@zzish/math-rich-input 0.1.0

package/src/mimeTypeHelper.js

@@ -0,0 +1,201 @@
+const INLINE_LATEX_REG_EXP = /\$\$[\s\S]+?\$\$|\\\[[\s\S]+?\\\]|\\\([\s\S]+?\\\)|\$[^$\\]*(?:\\.[^$\\]*)*\$/g;
+const BLOCL_LATEX_REG_EXP = /\$\$[\s\S]+?\$\$|\\\[[\s\S]+?\\\]/g;
+
+export const MIME_TYPE_PLAIN_TEXT = "text/plain"
+export const MIME_TYPE_HTML = "text/html"
+export const MIME_TYPE_TEX = "application/x-tex"
+export const MIME_TYPE_ZZISH_HTML_MATH = "application/x-zzish-html-math"
+
+const LATEX_COMMAND_REG_EXP = /\\[a-zA-Z]/g
+
+const HTML_CODES_REG_EXP = /&(#36|quot|apos|lt|gt|amp);/
+const HTML_TAGS_REG_EXP = /<(br|basefont|hr|input|source|frame|param|area|meta|!--|col|link|option|base|img|wbr|!DOCTYPE).*?>|<(a|abbr|acronym|address|applet|article|aside|audio|b|bdi|bdo|big|blockquote|body|button|canvas|caption|center|cite|code|colgroup|command|datalist|dd|del|details|dfn|dialog|dir|div|dl|dt|em|embed|fieldset|figcaption|figure|font|footer|form|frameset|head|header|hgroup|h1|h2|h3|h4|h5|h6|html|i|iframe|ins|kbd|keygen|label|legend|li|map|mark|menu|meter|nav|noframes|noscript|object|ol|optgroup|output|p|pre|progress|q|rp|rt|ruby|s|samp|script|section|select|small|span|strike|strong|style|sub|summary|sup|table|tbody|td|textarea|tfoot|th|thead|time|title|tr|track|tt|u|ul|var|video).*?<\/\2>/i
+const HTML_MATH_TAGS_REC_EXP = /<math>/i
+
+const WORDS_REG_EXP = /^[a-zA-Z]*[,."';:?!]*$/i
+
+function stripDollars (stringToStrip) {
+  return stringToStrip[0] === "$" && stringToStrip[1] !== "$"
+    ? stringToStrip.slice(1, -1)
+    : stringToStrip.slice(2, -2)
+}
+
+export function isTextLatexMath(text) {
+  // We need at least two dollar signs to be latex
+  let i = text.indexOf("$")
+  if (i < 0)
+    return false
+  i = text.indexOf("$",i+1)
+  if (i < 0)
+    return false
+
+  // Lets try and extract the potentially latex nodes
+  let latexMatch = text.match(BLOCL_LATEX_REG_EXP)
+  if (latexMatch === null || latexMatch === undefined)
+    latexMatch = text.match(INLINE_LATEX_REG_EXP);
+
+  if (latexMatch === null || latexMatch === undefined || latexMatch.length === 0)
+    return false
+  
+  for (let i = 0; i < latexMatch.length; i++)
+    latexMatch[i] = stripDollars(latexMatch[i])
+
+    
+  // If all dollar signs are followed by a number, it probably is not latex
+  i = text.indexOf("$")
+  let probablyLatex = false
+  while (i >=0) {
+    if (i === text.length-1) {
+      probablyLatex = true
+      break
+    }
+    let c = text.charAt(i+1)
+    if (c < "0" ||  c > "9") {
+      probablyLatex = true
+      break
+    }
+    i = text.indexOf("$",i+1)
+  }
+
+  // Note that there are other uses of a dollar sign, that couold potentially be
+  // entered.  For example, if someone created a question about the use of the 
+  // $ sign in excel cells or the use of the $ sign in regular expressions.
+
+  if (probablyLatex) {
+    // At this point it probably is latex, but we can check inside the latex
+    // elements for things that make it not valid latex
+
+    // Note that invalid latex does not necessarily mean it is not INTENDED
+    // to be latex. Normal text is valid latex, and invalid latex usually reflects
+    // an error in intended latex. Thus it is hard to come up for a test that indicates
+    // that text inside a pair of $ symbols is not latex
+
+    // for (let i = 0; i < latexMatch.length; i++) {
+    //   if (!isValidLatex(latexMatch[i]))
+    //     return false
+    // }
+
+    // If all the latex elements are empty, then we will treat it as plain text
+    let allEmpty = true
+    for (let i = 0; i < latexMatch.length; i++) {
+      if (latexMatch[i] !== "")
+        allEmpty = false
+    }
+    if (allEmpty)
+      return false
+
+    // If an element between two dollars is two or more words and only words
+    // and at least one word has three characters then we will treat it as text
+    for (let i = 0; i < latexMatch.length; i++) {
+      let words = latexMatch[i].split(" ")
+      if (words.length >= 2) {
+        let allWords = true
+        let longWord = false
+        for (let j = 0; j < words.length; j++) {
+          // console.log(j+":"+words[j])
+          if (words[j].length > 0 &&
+              WORDS_REG_EXP.test(words[j]) === false) {
+            // console.log("Not: "+words[j])
+            allWords = false
+            break
+          }
+          if (words[j].length >= 3)
+            longWord = true
+          }
+        if (allWords && longWord)
+          return false
+      }
+    }
+
+    return true
+  } else {
+    // At this point it probably is not latex, but we can check inside the latex
+    // elements for things that make it likely to be latex, such as latex commands
+
+    for (let i = 0; i < latexMatch.length; i++) {
+      if (latexMatch[i].match(LATEX_COMMAND_REG_EXP) !== null)
+        return true
+    }
+
+    return false
+  }
+}
+
+export function isTextHtml(text) {
+  if (HTML_CODES_REG_EXP.test(text))
+    return true
+  let i = text.indexOf("<")
+  if (i < 0)
+    return false
+  
+  let containsHtmlTags = HTML_TAGS_REG_EXP.test(text)
+  return containsHtmlTags
+}
+
+export function isTextHtmlMath(text) {
+  let containsMathTags = HTML_MATH_TAGS_REC_EXP.test(text)
+  return containsMathTags
+}
+
+export function determineMimeType(text) {
+  if (isTextHtmlMath(text))
+    return MIME_TYPE_ZZISH_HTML_MATH
+  
+  if (isTextHtml(text))
+    return MIME_TYPE_HTML
+
+  if (isTextLatexMath(text))
+    return MIME_TYPE_TEX
+
+  return MIME_TYPE_PLAIN_TEXT
+}
+
+
+// test = (text, mimeType) => {
+//   let result = determineMimeType(text)
+//   if (result === mimeType)
+//     return <p style={{color:"green"}}>{text}: {result}</p>
+//   else
+//     return <p style={{color:"red"}}>{text}: {result}</p>
+// }
+
+/*
+        {this.test("a", "text/plain")}
+        {this.test("hello world", "text/plain")}
+        {this.test("hello <b>world</b>", "text/html")}
+        {this.test("hello <math>e=mc^2</math>", "application/x-zzish-html-math")}
+        {this.test("hello <MATH>e=mc^2</MATH>", "application/x-zzish-html-math")}
+        {this.test("$a$", "application/x-tex")}
+        {this.test("$$a$$", "application/x-tex")}
+        {this.test("$$a$$ or $$b$$", "application/x-tex")}
+        {this.test("$1$", "application/x-tex")}
+        {this.test("$1.00$", "application/x-tex")}
+        {this.test("$+1.00$", "application/x-tex")}
+        {this.test("$-1.00$", "application/x-tex")}
+        {this.test("$1.00 $", "application/x-tex")}
+        {this.test("$ 1.00$", "application/x-tex")}
+        {this.test("$ 1.00 $", "application/x-tex")}
+        {this.test("If a is exactly $1.00$ then", "application/x-tex")}
+        {this.test("$1.00 $2.00", "text/plain")}
+        {this.test("Yes, $1.00 < $2.00", "text/plain")}
+        {this.test("If $x$ is big", "application/x-tex")}
+        {this.test("If I have $1.00", "text/plain")}
+        {this.test("If sugar costs $1.00 and tea costs $2.00", "text/plain")}
+        {this.test("If sugar costs $1.00 and tea \\times costs $2.00", "application/x-tex")}
+        {this.test("If sugar costs $1.00 and tea \\ times costs $2.00", "text/plain")}
+        {this.test("I love $$. What about you?", "text/plain")}
+        {this.test("I love $$$. What about you?", "text/plain")}
+        {this.test("Here is a $ and here is another $", "text/plain")}
+        {this.test("Here is a $ and here is, well , another $", "text/plain")}
+        {this.test("$a bc d$", "application/x-tex")}
+        {this.test("$abcd$", "application/x-tex")}
+        {this.test("$ab cd$", "application/x-tex")}
+        {this.test("$a bcd e$", "text/plain")}
+        {this.test("$a bcd, e$", "text/plain")}
+        {this.test("$a + bcd - e$", "application/x-tex")}
+        {this.test("$a bcd, e$", "text/plain")}
+        {this.test("$a bcd! e$", "text/plain")}
+        {this.test("$a bcd; e$", "text/plain")}
+        {this.test("$a bcd. e$", "text/plain")}
+        {this.test("$a bcd> e$", "application/x-tex")}
+*/