From ac611d51c9de9aa719b42b9463e6f28d6e7d74a4 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Thu, 2 Oct 2014 16:00:13 -0700 Subject: Use integers instead of strings for tags. Use === whenever possible to compare them. --- js/stmd.js | 238 ++++++++++++++++++++++++++++++++++--------------------------- 1 file changed, 131 insertions(+), 107 deletions(-) diff --git a/js/stmd.js b/js/stmd.js index efccad8..23caf31 100755 --- a/js/stmd.js +++ b/js/stmd.js @@ -2137,6 +2137,30 @@ zwj: '‍', zwnj: '‌' }; + // Constants for inline and block types: + + var I_STR = 1; + var I_SOFT_BREAK = 2; + var I_HARD_BREAK = 3; + var I_EMPH = 4; + var I_STRONG = 5; + var I_HTML = 6; + var I_LINK = 7; + var I_IMAGE = 8; + var I_CODE = 9; + var B_DOCUMENT = 10; + var B_PARAGRAPH = 11; + var B_BLOCK_QUOTE = 12; + var B_LIST_ITEM = 13; + var B_LIST = 14; + var B_ATX_HEADER = 15; + var B_SETEXT_HEADER = 16; + var B_INDENTED_CODE = 17; + var B_FENCED_CODE = 18; + var B_HTML_BLOCK = 19; + var B_REFERENCE_DEF = 20; + var B_HORIZONTAL_RULE = 21; + // Constants for character codes: var C_NEWLINE = 10; @@ -2273,7 +2297,7 @@ // Convert tabs to spaces on each line using a 4-space tab stop. var detabLine = function(text) { - if (text.indexOf('\t') == -1) { + if (text.indexOf('\t') === -1) { return text; } else { var lastStop = 0; @@ -2335,8 +2359,8 @@ var foundCode = false; var match; while (!foundCode && (match = this.match(/`+/m))) { - if (match == ticks) { - inlines.push({ t: 'Code', c: this.subject.slice(afterOpenTicks, + if (match === ticks) { + inlines.push({ t: I_CODE, c: this.subject.slice(afterOpenTicks, this.pos - ticks.length) .replace(/[ \n]+/g,' ') .trim() }); @@ -2345,7 +2369,7 @@ } // If we got here, we didn't match a closing backtick sequence. this.pos = afterOpenTicks; - inlines.push({ t: 'Str', c: ticks }); + inlines.push({ t: I_STR, c: ticks }); return true; }; @@ -2358,13 +2382,13 @@ if (subj.charCodeAt(pos) === C_BACKSLASH) { if (subj.charAt(pos + 1) === '\n') { this.pos = this.pos + 2; - inlines.push({ t: 'Hardbreak' }); + inlines.push({ t: I_HARD_BREAK }); } else if (reEscapable.test(subj.charAt(pos + 1))) { this.pos = this.pos + 2; - inlines.push({ t: 'Str', c: subj.charAt(pos + 1) }); + inlines.push({ t: I_STR, c: subj.charAt(pos + 1) }); } else { this.pos++; - inlines.push({t: 'Str', c: '\\'}); + inlines.push({t: I_STR, c: '\\'}); } return true; } else { @@ -2379,15 +2403,15 @@ if ((m = this.match(/^<([a-zA-Z0-9.!#$%&'*+\/=?^_`{|}~-]+@[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?(?:\.[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)*)>/))) { // email autolink dest = m.slice(1,-1); inlines.push( - {t: 'Link', - label: [{ t: 'Str', c: dest }], + {t: I_LINK, + label: [{ t: I_STR, c: dest }], destination: 'mailto:' + encodeURI(unescape(dest)) }); return true; } else if ((m = this.match(/^<(?:coap|doi|javascript|aaa|aaas|about|acap|cap|cid|crid|data|dav|dict|dns|file|ftp|geo|go|gopher|h323|http|https|iax|icap|im|imap|info|ipp|iris|iris.beep|iris.xpc|iris.xpcs|iris.lwz|ldap|mailto|mid|msrp|msrps|mtqp|mupdate|news|nfs|ni|nih|nntp|opaquelocktoken|pop|pres|rtsp|service|session|shttp|sieve|sip|sips|sms|snmp|soap.beep|soap.beeps|tag|tel|telnet|tftp|thismessage|tn3270|tip|tv|urn|vemmi|ws|wss|xcon|xcon-userid|xmlrpc.beep|xmlrpc.beeps|xmpp|z39.50r|z39.50s|adiumxtra|afp|afs|aim|apt|attachment|aw|beshare|bitcoin|bolo|callto|chrome|chrome-extension|com-eventbrite-attendee|content|cvs|dlna-playsingle|dlna-playcontainer|dtn|dvb|ed2k|facetime|feed|finger|fish|gg|git|gizmoproject|gtalk|hcp|icon|ipn|irc|irc6|ircs|itms|jar|jms|keyparc|lastfm|ldaps|magnet|maps|market|message|mms|ms-help|msnim|mumble|mvn|notes|oid|palm|paparazzi|platform|proxy|psyc|query|res|resource|rmi|rsync|rtmp|secondlife|sftp|sgn|skype|smb|soldat|spotify|ssh|steam|svn|teamspeak|things|udp|unreal|ut2004|ventrilo|view-source|webcal|wtai|wyciwyg|xfire|xri|ymsgr):[^<>\x00-\x20]*>/i))) { dest = m.slice(1,-1); inlines.push({ - t: 'Link', - label: [{ t: 'Str', c: dest }], + t: I_LINK, + label: [{ t: I_STR, c: dest }], destination: encodeURI(unescape(dest)) }); return true; } else { @@ -2399,7 +2423,7 @@ var parseHtmlTag = function(inlines) { var m = this.match(reHtmlTag); if (m) { - inlines.push({ t: 'Html', c: m }); + inlines.push({ t: I_HTML, c: m }); return true; } else { return false; @@ -2444,15 +2468,15 @@ }; var Emph = function(ils) { - return {t: 'Emph', c: ils}; + return {t: I_EMPH, c: ils}; } var Strong = function(ils) { - return {t: 'Strong', c: ils}; + return {t: I_STRONG, c: ils}; } var Str = function(s) { - return {t: 'Str', c: s}; + return {t: I_STR, c: s}; } // Attempt to parse emphasis or strong emphasis. @@ -2776,7 +2800,7 @@ // if we got this far, we've parsed a label. // Try to parse an explicit link: [label](url "title") - if (this.peek() == C_OPEN_PAREN) { + if (this.peek() === C_OPEN_PAREN) { this.pos++; if (this.spnl() && ((dest = this.parseLinkDestination()) !== null) && @@ -2786,7 +2810,7 @@ (title = this.parseLinkTitle() || '') || true) && this.spnl() && this.match(/^\)/)) { - inlines.push({ t: 'Link', + inlines.push({ t: I_LINK, destination: dest, title: title, label: parseRawLabel(rawlabel) }); @@ -2802,7 +2826,7 @@ this.spnl(); var beforelabel = this.pos; n = this.parseLinkLabel(); - if (n == 2) { + if (n === 2) { // empty second label reflabel = rawlabel; } else if (n > 0) { @@ -2814,7 +2838,7 @@ // lookup rawlabel in refmap var link = this.refmap[normalizeReference(reflabel)]; if (link) { - inlines.push({t: 'Link', + inlines.push({t: I_LINK, destination: link.destination, title: link.title, label: parseRawLabel(rawlabel) }); @@ -2832,7 +2856,7 @@ var parseEntity = function(inlines) { var m; if ((m = this.match(reEntityHere))) { - inlines.push({ t: 'Str', c: entityToChar(m) }); + inlines.push({ t: I_STR, c: entityToChar(m) }); return true; } else { return false; @@ -2844,7 +2868,7 @@ var parseString = function(inlines) { var m; if ((m = this.match(reMain))) { - inlines.push({ t: 'Str', c: m }); + inlines.push({ t: I_STR, c: m }); return true; } else { return false; @@ -2857,9 +2881,9 @@ var m = this.match(/^ *\n/); if (m) { if (m.length > 2) { - inlines.push({ t: 'Hardbreak' }); + inlines.push({ t: I_HARD_BREAK }); } else if (m.length > 0) { - inlines.push({ t: 'Softbreak' }); + inlines.push({ t: I_SOFT_BREAK }); } return true; } @@ -2872,10 +2896,10 @@ if (this.match(/^!/)) { var link = this.parseLink(inlines); if (link) { - inlines[inlines.length - 1].t = 'Image'; + inlines[inlines.length - 1].t = I_IMAGE; return true; } else { - inlines.push({ t: 'Str', c: '!' }); + inlines.push({ t: I_STR, c: '!' }); return true; } } else { @@ -2994,7 +3018,7 @@ } if (!res) { this.pos += 1; - inlines.push({t: 'Str', c: String.fromCharCode(c)}); + inlines.push({t: I_STR, c: String.fromCharCode(c)}); } if (memoize) { @@ -3071,17 +3095,17 @@ // Returns true if parent block can contain child block. var canContain = function(parent_type, child_type) { - return ( parent_type == 'Document' || - parent_type == 'BlockQuote' || - parent_type == 'ListItem' || - (parent_type == 'List' && child_type == 'ListItem') ); + return ( parent_type === B_DOCUMENT || + parent_type === B_BLOCK_QUOTE || + parent_type === B_LIST_ITEM || + (parent_type === B_LIST && child_type === B_LIST_ITEM) ); }; // Returns true if block type can accept lines of text. var acceptsLines = function(block_type) { - return ( block_type == 'Paragraph' || - block_type == 'IndentedCode' || - block_type == 'FencedCode' ); + return ( block_type === B_PARAGRAPH || + block_type === B_INDENTED_CODE || + block_type === B_FENCED_CODE ); }; // Returns true if block ends with a blank line, descending if needed @@ -3090,7 +3114,7 @@ if (block.last_line_blank) { return true; } - if ((block.t == 'List' || block.t == 'ListItem') && block.children.length > 0) { + if ((block.t === B_LIST || block.t === B_LIST_ITEM) && block.children.length > 0) { return endsWithBlankLine(block.children[block.children.length - 1]); } else { return false; @@ -3105,7 +3129,7 @@ var b = block; var last_list = null; do { - if (b.t === 'List') { + if (b.t === B_LIST) { last_list = b; } b = b.parent; @@ -3234,7 +3258,7 @@ indent = first_nonspace - offset; switch (container.t) { - case 'BlockQuote': + case B_BLOCK_QUOTE: if (indent <= 3 && ln.charCodeAt(first_nonspace) === C_GREATERTHAN) { offset = first_nonspace + 1; if (ln.charCodeAt(offset) === C_SPACE) { @@ -3245,7 +3269,7 @@ } break; - case 'ListItem': + case B_LIST_ITEM: if (indent >= container.list_data.marker_offset + container.list_data.padding) { offset += container.list_data.marker_offset + @@ -3257,7 +3281,7 @@ } break; - case 'IndentedCode': + case B_INDENTED_CODE: if (indent >= CODE_INDENT) { offset += CODE_INDENT; } else if (blank) { @@ -3267,14 +3291,14 @@ } break; - case 'ATXHeader': - case 'SetextHeader': - case 'HorizontalRule': + case B_ATX_HEADER: + case B_SETEXT_HEADER: + case B_HORIZONTAL_RULE: // a header can never container > 1 line, so fail to match: all_matched = false; break; - case 'FencedCode': + case B_FENCED_CODE: // skip optional spaces of fence offset i = container.fence_offset; while (i > 0 && ln.charCodeAt(offset) === C_SPACE) { @@ -3283,13 +3307,13 @@ } break; - case 'HtmlBlock': + case B_HTML_BLOCK: if (blank) { all_matched = false; } break; - case 'Paragraph': + case B_PARAGRAPH: if (blank) { container.last_line_blank = true; all_matched = false; @@ -3328,9 +3352,9 @@ // Unless last matched container is a code block, try new container starts, // adding children to the last matched container: - while (container.t != 'FencedCode' && - container.t != 'IndentedCode' && - container.t != 'HtmlBlock' && + while (container.t != B_FENCED_CODE && + container.t != B_INDENTED_CODE && + container.t != B_HTML_BLOCK && // this is a little performance optimization: matchAt(/^[ #`~*+_=<>0-9-]/,ln,offset) !== null) { @@ -3346,10 +3370,10 @@ if (indent >= CODE_INDENT) { // indented code - if (this.tip.t != 'Paragraph' && !blank) { + if (this.tip.t != B_PARAGRAPH && !blank) { offset += CODE_INDENT; closeUnmatchedBlocks(this); - container = this.addChild('IndentedCode', line_number, offset); + container = this.addChild(B_INDENTED_CODE, line_number, offset); } else { // indent > 4 in a lazy paragraph continuation break; } @@ -3362,13 +3386,13 @@ offset++; } closeUnmatchedBlocks(this); - container = this.addChild('BlockQuote', line_number, offset); + container = this.addChild(B_BLOCK_QUOTE, line_number, offset); } else if ((match = ln.slice(first_nonspace).match(/^#{1,6}(?: +|$)/))) { // ATX header offset = first_nonspace + match[0].length; closeUnmatchedBlocks(this); - container = this.addChild('ATXHeader', line_number, first_nonspace); + container = this.addChild(B_ATX_HEADER, line_number, first_nonspace); container.level = match[0].trim().length; // number of #s // remove trailing ###s: container.strings = @@ -3379,7 +3403,7 @@ // fenced code block var fence_length = match[0].length; closeUnmatchedBlocks(this); - container = this.addChild('FencedCode', line_number, first_nonspace); + container = this.addChild(B_FENCED_CODE, line_number, first_nonspace); container.fence_length = fence_length; container.fence_char = match[0][0]; container.fence_offset = first_nonspace - offset; @@ -3389,23 +3413,23 @@ } else if (matchAt(reHtmlBlockOpen, ln, first_nonspace) !== null) { // html block closeUnmatchedBlocks(this); - container = this.addChild('HtmlBlock', line_number, first_nonspace); + container = this.addChild(B_HTML_BLOCK, line_number, first_nonspace); // note, we don't adjust offset because the tag is part of the text break; - } else if (container.t == 'Paragraph' && + } else if (container.t == B_PARAGRAPH && container.strings.length === 1 && ((match = ln.slice(first_nonspace).match(/^(?:=+|-+) *$/)))) { // setext header line closeUnmatchedBlocks(this); - container.t = 'SetextHeader'; // convert Paragraph to SetextHeader + container.t = B_SETEXT_HEADER; // convert Paragraph to SetextHeader container.level = match[0][0] === '=' ? 1 : 2; offset = ln.length; } else if (matchAt(reHrule, ln, first_nonspace) !== null) { // hrule closeUnmatchedBlocks(this); - container = this.addChild('HorizontalRule', line_number, first_nonspace); + container = this.addChild(B_HORIZONTAL_RULE, line_number, first_nonspace); offset = ln.length - 1; break; @@ -3416,14 +3440,14 @@ offset = first_nonspace + data.padding; // add the list if needed - if (container.t !== 'List' || + if (container.t !== B_LIST || !(listsMatch(container.list_data, data))) { - container = this.addChild('List', line_number, first_nonspace); + container = this.addChild(B_LIST, line_number, first_nonspace); container.list_data = data; } // add the list item - container = this.addChild('ListItem', line_number, first_nonspace); + container = this.addChild(B_LIST_ITEM, line_number, first_nonspace); container.list_data = data; } else { @@ -3453,7 +3477,7 @@ // First check for a lazy paragraph continuation: if (this.tip !== last_matched_container && !blank && - this.tip.t == 'Paragraph' && + this.tip.t == B_PARAGRAPH && this.tip.strings.length > 0) { // lazy paragraph continuation @@ -3470,9 +3494,9 @@ // lists or breaking out of lists. We also don't set last_line_blank // on an empty list item. container.last_line_blank = blank && - !(container.t == 'BlockQuote' || - container.t == 'FencedCode' || - (container.t == 'ListItem' && + !(container.t == B_BLOCK_QUOTE || + container.t == B_FENCED_CODE || + (container.t == B_LIST_ITEM && container.children.length === 0 && container.start_line == line_number)); @@ -3483,12 +3507,12 @@ } switch (container.t) { - case 'IndentedCode': - case 'HtmlBlock': + case B_INDENTED_CODE: + case B_HTML_BLOCK: this.addLine(ln, offset); break; - case 'FencedCode': + case B_FENCED_CODE: // check for closing code fence: match = (indent <= 3 && ln.charAt(first_nonspace) == container.fence_char && @@ -3501,9 +3525,9 @@ } break; - case 'ATXHeader': - case 'SetextHeader': - case 'HorizontalRule': + case B_ATX_HEADER: + case B_SETEXT_HEADER: + case B_HORIZONTAL_RULE: // nothing to do; we already added the contents. break; @@ -3512,10 +3536,10 @@ this.addLine(ln, first_nonspace); } else if (blank) { // do nothing - } else if (container.t != 'HorizontalRule' && - container.t != 'SetextHeader') { + } else if (container.t != B_HORIZONTAL_RULE && + container.t != B_SETEXT_HEADER) { // create paragraph container for line - container = this.addChild('Paragraph', line_number, first_nonspace); + container = this.addChild(B_PARAGRAPH, line_number, first_nonspace); this.addLine(ln, first_nonspace); } else { console.log("Line " + line_number.toString() + @@ -3546,7 +3570,7 @@ } switch (block.t) { - case 'Paragraph': + case B_PARAGRAPH: block.string_content = block.strings.join('\n').replace(/^ */m,''); // try parsing the beginning as link reference definitions: @@ -3555,23 +3579,23 @@ this.refmap))) { block.string_content = block.string_content.slice(pos); if (isBlank(block.string_content)) { - block.t = 'ReferenceDef'; + block.t = B_REFERENCE_DEF; break; } } break; - case 'ATXHeader': - case 'SetextHeader': - case 'HtmlBlock': + case B_ATX_HEADER: + case B_SETEXT_HEADER: + case B_HTML_BLOCK: block.string_content = block.strings.join('\n'); break; - case 'IndentedCode': + case B_INDENTED_CODE: block.string_content = block.strings.join('\n').replace(/(\n *)*$/,'\n'); break; - case 'FencedCode': + case B_FENCED_CODE: // first line becomes info string block.info = unescapeEntBS(block.strings[0].trim()); if (block.strings.length == 1) { @@ -3581,7 +3605,7 @@ } break; - case 'List': + case B_LIST: block.tight = true; // tight by default var numitems = block.children.length; @@ -3622,9 +3646,9 @@ // into inline content where appropriate. var processInlines = function(block) { switch(block.t) { - case 'Paragraph': - case 'SetextHeader': - case 'ATXHeader': + case B_PARAGRAPH: + case B_SETEXT_HEADER: + case B_ATX_HEADER: block.inline_content = this.inlineParser.parse(block.string_content.trim(), this.refmap); block.string_content = ""; @@ -3643,7 +3667,7 @@ // The main parsing function. Returns a parsed document AST. var parse = function(input) { - this.doc = makeBlock('Document', 1, 1); + this.doc = makeBlock(B_DOCUMENT, 1, 1); this.tip = this.doc; this.refmap = {}; var lines = input.replace(/\n$/,'').split(/\r\n|\n|\r/); @@ -3662,7 +3686,7 @@ // The DocParser object. function DocParser(){ return { - doc: makeBlock('Document', 1, 1), + doc: makeBlock(B_DOCUMENT, 1, 1), tip: this.doc, refmap: {}, inlineParser: new InlineParser(), @@ -3703,32 +3727,32 @@ var renderInline = function(inline) { var attrs; switch (inline.t) { - case 'Str': + case I_STR: return this.escape(inline.c); - case 'Softbreak': + case I_SOFT_BREAK: return this.softbreak; - case 'Hardbreak': + case I_HARD_BREAK: return inTags('br',[],"",true) + '\n'; - case 'Emph': + case I_EMPH: return inTags('em', [], this.renderInlines(inline.c)); - case 'Strong': + case I_STRONG: return inTags('strong', [], this.renderInlines(inline.c)); - case 'Html': + case I_HTML: return inline.c; - case 'Link': + case I_LINK: attrs = [['href', this.escape(inline.destination, true)]]; if (inline.title) { attrs.push(['title', this.escape(inline.title, true)]); } return inTags('a', attrs, this.renderInlines(inline.label)); - case 'Image': + case I_IMAGE: attrs = [['src', this.escape(inline.destination, true)], ['alt', this.escape(this.renderInlines(inline.label))]]; if (inline.title) { attrs.push(['title', this.escape(inline.title, true)]); } return inTags('img', attrs, "", true); - case 'Code': + case I_CODE: return inTags('code', [], this.escape(inline.c)); default: console.log("Unknown inline type " + inline.t); @@ -3751,48 +3775,48 @@ var attr; var info_words; switch (block.t) { - case 'Document': + case B_DOCUMENT: var whole_doc = this.renderBlocks(block.children); return (whole_doc === '' ? '' : whole_doc + '\n'); - case 'Paragraph': + case B_PARAGRAPH: if (in_tight_list) { return this.renderInlines(block.inline_content); } else { return inTags('p', [], this.renderInlines(block.inline_content)); } break; - case 'BlockQuote': + case B_BLOCK_QUOTE: var filling = this.renderBlocks(block.children); return inTags('blockquote', [], filling === '' ? this.innersep : this.innersep + filling + this.innersep); - case 'ListItem': + case B_LIST_ITEM: return inTags('li', [], this.renderBlocks(block.children, in_tight_list).trim()); - case 'List': + case B_LIST: tag = block.list_data.type == 'Bullet' ? 'ul' : 'ol'; attr = (!block.list_data.start || block.list_data.start == 1) ? [] : [['start', block.list_data.start.toString()]]; return inTags(tag, attr, this.innersep + this.renderBlocks(block.children, block.tight) + this.innersep); - case 'ATXHeader': - case 'SetextHeader': + case B_ATX_HEADER: + case B_SETEXT_HEADER: tag = 'h' + block.level; return inTags(tag, [], this.renderInlines(block.inline_content)); - case 'IndentedCode': + case B_INDENTED_CODE: return inTags('pre', [], inTags('code', [], this.escape(block.string_content))); - case 'FencedCode': + case B_FENCED_CODE: info_words = block.info.split(/ +/); attr = info_words.length === 0 || info_words[0].length === 0 ? [] : [['class','language-' + this.escape(info_words[0],true)]]; return inTags('pre', [], inTags('code', attr, this.escape(block.string_content))); - case 'HtmlBlock': + case B_HTML_BLOCK: return block.string_content; - case 'ReferenceDef': + case B_REFERENCE_DEF: return ""; - case 'HorizontalRule': + case B_HORIZONTAL_RULE: return inTags('hr',[],"",true); default: console.log("Unknown block type " + block.t); @@ -3804,7 +3828,7 @@ var renderBlocks = function(blocks, in_tight_list) { var result = []; for (var i=0; i < blocks.length; i++) { - if (blocks[i].t !== 'ReferenceDef') { + if (blocks[i].t !== B_REFERENCE_DEF) { result.push(this.renderBlock(blocks[i], in_tight_list)); } } -- cgit v1.2.3