# "Math Processing Errors" on empty \overset and differences between preview and answer

This answer revision has an expression which displayed correctly in a prior Mathjax version, and also displays correctly in the edit window, but displays as [Math Processing Error] in the answer.

Revision Source

**Hint** $\rm\ 2^{\large\color{#c00}2} 3^{\large\color{#0a0}3} \mid 2^{\large\color{#c00}{2I}} 3^{\large\color{#0a0}{2J}} 5^{\large2K}\!\cdots \overset{}\Rightarrow \color{#c00}{I\ge 1},\ \color{#0a0}{J\ge 2}\ \$ by *existence* and *uniqueness* of prime factorizations.

Output

Hint $\rm\ 2^{\large\color{#c00}2} 3^{\large\color{#0a0}3} \mid 2^{\large\color{#c00}{2I}} 3^{\large\color{#0a0}{2J}} 5^{\large2K}\!\cdots \overset{}\Rightarrow \color{#c00}{I\ge 1},\ \color{#0a0}{J\ge 2}\ \$ by existence and uniqueness of prime factorizations.

Update I should clarify some points. The bug(s) appear to be triggered by vestigial empty \overset expressions (probably left behind when I was pulled away from the computer in the middle of trying out different presentations). Since these expressions worked fine in the past, this may or may not indicate a bug in changes made in recent versions.

However, the fact that this error is raised only in the answer, and not in the preview, is surely another bug, since the preview rendering should correspond 100% to the answer.

• Curiously, the answer displays correctly on Meta's sandbox. May 12 '16 at 4:07
• I don't see processing errors on neither post. (Linux 64bit, FF 46.0).
– Asaf Karagila Mod
May 12 '16 at 7:26
• @AsafKaragila Error occurs in Win7 in both Firefox and Chrome. May 12 '16 at 13:58
• Also Firefox on Android 6.
– Asaf Karagila Mod
May 12 '16 at 14:00
• Processing error for Safari 9.1, MacOS. May 12 '16 at 14:18
• It looks like the fast-preview is having trouble with the \overset{}\Rightarrow construct. Try turning off the fast preview in the MathJax contextual menu (Math Settings -> Math Renderer -> fast preview), and see if that improves the situation for you. May 15 '16 at 12:40
• Note that the new preview editing code turns off the fast-preview during editing (since that just slows down the refresh in the editor), and that would account for the difference in behavior between the editor and the main question. May 15 '16 at 12:57
• I have opened an issue tracker for the problem. May 15 '16 at 13:11

The culprit in the first example seems to be the \overset{}\Rightarrow construct.

• With it (should produce a [Math Processing Error] until fixed)

$\rm\ 2^{\large\color{#c00}2} 3^{\large\color{#0a0}3} \mid 2^{\large\color{#c00}{2I}} 3^{\large\color{#0a0}{2J}} 5^{\large2K}\!\cdots \overset{}\Rightarrow \color{#c00}{I\ge 1},\ \color{#0a0}{J\ge 2}\ \$

• Replaced with \Rightarrow (should render as expected)

$\rm\ 2^{\large\color{#c00}2} 3^{\large\color{#0a0}3} \mid 2^{\large\color{#c00}{2I}} 3^{\large\color{#0a0}{2J}} 5^{\large2K}\!\cdots \Rightarrow \color{#c00}{I\ge 1},\ \color{#0a0}{J\ge 2}\ \$

I'm willing to bet that the \overset{}\equiv construct is the culprit in the second example.

Checking the error messages (MathJax context menu ⇒ Show Math As ⇒ Error Message) on the problematic formula gives

Error: Cannot read property 'style' of null

Debugging tips: use 'unpacked/MathJax.js', inspect 'MathJax.Hub.lastError' in the browser console

Doing something to the MathJax rendering (e.g., changing the Math Renderer) seems to make the error temporarily go away on the user's end, though re-loading the page seems to cause the error to return.

• Out of idle curiosity: what's the point of the inclusion to begin with? Vertical spacing or something else?
– quid Mod
May 12 '16 at 19:01
• @quid It seems to be a tiny bit of vertical spacing, but it is the surrounding text that appears to be affected. (Animated gif) May 12 '16 at 19:22
• Ah, no wonder then I could not discern any difference at all in the snippet here.
– quid Mod
May 12 '16 at 22:06
• @quid That's not the reason. Rather, there was originally something in the overset, but it was temporarily removed, and I got pulled away before I could finish. I often test different ways of presenting things and sometimes I run out of time to polish it best, possibly leaving vestigial MJ like that. I mention it here in case it may be indicative of a more serious MJ bug. May 13 '16 at 13:07
• @quid The example from the animated gif may have been poorly chosen (though not intentionally so). An example with additional text surrounding it resulted in no discernible differences between \overset{}\Rightarrow, \Rightarrow and \overset{\,}\Rightarrow. May 13 '16 at 15:17