Why are MathJax images not displaying on my webpage?

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
Sign up Log in
340 replies · 54K views
Physics news on Phys.org
Well I tried that and pressed the debug button, and although I haven't a clue as to what I was doing or did -- it worked.

At the moment I can open and close IE8 and not see any errors and Tex works again.

It remains to be seen whether this fix is permanent on restart.

I would be interest in what 'debugging' achieves though.

Thanks a bunch Vela.
 
I am aware that the reason it now works is more likely to be some positive change made at the PF source end and that my random pressing buttons was just a coincidence.

Particularly as the fix has survived several reboots.

So thank you whoever finally got it right, I asssume someone is monitoring this thread for feedback.

There have obviously been lots of changes tried out over the last few days. PF speed and response has been highly variable.

It all seems tickety boo now.
 
I can't remember if it was already reported or not. There are difference between engines. In some old threads LaTeX is now not displayed correctly:

https://www.physicsforums.com/showthread.php?t=124602

I haven't checked yet why, and I will probably take care about this particular thread, but it is likely that similar problems are present in different threads.

Edit: new engine doesn't like unmatching pairs of \left[ & \right] - if there are both, they are rendered correctly, but just \right] triggers an error.
 
Last edited:
Good point. I don't know LaTeX well enough, but I guess k_i_j is ambiguous, should it be [itex]k_{ij}[/itex] or [itex]k_{i_j}[/itex]?

I haven't checked everything, just skimmed, could be there are more errors.
 
Borek said:
Good point. I don't know LaTeX well enough, but I guess k_i_j is ambiguous, should it be [itex]k_{ij}[/itex] or [itex]k_{i_j}[/itex]?

I haven't checked everything, just skimmed, could be there are more errors.

There's a third choice: [itex]{k_i}_j[/itex] which I suspect was intended. :wink:
 
No matter what was intended, it was wrong, as (aq) should be not a subscript, that's a common mistake. If memory serves me well [itex]H_2SO_{4}(aq)[/itex] is the correct notation (although I am not sure at the moment whether it should or shouldn't be italicized).
 
Borek said:
No matter what was intended, it was wrong, as (aq) should be not a subscript, that's a common mistake. If memory serves me well [itex]H_2SO_{4}(aq)[/itex] is the correct notation (although I am not sure at the moment whether it should or shouldn't be italicized).

Hmm, from wikipedia (as far as that is reliable): "An aqueous solution is a solution in which the solvent is water. It is usually shown in chemical equations by appending aq as a subscript to the relevant formula."

Since it is an attribute to the chemical compound, it seems logical to me that it would be a subscript.

Where is the "common mistake"?
 
Wikipedia is wrong and repeats a common mistake that state of aggregation should be written as a subscript. It shouldn't be. There is only one decisive source of information here, http://www.iupac.org/reports/provisional/abstract05/GreenBook051206_prs.pdf , General Chemistry section, Other symbols and conventions in chemistry (2.10.1), (vi) States of aggregation.

Note: could be there is already a newer version of Green Book, I just copied/pasted what I posted several years ago at CF.
 
Last edited by a moderator:
Borek said:
Wikipedia is wrong and repeats a common mistake that state of aggregation should be written as a subscript. It shouldn't be. There is only one decisive source of information here, http://www.iupac.org/reports/provisional/abstract05/GreenBook051206_prs.pdf , General Chemistry section, Other symbols and conventions in chemistry (2.10.1), (vi) States of aggregation.

Note: could be there is already a newer version of Green Book, I just copied/pasted what I posted several years ago at CF.

All right. I accept that! :smile:

Btw, wikipedia now says: "An aqueous solution is a solution in which the solvent is water. It is usually shown in chemical equations by appending aq to the relevant formula." :wink:

(Next time wikipedia will be right!)
 
Last edited by a moderator:
I like Serena said:
Btw, wikipedia now says: "An aqueous solution is a solution in which the solvent is water. It is usually shown in chemical equations by appending aq to the relevant formula." :wink:

(Next time wikipedia will be right!)


PF members in action. Outstanding. :cool:
 
In this thread, https://www.physicsforums.com/showthread.php?t=510385, does anyone know what might have triggered the appearance of the red part? I double checked and I didn't forget to write "\par" or anything. I even made a copy paste of the "{\partial f}" that appears in the LHS of the equation and the red part still appears in the LHS. The code I used is

\sum_i\frac{\partial f}{\partial x^i}\frac{d\varphi^i_x(t)}{dt}(0)=\sum_i\frac{\partial f}{\partial x^i}X^i(x)
 
quasar987 said:
In this thread, https://www.physicsforums.com/showthread.php?t=510385, does anyone know what might have triggered the appearance of the red part? I double checked and I didn't forget to write "\par" or anything. I even made a copy paste of the "{\partial f}" that appears in the LHS of the equation and the red part still appears in the LHS. The code I used is

\sum_i\frac{\partial f}{\partial x^i}\frac{d\varphi^i_x(t)}{dt}(0)=\sum_i\frac{\partial f}{\partial x^i}X^i(x)

Every 50 characters, the forum software automatically inserts a space. In this case, the \partial gets split into \par tial which doesn't exist. So it's better to leave some spaces and use

\sum_i\frac{\partial f}{\partial x^i}\frac{d\varphi^i_x(t)}{dt}(0) = \sum_i\frac{\partial f}{\partial x^i}X^i(x)

for example. This yields

[tex]\sum_i\frac{\partial f}{\partial x^i}\frac{d\varphi^i_x(t)}{dt}(0) = \sum_i\frac{\partial f}{\partial x^i}X^i(x)[/tex]
 
vela said:
Any decision on new delimiters?
I'm bumping this question. Greg, did you decide that you don't want any new delimiters for LaTeX?

The default simple delimiters are:
$...$ for itex
$$...$$ for tex

This option is the only one that would make it possible to copy stuff without change between forum posts and LaTeX documents. However, most people feel that it's much more important to be able to type a dollar sign in posts about money without preceding it with a backslash or something. So I came up with two alternatives.

The simple option is:
££...££ for itex
$$...$$ for tex

Almost everyone has # and/or £ on the 3 key, and $ on the 4 key. Most (all?) of the ones who don't can still type those symbols easily.

The complicated option is:
££...££, ##...##, §§...§§, ··...·· for itex.
$$...$$, ¢¢...¢¢ for tex.

The §,· and ¢ characters are minor improvements for Germans, Spaniards, and French Canadians respectively. The only advantage it has for them is that it allows them to use either Shift+3 for itex and Shift+4 for tex, or Alt Gr+3 for itex and Alt Gr+4 for tex, instead of e.g. Alt Gr+3 for itex and Shift+4 for tex.

Keyboard layouts at Wikipedia.

Note that one new "delimiter" (the option to use no delimiters at all in certain situations) has already been implemented. MathJax interprets stuff between "begin" and "end" statements as LaTeX, for example: \begin{align}\Lambda =\gamma\begin{pmatrix}1 & -v\\ -v & 1\end{pmatrix}, & & \gamma =\frac{1}{\sqrt{1-v^2}}.\end{align} This can be disabled if you prefer to force everyone to use itex and tex.
 
The trouble with $...$ is that vBulletin parses posts using a regular expression engine. LaTeX syntax is too flexible to allow $...$ to be parsed using regular expressions. It requires a finite state machine instead.

Here's a simple explanation why: The best a regular expression algorithm can do is search for two $ signs with something in between. That is, a regular expression such as (Perl syntax)

\$(.*)\$

The reason you have to grab any sort of text that might appear between $ signs is because LaTeX syntax itself allows any sorts of characters to appear between $ signs. In particular, it allows spaces, periods, backslashes, even additional $ signs.

So, if someone is writing a post about money, and talks about $200 at one point, and then later mentions $5.35, then the vBulletin parser would grab the string

"200 at one point, and then later mentions "

and send it off to the LaTeX parser. Whoops.

The reason LaTeX itself can deal with the $ signs is because it doesn't use regular expressions.

Note, I have actually written an implementation of $...$ for vBulletin, but in order to work I added the additional requirement that the characters immediately adjacent to the $...$ cannot be spaces (if they are spaces, it interprets them as dollar signs). In the end I scrapped the idea, because I thought it would be too confusing to most people, especially given the disastrous attempts at LaTeX that we often see here.
 
Ben Niehoff said:
THere's a simple explanation why: The best a regular expression algorithm can do is search for two $ signs with something in between. That is, a regular expression such as (Perl syntax)

\$(.*)\$

What about:

\$([^\$]*)\$

or

\$(.*?)\$

I've learned to always use pretty strict expressions in Perl to avoid such problems.
 
I like Serena said:
What about:

\$([^\$]*)\$

or

\$(.*?)\$

I've learned to always use pretty strict expressions in Perl to avoid such problems.

Both of those would still fail on the example I gave (a sentence with two dollar amounts in it).
 
Greg isn't going to write any regular expressions. All he's going to do (if he chooses to do anything) is to find the line in the MathJax configuration file where the optional delimiters are specified, and type them in there. I think this makes it 100% certain that we would have the problem that Ben describes if we choose to use single dollar signs as LaTeX delimiters. I also think I read somewhere that the users can type \$3.50 when they want $3.50 to be displayed, but new members will still be pretty confused.

That's why I suggested that we use ££ instead of $.
 
Why not choose \$ as the LaTeX delimiter? Then there is no collision with ordinary dollar signs.

Of course, it means you still can't copy and paste directly from LaTeX documents, but with your other solutions, you can't do that anyway.

Edit: Of course, then we'd have trouble writing Perl regular expressions in our posts! How does MathJax work; i.e., when does it run in the processing sequence? It seems to be a Javascript that runs on the completed page, which is long after the vBulletin parser processes the posts. In that case, I think we're stuck with 'itex' and 'tex'.

I would recommend against making the delimiter some random symbol. The $ sign is standardized already, but if it can't be the $ sign, it shouldn't be some other symbol.
 
Ben Niehoff said:
Why not choose \$ as the LaTeX delimiter? Then there is no collision with ordinary dollar signs.
It's significantly harder to to type than ££.
(Edit: If you're that attached to standard usage, doesn't it also bother you that your suggestion is a complete reversal of the default "$ for itex", "\$ for dollar sign"?)

Ben Niehoff said:
It seems to be a Javascript that runs on the completed page, which is long after the vBulletin parser processes the posts. In that case, I think we're stuck with 'itex' and 'tex'.
As I said, all that's required is a simple edit of a MathJax configuration file. (OK, maybe not if you specifically want to use \$ as a delimiter).
 
Last edited:
Ben Niehoff said:
I would recommend against making the delimiter some random symbol. The $ sign is standardized already, but if it can't be the $ sign, it shouldn't be some other symbol.
Why not? What's wrong with using something like ##?
 
Fredrik said:
The simple option is:
££...££ for itex
$$...$$ for tex

Almost everyone has # and/or £ on the 3 key, and $ on the 4 key. Most (all?) of the ones who don't can still type those symbols easily.

The complicated option is:
££...££, ##...##, §§...§§, ··...·· for itex.
$$...$$, ¢¢...¢¢ for tex.

The §,· and ¢ characters are minor improvements for Germans, Spaniards, and French Canadians respectively. The only advantage it has for them is that it allows them to use either Shift+3 for itex and Shift+4 for tex, or Alt Gr+3 for itex and Alt Gr+4 for tex, instead of e.g. Alt Gr+3 for itex and Shift+4 for tex.

Keyboard layouts at Wikipedia.
I had another look a the keyboard layout page. I had forgotten that ££ is a bad option for people with a US keyboard. ## is better for them. But most of the rest of the world should prefer ££ (based on the location on the keyboard), so here's a third option:
##...## or ££...££ for itex
$$...$$ for tex
 

Similar threads

LaTeX Uncovering the Source: How to View Latex Code in Mathjax Images
  • Jameson
  • · Replies 4 ·
  • MATLAB, Maple, Mathematica, LaTeX
Replies
4