Add '$$' at the beginning and end of empty latex equations to have them render correctly #8213
Conversation
equations to have them render correctly.
Codecov Report
@@ Coverage Diff @@
## master #8213 +/- ##
==========================================
- Coverage 58.79% 58.73% -0.07%
==========================================
Files 504 505 +1
Lines 23090 23109 +19
Branches 3723 3731 +8
==========================================
- Hits 13576 13572 -4
- Misses 8645 8656 +11
- Partials 869 881 +12
Continue to review full report at Codecov.
|
| return input; | ||
| } | ||
|
|
||
| private getAllIndexesOf(arr: string, value: string): number[] { |
There was a problem hiding this comment.
getAllIndexesOf [](start = 12, length = 15)
I think you could likely do this with regexs?
const beginReg = /\begin/g;
const endReg = /\end/g;
while (result = beginReg.exec(input)) {
result.index is index of match
beginReg.lastIndex is next match
There was a problem hiding this comment.
Did you take a look at how Jupyter does this? The equation matching looks more complex to me.
https://github.com/jupyter/notebook/blob/ff29ba8ab006e59bc264e0ea33084755533535c8/docs/source/examples/Notebook/Typesetting%20Equations.ipynb
In reply to: 339142495 [](ancestors = 339142495)
There was a problem hiding this comment.
Here is what you want.
https://github.com/jupyter/notebook/blob/d60445e92666a0c5651a33e180c92051d94a7d44/notebook/static/notebook/js/mathjaxutils.js
Looks like we might need to check licensing as well.
In reply to: 339245020 [](ancestors = 339245020,339142495)
Started using Regular Expressions instead of indexOf Added unit tests
| const beginIndexes = getAllIndexesOfRegex(input, /\\begin/g); | ||
| const endIndexes = getAllIndexesOfRegex(input, /\\end/g); | ||
|
|
||
| for (let i = 0; i < beginIndexes.length; i += 1) { |
There was a problem hiding this comment.
I think this would have issues with mismatched begin / end pairs. The code is assuming that an equal number of \begin and \end matches were found, but you could have valid markdown with just those strings in it.
I think it's worth taking a look at this regex that I linked to see if it breaks out latex bits from a markdown block.
it now includes the opening anc closing brackets, with whatever name the user chooses. Also, the code will run only if the amount of indexes are the same.
| // Copyright (c) Microsoft Corporation. All rights reserved. | ||
| // Licensed under the MIT License. | ||
| // Adds '$$' to latex formulas that don't have a '$', allowing users to input the formula directly. | ||
| function fixLatexEquations(input: string): string { |
There was a problem hiding this comment.
| function fixLatexEquations(input: string): string { | |
| export function fixLatexEquations(input: string): string { |
| return indexes; | ||
| } | ||
|
|
||
| export { fixLatexEquations }; |
There was a problem hiding this comment.
This is not required in typescript.
For #7992
package-lock.jsonhas been regenerated by runningnpm install(if dependencies have changed)