + 0x12: 0x17a09, // draggable + 0x14: 0x2f105, // video + 0x15: 0x2b102, // mn + 0x16: 0x38704, // menu + 0x17: 0x2cf06, // poster + 0x19: 0xf606, // footer + 0x1a: 0x2a806, // method + 0x1b: 0x2b808, // datetime + 0x1c: 0x19507, // onabort + 0x1d: 0x460e, // updateviacache + 0x1e: 0xff05, // async + 0x1f: 0x49d06, // onload + 0x21: 0x11908, // oncancel + 0x22: 0x62908, // onseeked + 0x23: 0x30205, // image + 0x24: 0x5d812, // onrejectionhandled + 0x26: 0x17404, // link + 0x27: 0x51d06, // output + 0x28: 0x33104, // head + 0x29: 0x4ff0c, // onmouseleave + 0x2a: 0x57f07, // onpaste + 0x2b: 0x5a409, // onplaying + 0x2c: 0x1c407, // colspan + 0x2f: 0x1bf05, // color + 0x30: 0x5f504, // size + 0x31: 0x2e80a, // http-equiv + 0x33: 0x601, // i + 0x34: 0x5590a, // onpagehide + 0x35: 0x68c14, // onunhandledrejection + 0x37: 0x42a07, // onerror + 0x3a: 0x3b08, // basefont + 0x3f: 0x1303, // nav + 0x40: 0x17704, // kind + 0x41: 0x35708, // readonly + 0x42: 0x30806, // mglyph + 0x44: 0xb202, // li + 0x46: 0x2d506, // hidden + 0x47: 0x70e03, // svg + 0x48: 0x58304, // step + 0x49: 0x23f09, // integrity + 0x4a: 0x58606, // public + 0x4c: 0x1ab03, // col + 0x4d: 0x1870a, // blockquote + 0x4e: 0x34f02, // h5 + 0x50: 0x5b908, // progress + 0x51: 0x5f505, // sizes + 0x52: 0x34502, // h4 + 0x56: 0x33005, // thead + 0x57: 0xd607, // keytype + 0x58: 0x5b70a, // onprogress + 0x59: 0x44b09, // inputmode + 0x5a: 0x3b109, // ondragend + 0x5d: 0x3a205, // oncut + 0x5e: 0x43706, // spacer + 0x5f: 0x1ab08, // colgroup + 0x62: 0x16502, // is + 0x65: 0x3c02, // as + 0x66: 0x54809, // onoffline + 0x67: 0x33706, // sorted + 0x69: 0x48d10, // onlanguagechange + 0x6c: 0x43d0c, // onhashchange + 0x6d: 0x9604, // name + 0x6e: 0xf505, // tfoot + 0x6f: 0x56104, // desc + 0x70: 0x33d03, // max + 0x72: 0x1ea06, // coords + 0x73: 0x30d02, // h3 + 0x74: 0x6e70e, // onbeforeunload + 0x75: 0x9c04, // rows + 0x76: 0x63c06, // select + 0x77: 0x9805, // meter + 0x78: 0x38b06, // itemid + 0x79: 0x53c0c, // onmousewheel + 0x7a: 0x5c006, // srcdoc + 0x7d: 0x1ba05, // track + 0x7f: 0x31f08, // itemtype + 0x82: 0xa402, // mo + 0x83: 0x41b08, // onchange + 0x84: 0x33107, // headers + 0x85: 0x5cc0c, // onratechange + 0x86: 0x60819, // onsecuritypolicyviolation + 0x88: 0x4a508, // datalist + 0x89: 0x4e80b, // onmousedown + 0x8a: 0x1ef04, // slot + 0x8b: 0x4b010, // onloadedmetadata + 0x8c: 0x1a06, // accept + 0x8d: 0x26806, // object + 0x91: 0x6b30e, // onvolumechange + 0x92: 0x2107, // charset + 0x93: 0x27613, // onautocompleteerror + 0x94: 0xc113, // allowpaymentrequest + 0x95: 0x2804, // body + 0x96: 0x10a07, // default + 0x97: 0x63c08, // selected + 0x98: 0x21e04, // face + 0x99: 0x1e505, // shape + 0x9b: 0x68408, // ontoggle + 0x9e: 0x64b02, // dt + 0x9f: 0xb604, // mark + 0xa1: 0xb01, // u + 0xa4: 0x6ab08, // onunload + 0xa5: 0x5d04, // loop + 0xa6: 0x16408, // disabled + 0xaa: 0x42307, // onended + 0xab: 0xb00a, // malignmark + 0xad: 0x67b09, // onsuspend + 0xae: 0x35105, // mtext + 0xaf: 0x64f06, // onsort + 0xb0: 0x19d08, // itemprop + 0xb3: 0x67109, // itemscope + 0xb4: 0x17305, // blink + 0xb6: 0x3b106, // ondrag + 0xb7: 0xa702, // ul + 0xb8: 0x26e04, // form + 0xb9: 0x12907, // sandbox + 0xba: 0x8b05, // frame + 0xbb: 0x1505, // value + 0xbc: 0x66209, // onstorage + 0xbf: 0xaa07, // acronym + 0xc0: 0x19a02, // rt + 0xc2: 0x202, // br + 0xc3: 0x22608, // fieldset + 0xc4: 0x2900d, // typemustmatch + 0xc5: 0xa208, // nomodule + 0xc6: 0x6c07, // noembed + 0xc7: 0x69e0d, // onbeforeprint + 0xc8: 0x19106, // button + 0xc9: 0x2f507, // onclick + 0xca: 0x70407, // summary + 0xcd: 0xfb04, // ruby + 0xce: 0x56405, // class + 0xcf: 0x3f40b, // ondragstart + 0xd0: 0x23107, // caption + 0xd4: 0xdd0e, // allowusermedia + 0xd5: 0x4cf0b, // onloadstart + 0xd9: 0x16b03, // div + 0xda: 0x4a904, // list + 0xdb: 0x32e04, // math + 0xdc: 0x44b05, // input + 0xdf: 0x3ea0a, // ondragover + 0xe0: 0x2de02, // h2 + 0xe2: 0x1b209, // plaintext + 0xe4: 0x4f30c, // onmouseenter + 0xe7: 0x47907, // checked + 0xe8: 0x47003, // pre + 0xea: 0x35f08, // multiple + 0xeb: 0xba03, // bdi + 0xec: 0x33d09, // maxlength + 0xed: 0xcf01, // q + 0xee: 0x61f0a, // onauxclick + 0xf0: 0x57c03, // wbr + 0xf2: 0x3b04, // base + 0xf3: 0x6e306, // option + 0xf5: 0x41310, // ondurationchange + 0xf7: 0x8908, // noframes + 0xf9: 0x40508, // dropzone + 0xfb: 0x67505, // scope + 0xfc: 0x8008, // reversed + 0xfd: 0x3ba0b, // ondragenter + 0xfe: 0x3fa05, // start + 0xff: 0x12f03, // xmp + 0x100: 0x5f907, // srclang + 0x101: 0x30703, // img + 0x104: 0x101, // b + 0x105: 0x25403, // for + 0x106: 0x10705, // aside + 0x107: 0x44907, // oninput + 0x108: 0x35604, // area + 0x109: 0x2a40a, // formmethod + 0x10a: 0x72604, // wrap + 0x10c: 0x23c02, // rp + 0x10d: 0x46b0a, // onkeypress + 0x10e: 0x6802, // tt + 0x110: 0x34702, // mi + 0x111: 0x36705, // muted + 0x112: 0xf303, // alt + 0x113: 0x5c504, // code + 0x114: 0x6e02, // em + 0x115: 0x3c50a, // ondragexit + 0x117: 0x9f04, // span + 0x119: 0x6d708, // manifest + 0x11a: 0x38708, // menuitem + 0x11b: 0x58b07, // content + 0x11d: 0x6c109, // onwaiting + 0x11f: 0x4c609, // onloadend + 0x121: 0x37e0d, // oncontextmenu + 0x123: 0x56d06, // onblur + 0x124: 0x3fc07, // article + 0x125: 0x9303, // dir + 0x126: 0xef04, // ping + 0x127: 0x24c08, // required + 0x128: 0x45509, // oninvalid + 0x129: 0xb105, // align + 0x12b: 0x58a04, // icon + 0x12c: 0x64d02, // h6 + 0x12d: 0x1c404, // cols + 0x12e: 0x22e0a, // figcaption + 0x12f: 0x45e09, // onkeydown + 0x130: 0x66b08, // onsubmit + 0x131: 0x14d09, // oncanplay + 0x132: 0x70b03, // sup + 0x133: 0xc01, // p + 0x135: 0x40a09, // onemptied + 0x136: 0x39106, // oncopy + 0x137: 0x19c04, // cite + 0x138: 0x3a70a, // ondblclick + 0x13a: 0x50b0b, // onmousemove + 0x13c: 0x66d03, // sub + 0x13d: 0x48703, // rel + 0x13e: 0x5f08, // optgroup + 0x142: 0x9c07, // rowspan + 0x143: 0x37806, // source + 0x144: 0x21608, // noscript + 0x145: 0x1a304, // open + 0x146: 0x20403, // ins + 0x147: 0x2540d, // foreignObject + 0x148: 0x5ad0a, // onpopstate + 0x14a: 0x28d07, // enctype + 0x14b: 0x2760e, // onautocomplete + 0x14c: 0x35208, // textarea + 0x14e: 0x2780c, // autocomplete + 0x14f: 0x15702, // hr + 0x150: 0x1de08, // controls + 0x151: 0x10902, // id + 0x153: 0x2360c, // onafterprint + 0x155: 0x2610d, // foreignobject + 0x156: 0x32707, // marquee + 0x157: 0x59a07, // onpause + 0x158: 0x5e602, // dl + 0x159: 0x5206, // height + 0x15a: 0x34703, // min + 0x15b: 0x9307, // dirname + 0x15c: 0x1f209, // translate + 0x15d: 0x5604, // html + 0x15e: 0x34709, // minlength + 0x15f: 0x48607, // preload + 0x160: 0x71408, // template + 0x161: 0x3df0b, // ondragleave + 0x162: 0x3a02, // rb + 0x164: 0x5c003, // src + 0x165: 0x6dd06, // strong + 0x167: 0x7804, // samp + 0x168: 0x6f307, // address + 0x169: 0x55108, // ononline + 0x16b: 0x1310b, // placeholder + 0x16c: 0x2c406, // target + 0x16d: 0x20605, // small + 0x16e: 0x6ca07, // onwheel + 0x16f: 0x1c90a, // annotation + 0x170: 0x4740a, // spellcheck + 0x171: 0x7207, // details + 0x172: 0x10306, // canvas + 0x173: 0x12109, // autofocus + 0x174: 0xc05, // param + 0x176: 0x46308, // download + 0x177: 0x45203, // del + 0x178: 0x36c07, // onclose + 0x179: 0xb903, // kbd + 0x17a: 0x31906, // applet + 0x17b: 0x2e004, // href + 0x17c: 0x5f108, // onresize + 0x17e: 0x49d0c, // onloadeddata + 0x180: 0xcc02, // tr + 0x181: 0x2c00a, // formtarget + 0x182: 0x11005, // title + 0x183: 0x6ff05, // style + 0x184: 0xd206, // strike + 0x185: 0x59e06, // usemap + 0x186: 0x2fc06, // iframe + 0x187: 0x1004, // main + 0x189: 0x7b07, // picture + 0x18c: 0x31605, // ismap + 0x18e: 0x4a504, // data + 0x18f: 0x5905, // label + 0x191: 0x3d10e, // referrerpolicy + 0x192: 0x15602, // th + 0x194: 0x53606, // prompt + 0x195: 0x56807, // section + 0x197: 0x6d107, // optimum + 0x198: 0x2db04, // high + 0x199: 0x15c02, // h1 + 0x19a: 0x65909, // onstalled + 0x19b: 0x16d03, // var + 0x19c: 0x4204, // time + 0x19e: 0x67402, // ms + 0x19f: 0x33106, // header + 0x1a0: 0x4da09, // onmessage + 0x1a1: 0x1a605, // nonce + 0x1a2: 0x26e0a, // formaction + 0x1a3: 0x22006, // center + 0x1a4: 0x3704, // nobr + 0x1a5: 0x59505, // table + 0x1a6: 0x4a907, // listing + 0x1a7: 0x18106, // legend + 0x1a9: 0x29b09, // challenge + 0x1aa: 0x24806, // figure + 0x1ab: 0xe605, // media + 0x1ae: 0xd904, // type + 0x1af: 0x3f04, // font + 0x1b0: 0x4da0e, // onmessageerror + 0x1b1: 0x37108, // seamless + 0x1b2: 0x8703, // dfn + 0x1b3: 0x5c705, // defer + 0x1b4: 0xc303, // low + 0x1b5: 0x19a03, // rtc + 0x1b6: 0x5230b, // onmouseover + 0x1b7: 0x2b20a, // novalidate + 0x1b8: 0x71c0a, // workertype + 0x1ba: 0x3cd07, // itemref + 0x1bd: 0x1, // a + 0x1be: 0x31803, // map + 0x1bf: 0x400c, // ontimeupdate + 0x1c0: 0x15e07, // bgsound + 0x1c1: 0x3206, // keygen + 0x1c2: 0x2705, // tbody + 0x1c5: 0x64406, // onshow + 0x1c7: 0x2501, // s + 0x1c8: 0x6607, // pattern + 0x1cc: 0x14d10, // oncanplaythrough + 0x1ce: 0x2d702, // dd + 0x1cf: 0x6f906, // srcset + 0x1d0: 0x17003, // big + 0x1d2: 0x65108, // sortable + 0x1d3: 0x48007, // onkeyup + 0x1d5: 0x5a406, // onplay + 0x1d7: 0x4b804, // meta + 0x1d8: 0x40306, // ondrop + 0x1da: 0x60008, // onscroll + 0x1db: 0x1fb0b, // crossorigin + 0x1dc: 0x5730a, // onpageshow + 0x1dd: 0x4, // abbr + 0x1de: 0x9202, // td + 0x1df: 0x58b0f, // contenteditable + 0x1e0: 0x27206, // action + 0x1e1: 0x1400b, // playsinline + 0x1e2: 0x43107, // onfocus + 0x1e3: 0x2e008, // hreflang + 0x1e5: 0x5160a, // onmouseout + 0x1e6: 0x5ea07, // onreset + 0x1e7: 0x13c08, // autoplay + 0x1e8: 0x63109, // onseeking + 0x1ea: 0x67506, // scoped + 0x1ec: 0x30a, // radiogroup + 0x1ee: 0x3800b, // contextmenu + 0x1ef: 0x52e09, // onmouseup + 0x1f1: 0x2ca06, // hgroup + 0x1f2: 0x2080f, // allowfullscreen + 0x1f3: 0x4be08, // tabindex + 0x1f6: 0x30f07, // isindex + 0x1f7: 0x1a0e, // accept-charset + 0x1f8: 0x2ae0e, // formnovalidate + 0x1fb: 0x1c90e, // annotation-xml + 0x1fc: 0x6e05, // embed + 0x1fd: 0x21806, // script + 0x1fe: 0xbb06, // dialog + 0x1ff: 0x1d707, // command +} + +const atomText = "abbradiogrouparamainavalueaccept-charsetbodyaccesskeygenobrb" + + "asefontimeupdateviacacheightmlabelooptgroupatternoembedetail" + + "sampictureversedfnoframesetdirnameterowspanomoduleacronymali" + + "gnmarkbdialogallowpaymentrequestrikeytypeallowusermediagroup" + + "ingaltfooterubyasyncanvasidefaultitleaudioncancelautofocusan" + + "dboxmplaceholderautoplaysinlinebdoncanplaythrough1bgsoundisa" + + "bledivarbigblinkindraggablegendblockquotebuttonabortcitempro" + + "penoncecolgrouplaintextrackcolorcolspannotation-xmlcommandco" + + "ntrolshapecoordslotranslatecrossoriginsmallowfullscreenoscri" + + "ptfacenterfieldsetfigcaptionafterprintegrityfigurequiredfore" + + "ignObjectforeignobjectformactionautocompleteerrorformenctype" + + "mustmatchallengeformmethodformnovalidatetimeformtargethgroup" + + "osterhiddenhigh2hreflanghttp-equivideonclickiframeimageimgly" + + "ph3isindexismappletitemtypemarqueematheadersortedmaxlength4m" + + "inlength5mtextareadonlymultiplemutedoncloseamlessourceoncont" + + "extmenuitemidoncopyoncuechangeoncutondblclickondragendondrag" + + "enterondragexitemreferrerpolicyondragleaveondragoverondragst" + + "articleondropzonemptiedondurationchangeonendedonerroronfocus" + + "paceronhashchangeoninputmodeloninvalidonkeydownloadonkeypres" + + "spellcheckedonkeyupreloadonlanguagechangeonloadeddatalisting" + + "onloadedmetadatabindexonloadendonloadstartonmessageerroronmo" + + "usedownonmouseenteronmouseleaveonmousemoveonmouseoutputonmou" + + "seoveronmouseupromptonmousewheelonofflineononlineonpagehides" + + "classectionbluronpageshowbronpastepublicontenteditableonpaus" + + "emaponplayingonpopstateonprogressrcdocodeferonratechangeonre" + + "jectionhandledonresetonresizesrclangonscrollonsecuritypolicy" + + "violationauxclickonseekedonseekingonselectedonshowidth6onsor" + + "tableonstalledonstorageonsubmitemscopedonsuspendontoggleonun" + + "handledrejectionbeforeprintonunloadonvolumechangeonwaitingon" + + "wheeloptimumanifestrongoptionbeforeunloaddressrcsetstylesumm" + + "arysupsvgsystemplateworkertypewrap" diff --git a/vendor/golang.org/x/net/html/const.go b/vendor/golang.org/x/net/html/const.go new file mode 100644 index 0000000..ff7acf2 --- /dev/null +++ b/vendor/golang.org/x/net/html/const.go @@ -0,0 +1,111 @@ +// Copyright 2011 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package html + +// Section 12.2.4.2 of the HTML5 specification says "The following elements +// have varying levels of special parsing rules". +// https://html.spec.whatwg.org/multipage/syntax.html#the-stack-of-open-elements +var isSpecialElementMap = map[string]bool{ + "address": true, + "applet": true, + "area": true, + "article": true, + "aside": true, + "base": true, + "basefont": true, + "bgsound": true, + "blockquote": true, + "body": true, + "br": true, + "button": true, + "caption": true, + "center": true, + "col": true, + "colgroup": true, + "dd": true, + "details": true, + "dir": true, + "div": true, + "dl": true, + "dt": true, + "embed": true, + "fieldset": true, + "figcaption": true, + "figure": true, + "footer": true, + "form": true, + "frame": true, + "frameset": true, + "h1": true, + "h2": true, + "h3": true, + "h4": true, + "h5": true, + "h6": true, + "head": true, + "header": true, + "hgroup": true, + "hr": true, + "html": true, + "iframe": true, + "img": true, + "input": true, + "keygen": true, // "keygen" has been removed from the spec, but are kept here for backwards compatibility. + "li": true, + "link": true, + "listing": true, + "main": true, + "marquee": true, + "menu": true, + "meta": true, + "nav": true, + "noembed": true, + "noframes": true, + "noscript": true, + "object": true, + "ol": true, + "p": true, + "param": true, + "plaintext": true, + "pre": true, + "script": true, + "section": true, + "select": true, + "source": true, + "style": true, + "summary": true, + "table": true, + "tbody": true, + "td": true, + "template": true, + "textarea": true, + "tfoot": true, + "th": true, + "thead": true, + "title": true, + "tr": true, + "track": true, + "ul": true, + "wbr": true, + "xmp": true, +} + +func isSpecialElement(element *Node) bool { + switch element.Namespace { + case "", "html": + return isSpecialElementMap[element.Data] + case "math": + switch element.Data { + case "mi", "mo", "mn", "ms", "mtext", "annotation-xml": + return true + } + case "svg": + switch element.Data { + case "foreignObject", "desc", "title": + return true + } + } + return false +} diff --git a/vendor/golang.org/x/net/html/doc.go b/vendor/golang.org/x/net/html/doc.go new file mode 100644 index 0000000..3a7e5ab --- /dev/null +++ b/vendor/golang.org/x/net/html/doc.go @@ -0,0 +1,127 @@ +// Copyright 2010 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +/* +Package html implements an HTML5-compliant tokenizer and parser. + +Tokenization is done by creating a Tokenizer for an io.Reader r. It is the +caller's responsibility to ensure that r provides UTF-8 encoded HTML. + + z := html.NewTokenizer(r) + +Given a Tokenizer z, the HTML is tokenized by repeatedly calling z.Next(), +which parses the next token and returns its type, or an error: + + for { + tt := z.Next() + if tt == html.ErrorToken { + // ... + return ... + } + // Process the current token. + } + +There are two APIs for retrieving the current token. The high-level API is to +call Token; the low-level API is to call Text or TagName / TagAttr. Both APIs +allow optionally calling Raw after Next but before Token, Text, TagName, or +TagAttr. In EBNF notation, the valid call sequence per token is: + + Next {Raw} [ Token | Text | TagName {TagAttr} ] + +Token returns an independent data structure that completely describes a token. +Entities (such as "<") are unescaped, tag names and attribute keys are +lower-cased, and attributes are collected into a []Attribute. For example: + + for { + if z.Next() == html.ErrorToken { + // Returning io.EOF indicates success. + return z.Err() + } + emitToken(z.Token()) + } + +The low-level API performs fewer allocations and copies, but the contents of +the []byte values returned by Text, TagName and TagAttr may change on the next +call to Next. For example, to extract an HTML page's anchor text: + + depth := 0 + for { + tt := z.Next() + switch tt { + case html.ErrorToken: + return z.Err() + case html.TextToken: + if depth > 0 { + // emitBytes should copy the []byte it receives, + // if it doesn't process it immediately. + emitBytes(z.Text()) + } + case html.StartTagToken, html.EndTagToken: + tn, _ := z.TagName() + if len(tn) == 1 && tn[0] == 'a' { + if tt == html.StartTagToken { + depth++ + } else { + depth-- + } + } + } + } + +Parsing is done by calling Parse with an io.Reader, which returns the root of +the parse tree (the document element) as a *Node. It is the caller's +responsibility to ensure that the Reader provides UTF-8 encoded HTML. For +example, to process each anchor node in depth-first order: + + doc, err := html.Parse(r) + if err != nil { + // ... + } + var f func(*html.Node) + f = func(n *html.Node) { + if n.Type == html.ElementNode && n.Data == "a" { + // Do something with n... + } + for c := n.FirstChild; c != nil; c = c.NextSibling { + f(c) + } + } + f(doc) + +The relevant specifications include: +https://html.spec.whatwg.org/multipage/syntax.html and +https://html.spec.whatwg.org/multipage/syntax.html#tokenization + +# Security Considerations + +Care should be taken when parsing and interpreting HTML, whether full documents +or fragments, within the framework of the HTML specification, especially with +regard to untrusted inputs. + +This package provides both a tokenizer and a parser, which implement the +tokenization, and tokenization and tree construction stages of the WHATWG HTML +parsing specification respectively. While the tokenizer parses and normalizes +individual HTML tokens, only the parser constructs the DOM tree from the +tokenized HTML, as described in the tree construction stage of the +specification, dynamically modifying or extending the document's DOM tree. + +If your use case requires semantically well-formed HTML documents, as defined by +the WHATWG specification, the parser should be used rather than the tokenizer. + +In security contexts, if trust decisions are being made using the tokenized or +parsed content, the input must be re-serialized (for instance by using Render or +Token.String) in order for those trust decisions to hold, as the process of +tokenization or parsing may alter the content. +*/ +package html // import "golang.org/x/net/html" + +// The tokenization algorithm implemented by this package is not a line-by-line +// transliteration of the relatively verbose state-machine in the WHATWG +// specification. A more direct approach is used instead, where the program +// counter implies the state, such as whether it is tokenizing a tag or a text +// node. Specification compliance is verified by checking expected and actual +// outputs over a test suite rather than aiming for algorithmic fidelity. + +// TODO(nigeltao): Does a DOM API belong in this package or a separate one? +// TODO(nigeltao): How does parsing interact with a JavaScript engine? diff --git a/vendor/golang.org/x/net/html/doctype.go b/vendor/golang.org/x/net/html/doctype.go new file mode 100644 index 0000000..c484e5a --- /dev/null +++ b/vendor/golang.org/x/net/html/doctype.go @@ -0,0 +1,156 @@ +// Copyright 2011 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package html + +import ( + "strings" +) + +// parseDoctype parses the data from a DoctypeToken into a name, +// public identifier, and system identifier. It returns a Node whose Type +// is DoctypeNode, whose Data is the name, and which has attributes +// named "system" and "public" for the two identifiers if they were present. +// quirks is whether the document should be parsed in "quirks mode". +func parseDoctype(s string) (n *Node, quirks bool) { + n = &Node{Type: DoctypeNode} + + // Find the name. + space := strings.IndexAny(s, whitespace) + if space == -1 { + space = len(s) + } + n.Data = s[:space] + // The comparison to "html" is case-sensitive. + if n.Data != "html" { + quirks = true + } + n.Data = strings.ToLower(n.Data) + s = strings.TrimLeft(s[space:], whitespace) + + if len(s) < 6 { + // It can't start with "PUBLIC" or "SYSTEM". + // Ignore the rest of the string. + return n, quirks || s != "" + } + + key := strings.ToLower(s[:6]) + s = s[6:] + for key == "public" || key == "system" { + s = strings.TrimLeft(s, whitespace) + if s == "" { + break + } + quote := s[0] + if quote != '"' && quote != '\'' { + break + } + s = s[1:] + q := strings.IndexRune(s, rune(quote)) + var id string + if q == -1 { + id = s + s = "" + } else { + id = s[:q] + s = s[q+1:] + } + n.Attr = append(n.Attr, Attribute{Key: key, Val: id}) + if key == "public" { + key = "system" + } else { + key = "" + } + } + + if key != "" || s != "" { + quirks = true + } else if len(n.Attr) > 0 { + if n.Attr[0].Key == "public" { + public := strings.ToLower(n.Attr[0].Val) + switch public { + case "-//w3o//dtd w3 html strict 3.0//en//", "-/w3d/dtd html 4.0 transitional/en", "html": + quirks = true + default: + for _, q := range quirkyIDs { + if strings.HasPrefix(public, q) { + quirks = true + break + } + } + } + // The following two public IDs only cause quirks mode if there is no system ID. + if len(n.Attr) == 1 && (strings.HasPrefix(public, "-//w3c//dtd html 4.01 frameset//") || + strings.HasPrefix(public, "-//w3c//dtd html 4.01 transitional//")) { + quirks = true + } + } + if lastAttr := n.Attr[len(n.Attr)-1]; lastAttr.Key == "system" && + strings.ToLower(lastAttr.Val) == "http://www.ibm.com/data/dtd/v11/ibmxhtml1-transitional.dtd" { + quirks = true + } + } + + return n, quirks +} + +// quirkyIDs is a list of public doctype identifiers that cause a document +// to be interpreted in quirks mode. The identifiers should be in lower case. +var quirkyIDs = []string{ + "+//silmaril//dtd html pro v0r11 19970101//", + "-//advasoft ltd//dtd html 3.0 aswedit + extensions//", + "-//as//dtd html 3.0 aswedit + extensions//", + "-//ietf//dtd html 2.0 level 1//", + "-//ietf//dtd html 2.0 level 2//", + "-//ietf//dtd html 2.0 strict level 1//", + "-//ietf//dtd html 2.0 strict level 2//", + "-//ietf//dtd html 2.0 strict//", + "-//ietf//dtd html 2.0//", + "-//ietf//dtd html 2.1e//", + "-//ietf//dtd html 3.0//", + "-//ietf//dtd html 3.2 final//", + "-//ietf//dtd html 3.2//", + "-//ietf//dtd html 3//", + "-//ietf//dtd html level 0//", + "-//ietf//dtd html level 1//", + "-//ietf//dtd html level 2//", + "-//ietf//dtd html level 3//", + "-//ietf//dtd html strict level 0//", + "-//ietf//dtd html strict level 1//", + "-//ietf//dtd html strict level 2//", + "-//ietf//dtd html strict level 3//", + "-//ietf//dtd html strict//", + "-//ietf//dtd html//", + "-//metrius//dtd metrius presentational//", + "-//microsoft//dtd internet explorer 2.0 html strict//", + "-//microsoft//dtd internet explorer 2.0 html//", + "-//microsoft//dtd internet explorer 2.0 tables//", + "-//microsoft//dtd internet explorer 3.0 html strict//", + "-//microsoft//dtd internet explorer 3.0 html//", + "-//microsoft//dtd internet explorer 3.0 tables//", + "-//netscape comm. corp.//dtd html//", + "-//netscape comm. corp.//dtd strict html//", + "-//o'reilly and associates//dtd html 2.0//", + "-//o'reilly and associates//dtd html extended 1.0//", + "-//o'reilly and associates//dtd html extended relaxed 1.0//", + "-//softquad software//dtd hotmetal pro 6.0::19990601::extensions to html 4.0//", + "-//softquad//dtd hotmetal pro 4.0::19971010::extensions to html 4.0//", + "-//spyglass//dtd html 2.0 extended//", + "-//sq//dtd html 2.0 hotmetal + extensions//", + "-//sun microsystems corp.//dtd hotjava html//", + "-//sun microsystems corp.//dtd hotjava strict html//", + "-//w3c//dtd html 3 1995-03-24//", + "-//w3c//dtd html 3.2 draft//", + "-//w3c//dtd html 3.2 final//", + "-//w3c//dtd html 3.2//", + "-//w3c//dtd html 3.2s draft//", + "-//w3c//dtd html 4.0 frameset//", + "-//w3c//dtd html 4.0 transitional//", + "-//w3c//dtd html experimental 19960712//", + "-//w3c//dtd html experimental 970421//", + "-//w3c//dtd w3 html//", + "-//w3o//dtd w3 html 3.0//", + "-//webtechs//dtd mozilla html 2.0//", + "-//webtechs//dtd mozilla html//", +} diff --git a/vendor/golang.org/x/net/html/entity.go b/vendor/golang.org/x/net/html/entity.go new file mode 100644 index 0000000..b628880 --- /dev/null +++ b/vendor/golang.org/x/net/html/entity.go @@ -0,0 +1,2253 @@ +// Copyright 2010 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package html + +// All entities that do not end with ';' are 6 or fewer bytes long. +const longestEntityWithoutSemicolon = 6 + +// entity is a map from HTML entity names to their values. The semicolon matters: +// https://html.spec.whatwg.org/multipage/syntax.html#named-character-references +// lists both "amp" and "amp;" as two separate entries. +// +// Note that the HTML5 list is larger than the HTML4 list at +// http://www.w3.org/TR/html4/sgml/entities.html +var entity = map[string]rune{ + "AElig;": '\U000000C6', + "AMP;": '\U00000026', + "Aacute;": '\U000000C1', + "Abreve;": '\U00000102', + "Acirc;": '\U000000C2', + "Acy;": '\U00000410', + "Afr;": '\U0001D504', + "Agrave;": '\U000000C0', + "Alpha;": '\U00000391', + "Amacr;": '\U00000100', + "And;": '\U00002A53', + "Aogon;": '\U00000104', + "Aopf;": '\U0001D538', + "ApplyFunction;": '\U00002061', + "Aring;": '\U000000C5', + "Ascr;": '\U0001D49C', + "Assign;": '\U00002254', + "Atilde;": '\U000000C3', + "Auml;": '\U000000C4', + "Backslash;": '\U00002216', + "Barv;": '\U00002AE7', + "Barwed;": '\U00002306', + "Bcy;": '\U00000411', + "Because;": '\U00002235', + "Bernoullis;": '\U0000212C', + "Beta;": '\U00000392', + "Bfr;": '\U0001D505', + "Bopf;": '\U0001D539', + "Breve;": '\U000002D8', + "Bscr;": '\U0000212C', + "Bumpeq;": '\U0000224E', + "CHcy;": '\U00000427', + "COPY;": '\U000000A9', + "Cacute;": '\U00000106', + "Cap;": '\U000022D2', + "CapitalDifferentialD;": '\U00002145', + "Cayleys;": '\U0000212D', + "Ccaron;": '\U0000010C', + "Ccedil;": '\U000000C7', + "Ccirc;": '\U00000108', + "Cconint;": '\U00002230', + "Cdot;": '\U0000010A', + "Cedilla;": '\U000000B8', + "CenterDot;": '\U000000B7', + "Cfr;": '\U0000212D', + "Chi;": '\U000003A7', + "CircleDot;": '\U00002299', + "CircleMinus;": '\U00002296', + "CirclePlus;": '\U00002295', + "CircleTimes;": '\U00002297', + "ClockwiseContourIntegral;": '\U00002232', + "CloseCurlyDoubleQuote;": '\U0000201D', + "CloseCurlyQuote;": '\U00002019', + "Colon;": '\U00002237', + "Colone;": '\U00002A74', + "Congruent;": '\U00002261', + "Conint;": '\U0000222F', + "ContourIntegral;": '\U0000222E', + "Copf;": '\U00002102', + "Coproduct;": '\U00002210', + "CounterClockwiseContourIntegral;": '\U00002233', + "Cross;": '\U00002A2F', + "Cscr;": '\U0001D49E', + "Cup;": '\U000022D3', + "CupCap;": '\U0000224D', + "DD;": '\U00002145', + "DDotrahd;": '\U00002911', + "DJcy;": '\U00000402', + "DScy;": '\U00000405', + "DZcy;": '\U0000040F', + "Dagger;": '\U00002021', + "Darr;": '\U000021A1', + "Dashv;": '\U00002AE4', + "Dcaron;": '\U0000010E', + "Dcy;": '\U00000414', + "Del;": '\U00002207', + "Delta;": '\U00000394', + "Dfr;": '\U0001D507', + "DiacriticalAcute;": '\U000000B4', + "DiacriticalDot;": '\U000002D9', + "DiacriticalDoubleAcute;": '\U000002DD', + "DiacriticalGrave;": '\U00000060', + "DiacriticalTilde;": '\U000002DC', + "Diamond;": '\U000022C4', + "DifferentialD;": '\U00002146', + "Dopf;": '\U0001D53B', + "Dot;": '\U000000A8', + "DotDot;": '\U000020DC', + "DotEqual;": '\U00002250', + "DoubleContourIntegral;": '\U0000222F', + "DoubleDot;": '\U000000A8', + "DoubleDownArrow;": '\U000021D3', + "DoubleLeftArrow;": '\U000021D0', + "DoubleLeftRightArrow;": '\U000021D4', + "DoubleLeftTee;": '\U00002AE4', + "DoubleLongLeftArrow;": '\U000027F8', + "DoubleLongLeftRightArrow;": '\U000027FA', + "DoubleLongRightArrow;": '\U000027F9', + "DoubleRightArrow;": '\U000021D2', + "DoubleRightTee;": '\U000022A8', + "DoubleUpArrow;": '\U000021D1', + "DoubleUpDownArrow;": '\U000021D5', + "DoubleVerticalBar;": '\U00002225', + "DownArrow;": '\U00002193', + "DownArrowBar;": '\U00002913', + "DownArrowUpArrow;": '\U000021F5', + "DownBreve;": '\U00000311', + "DownLeftRightVector;": '\U00002950', + "DownLeftTeeVector;": '\U0000295E', + "DownLeftVector;": '\U000021BD', + "DownLeftVectorBar;": '\U00002956', + "DownRightTeeVector;": '\U0000295F', + "DownRightVector;": '\U000021C1', + "DownRightVectorBar;": '\U00002957', + "DownTee;": '\U000022A4', + "DownTeeArrow;": '\U000021A7', + "Downarrow;": '\U000021D3', + "Dscr;": '\U0001D49F', + "Dstrok;": '\U00000110', + "ENG;": '\U0000014A', + "ETH;": '\U000000D0', + "Eacute;": '\U000000C9', + "Ecaron;": '\U0000011A', + "Ecirc;": '\U000000CA', + "Ecy;": '\U0000042D', + "Edot;": '\U00000116', + "Efr;": '\U0001D508', + "Egrave;": '\U000000C8', + "Element;": '\U00002208', + "Emacr;": '\U00000112', + "EmptySmallSquare;": '\U000025FB', + "EmptyVerySmallSquare;": '\U000025AB', + "Eogon;": '\U00000118', + "Eopf;": '\U0001D53C', + "Epsilon;": '\U00000395', + "Equal;": '\U00002A75', + "EqualTilde;": '\U00002242', + "Equilibrium;": '\U000021CC', + "Escr;": '\U00002130', + "Esim;": '\U00002A73', + "Eta;": '\U00000397', + "Euml;": '\U000000CB', + "Exists;": '\U00002203', + "ExponentialE;": '\U00002147', + "Fcy;": '\U00000424', + "Ffr;": '\U0001D509', + "FilledSmallSquare;": '\U000025FC', + "FilledVerySmallSquare;": '\U000025AA', + "Fopf;": '\U0001D53D', + "ForAll;": '\U00002200', + "Fouriertrf;": '\U00002131', + "Fscr;": '\U00002131', + "GJcy;": '\U00000403', + "GT;": '\U0000003E', + "Gamma;": '\U00000393', + "Gammad;": '\U000003DC', + "Gbreve;": '\U0000011E', + "Gcedil;": '\U00000122', + "Gcirc;": '\U0000011C', + "Gcy;": '\U00000413', + "Gdot;": '\U00000120', + "Gfr;": '\U0001D50A', + "Gg;": '\U000022D9', + "Gopf;": '\U0001D53E', + "GreaterEqual;": '\U00002265', + "GreaterEqualLess;": '\U000022DB', + "GreaterFullEqual;": '\U00002267', + "GreaterGreater;": '\U00002AA2', + "GreaterLess;": '\U00002277', + "GreaterSlantEqual;": '\U00002A7E', + "GreaterTilde;": '\U00002273', + "Gscr;": '\U0001D4A2', + "Gt;": '\U0000226B', + "HARDcy;": '\U0000042A', + "Hacek;": '\U000002C7', + "Hat;": '\U0000005E', + "Hcirc;": '\U00000124', + "Hfr;": '\U0000210C', + "HilbertSpace;": '\U0000210B', + "Hopf;": '\U0000210D', + "HorizontalLine;": '\U00002500', + "Hscr;": '\U0000210B', + "Hstrok;": '\U00000126', + "HumpDownHump;": '\U0000224E', + "HumpEqual;": '\U0000224F', + "IEcy;": '\U00000415', + "IJlig;": '\U00000132', + "IOcy;": '\U00000401', + "Iacute;": '\U000000CD', + "Icirc;": '\U000000CE', + "Icy;": '\U00000418', + "Idot;": '\U00000130', + "Ifr;": '\U00002111', + "Igrave;": '\U000000CC', + "Im;": '\U00002111', + "Imacr;": '\U0000012A', + "ImaginaryI;": '\U00002148', + "Implies;": '\U000021D2', + "Int;": '\U0000222C', + "Integral;": '\U0000222B', + "Intersection;": '\U000022C2', + "InvisibleComma;": '\U00002063', + "InvisibleTimes;": '\U00002062', + "Iogon;": '\U0000012E', + "Iopf;": '\U0001D540', + "Iota;": '\U00000399', + "Iscr;": '\U00002110', + "Itilde;": '\U00000128', + "Iukcy;": '\U00000406', + "Iuml;": '\U000000CF', + "Jcirc;": '\U00000134', + "Jcy;": '\U00000419', + "Jfr;": '\U0001D50D', + "Jopf;": '\U0001D541', + "Jscr;": '\U0001D4A5', + "Jsercy;": '\U00000408', + "Jukcy;": '\U00000404', + "KHcy;": '\U00000425', + "KJcy;": '\U0000040C', + "Kappa;": '\U0000039A', + "Kcedil;": '\U00000136', + "Kcy;": '\U0000041A', + "Kfr;": '\U0001D50E', + "Kopf;": '\U0001D542', + "Kscr;": '\U0001D4A6', + "LJcy;": '\U00000409', + "LT;": '\U0000003C', + "Lacute;": '\U00000139', + "Lambda;": '\U0000039B', + "Lang;": '\U000027EA', + "Laplacetrf;": '\U00002112', + "Larr;": '\U0000219E', + "Lcaron;": '\U0000013D', + "Lcedil;": '\U0000013B', + "Lcy;": '\U0000041B', + "LeftAngleBracket;": '\U000027E8', + "LeftArrow;": '\U00002190', + "LeftArrowBar;": '\U000021E4', + "LeftArrowRightArrow;": '\U000021C6', + "LeftCeiling;": '\U00002308', + "LeftDoubleBracket;": '\U000027E6', + "LeftDownTeeVector;": '\U00002961', + "LeftDownVector;": '\U000021C3', + "LeftDownVectorBar;": '\U00002959', + "LeftFloor;": '\U0000230A', + "LeftRightArrow;": '\U00002194', + "LeftRightVector;": '\U0000294E', + "LeftTee;": '\U000022A3', + "LeftTeeArrow;": '\U000021A4', + "LeftTeeVector;": '\U0000295A', + "LeftTriangle;": '\U000022B2', + "LeftTriangleBar;": '\U000029CF', + "LeftTriangleEqual;": '\U000022B4', + "LeftUpDownVector;": '\U00002951', + "LeftUpTeeVector;": '\U00002960', + "LeftUpVector;": '\U000021BF', + "LeftUpVectorBar;": '\U00002958', + "LeftVector;": '\U000021BC', + "LeftVectorBar;": '\U00002952', + "Leftarrow;": '\U000021D0', + "Leftrightarrow;": '\U000021D4', + "LessEqualGreater;": '\U000022DA', + "LessFullEqual;": '\U00002266', + "LessGreater;": '\U00002276', + "LessLess;": '\U00002AA1', + "LessSlantEqual;": '\U00002A7D', + "LessTilde;": '\U00002272', + "Lfr;": '\U0001D50F', + "Ll;": '\U000022D8', + "Lleftarrow;": '\U000021DA', + "Lmidot;": '\U0000013F', + "LongLeftArrow;": '\U000027F5', + "LongLeftRightArrow;": '\U000027F7', + "LongRightArrow;": '\U000027F6', + "Longleftarrow;": '\U000027F8', + "Longleftrightarrow;": '\U000027FA', + "Longrightarrow;": '\U000027F9', + "Lopf;": '\U0001D543', + "LowerLeftArrow;": '\U00002199', + "LowerRightArrow;": '\U00002198', + "Lscr;": '\U00002112', + "Lsh;": '\U000021B0', + "Lstrok;": '\U00000141', + "Lt;": '\U0000226A', + "Map;": '\U00002905', + "Mcy;": '\U0000041C', + "MediumSpace;": '\U0000205F', + "Mellintrf;": '\U00002133', + "Mfr;": '\U0001D510', + "MinusPlus;": '\U00002213', + "Mopf;": '\U0001D544', + "Mscr;": '\U00002133', + "Mu;": '\U0000039C', + "NJcy;": '\U0000040A', + "Nacute;": '\U00000143', + "Ncaron;": '\U00000147', + "Ncedil;": '\U00000145', + "Ncy;": '\U0000041D', + "NegativeMediumSpace;": '\U0000200B', + "NegativeThickSpace;": '\U0000200B', + "NegativeThinSpace;": '\U0000200B', + "NegativeVeryThinSpace;": '\U0000200B', + "NestedGreaterGreater;": '\U0000226B', + "NestedLessLess;": '\U0000226A', + "NewLine;": '\U0000000A', + "Nfr;": '\U0001D511', + "NoBreak;": '\U00002060', + "NonBreakingSpace;": '\U000000A0', + "Nopf;": '\U00002115', + "Not;": '\U00002AEC', + "NotCongruent;": '\U00002262', + "NotCupCap;": '\U0000226D', + "NotDoubleVerticalBar;": '\U00002226', + "NotElement;": '\U00002209', + "NotEqual;": '\U00002260', + "NotExists;": '\U00002204', + "NotGreater;": '\U0000226F', + "NotGreaterEqual;": '\U00002271', + "NotGreaterLess;": '\U00002279', + "NotGreaterTilde;": '\U00002275', + "NotLeftTriangle;": '\U000022EA', + "NotLeftTriangleEqual;": '\U000022EC', + "NotLess;": '\U0000226E', + "NotLessEqual;": '\U00002270', + "NotLessGreater;": '\U00002278', + "NotLessTilde;": '\U00002274', + "NotPrecedes;": '\U00002280', + "NotPrecedesSlantEqual;": '\U000022E0', + "NotReverseElement;": '\U0000220C', + "NotRightTriangle;": '\U000022EB', + "NotRightTriangleEqual;": '\U000022ED', + "NotSquareSubsetEqual;": '\U000022E2', + "NotSquareSupersetEqual;": '\U000022E3', + "NotSubsetEqual;": '\U00002288', + "NotSucceeds;": '\U00002281', + "NotSucceedsSlantEqual;": '\U000022E1', + "NotSupersetEqual;": '\U00002289', + "NotTilde;": '\U00002241', + "NotTildeEqual;": '\U00002244', + "NotTildeFullEqual;": '\U00002247', + "NotTildeTilde;": '\U00002249', + "NotVerticalBar;": '\U00002224', + "Nscr;": '\U0001D4A9', + "Ntilde;": '\U000000D1', + "Nu;": '\U0000039D', + "OElig;": '\U00000152', + "Oacute;": '\U000000D3', + "Ocirc;": '\U000000D4', + "Ocy;": '\U0000041E', + "Odblac;": '\U00000150', + "Ofr;": '\U0001D512', + "Ograve;": '\U000000D2', + "Omacr;": '\U0000014C', + "Omega;": '\U000003A9', + "Omicron;": '\U0000039F', + "Oopf;": '\U0001D546', + "OpenCurlyDoubleQuote;": '\U0000201C', + "OpenCurlyQuote;": '\U00002018', + "Or;": '\U00002A54', + "Oscr;": '\U0001D4AA', + "Oslash;": '\U000000D8', + "Otilde;": '\U000000D5', + "Otimes;": '\U00002A37', + "Ouml;": '\U000000D6', + "OverBar;": '\U0000203E', + "OverBrace;": '\U000023DE', + "OverBracket;": '\U000023B4', + "OverParenthesis;": '\U000023DC', + "PartialD;": '\U00002202', + "Pcy;": '\U0000041F', + "Pfr;": '\U0001D513', + "Phi;": '\U000003A6', + "Pi;": '\U000003A0', + "PlusMinus;": '\U000000B1', + "Poincareplane;": '\U0000210C', + "Popf;": '\U00002119', + "Pr;": '\U00002ABB', + "Precedes;": '\U0000227A', + "PrecedesEqual;": '\U00002AAF', + "PrecedesSlantEqual;": '\U0000227C', + "PrecedesTilde;": '\U0000227E', + "Prime;": '\U00002033', + "Product;": '\U0000220F', + "Proportion;": '\U00002237', + "Proportional;": '\U0000221D', + "Pscr;": '\U0001D4AB', + "Psi;": '\U000003A8', + "QUOT;": '\U00000022', + "Qfr;": '\U0001D514', + "Qopf;": '\U0000211A', + "Qscr;": '\U0001D4AC', + "RBarr;": '\U00002910', + "REG;": '\U000000AE', + "Racute;": '\U00000154', + "Rang;": '\U000027EB', + "Rarr;": '\U000021A0', + "Rarrtl;": '\U00002916', + "Rcaron;": '\U00000158', + "Rcedil;": '\U00000156', + "Rcy;": '\U00000420', + "Re;": '\U0000211C', + "ReverseElement;": '\U0000220B', + "ReverseEquilibrium;": '\U000021CB', + "ReverseUpEquilibrium;": '\U0000296F', + "Rfr;": '\U0000211C', + "Rho;": '\U000003A1', + "RightAngleBracket;": '\U000027E9', + "RightArrow;": '\U00002192', + "RightArrowBar;": '\U000021E5', + "RightArrowLeftArrow;": '\U000021C4', + "RightCeiling;": '\U00002309', + "RightDoubleBracket;": '\U000027E7', + "RightDownTeeVector;": '\U0000295D', + "RightDownVector;": '\U000021C2', + "RightDownVectorBar;": '\U00002955', + "RightFloor;": '\U0000230B', + "RightTee;": '\U000022A2', + "RightTeeArrow;": '\U000021A6', + "RightTeeVector;": '\U0000295B', + "RightTriangle;": '\U000022B3', + "RightTriangleBar;": '\U000029D0', + "RightTriangleEqual;": '\U000022B5', + "RightUpDownVector;": '\U0000294F', + "RightUpTeeVector;": '\U0000295C', + "RightUpVector;": '\U000021BE', + "RightUpVectorBar;": '\U00002954', + "RightVector;": '\U000021C0', + "RightVectorBar;": '\U00002953', + "Rightarrow;": '\U000021D2', + "Ropf;": '\U0000211D', + "RoundImplies;": '\U00002970', + "Rrightarrow;": '\U000021DB', + "Rscr;": '\U0000211B', + "Rsh;": '\U000021B1', + "RuleDelayed;": '\U000029F4', + "SHCHcy;": '\U00000429', + "SHcy;": '\U00000428', + "SOFTcy;": '\U0000042C', + "Sacute;": '\U0000015A', + "Sc;": '\U00002ABC', + "Scaron;": '\U00000160', + "Scedil;": '\U0000015E', + "Scirc;": '\U0000015C', + "Scy;": '\U00000421', + "Sfr;": '\U0001D516', + "ShortDownArrow;": '\U00002193', + "ShortLeftArrow;": '\U00002190', + "ShortRightArrow;": '\U00002192', + "ShortUpArrow;": '\U00002191', + "Sigma;": '\U000003A3', + "SmallCircle;": '\U00002218', + "Sopf;": '\U0001D54A', + "Sqrt;": '\U0000221A', + "Square;": '\U000025A1', + "SquareIntersection;": '\U00002293', + "SquareSubset;": '\U0000228F', + "SquareSubsetEqual;": '\U00002291', + "SquareSuperset;": '\U00002290', + "SquareSupersetEqual;": '\U00002292', + "SquareUnion;": '\U00002294', + "Sscr;": '\U0001D4AE', + "Star;": '\U000022C6', + "Sub;": '\U000022D0', + "Subset;": '\U000022D0', + "SubsetEqual;": '\U00002286', + "Succeeds;": '\U0000227B', + "SucceedsEqual;": '\U00002AB0', + "SucceedsSlantEqual;": '\U0000227D', + "SucceedsTilde;": '\U0000227F', + "SuchThat;": '\U0000220B', + "Sum;": '\U00002211', + "Sup;": '\U000022D1', + "Superset;": '\U00002283', + "SupersetEqual;": '\U00002287', + "Supset;": '\U000022D1', + "THORN;": '\U000000DE', + "TRADE;": '\U00002122', + "TSHcy;": '\U0000040B', + "TScy;": '\U00000426', + "Tab;": '\U00000009', + "Tau;": '\U000003A4', + "Tcaron;": '\U00000164', + "Tcedil;": '\U00000162', + "Tcy;": '\U00000422', + "Tfr;": '\U0001D517', + "Therefore;": '\U00002234', + "Theta;": '\U00000398', + "ThinSpace;": '\U00002009', + "Tilde;": '\U0000223C', + "TildeEqual;": '\U00002243', + "TildeFullEqual;": '\U00002245', + "TildeTilde;": '\U00002248', + "Topf;": '\U0001D54B', + "TripleDot;": '\U000020DB', + "Tscr;": '\U0001D4AF', + "Tstrok;": '\U00000166', + "Uacute;": '\U000000DA', + "Uarr;": '\U0000219F', + "Uarrocir;": '\U00002949', + "Ubrcy;": '\U0000040E', + "Ubreve;": '\U0000016C', + "Ucirc;": '\U000000DB', + "Ucy;": '\U00000423', + "Udblac;": '\U00000170', + "Ufr;": '\U0001D518', + "Ugrave;": '\U000000D9', + "Umacr;": '\U0000016A', + "UnderBar;": '\U0000005F', + "UnderBrace;": '\U000023DF', + "UnderBracket;": '\U000023B5', + "UnderParenthesis;": '\U000023DD', + "Union;": '\U000022C3', + "UnionPlus;": '\U0000228E', + "Uogon;": '\U00000172', + "Uopf;": '\U0001D54C', + "UpArrow;": '\U00002191', + "UpArrowBar;": '\U00002912', + "UpArrowDownArrow;": '\U000021C5', + "UpDownArrow;": '\U00002195', + "UpEquilibrium;": '\U0000296E', + "UpTee;": '\U000022A5', + "UpTeeArrow;": '\U000021A5', + "Uparrow;": '\U000021D1', + "Updownarrow;": '\U000021D5', + "UpperLeftArrow;": '\U00002196', + "UpperRightArrow;": '\U00002197', + "Upsi;": '\U000003D2', + "Upsilon;": '\U000003A5', + "Uring;": '\U0000016E', + "Uscr;": '\U0001D4B0', + "Utilde;": '\U00000168', + "Uuml;": '\U000000DC', + "VDash;": '\U000022AB', + "Vbar;": '\U00002AEB', + "Vcy;": '\U00000412', + "Vdash;": '\U000022A9', + "Vdashl;": '\U00002AE6', + "Vee;": '\U000022C1', + "Verbar;": '\U00002016', + "Vert;": '\U00002016', + "VerticalBar;": '\U00002223', + "VerticalLine;": '\U0000007C', + "VerticalSeparator;": '\U00002758', + "VerticalTilde;": '\U00002240', + "VeryThinSpace;": '\U0000200A', + "Vfr;": '\U0001D519', + "Vopf;": '\U0001D54D', + "Vscr;": '\U0001D4B1', + "Vvdash;": '\U000022AA', + "Wcirc;": '\U00000174', + "Wedge;": '\U000022C0', + "Wfr;": '\U0001D51A', + "Wopf;": '\U0001D54E', + "Wscr;": '\U0001D4B2', + "Xfr;": '\U0001D51B', + "Xi;": '\U0000039E', + "Xopf;": '\U0001D54F', + "Xscr;": '\U0001D4B3', + "YAcy;": '\U0000042F', + "YIcy;": '\U00000407', + "YUcy;": '\U0000042E', + "Yacute;": '\U000000DD', + "Ycirc;": '\U00000176', + "Ycy;": '\U0000042B', + "Yfr;": '\U0001D51C', + "Yopf;": '\U0001D550', + "Yscr;": '\U0001D4B4', + "Yuml;": '\U00000178', + "ZHcy;": '\U00000416', + "Zacute;": '\U00000179', + "Zcaron;": '\U0000017D', + "Zcy;": '\U00000417', + "Zdot;": '\U0000017B', + "ZeroWidthSpace;": '\U0000200B', + "Zeta;": '\U00000396', + "Zfr;": '\U00002128', + "Zopf;": '\U00002124', + "Zscr;": '\U0001D4B5', + "aacute;": '\U000000E1', + "abreve;": '\U00000103', + "ac;": '\U0000223E', + "acd;": '\U0000223F', + "acirc;": '\U000000E2', + "acute;": '\U000000B4', + "acy;": '\U00000430', + "aelig;": '\U000000E6', + "af;": '\U00002061', + "afr;": '\U0001D51E', + "agrave;": '\U000000E0', + "alefsym;": '\U00002135', + "aleph;": '\U00002135', + "alpha;": '\U000003B1', + "amacr;": '\U00000101', + "amalg;": '\U00002A3F', + "amp;": '\U00000026', + "and;": '\U00002227', + "andand;": '\U00002A55', + "andd;": '\U00002A5C', + "andslope;": '\U00002A58', + "andv;": '\U00002A5A', + "ang;": '\U00002220', + "ange;": '\U000029A4', + "angle;": '\U00002220', + "angmsd;": '\U00002221', + "angmsdaa;": '\U000029A8', + "angmsdab;": '\U000029A9', + "angmsdac;": '\U000029AA', + "angmsdad;": '\U000029AB', + "angmsdae;": '\U000029AC', + "angmsdaf;": '\U000029AD', + "angmsdag;": '\U000029AE', + "angmsdah;": '\U000029AF', + "angrt;": '\U0000221F', + "angrtvb;": '\U000022BE', + "angrtvbd;": '\U0000299D', + "angsph;": '\U00002222', + "angst;": '\U000000C5', + "angzarr;": '\U0000237C', + "aogon;": '\U00000105', + "aopf;": '\U0001D552', + "ap;": '\U00002248', + "apE;": '\U00002A70', + "apacir;": '\U00002A6F', + "ape;": '\U0000224A', + "apid;": '\U0000224B', + "apos;": '\U00000027', + "approx;": '\U00002248', + "approxeq;": '\U0000224A', + "aring;": '\U000000E5', + "ascr;": '\U0001D4B6', + "ast;": '\U0000002A', + "asymp;": '\U00002248', + "asympeq;": '\U0000224D', + "atilde;": '\U000000E3', + "auml;": '\U000000E4', + "awconint;": '\U00002233', + "awint;": '\U00002A11', + "bNot;": '\U00002AED', + "backcong;": '\U0000224C', + "backepsilon;": '\U000003F6', + "backprime;": '\U00002035', + "backsim;": '\U0000223D', + "backsimeq;": '\U000022CD', + "barvee;": '\U000022BD', + "barwed;": '\U00002305', + "barwedge;": '\U00002305', + "bbrk;": '\U000023B5', + "bbrktbrk;": '\U000023B6', + "bcong;": '\U0000224C', + "bcy;": '\U00000431', + "bdquo;": '\U0000201E', + "becaus;": '\U00002235', + "because;": '\U00002235', + "bemptyv;": '\U000029B0', + "bepsi;": '\U000003F6', + "bernou;": '\U0000212C', + "beta;": '\U000003B2', + "beth;": '\U00002136', + "between;": '\U0000226C', + "bfr;": '\U0001D51F', + "bigcap;": '\U000022C2', + "bigcirc;": '\U000025EF', + "bigcup;": '\U000022C3', + "bigodot;": '\U00002A00', + "bigoplus;": '\U00002A01', + "bigotimes;": '\U00002A02', + "bigsqcup;": '\U00002A06', + "bigstar;": '\U00002605', + "bigtriangledown;": '\U000025BD', + "bigtriangleup;": '\U000025B3', + "biguplus;": '\U00002A04', + "bigvee;": '\U000022C1', + "bigwedge;": '\U000022C0', + "bkarow;": '\U0000290D', + "blacklozenge;": '\U000029EB', + "blacksquare;": '\U000025AA', + "blacktriangle;": '\U000025B4', + "blacktriangledown;": '\U000025BE', + "blacktriangleleft;": '\U000025C2', + "blacktriangleright;": '\U000025B8', + "blank;": '\U00002423', + "blk12;": '\U00002592', + "blk14;": '\U00002591', + "blk34;": '\U00002593', + "block;": '\U00002588', + "bnot;": '\U00002310', + "bopf;": '\U0001D553', + "bot;": '\U000022A5', + "bottom;": '\U000022A5', + "bowtie;": '\U000022C8', + "boxDL;": '\U00002557', + "boxDR;": '\U00002554', + "boxDl;": '\U00002556', + "boxDr;": '\U00002553', + "boxH;": '\U00002550', + "boxHD;": '\U00002566', + "boxHU;": '\U00002569', + "boxHd;": '\U00002564', + "boxHu;": '\U00002567', + "boxUL;": '\U0000255D', + "boxUR;": '\U0000255A', + "boxUl;": '\U0000255C', + "boxUr;": '\U00002559', + "boxV;": '\U00002551', + "boxVH;": '\U0000256C', + "boxVL;": '\U00002563', + "boxVR;": '\U00002560', + "boxVh;": '\U0000256B', + "boxVl;": '\U00002562', + "boxVr;": '\U0000255F', + "boxbox;": '\U000029C9', + "boxdL;": '\U00002555', + "boxdR;": '\U00002552', + "boxdl;": '\U00002510', + "boxdr;": '\U0000250C', + "boxh;": '\U00002500', + "boxhD;": '\U00002565', + "boxhU;": '\U00002568', + "boxhd;": '\U0000252C', + "boxhu;": '\U00002534', + "boxminus;": '\U0000229F', + "boxplus;": '\U0000229E', + "boxtimes;": '\U000022A0', + "boxuL;": '\U0000255B', + "boxuR;": '\U00002558', + "boxul;": '\U00002518', + "boxur;": '\U00002514', + "boxv;": '\U00002502', + "boxvH;": '\U0000256A', + "boxvL;": '\U00002561', + "boxvR;": '\U0000255E', + "boxvh;": '\U0000253C', + "boxvl;": '\U00002524', + "boxvr;": '\U0000251C', + "bprime;": '\U00002035', + "breve;": '\U000002D8', + "brvbar;": '\U000000A6', + "bscr;": '\U0001D4B7', + "bsemi;": '\U0000204F', + "bsim;": '\U0000223D', + "bsime;": '\U000022CD', + "bsol;": '\U0000005C', + "bsolb;": '\U000029C5', + "bsolhsub;": '\U000027C8', + "bull;": '\U00002022', + "bullet;": '\U00002022', + "bump;": '\U0000224E', + "bumpE;": '\U00002AAE', + "bumpe;": '\U0000224F', + "bumpeq;": '\U0000224F', + "cacute;": '\U00000107', + "cap;": '\U00002229', + "capand;": '\U00002A44', + "capbrcup;": '\U00002A49', + "capcap;": '\U00002A4B', + "capcup;": '\U00002A47', + "capdot;": '\U00002A40', + "caret;": '\U00002041', + "caron;": '\U000002C7', + "ccaps;": '\U00002A4D', + "ccaron;": '\U0000010D', + "ccedil;": '\U000000E7', + "ccirc;": '\U00000109', + "ccups;": '\U00002A4C', + "ccupssm;": '\U00002A50', + "cdot;": '\U0000010B', + "cedil;": '\U000000B8', + "cemptyv;": '\U000029B2', + "cent;": '\U000000A2', + "centerdot;": '\U000000B7', + "cfr;": '\U0001D520', + "chcy;": '\U00000447', + "check;": '\U00002713', + "checkmark;": '\U00002713', + "chi;": '\U000003C7', + "cir;": '\U000025CB', + "cirE;": '\U000029C3', + "circ;": '\U000002C6', + "circeq;": '\U00002257', + "circlearrowleft;": '\U000021BA', + "circlearrowright;": '\U000021BB', + "circledR;": '\U000000AE', + "circledS;": '\U000024C8', + "circledast;": '\U0000229B', + "circledcirc;": '\U0000229A', + "circleddash;": '\U0000229D', + "cire;": '\U00002257', + "cirfnint;": '\U00002A10', + "cirmid;": '\U00002AEF', + "cirscir;": '\U000029C2', + "clubs;": '\U00002663', + "clubsuit;": '\U00002663', + "colon;": '\U0000003A', + "colone;": '\U00002254', + "coloneq;": '\U00002254', + "comma;": '\U0000002C', + "commat;": '\U00000040', + "comp;": '\U00002201', + "compfn;": '\U00002218', + "complement;": '\U00002201', + "complexes;": '\U00002102', + "cong;": '\U00002245', + "congdot;": '\U00002A6D', + "conint;": '\U0000222E', + "copf;": '\U0001D554', + "coprod;": '\U00002210', + "copy;": '\U000000A9', + "copysr;": '\U00002117', + "crarr;": '\U000021B5', + "cross;": '\U00002717', + "cscr;": '\U0001D4B8', + "csub;": '\U00002ACF', + "csube;": '\U00002AD1', + "csup;": '\U00002AD0', + "csupe;": '\U00002AD2', + "ctdot;": '\U000022EF', + "cudarrl;": '\U00002938', + "cudarrr;": '\U00002935', + "cuepr;": '\U000022DE', + "cuesc;": '\U000022DF', + "cularr;": '\U000021B6', + "cularrp;": '\U0000293D', + "cup;": '\U0000222A', + "cupbrcap;": '\U00002A48', + "cupcap;": '\U00002A46', + "cupcup;": '\U00002A4A', + "cupdot;": '\U0000228D', + "cupor;": '\U00002A45', + "curarr;": '\U000021B7', + "curarrm;": '\U0000293C', + "curlyeqprec;": '\U000022DE', + "curlyeqsucc;": '\U000022DF', + "curlyvee;": '\U000022CE', + "curlywedge;": '\U000022CF', + "curren;": '\U000000A4', + "curvearrowleft;": '\U000021B6', + "curvearrowright;": '\U000021B7', + "cuvee;": '\U000022CE', + "cuwed;": '\U000022CF', + "cwconint;": '\U00002232', + "cwint;": '\U00002231', + "cylcty;": '\U0000232D', + "dArr;": '\U000021D3', + "dHar;": '\U00002965', + "dagger;": '\U00002020', + "daleth;": '\U00002138', + "darr;": '\U00002193', + "dash;": '\U00002010', + "dashv;": '\U000022A3', + "dbkarow;": '\U0000290F', + "dblac;": '\U000002DD', + "dcaron;": '\U0000010F', + "dcy;": '\U00000434', + "dd;": '\U00002146', + "ddagger;": '\U00002021', + "ddarr;": '\U000021CA', + "ddotseq;": '\U00002A77', + "deg;": '\U000000B0', + "delta;": '\U000003B4', + "demptyv;": '\U000029B1', + "dfisht;": '\U0000297F', + "dfr;": '\U0001D521', + "dharl;": '\U000021C3', + "dharr;": '\U000021C2', + "diam;": '\U000022C4', + "diamond;": '\U000022C4', + "diamondsuit;": '\U00002666', + "diams;": '\U00002666', + "die;": '\U000000A8', + "digamma;": '\U000003DD', + "disin;": '\U000022F2', + "div;": '\U000000F7', + "divide;": '\U000000F7', + "divideontimes;": '\U000022C7', + "divonx;": '\U000022C7', + "djcy;": '\U00000452', + "dlcorn;": '\U0000231E', + "dlcrop;": '\U0000230D', + "dollar;": '\U00000024', + "dopf;": '\U0001D555', + "dot;": '\U000002D9', + "doteq;": '\U00002250', + "doteqdot;": '\U00002251', + "dotminus;": '\U00002238', + "dotplus;": '\U00002214', + "dotsquare;": '\U000022A1', + "doublebarwedge;": '\U00002306', + "downarrow;": '\U00002193', + "downdownarrows;": '\U000021CA', + "downharpoonleft;": '\U000021C3', + "downharpoonright;": '\U000021C2', + "drbkarow;": '\U00002910', + "drcorn;": '\U0000231F', + "drcrop;": '\U0000230C', + "dscr;": '\U0001D4B9', + "dscy;": '\U00000455', + "dsol;": '\U000029F6', + "dstrok;": '\U00000111', + "dtdot;": '\U000022F1', + "dtri;": '\U000025BF', + "dtrif;": '\U000025BE', + "duarr;": '\U000021F5', + "duhar;": '\U0000296F', + "dwangle;": '\U000029A6', + "dzcy;": '\U0000045F', + "dzigrarr;": '\U000027FF', + "eDDot;": '\U00002A77', + "eDot;": '\U00002251', + "eacute;": '\U000000E9', + "easter;": '\U00002A6E', + "ecaron;": '\U0000011B', + "ecir;": '\U00002256', + "ecirc;": '\U000000EA', + "ecolon;": '\U00002255', + "ecy;": '\U0000044D', + "edot;": '\U00000117', + "ee;": '\U00002147', + "efDot;": '\U00002252', + "efr;": '\U0001D522', + "eg;": '\U00002A9A', + "egrave;": '\U000000E8', + "egs;": '\U00002A96', + "egsdot;": '\U00002A98', + "el;": '\U00002A99', + "elinters;": '\U000023E7', + "ell;": '\U00002113', + "els;": '\U00002A95', + "elsdot;": '\U00002A97', + "emacr;": '\U00000113', + "empty;": '\U00002205', + "emptyset;": '\U00002205', + "emptyv;": '\U00002205', + "emsp;": '\U00002003', + "emsp13;": '\U00002004', + "emsp14;": '\U00002005', + "eng;": '\U0000014B', + "ensp;": '\U00002002', + "eogon;": '\U00000119', + "eopf;": '\U0001D556', + "epar;": '\U000022D5', + "eparsl;": '\U000029E3', + "eplus;": '\U00002A71', + "epsi;": '\U000003B5', + "epsilon;": '\U000003B5', + "epsiv;": '\U000003F5', + "eqcirc;": '\U00002256', + "eqcolon;": '\U00002255', + "eqsim;": '\U00002242', + "eqslantgtr;": '\U00002A96', + "eqslantless;": '\U00002A95', + "equals;": '\U0000003D', + "equest;": '\U0000225F', + "equiv;": '\U00002261', + "equivDD;": '\U00002A78', + "eqvparsl;": '\U000029E5', + "erDot;": '\U00002253', + "erarr;": '\U00002971', + "escr;": '\U0000212F', + "esdot;": '\U00002250', + "esim;": '\U00002242', + "eta;": '\U000003B7', + "eth;": '\U000000F0', + "euml;": '\U000000EB', + "euro;": '\U000020AC', + "excl;": '\U00000021', + "exist;": '\U00002203', + "expectation;": '\U00002130', + "exponentiale;": '\U00002147', + "fallingdotseq;": '\U00002252', + "fcy;": '\U00000444', + "female;": '\U00002640', + "ffilig;": '\U0000FB03', + "fflig;": '\U0000FB00', + "ffllig;": '\U0000FB04', + "ffr;": '\U0001D523', + "filig;": '\U0000FB01', + "flat;": '\U0000266D', + "fllig;": '\U0000FB02', + "fltns;": '\U000025B1', + "fnof;": '\U00000192', + "fopf;": '\U0001D557', + "forall;": '\U00002200', + "fork;": '\U000022D4', + "forkv;": '\U00002AD9', + "fpartint;": '\U00002A0D', + "frac12;": '\U000000BD', + "frac13;": '\U00002153', + "frac14;": '\U000000BC', + "frac15;": '\U00002155', + "frac16;": '\U00002159', + "frac18;": '\U0000215B', + "frac23;": '\U00002154', + "frac25;": '\U00002156', + "frac34;": '\U000000BE', + "frac35;": '\U00002157', + "frac38;": '\U0000215C', + "frac45;": '\U00002158', + "frac56;": '\U0000215A', + "frac58;": '\U0000215D', + "frac78;": '\U0000215E', + "frasl;": '\U00002044', + "frown;": '\U00002322', + "fscr;": '\U0001D4BB', + "gE;": '\U00002267', + "gEl;": '\U00002A8C', + "gacute;": '\U000001F5', + "gamma;": '\U000003B3', + "gammad;": '\U000003DD', + "gap;": '\U00002A86', + "gbreve;": '\U0000011F', + "gcirc;": '\U0000011D', + "gcy;": '\U00000433', + "gdot;": '\U00000121', + "ge;": '\U00002265', + "gel;": '\U000022DB', + "geq;": '\U00002265', + "geqq;": '\U00002267', + "geqslant;": '\U00002A7E', + "ges;": '\U00002A7E', + "gescc;": '\U00002AA9', + "gesdot;": '\U00002A80', + "gesdoto;": '\U00002A82', + "gesdotol;": '\U00002A84', + "gesles;": '\U00002A94', + "gfr;": '\U0001D524', + "gg;": '\U0000226B', + "ggg;": '\U000022D9', + "gimel;": '\U00002137', + "gjcy;": '\U00000453', + "gl;": '\U00002277', + "glE;": '\U00002A92', + "gla;": '\U00002AA5', + "glj;": '\U00002AA4', + "gnE;": '\U00002269', + "gnap;": '\U00002A8A', + "gnapprox;": '\U00002A8A', + "gne;": '\U00002A88', + "gneq;": '\U00002A88', + "gneqq;": '\U00002269', + "gnsim;": '\U000022E7', + "gopf;": '\U0001D558', + "grave;": '\U00000060', + "gscr;": '\U0000210A', + "gsim;": '\U00002273', + "gsime;": '\U00002A8E', + "gsiml;": '\U00002A90', + "gt;": '\U0000003E', + "gtcc;": '\U00002AA7', + "gtcir;": '\U00002A7A', + "gtdot;": '\U000022D7', + "gtlPar;": '\U00002995', + "gtquest;": '\U00002A7C', + "gtrapprox;": '\U00002A86', + "gtrarr;": '\U00002978', + "gtrdot;": '\U000022D7', + "gtreqless;": '\U000022DB', + "gtreqqless;": '\U00002A8C', + "gtrless;": '\U00002277', + "gtrsim;": '\U00002273', + "hArr;": '\U000021D4', + "hairsp;": '\U0000200A', + "half;": '\U000000BD', + "hamilt;": '\U0000210B', + "hardcy;": '\U0000044A', + "harr;": '\U00002194', + "harrcir;": '\U00002948', + "harrw;": '\U000021AD', + "hbar;": '\U0000210F', + "hcirc;": '\U00000125', + "hearts;": '\U00002665', + "heartsuit;": '\U00002665', + "hellip;": '\U00002026', + "hercon;": '\U000022B9', + "hfr;": '\U0001D525', + "hksearow;": '\U00002925', + "hkswarow;": '\U00002926', + "hoarr;": '\U000021FF', + "homtht;": '\U0000223B', + "hookleftarrow;": '\U000021A9', + "hookrightarrow;": '\U000021AA', + "hopf;": '\U0001D559', + "horbar;": '\U00002015', + "hscr;": '\U0001D4BD', + "hslash;": '\U0000210F', + "hstrok;": '\U00000127', + "hybull;": '\U00002043', + "hyphen;": '\U00002010', + "iacute;": '\U000000ED', + "ic;": '\U00002063', + "icirc;": '\U000000EE', + "icy;": '\U00000438', + "iecy;": '\U00000435', + "iexcl;": '\U000000A1', + "iff;": '\U000021D4', + "ifr;": '\U0001D526', + "igrave;": '\U000000EC', + "ii;": '\U00002148', + "iiiint;": '\U00002A0C', + "iiint;": '\U0000222D', + "iinfin;": '\U000029DC', + "iiota;": '\U00002129', + "ijlig;": '\U00000133', + "imacr;": '\U0000012B', + "image;": '\U00002111', + "imagline;": '\U00002110', + "imagpart;": '\U00002111', + "imath;": '\U00000131', + "imof;": '\U000022B7', + "imped;": '\U000001B5', + "in;": '\U00002208', + "incare;": '\U00002105', + "infin;": '\U0000221E', + "infintie;": '\U000029DD', + "inodot;": '\U00000131', + "int;": '\U0000222B', + "intcal;": '\U000022BA', + "integers;": '\U00002124', + "intercal;": '\U000022BA', + "intlarhk;": '\U00002A17', + "intprod;": '\U00002A3C', + "iocy;": '\U00000451', + "iogon;": '\U0000012F', + "iopf;": '\U0001D55A', + "iota;": '\U000003B9', + "iprod;": '\U00002A3C', + "iquest;": '\U000000BF', + "iscr;": '\U0001D4BE', + "isin;": '\U00002208', + "isinE;": '\U000022F9', + "isindot;": '\U000022F5', + "isins;": '\U000022F4', + "isinsv;": '\U000022F3', + "isinv;": '\U00002208', + "it;": '\U00002062', + "itilde;": '\U00000129', + "iukcy;": '\U00000456', + "iuml;": '\U000000EF', + "jcirc;": '\U00000135', + "jcy;": '\U00000439', + "jfr;": '\U0001D527', + "jmath;": '\U00000237', + "jopf;": '\U0001D55B', + "jscr;": '\U0001D4BF', + "jsercy;": '\U00000458', + "jukcy;": '\U00000454', + "kappa;": '\U000003BA', + "kappav;": '\U000003F0', + "kcedil;": '\U00000137', + "kcy;": '\U0000043A', + "kfr;": '\U0001D528', + "kgreen;": '\U00000138', + "khcy;": '\U00000445', + "kjcy;": '\U0000045C', + "kopf;": '\U0001D55C', + "kscr;": '\U0001D4C0', + "lAarr;": '\U000021DA', + "lArr;": '\U000021D0', + "lAtail;": '\U0000291B', + "lBarr;": '\U0000290E', + "lE;": '\U00002266', + "lEg;": '\U00002A8B', + "lHar;": '\U00002962', + "lacute;": '\U0000013A', + "laemptyv;": '\U000029B4', + "lagran;": '\U00002112', + "lambda;": '\U000003BB', + "lang;": '\U000027E8', + "langd;": '\U00002991', + "langle;": '\U000027E8', + "lap;": '\U00002A85', + "laquo;": '\U000000AB', + "larr;": '\U00002190', + "larrb;": '\U000021E4', + "larrbfs;": '\U0000291F', + "larrfs;": '\U0000291D', + "larrhk;": '\U000021A9', + "larrlp;": '\U000021AB', + "larrpl;": '\U00002939', + "larrsim;": '\U00002973', + "larrtl;": '\U000021A2', + "lat;": '\U00002AAB', + "latail;": '\U00002919', + "late;": '\U00002AAD', + "lbarr;": '\U0000290C', + "lbbrk;": '\U00002772', + "lbrace;": '\U0000007B', + "lbrack;": '\U0000005B', + "lbrke;": '\U0000298B', + "lbrksld;": '\U0000298F', + "lbrkslu;": '\U0000298D', + "lcaron;": '\U0000013E', + "lcedil;": '\U0000013C', + "lceil;": '\U00002308', + "lcub;": '\U0000007B', + "lcy;": '\U0000043B', + "ldca;": '\U00002936', + "ldquo;": '\U0000201C', + "ldquor;": '\U0000201E', + "ldrdhar;": '\U00002967', + "ldrushar;": '\U0000294B', + "ldsh;": '\U000021B2', + "le;": '\U00002264', + "leftarrow;": '\U00002190', + "leftarrowtail;": '\U000021A2', + "leftharpoondown;": '\U000021BD', + "leftharpoonup;": '\U000021BC', + "leftleftarrows;": '\U000021C7', + "leftrightarrow;": '\U00002194', + "leftrightarrows;": '\U000021C6', + "leftrightharpoons;": '\U000021CB', + "leftrightsquigarrow;": '\U000021AD', + "leftthreetimes;": '\U000022CB', + "leg;": '\U000022DA', + "leq;": '\U00002264', + "leqq;": '\U00002266', + "leqslant;": '\U00002A7D', + "les;": '\U00002A7D', + "lescc;": '\U00002AA8', + "lesdot;": '\U00002A7F', + "lesdoto;": '\U00002A81', + "lesdotor;": '\U00002A83', + "lesges;": '\U00002A93', + "lessapprox;": '\U00002A85', + "lessdot;": '\U000022D6', + "lesseqgtr;": '\U000022DA', + "lesseqqgtr;": '\U00002A8B', + "lessgtr;": '\U00002276', + "lesssim;": '\U00002272', + "lfisht;": '\U0000297C', + "lfloor;": '\U0000230A', + "lfr;": '\U0001D529', + "lg;": '\U00002276', + "lgE;": '\U00002A91', + "lhard;": '\U000021BD', + "lharu;": '\U000021BC', + "lharul;": '\U0000296A', + "lhblk;": '\U00002584', + "ljcy;": '\U00000459', + "ll;": '\U0000226A', + "llarr;": '\U000021C7', + "llcorner;": '\U0000231E', + "llhard;": '\U0000296B', + "lltri;": '\U000025FA', + "lmidot;": '\U00000140', + "lmoust;": '\U000023B0', + "lmoustache;": '\U000023B0', + "lnE;": '\U00002268', + "lnap;": '\U00002A89', + "lnapprox;": '\U00002A89', + "lne;": '\U00002A87', + "lneq;": '\U00002A87', + "lneqq;": '\U00002268', + "lnsim;": '\U000022E6', + "loang;": '\U000027EC', + "loarr;": '\U000021FD', + "lobrk;": '\U000027E6', + "longleftarrow;": '\U000027F5', + "longleftrightarrow;": '\U000027F7', + "longmapsto;": '\U000027FC', + "longrightarrow;": '\U000027F6', + "looparrowleft;": '\U000021AB', + "looparrowright;": '\U000021AC', + "lopar;": '\U00002985', + "lopf;": '\U0001D55D', + "loplus;": '\U00002A2D', + "lotimes;": '\U00002A34', + "lowast;": '\U00002217', + "lowbar;": '\U0000005F', + "loz;": '\U000025CA', + "lozenge;": '\U000025CA', + "lozf;": '\U000029EB', + "lpar;": '\U00000028', + "lparlt;": '\U00002993', + "lrarr;": '\U000021C6', + "lrcorner;": '\U0000231F', + "lrhar;": '\U000021CB', + "lrhard;": '\U0000296D', + "lrm;": '\U0000200E', + "lrtri;": '\U000022BF', + "lsaquo;": '\U00002039', + "lscr;": '\U0001D4C1', + "lsh;": '\U000021B0', + "lsim;": '\U00002272', + "lsime;": '\U00002A8D', + "lsimg;": '\U00002A8F', + "lsqb;": '\U0000005B', + "lsquo;": '\U00002018', + "lsquor;": '\U0000201A', + "lstrok;": '\U00000142', + "lt;": '\U0000003C', + "ltcc;": '\U00002AA6', + "ltcir;": '\U00002A79', + "ltdot;": '\U000022D6', + "lthree;": '\U000022CB', + "ltimes;": '\U000022C9', + "ltlarr;": '\U00002976', + "ltquest;": '\U00002A7B', + "ltrPar;": '\U00002996', + "ltri;": '\U000025C3', + "ltrie;": '\U000022B4', + "ltrif;": '\U000025C2', + "lurdshar;": '\U0000294A', + "luruhar;": '\U00002966', + "mDDot;": '\U0000223A', + "macr;": '\U000000AF', + "male;": '\U00002642', + "malt;": '\U00002720', + "maltese;": '\U00002720', + "map;": '\U000021A6', + "mapsto;": '\U000021A6', + "mapstodown;": '\U000021A7', + "mapstoleft;": '\U000021A4', + "mapstoup;": '\U000021A5', + "marker;": '\U000025AE', + "mcomma;": '\U00002A29', + "mcy;": '\U0000043C', + "mdash;": '\U00002014', + "measuredangle;": '\U00002221', + "mfr;": '\U0001D52A', + "mho;": '\U00002127', + "micro;": '\U000000B5', + "mid;": '\U00002223', + "midast;": '\U0000002A', + "midcir;": '\U00002AF0', + "middot;": '\U000000B7', + "minus;": '\U00002212', + "minusb;": '\U0000229F', + "minusd;": '\U00002238', + "minusdu;": '\U00002A2A', + "mlcp;": '\U00002ADB', + "mldr;": '\U00002026', + "mnplus;": '\U00002213', + "models;": '\U000022A7', + "mopf;": '\U0001D55E', + "mp;": '\U00002213', + "mscr;": '\U0001D4C2', + "mstpos;": '\U0000223E', + "mu;": '\U000003BC', + "multimap;": '\U000022B8', + "mumap;": '\U000022B8', + "nLeftarrow;": '\U000021CD', + "nLeftrightarrow;": '\U000021CE', + "nRightarrow;": '\U000021CF', + "nVDash;": '\U000022AF', + "nVdash;": '\U000022AE', + "nabla;": '\U00002207', + "nacute;": '\U00000144', + "nap;": '\U00002249', + "napos;": '\U00000149', + "napprox;": '\U00002249', + "natur;": '\U0000266E', + "natural;": '\U0000266E', + "naturals;": '\U00002115', + "nbsp;": '\U000000A0', + "ncap;": '\U00002A43', + "ncaron;": '\U00000148', + "ncedil;": '\U00000146', + "ncong;": '\U00002247', + "ncup;": '\U00002A42', + "ncy;": '\U0000043D', + "ndash;": '\U00002013', + "ne;": '\U00002260', + "neArr;": '\U000021D7', + "nearhk;": '\U00002924', + "nearr;": '\U00002197', + "nearrow;": '\U00002197', + "nequiv;": '\U00002262', + "nesear;": '\U00002928', + "nexist;": '\U00002204', + "nexists;": '\U00002204', + "nfr;": '\U0001D52B', + "nge;": '\U00002271', + "ngeq;": '\U00002271', + "ngsim;": '\U00002275', + "ngt;": '\U0000226F', + "ngtr;": '\U0000226F', + "nhArr;": '\U000021CE', + "nharr;": '\U000021AE', + "nhpar;": '\U00002AF2', + "ni;": '\U0000220B', + "nis;": '\U000022FC', + "nisd;": '\U000022FA', + "niv;": '\U0000220B', + "njcy;": '\U0000045A', + "nlArr;": '\U000021CD', + "nlarr;": '\U0000219A', + "nldr;": '\U00002025', + "nle;": '\U00002270', + "nleftarrow;": '\U0000219A', + "nleftrightarrow;": '\U000021AE', + "nleq;": '\U00002270', + "nless;": '\U0000226E', + "nlsim;": '\U00002274', + "nlt;": '\U0000226E', + "nltri;": '\U000022EA', + "nltrie;": '\U000022EC', + "nmid;": '\U00002224', + "nopf;": '\U0001D55F', + "not;": '\U000000AC', + "notin;": '\U00002209', + "notinva;": '\U00002209', + "notinvb;": '\U000022F7', + "notinvc;": '\U000022F6', + "notni;": '\U0000220C', + "notniva;": '\U0000220C', + "notnivb;": '\U000022FE', + "notnivc;": '\U000022FD', + "npar;": '\U00002226', + "nparallel;": '\U00002226', + "npolint;": '\U00002A14', + "npr;": '\U00002280', + "nprcue;": '\U000022E0', + "nprec;": '\U00002280', + "nrArr;": '\U000021CF', + "nrarr;": '\U0000219B', + "nrightarrow;": '\U0000219B', + "nrtri;": '\U000022EB', + "nrtrie;": '\U000022ED', + "nsc;": '\U00002281', + "nsccue;": '\U000022E1', + "nscr;": '\U0001D4C3', + "nshortmid;": '\U00002224', + "nshortparallel;": '\U00002226', + "nsim;": '\U00002241', + "nsime;": '\U00002244', + "nsimeq;": '\U00002244', + "nsmid;": '\U00002224', + "nspar;": '\U00002226', + "nsqsube;": '\U000022E2', + "nsqsupe;": '\U000022E3', + "nsub;": '\U00002284', + "nsube;": '\U00002288', + "nsubseteq;": '\U00002288', + "nsucc;": '\U00002281', + "nsup;": '\U00002285', + "nsupe;": '\U00002289', + "nsupseteq;": '\U00002289', + "ntgl;": '\U00002279', + "ntilde;": '\U000000F1', + "ntlg;": '\U00002278', + "ntriangleleft;": '\U000022EA', + "ntrianglelefteq;": '\U000022EC', + "ntriangleright;": '\U000022EB', + "ntrianglerighteq;": '\U000022ED', + "nu;": '\U000003BD', + "num;": '\U00000023', + "numero;": '\U00002116', + "numsp;": '\U00002007', + "nvDash;": '\U000022AD', + "nvHarr;": '\U00002904', + "nvdash;": '\U000022AC', + "nvinfin;": '\U000029DE', + "nvlArr;": '\U00002902', + "nvrArr;": '\U00002903', + "nwArr;": '\U000021D6', + "nwarhk;": '\U00002923', + "nwarr;": '\U00002196', + "nwarrow;": '\U00002196', + "nwnear;": '\U00002927', + "oS;": '\U000024C8', + "oacute;": '\U000000F3', + "oast;": '\U0000229B', + "ocir;": '\U0000229A', + "ocirc;": '\U000000F4', + "ocy;": '\U0000043E', + "odash;": '\U0000229D', + "odblac;": '\U00000151', + "odiv;": '\U00002A38', + "odot;": '\U00002299', + "odsold;": '\U000029BC', + "oelig;": '\U00000153', + "ofcir;": '\U000029BF', + "ofr;": '\U0001D52C', + "ogon;": '\U000002DB', + "ograve;": '\U000000F2', + "ogt;": '\U000029C1', + "ohbar;": '\U000029B5', + "ohm;": '\U000003A9', + "oint;": '\U0000222E', + "olarr;": '\U000021BA', + "olcir;": '\U000029BE', + "olcross;": '\U000029BB', + "oline;": '\U0000203E', + "olt;": '\U000029C0', + "omacr;": '\U0000014D', + "omega;": '\U000003C9', + "omicron;": '\U000003BF', + "omid;": '\U000029B6', + "ominus;": '\U00002296', + "oopf;": '\U0001D560', + "opar;": '\U000029B7', + "operp;": '\U000029B9', + "oplus;": '\U00002295', + "or;": '\U00002228', + "orarr;": '\U000021BB', + "ord;": '\U00002A5D', + "order;": '\U00002134', + "orderof;": '\U00002134', + "ordf;": '\U000000AA', + "ordm;": '\U000000BA', + "origof;": '\U000022B6', + "oror;": '\U00002A56', + "orslope;": '\U00002A57', + "orv;": '\U00002A5B', + "oscr;": '\U00002134', + "oslash;": '\U000000F8', + "osol;": '\U00002298', + "otilde;": '\U000000F5', + "otimes;": '\U00002297', + "otimesas;": '\U00002A36', + "ouml;": '\U000000F6', + "ovbar;": '\U0000233D', + "par;": '\U00002225', + "para;": '\U000000B6', + "parallel;": '\U00002225', + "parsim;": '\U00002AF3', + "parsl;": '\U00002AFD', + "part;": '\U00002202', + "pcy;": '\U0000043F', + "percnt;": '\U00000025', + "period;": '\U0000002E', + "permil;": '\U00002030', + "perp;": '\U000022A5', + "pertenk;": '\U00002031', + "pfr;": '\U0001D52D', + "phi;": '\U000003C6', + "phiv;": '\U000003D5', + "phmmat;": '\U00002133', + "phone;": '\U0000260E', + "pi;": '\U000003C0', + "pitchfork;": '\U000022D4', + "piv;": '\U000003D6', + "planck;": '\U0000210F', + "planckh;": '\U0000210E', + "plankv;": '\U0000210F', + "plus;": '\U0000002B', + "plusacir;": '\U00002A23', + "plusb;": '\U0000229E', + "pluscir;": '\U00002A22', + "plusdo;": '\U00002214', + "plusdu;": '\U00002A25', + "pluse;": '\U00002A72', + "plusmn;": '\U000000B1', + "plussim;": '\U00002A26', + "plustwo;": '\U00002A27', + "pm;": '\U000000B1', + "pointint;": '\U00002A15', + "popf;": '\U0001D561', + "pound;": '\U000000A3', + "pr;": '\U0000227A', + "prE;": '\U00002AB3', + "prap;": '\U00002AB7', + "prcue;": '\U0000227C', + "pre;": '\U00002AAF', + "prec;": '\U0000227A', + "precapprox;": '\U00002AB7', + "preccurlyeq;": '\U0000227C', + "preceq;": '\U00002AAF', + "precnapprox;": '\U00002AB9', + "precneqq;": '\U00002AB5', + "precnsim;": '\U000022E8', + "precsim;": '\U0000227E', + "prime;": '\U00002032', + "primes;": '\U00002119', + "prnE;": '\U00002AB5', + "prnap;": '\U00002AB9', + "prnsim;": '\U000022E8', + "prod;": '\U0000220F', + "profalar;": '\U0000232E', + "profline;": '\U00002312', + "profsurf;": '\U00002313', + "prop;": '\U0000221D', + "propto;": '\U0000221D', + "prsim;": '\U0000227E', + "prurel;": '\U000022B0', + "pscr;": '\U0001D4C5', + "psi;": '\U000003C8', + "puncsp;": '\U00002008', + "qfr;": '\U0001D52E', + "qint;": '\U00002A0C', + "qopf;": '\U0001D562', + "qprime;": '\U00002057', + "qscr;": '\U0001D4C6', + "quaternions;": '\U0000210D', + "quatint;": '\U00002A16', + "quest;": '\U0000003F', + "questeq;": '\U0000225F', + "quot;": '\U00000022', + "rAarr;": '\U000021DB', + "rArr;": '\U000021D2', + "rAtail;": '\U0000291C', + "rBarr;": '\U0000290F', + "rHar;": '\U00002964', + "racute;": '\U00000155', + "radic;": '\U0000221A', + "raemptyv;": '\U000029B3', + "rang;": '\U000027E9', + "rangd;": '\U00002992', + "range;": '\U000029A5', + "rangle;": '\U000027E9', + "raquo;": '\U000000BB', + "rarr;": '\U00002192', + "rarrap;": '\U00002975', + "rarrb;": '\U000021E5', + "rarrbfs;": '\U00002920', + "rarrc;": '\U00002933', + "rarrfs;": '\U0000291E', + "rarrhk;": '\U000021AA', + "rarrlp;": '\U000021AC', + "rarrpl;": '\U00002945', + "rarrsim;": '\U00002974', + "rarrtl;": '\U000021A3', + "rarrw;": '\U0000219D', + "ratail;": '\U0000291A', + "ratio;": '\U00002236', + "rationals;": '\U0000211A', + "rbarr;": '\U0000290D', + "rbbrk;": '\U00002773', + "rbrace;": '\U0000007D', + "rbrack;": '\U0000005D', + "rbrke;": '\U0000298C', + "rbrksld;": '\U0000298E', + "rbrkslu;": '\U00002990', + "rcaron;": '\U00000159', + "rcedil;": '\U00000157', + "rceil;": '\U00002309', + "rcub;": '\U0000007D', + "rcy;": '\U00000440', + "rdca;": '\U00002937', + "rdldhar;": '\U00002969', + "rdquo;": '\U0000201D', + "rdquor;": '\U0000201D', + "rdsh;": '\U000021B3', + "real;": '\U0000211C', + "realine;": '\U0000211B', + "realpart;": '\U0000211C', + "reals;": '\U0000211D', + "rect;": '\U000025AD', + "reg;": '\U000000AE', + "rfisht;": '\U0000297D', + "rfloor;": '\U0000230B', + "rfr;": '\U0001D52F', + "rhard;": '\U000021C1', + "rharu;": '\U000021C0', + "rharul;": '\U0000296C', + "rho;": '\U000003C1', + "rhov;": '\U000003F1', + "rightarrow;": '\U00002192', + "rightarrowtail;": '\U000021A3', + "rightharpoondown;": '\U000021C1', + "rightharpoonup;": '\U000021C0', + "rightleftarrows;": '\U000021C4', + "rightleftharpoons;": '\U000021CC', + "rightrightarrows;": '\U000021C9', + "rightsquigarrow;": '\U0000219D', + "rightthreetimes;": '\U000022CC', + "ring;": '\U000002DA', + "risingdotseq;": '\U00002253', + "rlarr;": '\U000021C4', + "rlhar;": '\U000021CC', + "rlm;": '\U0000200F', + "rmoust;": '\U000023B1', + "rmoustache;": '\U000023B1', + "rnmid;": '\U00002AEE', + "roang;": '\U000027ED', + "roarr;": '\U000021FE', + "robrk;": '\U000027E7', + "ropar;": '\U00002986', + "ropf;": '\U0001D563', + "roplus;": '\U00002A2E', + "rotimes;": '\U00002A35', + "rpar;": '\U00000029', + "rpargt;": '\U00002994', + "rppolint;": '\U00002A12', + "rrarr;": '\U000021C9', + "rsaquo;": '\U0000203A', + "rscr;": '\U0001D4C7', + "rsh;": '\U000021B1', + "rsqb;": '\U0000005D', + "rsquo;": '\U00002019', + "rsquor;": '\U00002019', + "rthree;": '\U000022CC', + "rtimes;": '\U000022CA', + "rtri;": '\U000025B9', + "rtrie;": '\U000022B5', + "rtrif;": '\U000025B8', + "rtriltri;": '\U000029CE', + "ruluhar;": '\U00002968', + "rx;": '\U0000211E', + "sacute;": '\U0000015B', + "sbquo;": '\U0000201A', + "sc;": '\U0000227B', + "scE;": '\U00002AB4', + "scap;": '\U00002AB8', + "scaron;": '\U00000161', + "sccue;": '\U0000227D', + "sce;": '\U00002AB0', + "scedil;": '\U0000015F', + "scirc;": '\U0000015D', + "scnE;": '\U00002AB6', + "scnap;": '\U00002ABA', + "scnsim;": '\U000022E9', + "scpolint;": '\U00002A13', + "scsim;": '\U0000227F', + "scy;": '\U00000441', + "sdot;": '\U000022C5', + "sdotb;": '\U000022A1', + "sdote;": '\U00002A66', + "seArr;": '\U000021D8', + "searhk;": '\U00002925', + "searr;": '\U00002198', + "searrow;": '\U00002198', + "sect;": '\U000000A7', + "semi;": '\U0000003B', + "seswar;": '\U00002929', + "setminus;": '\U00002216', + "setmn;": '\U00002216', + "sext;": '\U00002736', + "sfr;": '\U0001D530', + "sfrown;": '\U00002322', + "sharp;": '\U0000266F', + "shchcy;": '\U00000449', + "shcy;": '\U00000448', + "shortmid;": '\U00002223', + "shortparallel;": '\U00002225', + "shy;": '\U000000AD', + "sigma;": '\U000003C3', + "sigmaf;": '\U000003C2', + "sigmav;": '\U000003C2', + "sim;": '\U0000223C', + "simdot;": '\U00002A6A', + "sime;": '\U00002243', + "simeq;": '\U00002243', + "simg;": '\U00002A9E', + "simgE;": '\U00002AA0', + "siml;": '\U00002A9D', + "simlE;": '\U00002A9F', + "simne;": '\U00002246', + "simplus;": '\U00002A24', + "simrarr;": '\U00002972', + "slarr;": '\U00002190', + "smallsetminus;": '\U00002216', + "smashp;": '\U00002A33', + "smeparsl;": '\U000029E4', + "smid;": '\U00002223', + "smile;": '\U00002323', + "smt;": '\U00002AAA', + "smte;": '\U00002AAC', + "softcy;": '\U0000044C', + "sol;": '\U0000002F', + "solb;": '\U000029C4', + "solbar;": '\U0000233F', + "sopf;": '\U0001D564', + "spades;": '\U00002660', + "spadesuit;": '\U00002660', + "spar;": '\U00002225', + "sqcap;": '\U00002293', + "sqcup;": '\U00002294', + "sqsub;": '\U0000228F', + "sqsube;": '\U00002291', + "sqsubset;": '\U0000228F', + "sqsubseteq;": '\U00002291', + "sqsup;": '\U00002290', + "sqsupe;": '\U00002292', + "sqsupset;": '\U00002290', + "sqsupseteq;": '\U00002292', + "squ;": '\U000025A1', + "square;": '\U000025A1', + "squarf;": '\U000025AA', + "squf;": '\U000025AA', + "srarr;": '\U00002192', + "sscr;": '\U0001D4C8', + "ssetmn;": '\U00002216', + "ssmile;": '\U00002323', + "sstarf;": '\U000022C6', + "star;": '\U00002606', + "starf;": '\U00002605', + "straightepsilon;": '\U000003F5', + "straightphi;": '\U000003D5', + "strns;": '\U000000AF', + "sub;": '\U00002282', + "subE;": '\U00002AC5', + "subdot;": '\U00002ABD', + "sube;": '\U00002286', + "subedot;": '\U00002AC3', + "submult;": '\U00002AC1', + "subnE;": '\U00002ACB', + "subne;": '\U0000228A', + "subplus;": '\U00002ABF', + "subrarr;": '\U00002979', + "subset;": '\U00002282', + "subseteq;": '\U00002286', + "subseteqq;": '\U00002AC5', + "subsetneq;": '\U0000228A', + "subsetneqq;": '\U00002ACB', + "subsim;": '\U00002AC7', + "subsub;": '\U00002AD5', + "subsup;": '\U00002AD3', + "succ;": '\U0000227B', + "succapprox;": '\U00002AB8', + "succcurlyeq;": '\U0000227D', + "succeq;": '\U00002AB0', + "succnapprox;": '\U00002ABA', + "succneqq;": '\U00002AB6', + "succnsim;": '\U000022E9', + "succsim;": '\U0000227F', + "sum;": '\U00002211', + "sung;": '\U0000266A', + "sup;": '\U00002283', + "sup1;": '\U000000B9', + "sup2;": '\U000000B2', + "sup3;": '\U000000B3', + "supE;": '\U00002AC6', + "supdot;": '\U00002ABE', + "supdsub;": '\U00002AD8', + "supe;": '\U00002287', + "supedot;": '\U00002AC4', + "suphsol;": '\U000027C9', + "suphsub;": '\U00002AD7', + "suplarr;": '\U0000297B', + "supmult;": '\U00002AC2', + "supnE;": '\U00002ACC', + "supne;": '\U0000228B', + "supplus;": '\U00002AC0', + "supset;": '\U00002283', + "supseteq;": '\U00002287', + "supseteqq;": '\U00002AC6', + "supsetneq;": '\U0000228B', + "supsetneqq;": '\U00002ACC', + "supsim;": '\U00002AC8', + "supsub;": '\U00002AD4', + "supsup;": '\U00002AD6', + "swArr;": '\U000021D9', + "swarhk;": '\U00002926', + "swarr;": '\U00002199', + "swarrow;": '\U00002199', + "swnwar;": '\U0000292A', + "szlig;": '\U000000DF', + "target;": '\U00002316', + "tau;": '\U000003C4', + "tbrk;": '\U000023B4', + "tcaron;": '\U00000165', + "tcedil;": '\U00000163', + "tcy;": '\U00000442', + "tdot;": '\U000020DB', + "telrec;": '\U00002315', + "tfr;": '\U0001D531', + "there4;": '\U00002234', + "therefore;": '\U00002234', + "theta;": '\U000003B8', + "thetasym;": '\U000003D1', + "thetav;": '\U000003D1', + "thickapprox;": '\U00002248', + "thicksim;": '\U0000223C', + "thinsp;": '\U00002009', + "thkap;": '\U00002248', + "thksim;": '\U0000223C', + "thorn;": '\U000000FE', + "tilde;": '\U000002DC', + "times;": '\U000000D7', + "timesb;": '\U000022A0', + "timesbar;": '\U00002A31', + "timesd;": '\U00002A30', + "tint;": '\U0000222D', + "toea;": '\U00002928', + "top;": '\U000022A4', + "topbot;": '\U00002336', + "topcir;": '\U00002AF1', + "topf;": '\U0001D565', + "topfork;": '\U00002ADA', + "tosa;": '\U00002929', + "tprime;": '\U00002034', + "trade;": '\U00002122', + "triangle;": '\U000025B5', + "triangledown;": '\U000025BF', + "triangleleft;": '\U000025C3', + "trianglelefteq;": '\U000022B4', + "triangleq;": '\U0000225C', + "triangleright;": '\U000025B9', + "trianglerighteq;": '\U000022B5', + "tridot;": '\U000025EC', + "trie;": '\U0000225C', + "triminus;": '\U00002A3A', + "triplus;": '\U00002A39', + "trisb;": '\U000029CD', + "tritime;": '\U00002A3B', + "trpezium;": '\U000023E2', + "tscr;": '\U0001D4C9', + "tscy;": '\U00000446', + "tshcy;": '\U0000045B', + "tstrok;": '\U00000167', + "twixt;": '\U0000226C', + "twoheadleftarrow;": '\U0000219E', + "twoheadrightarrow;": '\U000021A0', + "uArr;": '\U000021D1', + "uHar;": '\U00002963', + "uacute;": '\U000000FA', + "uarr;": '\U00002191', + "ubrcy;": '\U0000045E', + "ubreve;": '\U0000016D', + "ucirc;": '\U000000FB', + "ucy;": '\U00000443', + "udarr;": '\U000021C5', + "udblac;": '\U00000171', + "udhar;": '\U0000296E', + "ufisht;": '\U0000297E', + "ufr;": '\U0001D532', + "ugrave;": '\U000000F9', + "uharl;": '\U000021BF', + "uharr;": '\U000021BE', + "uhblk;": '\U00002580', + "ulcorn;": '\U0000231C', + "ulcorner;": '\U0000231C', + "ulcrop;": '\U0000230F', + "ultri;": '\U000025F8', + "umacr;": '\U0000016B', + "uml;": '\U000000A8', + "uogon;": '\U00000173', + "uopf;": '\U0001D566', + "uparrow;": '\U00002191', + "updownarrow;": '\U00002195', + "upharpoonleft;": '\U000021BF', + "upharpoonright;": '\U000021BE', + "uplus;": '\U0000228E', + "upsi;": '\U000003C5', + "upsih;": '\U000003D2', + "upsilon;": '\U000003C5', + "upuparrows;": '\U000021C8', + "urcorn;": '\U0000231D', + "urcorner;": '\U0000231D', + "urcrop;": '\U0000230E', + "uring;": '\U0000016F', + "urtri;": '\U000025F9', + "uscr;": '\U0001D4CA', + "utdot;": '\U000022F0', + "utilde;": '\U00000169', + "utri;": '\U000025B5', + "utrif;": '\U000025B4', + "uuarr;": '\U000021C8', + "uuml;": '\U000000FC', + "uwangle;": '\U000029A7', + "vArr;": '\U000021D5', + "vBar;": '\U00002AE8', + "vBarv;": '\U00002AE9', + "vDash;": '\U000022A8', + "vangrt;": '\U0000299C', + "varepsilon;": '\U000003F5', + "varkappa;": '\U000003F0', + "varnothing;": '\U00002205', + "varphi;": '\U000003D5', + "varpi;": '\U000003D6', + "varpropto;": '\U0000221D', + "varr;": '\U00002195', + "varrho;": '\U000003F1', + "varsigma;": '\U000003C2', + "vartheta;": '\U000003D1', + "vartriangleleft;": '\U000022B2', + "vartriangleright;": '\U000022B3', + "vcy;": '\U00000432', + "vdash;": '\U000022A2', + "vee;": '\U00002228', + "veebar;": '\U000022BB', + "veeeq;": '\U0000225A', + "vellip;": '\U000022EE', + "verbar;": '\U0000007C', + "vert;": '\U0000007C', + "vfr;": '\U0001D533', + "vltri;": '\U000022B2', + "vopf;": '\U0001D567', + "vprop;": '\U0000221D', + "vrtri;": '\U000022B3', + "vscr;": '\U0001D4CB', + "vzigzag;": '\U0000299A', + "wcirc;": '\U00000175', + "wedbar;": '\U00002A5F', + "wedge;": '\U00002227', + "wedgeq;": '\U00002259', + "weierp;": '\U00002118', + "wfr;": '\U0001D534', + "wopf;": '\U0001D568', + "wp;": '\U00002118', + "wr;": '\U00002240', + "wreath;": '\U00002240', + "wscr;": '\U0001D4CC', + "xcap;": '\U000022C2', + "xcirc;": '\U000025EF', + "xcup;": '\U000022C3', + "xdtri;": '\U000025BD', + "xfr;": '\U0001D535', + "xhArr;": '\U000027FA', + "xharr;": '\U000027F7', + "xi;": '\U000003BE', + "xlArr;": '\U000027F8', + "xlarr;": '\U000027F5', + "xmap;": '\U000027FC', + "xnis;": '\U000022FB', + "xodot;": '\U00002A00', + "xopf;": '\U0001D569', + "xoplus;": '\U00002A01', + "xotime;": '\U00002A02', + "xrArr;": '\U000027F9', + "xrarr;": '\U000027F6', + "xscr;": '\U0001D4CD', + "xsqcup;": '\U00002A06', + "xuplus;": '\U00002A04', + "xutri;": '\U000025B3', + "xvee;": '\U000022C1', + "xwedge;": '\U000022C0', + "yacute;": '\U000000FD', + "yacy;": '\U0000044F', + "ycirc;": '\U00000177', + "ycy;": '\U0000044B', + "yen;": '\U000000A5', + "yfr;": '\U0001D536', + "yicy;": '\U00000457', + "yopf;": '\U0001D56A', + "yscr;": '\U0001D4CE', + "yucy;": '\U0000044E', + "yuml;": '\U000000FF', + "zacute;": '\U0000017A', + "zcaron;": '\U0000017E', + "zcy;": '\U00000437', + "zdot;": '\U0000017C', + "zeetrf;": '\U00002128', + "zeta;": '\U000003B6', + "zfr;": '\U0001D537', + "zhcy;": '\U00000436', + "zigrarr;": '\U000021DD', + "zopf;": '\U0001D56B', + "zscr;": '\U0001D4CF', + "zwj;": '\U0000200D', + "zwnj;": '\U0000200C', + "AElig": '\U000000C6', + "AMP": '\U00000026', + "Aacute": '\U000000C1', + "Acirc": '\U000000C2', + "Agrave": '\U000000C0', + "Aring": '\U000000C5', + "Atilde": '\U000000C3', + "Auml": '\U000000C4', + "COPY": '\U000000A9', + "Ccedil": '\U000000C7', + "ETH": '\U000000D0', + "Eacute": '\U000000C9', + "Ecirc": '\U000000CA', + "Egrave": '\U000000C8', + "Euml": '\U000000CB', + "GT": '\U0000003E', + "Iacute": '\U000000CD', + "Icirc": '\U000000CE', + "Igrave": '\U000000CC', + "Iuml": '\U000000CF', + "LT": '\U0000003C', + "Ntilde": '\U000000D1', + "Oacute": '\U000000D3', + "Ocirc": '\U000000D4', + "Ograve": '\U000000D2', + "Oslash": '\U000000D8', + "Otilde": '\U000000D5', + "Ouml": '\U000000D6', + "QUOT": '\U00000022', + "REG": '\U000000AE', + "THORN": '\U000000DE', + "Uacute": '\U000000DA', + "Ucirc": '\U000000DB', + "Ugrave": '\U000000D9', + "Uuml": '\U000000DC', + "Yacute": '\U000000DD', + "aacute": '\U000000E1', + "acirc": '\U000000E2', + "acute": '\U000000B4', + "aelig": '\U000000E6', + "agrave": '\U000000E0', + "amp": '\U00000026', + "aring": '\U000000E5', + "atilde": '\U000000E3', + "auml": '\U000000E4', + "brvbar": '\U000000A6', + "ccedil": '\U000000E7', + "cedil": '\U000000B8', + "cent": '\U000000A2', + "copy": '\U000000A9', + "curren": '\U000000A4', + "deg": '\U000000B0', + "divide": '\U000000F7', + "eacute": '\U000000E9', + "ecirc": '\U000000EA', + "egrave": '\U000000E8', + "eth": '\U000000F0', + "euml": '\U000000EB', + "frac12": '\U000000BD', + "frac14": '\U000000BC', + "frac34": '\U000000BE', + "gt": '\U0000003E', + "iacute": '\U000000ED', + "icirc": '\U000000EE', + "iexcl": '\U000000A1', + "igrave": '\U000000EC', + "iquest": '\U000000BF', + "iuml": '\U000000EF', + "laquo": '\U000000AB', + "lt": '\U0000003C', + "macr": '\U000000AF', + "micro": '\U000000B5', + "middot": '\U000000B7', + "nbsp": '\U000000A0', + "not": '\U000000AC', + "ntilde": '\U000000F1', + "oacute": '\U000000F3', + "ocirc": '\U000000F4', + "ograve": '\U000000F2', + "ordf": '\U000000AA', + "ordm": '\U000000BA', + "oslash": '\U000000F8', + "otilde": '\U000000F5', + "ouml": '\U000000F6', + "para": '\U000000B6', + "plusmn": '\U000000B1', + "pound": '\U000000A3', + "quot": '\U00000022', + "raquo": '\U000000BB', + "reg": '\U000000AE', + "sect": '\U000000A7', + "shy": '\U000000AD', + "sup1": '\U000000B9', + "sup2": '\U000000B2', + "sup3": '\U000000B3', + "szlig": '\U000000DF', + "thorn": '\U000000FE', + "times": '\U000000D7', + "uacute": '\U000000FA', + "ucirc": '\U000000FB', + "ugrave": '\U000000F9', + "uml": '\U000000A8', + "uuml": '\U000000FC', + "yacute": '\U000000FD', + "yen": '\U000000A5', + "yuml": '\U000000FF', +} + +// HTML entities that are two unicode codepoints. +var entity2 = map[string][2]rune{ + // TODO(nigeltao): Handle replacements that are wider than their names. + // "nLt;": {'\u226A', '\u20D2'}, + // "nGt;": {'\u226B', '\u20D2'}, + "NotEqualTilde;": {'\u2242', '\u0338'}, + "NotGreaterFullEqual;": {'\u2267', '\u0338'}, + "NotGreaterGreater;": {'\u226B', '\u0338'}, + "NotGreaterSlantEqual;": {'\u2A7E', '\u0338'}, + "NotHumpDownHump;": {'\u224E', '\u0338'}, + "NotHumpEqual;": {'\u224F', '\u0338'}, + "NotLeftTriangleBar;": {'\u29CF', '\u0338'}, + "NotLessLess;": {'\u226A', '\u0338'}, + "NotLessSlantEqual;": {'\u2A7D', '\u0338'}, + "NotNestedGreaterGreater;": {'\u2AA2', '\u0338'}, + "NotNestedLessLess;": {'\u2AA1', '\u0338'}, + "NotPrecedesEqual;": {'\u2AAF', '\u0338'}, + "NotRightTriangleBar;": {'\u29D0', '\u0338'}, + "NotSquareSubset;": {'\u228F', '\u0338'}, + "NotSquareSuperset;": {'\u2290', '\u0338'}, + "NotSubset;": {'\u2282', '\u20D2'}, + "NotSucceedsEqual;": {'\u2AB0', '\u0338'}, + "NotSucceedsTilde;": {'\u227F', '\u0338'}, + "NotSuperset;": {'\u2283', '\u20D2'}, + "ThickSpace;": {'\u205F', '\u200A'}, + "acE;": {'\u223E', '\u0333'}, + "bne;": {'\u003D', '\u20E5'}, + "bnequiv;": {'\u2261', '\u20E5'}, + "caps;": {'\u2229', '\uFE00'}, + "cups;": {'\u222A', '\uFE00'}, + "fjlig;": {'\u0066', '\u006A'}, + "gesl;": {'\u22DB', '\uFE00'}, + "gvertneqq;": {'\u2269', '\uFE00'}, + "gvnE;": {'\u2269', '\uFE00'}, + "lates;": {'\u2AAD', '\uFE00'}, + "lesg;": {'\u22DA', '\uFE00'}, + "lvertneqq;": {'\u2268', '\uFE00'}, + "lvnE;": {'\u2268', '\uFE00'}, + "nGg;": {'\u22D9', '\u0338'}, + "nGtv;": {'\u226B', '\u0338'}, + "nLl;": {'\u22D8', '\u0338'}, + "nLtv;": {'\u226A', '\u0338'}, + "nang;": {'\u2220', '\u20D2'}, + "napE;": {'\u2A70', '\u0338'}, + "napid;": {'\u224B', '\u0338'}, + "nbump;": {'\u224E', '\u0338'}, + "nbumpe;": {'\u224F', '\u0338'}, + "ncongdot;": {'\u2A6D', '\u0338'}, + "nedot;": {'\u2250', '\u0338'}, + "nesim;": {'\u2242', '\u0338'}, + "ngE;": {'\u2267', '\u0338'}, + "ngeqq;": {'\u2267', '\u0338'}, + "ngeqslant;": {'\u2A7E', '\u0338'}, + "nges;": {'\u2A7E', '\u0338'}, + "nlE;": {'\u2266', '\u0338'}, + "nleqq;": {'\u2266', '\u0338'}, + "nleqslant;": {'\u2A7D', '\u0338'}, + "nles;": {'\u2A7D', '\u0338'}, + "notinE;": {'\u22F9', '\u0338'}, + "notindot;": {'\u22F5', '\u0338'}, + "nparsl;": {'\u2AFD', '\u20E5'}, + "npart;": {'\u2202', '\u0338'}, + "npre;": {'\u2AAF', '\u0338'}, + "npreceq;": {'\u2AAF', '\u0338'}, + "nrarrc;": {'\u2933', '\u0338'}, + "nrarrw;": {'\u219D', '\u0338'}, + "nsce;": {'\u2AB0', '\u0338'}, + "nsubE;": {'\u2AC5', '\u0338'}, + "nsubset;": {'\u2282', '\u20D2'}, + "nsubseteqq;": {'\u2AC5', '\u0338'}, + "nsucceq;": {'\u2AB0', '\u0338'}, + "nsupE;": {'\u2AC6', '\u0338'}, + "nsupset;": {'\u2283', '\u20D2'}, + "nsupseteqq;": {'\u2AC6', '\u0338'}, + "nvap;": {'\u224D', '\u20D2'}, + "nvge;": {'\u2265', '\u20D2'}, + "nvgt;": {'\u003E', '\u20D2'}, + "nvle;": {'\u2264', '\u20D2'}, + "nvlt;": {'\u003C', '\u20D2'}, + "nvltrie;": {'\u22B4', '\u20D2'}, + "nvrtrie;": {'\u22B5', '\u20D2'}, + "nvsim;": {'\u223C', '\u20D2'}, + "race;": {'\u223D', '\u0331'}, + "smtes;": {'\u2AAC', '\uFE00'}, + "sqcaps;": {'\u2293', '\uFE00'}, + "sqcups;": {'\u2294', '\uFE00'}, + "varsubsetneq;": {'\u228A', '\uFE00'}, + "varsubsetneqq;": {'\u2ACB', '\uFE00'}, + "varsupsetneq;": {'\u228B', '\uFE00'}, + "varsupsetneqq;": {'\u2ACC', '\uFE00'}, + "vnsub;": {'\u2282', '\u20D2'}, + "vnsup;": {'\u2283', '\u20D2'}, + "vsubnE;": {'\u2ACB', '\uFE00'}, + "vsubne;": {'\u228A', '\uFE00'}, + "vsupnE;": {'\u2ACC', '\uFE00'}, + "vsupne;": {'\u228B', '\uFE00'}, +} diff --git a/vendor/golang.org/x/net/html/escape.go b/vendor/golang.org/x/net/html/escape.go new file mode 100644 index 0000000..04c6bec --- /dev/null +++ b/vendor/golang.org/x/net/html/escape.go @@ -0,0 +1,339 @@ +// Copyright 2010 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package html + +import ( + "bytes" + "strings" + "unicode/utf8" +) + +// These replacements permit compatibility with old numeric entities that +// assumed Windows-1252 encoding. +// https://html.spec.whatwg.org/multipage/syntax.html#consume-a-character-reference +var replacementTable = [...]rune{ + '\u20AC', // First entry is what 0x80 should be replaced with. + '\u0081', + '\u201A', + '\u0192', + '\u201E', + '\u2026', + '\u2020', + '\u2021', + '\u02C6', + '\u2030', + '\u0160', + '\u2039', + '\u0152', + '\u008D', + '\u017D', + '\u008F', + '\u0090', + '\u2018', + '\u2019', + '\u201C', + '\u201D', + '\u2022', + '\u2013', + '\u2014', + '\u02DC', + '\u2122', + '\u0161', + '\u203A', + '\u0153', + '\u009D', + '\u017E', + '\u0178', // Last entry is 0x9F. + // 0x00->'\uFFFD' is handled programmatically. + // 0x0D->'\u000D' is a no-op. +} + +// unescapeEntity reads an entity like "<" from b[src:] and writes the +// corresponding "<" to b[dst:], returning the incremented dst and src cursors. +// Precondition: b[src] == '&' && dst <= src. +// attribute should be true if parsing an attribute value. +func unescapeEntity(b []byte, dst, src int, attribute bool) (dst1, src1 int) { + // https://html.spec.whatwg.org/multipage/syntax.html#consume-a-character-reference + + // i starts at 1 because we already know that s[0] == '&'. + i, s := 1, b[src:] + + if len(s) <= 1 { + b[dst] = b[src] + return dst + 1, src + 1 + } + + if s[i] == '#' { + if len(s) <= 3 { // We need to have at least "&#.". + b[dst] = b[src] + return dst + 1, src + 1 + } + i++ + c := s[i] + hex := false + if c == 'x' || c == 'X' { + hex = true + i++ + } + + x := '\x00' + for i < len(s) { + c = s[i] + i++ + if hex { + if '0' <= c && c <= '9' { + x = 16*x + rune(c) - '0' + continue + } else if 'a' <= c && c <= 'f' { + x = 16*x + rune(c) - 'a' + 10 + continue + } else if 'A' <= c && c <= 'F' { + x = 16*x + rune(c) - 'A' + 10 + continue + } + } else if '0' <= c && c <= '9' { + x = 10*x + rune(c) - '0' + continue + } + if c != ';' { + i-- + } + break + } + + if i <= 3 { // No characters matched. + b[dst] = b[src] + return dst + 1, src + 1 + } + + if 0x80 <= x && x <= 0x9F { + // Replace characters from Windows-1252 with UTF-8 equivalents. + x = replacementTable[x-0x80] + } else if x == 0 || (0xD800 <= x && x <= 0xDFFF) || x > 0x10FFFF { + // Replace invalid characters with the replacement character. + x = '\uFFFD' + } + + return dst + utf8.EncodeRune(b[dst:], x), src + i + } + + // Consume the maximum number of characters possible, with the + // consumed characters matching one of the named references. + + for i < len(s) { + c := s[i] + i++ + // Lower-cased characters are more common in entities, so we check for them first. + if 'a' <= c && c <= 'z' || 'A' <= c && c <= 'Z' || '0' <= c && c <= '9' { + continue + } + if c != ';' { + i-- + } + break + } + + entityName := string(s[1:i]) + if entityName == "" { + // No-op. + } else if attribute && entityName[len(entityName)-1] != ';' && len(s) > i && s[i] == '=' { + // No-op. + } else if x := entity[entityName]; x != 0 { + return dst + utf8.EncodeRune(b[dst:], x), src + i + } else if x := entity2[entityName]; x[0] != 0 { + dst1 := dst + utf8.EncodeRune(b[dst:], x[0]) + return dst1 + utf8.EncodeRune(b[dst1:], x[1]), src + i + } else if !attribute { + maxLen := len(entityName) - 1 + if maxLen > longestEntityWithoutSemicolon { + maxLen = longestEntityWithoutSemicolon + } + for j := maxLen; j > 1; j-- { + if x := entity[entityName[:j]]; x != 0 { + return dst + utf8.EncodeRune(b[dst:], x), src + j + 1 + } + } + } + + dst1, src1 = dst+i, src+i + copy(b[dst:dst1], b[src:src1]) + return dst1, src1 +} + +// unescape unescapes b's entities in-place, so that "a<b" becomes "a' byte that, per above, we'd like to avoid escaping unless we have to. +// +// Studying the summary table (and T actions in its '>' column) closely, we +// only need to escape in states 43, 44, 49, 51 and 52. State 43 is at the +// start of the comment data. State 52 is after a '!'. The other three states +// are after a '-'. +// +// Our algorithm is thus to escape every '&' and to escape '>' if and only if: +// - The '>' is after a '!' or '-' (in the unescaped data) or +// - The '>' is at the start of the comment data (after the opening ""); err != nil { + return err + } + return nil + case DoctypeNode: + if _, err := w.WriteString("') + case RawNode: + _, err := w.WriteString(n.Data) + return err + default: + return errors.New("html: unknown node type") + } + + // Render the opening tag. + if err := w.WriteByte('<'); err != nil { + return err + } + if _, err := w.WriteString(n.Data); err != nil { + return err + } + for _, a := range n.Attr { + if err := w.WriteByte(' '); err != nil { + return err + } + if a.Namespace != "" { + if _, err := w.WriteString(a.Namespace); err != nil { + return err + } + if err := w.WriteByte(':'); err != nil { + return err + } + } + if _, err := w.WriteString(a.Key); err != nil { + return err + } + if _, err := w.WriteString(`="`); err != nil { + return err + } + if err := escape(w, a.Val); err != nil { + return err + } + if err := w.WriteByte('"'); err != nil { + return err + } + } + if voidElements[n.Data] { + if n.FirstChild != nil { + return fmt.Errorf("html: void element <%s> has child nodes", n.Data) + } + _, err := w.WriteString("/>") + return err + } + if err := w.WriteByte('>'); err != nil { + return err + } + + // Add initial newline where there is danger of a newline beging ignored. + if c := n.FirstChild; c != nil && c.Type == TextNode && strings.HasPrefix(c.Data, "\n") { + switch n.Data { + case "pre", "listing", "textarea": + if err := w.WriteByte('\n'); err != nil { + return err + } + } + } + + // Render any child nodes + if childTextNodesAreLiteral(n) { + for c := n.FirstChild; c != nil; c = c.NextSibling { + if c.Type == TextNode { + if _, err := w.WriteString(c.Data); err != nil { + return err + } + } else { + if err := render1(w, c); err != nil { + return err + } + } + } + if n.Data == "plaintext" { + // Don't render anything else. must be the + // last element in the file, with no closing tag. + return plaintextAbort + } + } else { + for c := n.FirstChild; c != nil; c = c.NextSibling { + if err := render1(w, c); err != nil { + return err + } + } + } + + // Render the </xxx> closing tag. + if _, err := w.WriteString("</"); err != nil { + return err + } + if _, err := w.WriteString(n.Data); err != nil { + return err + } + return w.WriteByte('>') +} + +func childTextNodesAreLiteral(n *Node) bool { + // Per WHATWG HTML 13.3, if the parent of the current node is a style, + // script, xmp, iframe, noembed, noframes, or plaintext element, and the + // current node is a text node, append the value of the node's data + // literally. The specification is not explicit about it, but we only + // enforce this if we are in the HTML namespace (i.e. when the namespace is + // ""). + // NOTE: we also always include noscript elements, although the + // specification states that they should only be rendered as such if + // scripting is enabled for the node (which is not something we track). + if n.Namespace != "" { + return false + } + switch n.Data { + case "iframe", "noembed", "noframes", "noscript", "plaintext", "script", "style", "xmp": + return true + default: + return false + } +} + +// writeQuoted writes s to w surrounded by quotes. Normally it will use double +// quotes, but if s contains a double quote, it will use single quotes. +// It is used for writing the identifiers in a doctype declaration. +// In valid HTML, they can't contain both types of quotes. +func writeQuoted(w writer, s string) error { + var q byte = '"' + if strings.Contains(s, `"`) { + q = '\'' + } + if err := w.WriteByte(q); err != nil { + return err + } + if _, err := w.WriteString(s); err != nil { + return err + } + if err := w.WriteByte(q); err != nil { + return err + } + return nil +} + +// Section 12.1.2, "Elements", gives this list of void elements. Void elements +// are those that can't have any contents. +var voidElements = map[string]bool{ + "area": true, + "base": true, + "br": true, + "col": true, + "embed": true, + "hr": true, + "img": true, + "input": true, + "keygen": true, // "keygen" has been removed from the spec, but are kept here for backwards compatibility. + "link": true, + "meta": true, + "param": true, + "source": true, + "track": true, + "wbr": true, +} diff --git a/vendor/golang.org/x/net/html/token.go b/vendor/golang.org/x/net/html/token.go new file mode 100644 index 0000000..3c57880 --- /dev/null +++ b/vendor/golang.org/x/net/html/token.go @@ -0,0 +1,1272 @@ +// Copyright 2010 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package html + +import ( + "bytes" + "errors" + "io" + "strconv" + "strings" + + "golang.org/x/net/html/atom" +) + +// A TokenType is the type of a Token. +type TokenType uint32 + +const ( + // ErrorToken means that an error occurred during tokenization. + ErrorToken TokenType = iota + // TextToken means a text node. + TextToken + // A StartTagToken looks like <a>. + StartTagToken + // An EndTagToken looks like </a>. + EndTagToken + // A SelfClosingTagToken tag looks like <br/>. + SelfClosingTagToken + // A CommentToken looks like <!--x-->. + CommentToken + // A DoctypeToken looks like <!DOCTYPE x> + DoctypeToken +) + +// ErrBufferExceeded means that the buffering limit was exceeded. +var ErrBufferExceeded = errors.New("max buffer exceeded") + +// String returns a string representation of the TokenType. +func (t TokenType) String() string { + switch t { + case ErrorToken: + return "Error" + case TextToken: + return "Text" + case StartTagToken: + return "StartTag" + case EndTagToken: + return "EndTag" + case SelfClosingTagToken: + return "SelfClosingTag" + case CommentToken: + return "Comment" + case DoctypeToken: + return "Doctype" + } + return "Invalid(" + strconv.Itoa(int(t)) + ")" +} + +// An Attribute is an attribute namespace-key-value triple. Namespace is +// non-empty for foreign attributes like xlink, Key is alphabetic (and hence +// does not contain escapable characters like '&', '<' or '>'), and Val is +// unescaped (it looks like "a<b" rather than "a&lt;b"). +// +// Namespace is only used by the parser, not the tokenizer. +type Attribute struct { + Namespace, Key, Val string +} + +// A Token consists of a TokenType and some Data (tag name for start and end +// tags, content for text, comments and doctypes). A tag Token may also contain +// a slice of Attributes. Data is unescaped for all Tokens (it looks like "a<b" +// rather than "a&lt;b"). For tag Tokens, DataAtom is the atom for Data, or +// zero if Data is not a known tag name. +type Token struct { + Type TokenType + DataAtom atom.Atom + Data string + Attr []Attribute +} + +// tagString returns a string representation of a tag Token's Data and Attr. +func (t Token) tagString() string { + if len(t.Attr) == 0 { + return t.Data + } + buf := bytes.NewBufferString(t.Data) + for _, a := range t.Attr { + buf.WriteByte(' ') + buf.WriteString(a.Key) + buf.WriteString(`="`) + escape(buf, a.Val) + buf.WriteByte('"') + } + return buf.String() +} + +// String returns a string representation of the Token. +func (t Token) String() string { + switch t.Type { + case ErrorToken: + return "" + case TextToken: + return EscapeString(t.Data) + case StartTagToken: + return "<" + t.tagString() + ">" + case EndTagToken: + return "</" + t.tagString() + ">" + case SelfClosingTagToken: + return "<" + t.tagString() + "/>" + case CommentToken: + return "<!--" + escapeCommentString(t.Data) + "-->" + case DoctypeToken: + return "<!DOCTYPE " + EscapeString(t.Data) + ">" + } + return "Invalid(" + strconv.Itoa(int(t.Type)) + ")" +} + +// span is a range of bytes in a Tokenizer's buffer. The start is inclusive, +// the end is exclusive. +type span struct { + start, end int +} + +// A Tokenizer returns a stream of HTML Tokens. +type Tokenizer struct { + // r is the source of the HTML text. + r io.Reader + // tt is the TokenType of the current token. + tt TokenType + // err is the first error encountered during tokenization. It is possible + // for tt != Error && err != nil to hold: this means that Next returned a + // valid token but the subsequent Next call will return an error token. + // For example, if the HTML text input was just "plain", then the first + // Next call would set z.err to io.EOF but return a TextToken, and all + // subsequent Next calls would return an ErrorToken. + // err is never reset. Once it becomes non-nil, it stays non-nil. + err error + // readErr is the error returned by the io.Reader r. It is separate from + // err because it is valid for an io.Reader to return (n int, err1 error) + // such that n > 0 && err1 != nil, and callers should always process the + // n > 0 bytes before considering the error err1. + readErr error + // buf[raw.start:raw.end] holds the raw bytes of the current token. + // buf[raw.end:] is buffered input that will yield future tokens. + raw span + buf []byte + // maxBuf limits the data buffered in buf. A value of 0 means unlimited. + maxBuf int + // buf[data.start:data.end] holds the raw bytes of the current token's data: + // a text token's text, a tag token's tag name, etc. + data span + // pendingAttr is the attribute key and value currently being tokenized. + // When complete, pendingAttr is pushed onto attr. nAttrReturned is + // incremented on each call to TagAttr. + pendingAttr [2]span + attr [][2]span + nAttrReturned int + // rawTag is the "script" in "</script>" that closes the next token. If + // non-empty, the subsequent call to Next will return a raw or RCDATA text + // token: one that treats "<p>" as text instead of an element. + // rawTag's contents are lower-cased. + rawTag string + // textIsRaw is whether the current text token's data is not escaped. + textIsRaw bool + // convertNUL is whether NUL bytes in the current token's data should + // be converted into \ufffd replacement characters. + convertNUL bool + // allowCDATA is whether CDATA sections are allowed in the current context. + allowCDATA bool +} + +// AllowCDATA sets whether or not the tokenizer recognizes <![CDATA[foo]]> as +// the text "foo". The default value is false, which means to recognize it as +// a bogus comment "<!-- [CDATA[foo]] -->" instead. +// +// Strictly speaking, an HTML5 compliant tokenizer should allow CDATA if and +// only if tokenizing foreign content, such as MathML and SVG. However, +// tracking foreign-contentness is difficult to do purely in the tokenizer, +// as opposed to the parser, due to HTML integration points: an <svg> element +// can contain a <foreignObject> that is foreign-to-SVG but not foreign-to- +// HTML. For strict compliance with the HTML5 tokenization algorithm, it is the +// responsibility of the user of a tokenizer to call AllowCDATA as appropriate. +// In practice, if using the tokenizer without caring whether MathML or SVG +// CDATA is text or comments, such as tokenizing HTML to find all the anchor +// text, it is acceptable to ignore this responsibility. +func (z *Tokenizer) AllowCDATA(allowCDATA bool) { + z.allowCDATA = allowCDATA +} + +// NextIsNotRawText instructs the tokenizer that the next token should not be +// considered as 'raw text'. Some elements, such as script and title elements, +// normally require the next token after the opening tag to be 'raw text' that +// has no child elements. For example, tokenizing "<title>a<b>c</b>d</title>" +// yields a start tag token for "<title>", a text token for "a<b>c</b>d", and +// an end tag token for "</title>". There are no distinct start tag or end tag +// tokens for the "<b>" and "</b>". +// +// This tokenizer implementation will generally look for raw text at the right +// times. Strictly speaking, an HTML5 compliant tokenizer should not look for +// raw text if in foreign content: <title> generally needs raw text, but a +// <title> inside an <svg> does not. Another example is that a <textarea> +// generally needs raw text, but a <textarea> is not allowed as an immediate +// child of a <select>; in normal parsing, a <textarea> implies </select>, but +// one cannot close the implicit element when parsing a <select>'s InnerHTML. +// Similarly to AllowCDATA, tracking the correct moment to override raw-text- +// ness is difficult to do purely in the tokenizer, as opposed to the parser. +// For strict compliance with the HTML5 tokenization algorithm, it is the +// responsibility of the user of a tokenizer to call NextIsNotRawText as +// appropriate. In practice, like AllowCDATA, it is acceptable to ignore this +// responsibility for basic usage. +// +// Note that this 'raw text' concept is different from the one offered by the +// Tokenizer.Raw method. +func (z *Tokenizer) NextIsNotRawText() { + z.rawTag = "" +} + +// Err returns the error associated with the most recent ErrorToken token. +// This is typically io.EOF, meaning the end of tokenization. +func (z *Tokenizer) Err() error { + if z.tt != ErrorToken { + return nil + } + return z.err +} + +// readByte returns the next byte from the input stream, doing a buffered read +// from z.r into z.buf if necessary. z.buf[z.raw.start:z.raw.end] remains a contiguous byte +// slice that holds all the bytes read so far for the current token. +// It sets z.err if the underlying reader returns an error. +// Pre-condition: z.err == nil. +func (z *Tokenizer) readByte() byte { + if z.raw.end >= len(z.buf) { + // Our buffer is exhausted and we have to read from z.r. Check if the + // previous read resulted in an error. + if z.readErr != nil { + z.err = z.readErr + return 0 + } + // We copy z.buf[z.raw.start:z.raw.end] to the beginning of z.buf. If the length + // z.raw.end - z.raw.start is more than half the capacity of z.buf, then we + // allocate a new buffer before the copy. + c := cap(z.buf) + d := z.raw.end - z.raw.start + var buf1 []byte + if 2*d > c { + buf1 = make([]byte, d, 2*c) + } else { + buf1 = z.buf[:d] + } + copy(buf1, z.buf[z.raw.start:z.raw.end]) + if x := z.raw.start; x != 0 { + // Adjust the data/attr spans to refer to the same contents after the copy. + z.data.start -= x + z.data.end -= x + z.pendingAttr[0].start -= x + z.pendingAttr[0].end -= x + z.pendingAttr[1].start -= x + z.pendingAttr[1].end -= x + for i := range z.attr { + z.attr[i][0].start -= x + z.attr[i][0].end -= x + z.attr[i][1].start -= x + z.attr[i][1].end -= x + } + } + z.raw.start, z.raw.end, z.buf = 0, d, buf1[:d] + // Now that we have copied the live bytes to the start of the buffer, + // we read from z.r into the remainder. + var n int + n, z.readErr = readAtLeastOneByte(z.r, buf1[d:cap(buf1)]) + if n == 0 { + z.err = z.readErr + return 0 + } + z.buf = buf1[:d+n] + } + x := z.buf[z.raw.end] + z.raw.end++ + if z.maxBuf > 0 && z.raw.end-z.raw.start >= z.maxBuf { + z.err = ErrBufferExceeded + return 0 + } + return x +} + +// Buffered returns a slice containing data buffered but not yet tokenized. +func (z *Tokenizer) Buffered() []byte { + return z.buf[z.raw.end:] +} + +// readAtLeastOneByte wraps an io.Reader so that reading cannot return (0, nil). +// It returns io.ErrNoProgress if the underlying r.Read method returns (0, nil) +// too many times in succession. +func readAtLeastOneByte(r io.Reader, b []byte) (int, error) { + for i := 0; i < 100; i++ { + if n, err := r.Read(b); n != 0 || err != nil { + return n, err + } + } + return 0, io.ErrNoProgress +} + +// skipWhiteSpace skips past any white space. +func (z *Tokenizer) skipWhiteSpace() { + if z.err != nil { + return + } + for { + c := z.readByte() + if z.err != nil { + return + } + switch c { + case ' ', '\n', '\r', '\t', '\f': + // No-op. + default: + z.raw.end-- + return + } + } +} + +// readRawOrRCDATA reads until the next "</foo>", where "foo" is z.rawTag and +// is typically something like "script" or "textarea". +func (z *Tokenizer) readRawOrRCDATA() { + if z.rawTag == "script" { + z.readScript() + z.textIsRaw = true + z.rawTag = "" + return + } +loop: + for { + c := z.readByte() + if z.err != nil { + break loop + } + if c != '<' { + continue loop + } + c = z.readByte() + if z.err != nil { + break loop + } + if c != '/' { + z.raw.end-- + continue loop + } + if z.readRawEndTag() || z.err != nil { + break loop + } + } + z.data.end = z.raw.end + // A textarea's or title's RCDATA can contain escaped entities. + z.textIsRaw = z.rawTag != "textarea" && z.rawTag != "title" + z.rawTag = "" +} + +// readRawEndTag attempts to read a tag like "</foo>", where "foo" is z.rawTag. +// If it succeeds, it backs up the input position to reconsume the tag and +// returns true. Otherwise it returns false. The opening "</" has already been +// consumed. +func (z *Tokenizer) readRawEndTag() bool { + for i := 0; i < len(z.rawTag); i++ { + c := z.readByte() + if z.err != nil { + return false + } + if c != z.rawTag[i] && c != z.rawTag[i]-('a'-'A') { + z.raw.end-- + return false + } + } + c := z.readByte() + if z.err != nil { + return false + } + switch c { + case ' ', '\n', '\r', '\t', '\f', '/', '>': + // The 3 is 2 for the leading "</" plus 1 for the trailing character c. + z.raw.end -= 3 + len(z.rawTag) + return true + } + z.raw.end-- + return false +} + +// readScript reads until the next </script> tag, following the byzantine +// rules for escaping/hiding the closing tag. +func (z *Tokenizer) readScript() { + defer func() { + z.data.end = z.raw.end + }() + var c byte + +scriptData: + c = z.readByte() + if z.err != nil { + return + } + if c == '<' { + goto scriptDataLessThanSign + } + goto scriptData + +scriptDataLessThanSign: + c = z.readByte() + if z.err != nil { + return + } + switch c { + case '/': + goto scriptDataEndTagOpen + case '!': + goto scriptDataEscapeStart + } + z.raw.end-- + goto scriptData + +scriptDataEndTagOpen: + if z.readRawEndTag() || z.err != nil { + return + } + goto scriptData + +scriptDataEscapeStart: + c = z.readByte() + if z.err != nil { + return + } + if c == '-' { + goto scriptDataEscapeStartDash + } + z.raw.end-- + goto scriptData + +scriptDataEscapeStartDash: + c = z.readByte() + if z.err != nil { + return + } + if c == '-' { + goto scriptDataEscapedDashDash + } + z.raw.end-- + goto scriptData + +scriptDataEscaped: + c = z.readByte() + if z.err != nil { + return + } + switch c { + case '-': + goto scriptDataEscapedDash + case '<': + goto scriptDataEscapedLessThanSign + } + goto scriptDataEscaped + +scriptDataEscapedDash: + c = z.readByte() + if z.err != nil { + return + } + switch c { + case '-': + goto scriptDataEscapedDashDash + case '<': + goto scriptDataEscapedLessThanSign + } + goto scriptDataEscaped + +scriptDataEscapedDashDash: + c = z.readByte() + if z.err != nil { + return + } + switch c { + case '-': + goto scriptDataEscapedDashDash + case '<': + goto scriptDataEscapedLessThanSign + case '>': + goto scriptData + } + goto scriptDataEscaped + +scriptDataEscapedLessThanSign: + c = z.readByte() + if z.err != nil { + return + } + if c == '/' { + goto scriptDataEscapedEndTagOpen + } + if 'a' <= c && c <= 'z' || 'A' <= c && c <= 'Z' { + goto scriptDataDoubleEscapeStart + } + z.raw.end-- + goto scriptData + +scriptDataEscapedEndTagOpen: + if z.readRawEndTag() || z.err != nil { + return + } + goto scriptDataEscaped + +scriptDataDoubleEscapeStart: + z.raw.end-- + for i := 0; i < len("script"); i++ { + c = z.readByte() + if z.err != nil { + return + } + if c != "script"[i] && c != "SCRIPT"[i] { + z.raw.end-- + goto scriptDataEscaped + } + } + c = z.readByte() + if z.err != nil { + return + } + switch c { + case ' ', '\n', '\r', '\t', '\f', '/', '>': + goto scriptDataDoubleEscaped + } + z.raw.end-- + goto scriptDataEscaped + +scriptDataDoubleEscaped: + c = z.readByte() + if z.err != nil { + return + } + switch c { + case '-': + goto scriptDataDoubleEscapedDash + case '<': + goto scriptDataDoubleEscapedLessThanSign + } + goto scriptDataDoubleEscaped + +scriptDataDoubleEscapedDash: + c = z.readByte() + if z.err != nil { + return + } + switch c { + case '-': + goto scriptDataDoubleEscapedDashDash + case '<': + goto scriptDataDoubleEscapedLessThanSign + } + goto scriptDataDoubleEscaped + +scriptDataDoubleEscapedDashDash: + c = z.readByte() + if z.err != nil { + return + } + switch c { + case '-': + goto scriptDataDoubleEscapedDashDash + case '<': + goto scriptDataDoubleEscapedLessThanSign + case '>': + goto scriptData + } + goto scriptDataDoubleEscaped + +scriptDataDoubleEscapedLessThanSign: + c = z.readByte() + if z.err != nil { + return + } + if c == '/' { + goto scriptDataDoubleEscapeEnd + } + z.raw.end-- + goto scriptDataDoubleEscaped + +scriptDataDoubleEscapeEnd: + if z.readRawEndTag() { + z.raw.end += len("</script>") + goto scriptDataEscaped + } + if z.err != nil { + return + } + goto scriptDataDoubleEscaped +} + +// readComment reads the next comment token starting with "<!--". The opening +// "<!--" has already been consumed. +func (z *Tokenizer) readComment() { + // When modifying this function, consider manually increasing the + // maxSuffixLen constant in func TestComments, from 6 to e.g. 9 or more. + // That increase should only be temporary, not committed, as it + // exponentially affects the test running time. + + z.data.start = z.raw.end + defer func() { + if z.data.end < z.data.start { + // It's a comment with no data, like <!-->. + z.data.end = z.data.start + } + }() + + var dashCount int + beginning := true + for { + c := z.readByte() + if z.err != nil { + z.data.end = z.calculateAbruptCommentDataEnd() + return + } + switch c { + case '-': + dashCount++ + continue + case '>': + if dashCount >= 2 || beginning { + z.data.end = z.raw.end - len("-->") + return + } + case '!': + if dashCount >= 2 { + c = z.readByte() + if z.err != nil { + z.data.end = z.calculateAbruptCommentDataEnd() + return + } else if c == '>' { + z.data.end = z.raw.end - len("--!>") + return + } else if c == '-' { + dashCount = 1 + beginning = false + continue + } + } + } + dashCount = 0 + beginning = false + } +} + +func (z *Tokenizer) calculateAbruptCommentDataEnd() int { + raw := z.Raw() + const prefixLen = len("<!--") + if len(raw) >= prefixLen { + raw = raw[prefixLen:] + if hasSuffix(raw, "--!") { + return z.raw.end - 3 + } else if hasSuffix(raw, "--") { + return z.raw.end - 2 + } else if hasSuffix(raw, "-") { + return z.raw.end - 1 + } + } + return z.raw.end +} + +func hasSuffix(b []byte, suffix string) bool { + if len(b) < len(suffix) { + return false + } + b = b[len(b)-len(suffix):] + for i := range b { + if b[i] != suffix[i] { + return false + } + } + return true +} + +// readUntilCloseAngle reads until the next ">". +func (z *Tokenizer) readUntilCloseAngle() { + z.data.start = z.raw.end + for { + c := z.readByte() + if z.err != nil { + z.data.end = z.raw.end + return + } + if c == '>' { + z.data.end = z.raw.end - len(">") + return + } + } +} + +// readMarkupDeclaration reads the next token starting with "<!". It might be +// a "<!--comment-->", a "<!DOCTYPE foo>", a "<![CDATA[section]]>" or +// "<!a bogus comment". The opening "<!" has already been consumed. +func (z *Tokenizer) readMarkupDeclaration() TokenType { + z.data.start = z.raw.end + var c [2]byte + for i := 0; i < 2; i++ { + c[i] = z.readByte() + if z.err != nil { + z.data.end = z.raw.end + return CommentToken + } + } + if c[0] == '-' && c[1] == '-' { + z.readComment() + return CommentToken + } + z.raw.end -= 2 + if z.readDoctype() { + return DoctypeToken + } + if z.allowCDATA && z.readCDATA() { + z.convertNUL = true + return TextToken + } + // It's a bogus comment. + z.readUntilCloseAngle() + return CommentToken +} + +// readDoctype attempts to read a doctype declaration and returns true if +// successful. The opening "<!" has already been consumed. +func (z *Tokenizer) readDoctype() bool { + const s = "DOCTYPE" + for i := 0; i < len(s); i++ { + c := z.readByte() + if z.err != nil { + z.data.end = z.raw.end + return false + } + if c != s[i] && c != s[i]+('a'-'A') { + // Back up to read the fragment of "DOCTYPE" again. + z.raw.end = z.data.start + return false + } + } + if z.skipWhiteSpace(); z.err != nil { + z.data.start = z.raw.end + z.data.end = z.raw.end + return true + } + z.readUntilCloseAngle() + return true +} + +// readCDATA attempts to read a CDATA section and returns true if +// successful. The opening "<!" has already been consumed. +func (z *Tokenizer) readCDATA() bool { + const s = "[CDATA[" + for i := 0; i < len(s); i++ { + c := z.readByte() + if z.err != nil { + z.data.end = z.raw.end + return false + } + if c != s[i] { + // Back up to read the fragment of "[CDATA[" again. + z.raw.end = z.data.start + return false + } + } + z.data.start = z.raw.end + brackets := 0 + for { + c := z.readByte() + if z.err != nil { + z.data.end = z.raw.end + return true + } + switch c { + case ']': + brackets++ + case '>': + if brackets >= 2 { + z.data.end = z.raw.end - len("]]>") + return true + } + brackets = 0 + default: + brackets = 0 + } + } +} + +// startTagIn returns whether the start tag in z.buf[z.data.start:z.data.end] +// case-insensitively matches any element of ss. +func (z *Tokenizer) startTagIn(ss ...string) bool { +loop: + for _, s := range ss { + if z.data.end-z.data.start != len(s) { + continue loop + } + for i := 0; i < len(s); i++ { + c := z.buf[z.data.start+i] + if 'A' <= c && c <= 'Z' { + c += 'a' - 'A' + } + if c != s[i] { + continue loop + } + } + return true + } + return false +} + +// readStartTag reads the next start tag token. The opening "<a" has already +// been consumed, where 'a' means anything in [A-Za-z]. +func (z *Tokenizer) readStartTag() TokenType { + z.readTag(true) + if z.err != nil { + return ErrorToken + } + // Several tags flag the tokenizer's next token as raw. + c, raw := z.buf[z.data.start], false + if 'A' <= c && c <= 'Z' { + c += 'a' - 'A' + } + switch c { + case 'i': + raw = z.startTagIn("iframe") + case 'n': + raw = z.startTagIn("noembed", "noframes", "noscript") + case 'p': + raw = z.startTagIn("plaintext") + case 's': + raw = z.startTagIn("script", "style") + case 't': + raw = z.startTagIn("textarea", "title") + case 'x': + raw = z.startTagIn("xmp") + } + if raw { + z.rawTag = strings.ToLower(string(z.buf[z.data.start:z.data.end])) + } + // Look for a self-closing token like "<br/>". + if z.err == nil && z.buf[z.raw.end-2] == '/' { + return SelfClosingTagToken + } + return StartTagToken +} + +// readTag reads the next tag token and its attributes. If saveAttr, those +// attributes are saved in z.attr, otherwise z.attr is set to an empty slice. +// The opening "<a" or "</a" has already been consumed, where 'a' means anything +// in [A-Za-z]. +func (z *Tokenizer) readTag(saveAttr bool) { + z.attr = z.attr[:0] + z.nAttrReturned = 0 + // Read the tag name and attribute key/value pairs. + z.readTagName() + if z.skipWhiteSpace(); z.err != nil { + return + } + for { + c := z.readByte() + if z.err != nil || c == '>' { + break + } + z.raw.end-- + z.readTagAttrKey() + z.readTagAttrVal() + // Save pendingAttr if saveAttr and that attribute has a non-empty key. + if saveAttr && z.pendingAttr[0].start != z.pendingAttr[0].end { + z.attr = append(z.attr, z.pendingAttr) + } + if z.skipWhiteSpace(); z.err != nil { + break + } + } +} + +// readTagName sets z.data to the "div" in "<div k=v>". The reader (z.raw.end) +// is positioned such that the first byte of the tag name (the "d" in "<div") +// has already been consumed. +func (z *Tokenizer) readTagName() { + z.data.start = z.raw.end - 1 + for { + c := z.readByte() + if z.err != nil { + z.data.end = z.raw.end + return + } + switch c { + case ' ', '\n', '\r', '\t', '\f': + z.data.end = z.raw.end - 1 + return + case '/', '>': + z.raw.end-- + z.data.end = z.raw.end + return + } + } +} + +// readTagAttrKey sets z.pendingAttr[0] to the "k" in "<div k=v>". +// Precondition: z.err == nil. +func (z *Tokenizer) readTagAttrKey() { + z.pendingAttr[0].start = z.raw.end + for { + c := z.readByte() + if z.err != nil { + z.pendingAttr[0].end = z.raw.end + return + } + switch c { + case '=': + if z.pendingAttr[0].start+1 == z.raw.end { + // WHATWG 13.2.5.32, if we see an equals sign before the attribute name + // begins, we treat it as a character in the attribute name and continue. + continue + } + fallthrough + case ' ', '\n', '\r', '\t', '\f', '/', '>': + // WHATWG 13.2.5.33 Attribute name state + // We need to reconsume the char in the after attribute name state to support the / character + z.raw.end-- + z.pendingAttr[0].end = z.raw.end + return + } + } +} + +// readTagAttrVal sets z.pendingAttr[1] to the "v" in "<div k=v>". +func (z *Tokenizer) readTagAttrVal() { + z.pendingAttr[1].start = z.raw.end + z.pendingAttr[1].end = z.raw.end + if z.skipWhiteSpace(); z.err != nil { + return + } + c := z.readByte() + if z.err != nil { + return + } + if c == '/' { + // WHATWG 13.2.5.34 After attribute name state + // U+002F SOLIDUS (/) - Switch to the self-closing start tag state. + return + } + if c != '=' { + z.raw.end-- + return + } + if z.skipWhiteSpace(); z.err != nil { + return + } + quote := z.readByte() + if z.err != nil { + return + } + switch quote { + case '>': + z.raw.end-- + return + + case '\'', '"': + z.pendingAttr[1].start = z.raw.end + for { + c := z.readByte() + if z.err != nil { + z.pendingAttr[1].end = z.raw.end + return + } + if c == quote { + z.pendingAttr[1].end = z.raw.end - 1 + return + } + } + + default: + z.pendingAttr[1].start = z.raw.end - 1 + for { + c := z.readByte() + if z.err != nil { + z.pendingAttr[1].end = z.raw.end + return + } + switch c { + case ' ', '\n', '\r', '\t', '\f': + z.pendingAttr[1].end = z.raw.end - 1 + return + case '>': + z.raw.end-- + z.pendingAttr[1].end = z.raw.end + return + } + } + } +} + +// Next scans the next token and returns its type. +func (z *Tokenizer) Next() TokenType { + z.raw.start = z.raw.end + z.data.start = z.raw.end + z.data.end = z.raw.end + if z.err != nil { + z.tt = ErrorToken + return z.tt + } + if z.rawTag != "" { + if z.rawTag == "plaintext" { + // Read everything up to EOF. + for z.err == nil { + z.readByte() + } + z.data.end = z.raw.end + z.textIsRaw = true + } else { + z.readRawOrRCDATA() + } + if z.data.end > z.data.start { + z.tt = TextToken + z.convertNUL = true + return z.tt + } + } + z.textIsRaw = false + z.convertNUL = false + +loop: + for { + c := z.readByte() + if z.err != nil { + break loop + } + if c != '<' { + continue loop + } + + // Check if the '<' we have just read is part of a tag, comment + // or doctype. If not, it's part of the accumulated text token. + c = z.readByte() + if z.err != nil { + break loop + } + var tokenType TokenType + switch { + case 'a' <= c && c <= 'z' || 'A' <= c && c <= 'Z': + tokenType = StartTagToken + case c == '/': + tokenType = EndTagToken + case c == '!' || c == '?': + // We use CommentToken to mean any of "<!--actual comments-->", + // "<!DOCTYPE declarations>" and "<?xml processing instructions?>". + tokenType = CommentToken + default: + // Reconsume the current character. + z.raw.end-- + continue + } + + // We have a non-text token, but we might have accumulated some text + // before that. If so, we return the text first, and return the non- + // text token on the subsequent call to Next. + if x := z.raw.end - len("<a"); z.raw.start < x { + z.raw.end = x + z.data.end = x + z.tt = TextToken + return z.tt + } + switch tokenType { + case StartTagToken: + z.tt = z.readStartTag() + return z.tt + case EndTagToken: + c = z.readByte() + if z.err != nil { + break loop + } + if c == '>' { + // "</>" does not generate a token at all. Generate an empty comment + // to allow passthrough clients to pick up the data using Raw. + // Reset the tokenizer state and start again. + z.tt = CommentToken + return z.tt + } + if 'a' <= c && c <= 'z' || 'A' <= c && c <= 'Z' { + z.readTag(false) + if z.err != nil { + z.tt = ErrorToken + } else { + z.tt = EndTagToken + } + return z.tt + } + z.raw.end-- + z.readUntilCloseAngle() + z.tt = CommentToken + return z.tt + case CommentToken: + if c == '!' { + z.tt = z.readMarkupDeclaration() + return z.tt + } + z.raw.end-- + z.readUntilCloseAngle() + z.tt = CommentToken + return z.tt + } + } + if z.raw.start < z.raw.end { + z.data.end = z.raw.end + z.tt = TextToken + return z.tt + } + z.tt = ErrorToken + return z.tt +} + +// Raw returns the unmodified text of the current token. Calling Next, Token, +// Text, TagName or TagAttr may change the contents of the returned slice. +// +// The token stream's raw bytes partition the byte stream (up until an +// ErrorToken). There are no overlaps or gaps between two consecutive token's +// raw bytes. One implication is that the byte offset of the current token is +// the sum of the lengths of all previous tokens' raw bytes. +func (z *Tokenizer) Raw() []byte { + return z.buf[z.raw.start:z.raw.end] +} + +// convertNewlines converts "\r" and "\r\n" in s to "\n". +// The conversion happens in place, but the resulting slice may be shorter. +func convertNewlines(s []byte) []byte { + for i, c := range s { + if c != '\r' { + continue + } + + src := i + 1 + if src >= len(s) || s[src] != '\n' { + s[i] = '\n' + continue + } + + dst := i + for src < len(s) { + if s[src] == '\r' { + if src+1 < len(s) && s[src+1] == '\n' { + src++ + } + s[dst] = '\n' + } else { + s[dst] = s[src] + } + src++ + dst++ + } + return s[:dst] + } + return s +} + +var ( + nul = []byte("\x00") + replacement = []byte("\ufffd") +) + +// Text returns the unescaped text of a text, comment or doctype token. The +// contents of the returned slice may change on the next call to Next. +func (z *Tokenizer) Text() []byte { + switch z.tt { + case TextToken, CommentToken, DoctypeToken: + s := z.buf[z.data.start:z.data.end] + z.data.start = z.raw.end + z.data.end = z.raw.end + s = convertNewlines(s) + if (z.convertNUL || z.tt == CommentToken) && bytes.Contains(s, nul) { + s = bytes.Replace(s, nul, replacement, -1) + } + if !z.textIsRaw { + s = unescape(s, false) + } + return s + } + return nil +} + +// TagName returns the lower-cased name of a tag token (the `img` out of +// `<IMG SRC="foo">`) and whether the tag has attributes. +// The contents of the returned slice may change on the next call to Next. +func (z *Tokenizer) TagName() (name []byte, hasAttr bool) { + if z.data.start < z.data.end { + switch z.tt { + case StartTagToken, EndTagToken, SelfClosingTagToken: + s := z.buf[z.data.start:z.data.end] + z.data.start = z.raw.end + z.data.end = z.raw.end + return lower(s), z.nAttrReturned < len(z.attr) + } + } + return nil, false +} + +// TagAttr returns the lower-cased key and unescaped value of the next unparsed +// attribute for the current tag token and whether there are more attributes. +// The contents of the returned slices may change on the next call to Next. +func (z *Tokenizer) TagAttr() (key, val []byte, moreAttr bool) { + if z.nAttrReturned < len(z.attr) { + switch z.tt { + case StartTagToken, SelfClosingTagToken: + x := z.attr[z.nAttrReturned] + z.nAttrReturned++ + key = z.buf[x[0].start:x[0].end] + val = z.buf[x[1].start:x[1].end] + return lower(key), unescape(convertNewlines(val), true), z.nAttrReturned < len(z.attr) + } + } + return nil, nil, false +} + +// Token returns the current Token. The result's Data and Attr values remain +// valid after subsequent Next calls. +func (z *Tokenizer) Token() Token { + t := Token{Type: z.tt} + switch z.tt { + case TextToken, CommentToken, DoctypeToken: + t.Data = string(z.Text()) + case StartTagToken, SelfClosingTagToken, EndTagToken: + name, moreAttr := z.TagName() + for moreAttr { + var key, val []byte + key, val, moreAttr = z.TagAttr() + t.Attr = append(t.Attr, Attribute{"", atom.String(key), string(val)}) + } + if a := atom.Lookup(name); a != 0 { + t.DataAtom, t.Data = a, a.String() + } else { + t.DataAtom, t.Data = 0, string(name) + } + } + return t +} + +// SetMaxBuf sets a limit on the amount of data buffered during tokenization. +// A value of 0 means unlimited. +func (z *Tokenizer) SetMaxBuf(n int) { + z.maxBuf = n +} + +// NewTokenizer returns a new HTML Tokenizer for the given Reader. +// The input is assumed to be UTF-8 encoded. +func NewTokenizer(r io.Reader) *Tokenizer { + return NewTokenizerFragment(r, "") +} + +// NewTokenizerFragment returns a new HTML Tokenizer for the given Reader, for +// tokenizing an existing element's InnerHTML fragment. contextTag is that +// element's tag, such as "div" or "iframe". +// +// For example, how the InnerHTML "a<b" is tokenized depends on whether it is +// for a <p> tag or a <script> tag. +// +// The input is assumed to be UTF-8 encoded. +func NewTokenizerFragment(r io.Reader, contextTag string) *Tokenizer { + z := &Tokenizer{ + r: r, + buf: make([]byte, 0, 4096), + } + if contextTag != "" { + switch s := strings.ToLower(contextTag); s { + case "iframe", "noembed", "noframes", "noscript", "plaintext", "script", "style", "title", "textarea", "xmp": + z.rawTag = s + } + } + return z +} -- cgit 1.4.1