I have some formulas generated on the fly. Mathjax does not turn double negatives to positives. For example, I want "1 - - 1" to become "1 + 1". Same for "1 + - 1 = 1 - 1".
Is there a way to force MathJax to do this?
Ofcourse I can do this step in the code but it makes my formula templates uglier.
Thanks
MathJax is not a computer algebra system, it is just a mathematical typesetting system. It has no understanding of the mathematics it typesets, and so can not perform transformations on the equations. You will have to do that yourself.
It is possible to register pre-filters for the TeX input processor, so you could use that approach to modify the TeX code before MathJax evaluates it. For example, place
<script type="text/x-mathjax-config">
MathJax.Hub.Register.StartupHook("TeX Jax Ready",function () {
MathJax.InputJax.TeX.prefilterHooks.Add(function (data) {
data.math = data.math.replace(/- *-/g,"+")
.replace(/- *+|+ *-/g,"-");
});
});
</script>
before the script that loads MathJax.js itself. That might do what you need.
Related
Does any one know how to force browser to display MathML code instead of equation?
PS: Rendering MathML to view as plain text gives the TeX output.
For example,
The axis on which the point (0,4) lie, is _____
Should be displayed as:
The axis on which the point <math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mo stretchy="false">(</mo><mn>0</mn><mo>,</mo><mn>4</mn><mo stretchy="false">)</mo></mrow><annotation encoding="application/x-tex">(0, 4)</annotation></semantics></math> lie, is _____
In most common configs, if your ouput is not directly mathML, mathjax stores mathml informations in the attribute data-mathml of a span tag wich wraps the mathJax element
This is what is displayed in the popup when you right click on a mathJax element : show math as -> MathMl Code
If your goal is to grab equations from html in mathml format, you can create a script which parse your document and get all data-mathml attributes.
There is many ways to achieve that, this is just an example you may have to adapt:
function grabMathMl(){
var spanMathMl = document.querySelectorAll(".MathJax");
let results = [];
let i = 0, ln = spanMathMl.length;
for ( i; i < ln; ++i){
if ( spanMathMl[i].hasAttribute("data-mathml") ){
results.push(spanMathMl[i].dataset.mathml);
// if you really want to replace content
spanMathMl[i].innerHTML = "<textarea>"+spanMathMl[i].dataset.mathml+"</textarea>";
}
}
return results;
}
// put this fonction in the mathJax queue for you have to wait until mathJax process is done
MathJax.Hub.Queue(function(){
let equations = grabMathMl();
//console.log (equations.toString());// your equations in mathml
});
<script>
</script>
<script type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.3/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script>
<div>$$\left|\int_a^b fg\right| \leq \left(\int_a^b
f^2\right)^{1/2}\left(\int_a^b g^2\right)^{1/2}.$$</div>
<div>
\begin{equation} x+1\over\sqrt{1-x^2} \end{equation}
</div>
Then in word, this link should interest you
https://superuser.com/questions/340650/type-math-formulas-in-microsoft-word-the-latex-way#802093
Sometimes in LaTeX I'll have a bunch of stuff written up with just e for Euler's constant, instead of using \mathrm{e} or making a macro for it, but I want it to display in roman font as constants should, so in LaTeX I'd do something like:
\DeclareSymbolFont{constants}{OT1}{cmr}{m}{n}
\DeclareMathSymbol{e}{\mathalpha}{constants}{`e}
which makes any e in math mode display as a constant (which is fine because I don't normally call any variables e).
My question is: how can I do this in MathJax? I don't want to go through all the stuff I've written up and change all the es to some macro; I want to configure MathJax to automatically display that character in roman font. So for example $abcde$ should display with abcd italic, but the e should be roman.
One way would be to add
<script type="text/x-mathjax-config">
(function () {
var MML;
MathJax.Hub.Register.StartupHook("mml Jax Ready",function () {
MML = MathJax.ElementJax.mml;
});
MathJax.Hub.Register.StartupHook("TeX Jax Ready",function () {
var TEX = MathJax.InputJax.TeX;
TEX.Definitions.special["e"] = "myConstant";
TEX.Parse.Augment({
myConstant: function (c) {
this.Push(this.mmlToken(MML.mi(c).With({mathvariant:MML.VARIANT.NORMAL})));
}
});
});
})();
</script>
just before the script tag that loads MathJax.js itself. This marks the letter "e" as a special character that causes your myConstant function to be called (being passed the "e", so you could attach other letters to myConstant to get them to be in Roman). The myConstant function creates an <mi> element with mathvariant="normal" containing the given character, and pushes that on the TeX processing stack.
I am new here... I am wondering if anyone could help point me in the right direction here.
I am looking to create a Chrome extension that searches a page for a number of different strings (one example: "(410)" or "(1040)" without the quotes) and highlight these so they're easier to see.
To explain a little further why I need this: I frequently work out of a queue with other coworkers, and there are specific things I need to focus on but I can ignore the rest of the questions on the page. So it would be helpful if my particular items were highlighted.
Thank you!
Edit: an example of how the source code works:
<td class="col-question">28 (510). <span id="ctl00_PlaceHolderMain_ctl01_ContentCheckList_ctl28_Label1" title=" <p>
<td class="col-question">49 (1150). <span id="ctl00_PlaceHolderMain_ctl01_ContentCheckList_ctl49_Label1" title="<p>
etc etc etc... there are around 100 numbers in parenthesis I would want highlighted. And probably another 100 that I wouldn't want highlighted.
Okay, to start off with I will show you how to inject the code into the page(s) you want, we will get to selecting the correct numbers in a bit. I will be using jQuery throughout this example, it isn't strictly necessary, but I feel that it may make it a bit easier.
First we declare a content script in our manifest as well as host permissions for the page you are injecting into:
"content_scripts": [
{
"matches": ["http://www.domain.com/page.html"],
"js": ["jquery.js","highlightNumbers.js"],
"css": ["highlight.css"]
}],
"permissions": ["http://www.domain.com/*"]
This will place our code in the page we are trying to change. Now you said that there are about 100 different numbers you would want to highlight and I will assume that these are specific numbers that don't match any patterns, so the only way to select all of them would be to make an explicit list of numbers to highlight.
highlightNumbers.js
// This array will contain all of the numbers you want to highlight
// in no particular order
var numberArray = [670,710,820,1000,...];
numberArray.forEach(function(v){
// Without knowing exactly what the page looks like I will just show you
// how to highlight just the numbers in question, but you could easily
// similarly highlight surrounding text as well
var num = "(" + v + ")";
// Select the '<td>' that contains the number we are looking for
var td = $('td.col-question:contains('+num+')');
// Make sure that this number exists
if(td.length > 0){
// Now that we have it we need to single out the number and replace it
var span = td.html().replace(num,'<span class="highlight-num">'+num+'</span>');
var n = td.html(span);
}
// Now instead of '(1000)' we have
// '<span class="highlight-num">(1000)</span>'
// We will color it in the css file
});
Now that we have singled out all of the numbers that are important, we need to color them. You can, of course, use whatever color you want, but for the sake of the example I will be using a bright green.
highlight.css
span.highlight-num{
background-color: rgb(100, 255, 71);
}
This should color all of the numbers that you put in the array in the js file. Let me know if there are any problems with it as I can't exactly test it.
I'm trying to author a few paragraphs with Jade, but finding it difficult when there are links inside a paragraph.
The best I can come up with, and I'm wondering if there's a way to do it with less markup:
p
span.
this is the start
of the para.
a(href="http://example.com") a link
span.
and this is the rest of
the paragraph.
As of jade 1.0 there's an easier way to deal with this, unfortunately I can't find it anywhere in the official documentation.
You can add inline elements with the following syntax:
#[a.someClass A Link!]
So, an example without going into multiple lines in a p, would be something like:
p: #[span this is the start of the para] #[a(href="http://example.com") a link] #[span and this is the rest of the paragraph]
You can also do nested inline elements:
p: This is a #[a(href="#") link with a nested #[span element]]
You can use a markdown filter and use markdown (and allowed HTML) to write your paragraph.
:markdown
this is the start of the para.
[a link](http://example.com)
and this is the rest of the paragraph.
Alternatively it seems like you can simply ouput HTML without any problems:
p
| this is the start of the para.
| a link
| and this is he rest of the paragraph
I wasn't aware of this myself and just tested it using the jade command line tool. It seems to work just fine.
EDIT:
It seems it can actually be done entirely in Jade as follows:
p
| this is the start of the para
a(href='http://example.com;) a link
| and this is the rest of the paragraph
Don't forget an extra space at the end of para (although you can't see it. and between | and. Otherwise it will look like this para.a linkand not para a link and
Another way to do it:
p
| this is the start of the para
a(href="http://example.com") a link
|
| this is the rest of the paragraph.
Another completely different approach, would be to create a filter, which has first stab at replacing links, and then renders with jade second
h1 happy days
:inline
p this can have [a link](http://going-nowhere.com/) in it
Renders:
<h1>happy days</h1><p>this can have <a href='http://going-nowhere.com/'>a link</a> in it</p>
Full working example: index.js (run with nodejs)
var f, jade;
jade = require('jade');
jade.filters.inline = function(txt) {
// simple regex to match links, might be better as parser, but seems overkill
txt = txt.replace(/\[(.+?)\]\((.+?)\)/, "<a href='$2'>$1</a>");
return jade.compile(txt)();
};
jadestring = ""+ // p.s. I hate javascript's non-handling of multiline strings
"h1 happy days\n"+
":inline\n"+
" p this can have [a link](http://going-nowhere.com/) in it"
f = jade.compile(jadestring);
console.log(f());
A more general solution would render mini sub-blocks of jade in a unique block (maybe identified by something like ${jade goes here}), so...
p some paragraph text where ${a(href="wherever.htm") the link} is embedded
This could be implemented in exactly the same way as above.
Working example of general solution:
var f, jade;
jade = require('jade');
jade.filters.inline = function(txt) {
txt = txt.replace(/\${(.+?)}/, function(a,b){
return jade.compile(b)();
});
return jade.compile(txt)();
};
jadestring = ""+ // p.s. I hate javascript's non-handling of multiline strings
"h1 happy days\n"+
":inline\n"+
" p this can have ${a(href='http://going-nowhere.com/') a link} in it"
f = jade.compile(jadestring);
console.log(f());
If your links come from a data source you can use:
ul
each val in results
p
| blah blah
a(href="#{val.url}") #{val.name}
| more blah
See interpolation
Edit: This feature was implemented and issue closed, see answer above.
I've posted an issue to get this feature added into Jade
https://github.com/visionmedia/jade/issues/936
Haven't had time to implement it though, more +1s may help !
This is the best I can come up with
-var a = function(href,text){ return "<a href='"+href+"'>"+text+"</a>" }
p this is an !{a("http://example.com/","embedded link")} in the paragraph
Renders...
<p>this is an <a href='http://example.com/'>embedded link</a> in the paragraph</p>
Works ok, but feels like a bit of a hack - there really should be a syntax for this!
I did not realize that jade requires a line per tag. I thought we can save space. Much better if this can be understood ul>li>a[class="emmet"]{text}
I had to add a period directly behind a link, like this:
This is your test [link].
I solved it like this:
label(for="eula").lbl.lbl-checkbox.lbl-eula #{i18n.signup.text.accept_eula}
| #{i18n.signup.links.eula}.
As suggested by Daniel Baulig, used below with dynamic params
| <a href=!{aData.link}>link</a>
Turns out there is (now at least) a perfectly simple option
p Convert a .fit file using Garmin Connect's export functionality.
p
| At Times Like These We Suggest Just Going:
a(ui-sref="app") HOME
|
Most simplest thing ever ;) but I was struggling with this myself for a few seconds. Anywho, you need to use an HTML entity for the "#" sign -> #
If you want to in include a link, let's say your/some email address use this:
p
a(href="mailto:me#myemail.com" target="_top") me#myemail.com
I'd like to know how to replace parts of a string in latex. Specifically I'm given a measurement (like 3pt, 10mm, etc) and I'd like to remove the units of that measurement (so 3pt-->3, 10mm-->10, etc).
The reason why I'd like a command to do this is in the following piece of code:
\newsavebox{\mybox}
\sbox{\mybox}{Hello World!}
\newlength{\myboxw}
\newlength{\myboxh}
\settowidth{\myboxw}{\usebox{\mybox}}
\settoheight{\myboxh}{\usebox{\mybox}}
\begin{picture}(\myboxw,\myboxh)
\end{picture}
Basically I create a savebox called mybox. I insert the words "Hello World" into mybox. I create a new length/width, called myboxw/h. I then get the width/height of mybox, and store this in myboxw/h. Then I set up a picture environment whose dimensions correspond to myboxw/h. The trouble is that myboxw is returning something of the form "132.56pt", while the input to the picture environment has to be dimensionless: "\begin{picture}{132.56, 132.56}".
So, I need a command which will strip the units of measurement from a string.
Thanks.
Use the following trick:
{
\catcode`p=12 \catcode`t=12
\gdef\removedim#1pt{#1}
}
Then write:
\edef\myboxwnopt{\expandafter\removedim\the\myboxw}
\edef\myboxhnopt{\expandafter\removedim\the\myboxh}
\begin{picture}(\myboxwnopt,\myboxhnopt)
\end{picture}
Consider the xstring package at https://www.ctan.org/pkg/xstring.
The LaTeX kernel - latex.ltx - already provides \strip#pt, which you can use to strip away any reference to a length. Additionally, there's no need to create a length for the width and/or height of a box; \wd<box> returns the width, while \ht<box> returns the height:
\documentclass{article}
\makeatletter
\let\stripdim\strip#pt % User interface for \strip#pt
\makeatother
\begin{document}
\newsavebox{\mybox}
\savebox{\mybox}{Hello World!}
\begin{picture}(\stripdim\wd\mybox,\stripdim\ht\mybox)
\put(0,0){Hello world}
\end{picture}
\end{document}