1 // Copyright 2011 The Go Authors. All rights reserved. 2 // Use of this source code is governed by a BSD-style 3 // license that can be found in the LICENSE file. 4 5 package template 6 7 import ( 8 "fmt" 9 ) 10 11 // context describes the state an HTML parser must be in when it reaches the 12 // portion of HTML produced by evaluating a particular template node. 13 // 14 // The zero value of type context is the start context for a template that 15 // produces an HTML fragment as defined at 16 // https://www.w3.org/TR/html5/syntax.html#the-end 17 // where the context element is null. 18 type context struct { 19 state state 20 delim delim 21 urlPart urlPart 22 jsCtx jsCtx 23 attr attr 24 element element 25 err *Error 26 } 27 28 func (c context) String() string { 29 var err error 30 if c.err != nil { 31 err = c.err 32 } 33 return fmt.Sprintf("{%v %v %v %v %v %v %v}", c.state, c.delim, c.urlPart, c.jsCtx, c.attr, c.element, err) 34 } 35 36 // eq reports whether two contexts are equal. 37 func (c context) eq(d context) bool { 38 return c.state == d.state && 39 c.delim == d.delim && 40 c.urlPart == d.urlPart && 41 c.jsCtx == d.jsCtx && 42 c.attr == d.attr && 43 c.element == d.element && 44 c.err == d.err 45 } 46 47 // mangle produces an identifier that includes a suffix that distinguishes it 48 // from template names mangled with different contexts. 49 func (c context) mangle(templateName string) string { 50 // The mangled name for the default context is the input templateName. 51 if c.state == stateText { 52 return templateName 53 } 54 s := templateName + "$htmltemplate_" + c.state.String() 55 if c.delim != delimNone { 56 s += "_" + c.delim.String() 57 } 58 if c.urlPart != urlPartNone { 59 s += "_" + c.urlPart.String() 60 } 61 if c.jsCtx != jsCtxRegexp { 62 s += "_" + c.jsCtx.String() 63 } 64 if c.attr != attrNone { 65 s += "_" + c.attr.String() 66 } 67 if c.element != elementNone { 68 s += "_" + c.element.String() 69 } 70 return s 71 } 72 73 // state describes a high-level HTML parser state. 74 // 75 // It bounds the top of the element stack, and by extension the HTML insertion 76 // mode, but also contains state that does not correspond to anything in the 77 // HTML5 parsing algorithm because a single token production in the HTML 78 // grammar may contain embedded actions in a template. For instance, the quoted 79 // HTML attribute produced by 80 // <div title="Hello {{.World}}"> 81 // is a single token in HTML's grammar but in a template spans several nodes. 82 type state uint8 83 84 //go:generate stringer -type state 85 86 const ( 87 // stateText is parsed character data. An HTML parser is in 88 // this state when its parse position is outside an HTML tag, 89 // directive, comment, and special element body. 90 stateText state = iota 91 // stateTag occurs before an HTML attribute or the end of a tag. 92 stateTag 93 // stateAttrName occurs inside an attribute name. 94 // It occurs between the ^'s in ` ^name^ = value`. 95 stateAttrName 96 // stateAfterName occurs after an attr name has ended but before any 97 // equals sign. It occurs between the ^'s in ` name^ ^= value`. 98 stateAfterName 99 // stateBeforeValue occurs after the equals sign but before the value. 100 // It occurs between the ^'s in ` name =^ ^value`. 101 stateBeforeValue 102 // stateHTMLCmt occurs inside an <!-- HTML comment -->. 103 stateHTMLCmt 104 // stateRCDATA occurs inside an RCDATA element (<textarea> or <title>) 105 // as described at https://www.w3.org/TR/html5/syntax.html#elements-0 106 stateRCDATA 107 // stateAttr occurs inside an HTML attribute whose content is text. 108 stateAttr 109 // stateURL occurs inside an HTML attribute whose content is a URL. 110 stateURL 111 // stateSrcset occurs inside an HTML srcset attribute. 112 stateSrcset 113 // stateJS occurs inside an event handler or script element. 114 stateJS 115 // stateJSDqStr occurs inside a JavaScript double quoted string. 116 stateJSDqStr 117 // stateJSSqStr occurs inside a JavaScript single quoted string. 118 stateJSSqStr 119 // stateJSRegexp occurs inside a JavaScript regexp literal. 120 stateJSRegexp 121 // stateJSBlockCmt occurs inside a JavaScript /* block comment */. 122 stateJSBlockCmt 123 // stateJSLineCmt occurs inside a JavaScript // line comment. 124 stateJSLineCmt 125 // stateCSS occurs inside a <style> element or style attribute. 126 stateCSS 127 // stateCSSDqStr occurs inside a CSS double quoted string. 128 stateCSSDqStr 129 // stateCSSSqStr occurs inside a CSS single quoted string. 130 stateCSSSqStr 131 // stateCSSDqURL occurs inside a CSS double quoted url("..."). 132 stateCSSDqURL 133 // stateCSSSqURL occurs inside a CSS single quoted url('...'). 134 stateCSSSqURL 135 // stateCSSURL occurs inside a CSS unquoted url(...). 136 stateCSSURL 137 // stateCSSBlockCmt occurs inside a CSS /* block comment */. 138 stateCSSBlockCmt 139 // stateCSSLineCmt occurs inside a CSS // line comment. 140 stateCSSLineCmt 141 // stateError is an infectious error state outside any valid 142 // HTML/CSS/JS construct. 143 stateError 144 ) 145 146 // isComment is true for any state that contains content meant for template 147 // authors & maintainers, not for end-users or machines. 148 func isComment(s state) bool { 149 switch s { 150 case stateHTMLCmt, stateJSBlockCmt, stateJSLineCmt, stateCSSBlockCmt, stateCSSLineCmt: 151 return true 152 } 153 return false 154 } 155 156 // isInTag return whether s occurs solely inside an HTML tag. 157 func isInTag(s state) bool { 158 switch s { 159 case stateTag, stateAttrName, stateAfterName, stateBeforeValue, stateAttr: 160 return true 161 } 162 return false 163 } 164 165 // delim is the delimiter that will end the current HTML attribute. 166 type delim uint8 167 168 //go:generate stringer -type delim 169 170 const ( 171 // delimNone occurs outside any attribute. 172 delimNone delim = iota 173 // delimDoubleQuote occurs when a double quote (") closes the attribute. 174 delimDoubleQuote 175 // delimSingleQuote occurs when a single quote (') closes the attribute. 176 delimSingleQuote 177 // delimSpaceOrTagEnd occurs when a space or right angle bracket (>) 178 // closes the attribute. 179 delimSpaceOrTagEnd 180 ) 181 182 // urlPart identifies a part in an RFC 3986 hierarchical URL to allow different 183 // encoding strategies. 184 type urlPart uint8 185 186 //go:generate stringer -type urlPart 187 188 const ( 189 // urlPartNone occurs when not in a URL, or possibly at the start: 190 // ^ in "^http://auth/path?k=v#frag". 191 urlPartNone urlPart = iota 192 // urlPartPreQuery occurs in the scheme, authority, or path; between the 193 // ^s in "h^ttp://auth/path^?k=v#frag". 194 urlPartPreQuery 195 // urlPartQueryOrFrag occurs in the query portion between the ^s in 196 // "http://auth/path?^k=v#frag^". 197 urlPartQueryOrFrag 198 // urlPartUnknown occurs due to joining of contexts both before and 199 // after the query separator. 200 urlPartUnknown 201 ) 202 203 // jsCtx determines whether a '/' starts a regular expression literal or a 204 // division operator. 205 type jsCtx uint8 206 207 //go:generate stringer -type jsCtx 208 209 const ( 210 // jsCtxRegexp occurs where a '/' would start a regexp literal. 211 jsCtxRegexp jsCtx = iota 212 // jsCtxDivOp occurs where a '/' would start a division operator. 213 jsCtxDivOp 214 // jsCtxUnknown occurs where a '/' is ambiguous due to context joining. 215 jsCtxUnknown 216 ) 217 218 // element identifies the HTML element when inside a start tag or special body. 219 // Certain HTML element (for example <script> and <style>) have bodies that are 220 // treated differently from stateText so the element type is necessary to 221 // transition into the correct context at the end of a tag and to identify the 222 // end delimiter for the body. 223 type element uint8 224 225 //go:generate stringer -type element 226 227 const ( 228 // elementNone occurs outside a special tag or special element body. 229 elementNone element = iota 230 // elementScript corresponds to the raw text <script> element 231 // with JS MIME type or no type attribute. 232 elementScript 233 // elementStyle corresponds to the raw text <style> element. 234 elementStyle 235 // elementTextarea corresponds to the RCDATA <textarea> element. 236 elementTextarea 237 // elementTitle corresponds to the RCDATA <title> element. 238 elementTitle 239 ) 240 241 //go:generate stringer -type attr 242 243 // attr identifies the current HTML attribute when inside the attribute, 244 // that is, starting from stateAttrName until stateTag/stateText (exclusive). 245 type attr uint8 246 247 const ( 248 // attrNone corresponds to a normal attribute or no attribute. 249 attrNone attr = iota 250 // attrScript corresponds to an event handler attribute. 251 attrScript 252 // attrScriptType corresponds to the type attribute in script HTML element 253 attrScriptType 254 // attrStyle corresponds to the style attribute whose value is CSS. 255 attrStyle 256 // attrURL corresponds to an attribute whose value is a URL. 257 attrURL 258 // attrSrcset corresponds to a srcset attribute. 259 attrSrcset 260 ) 261