Added support for markdown in heading cells when they are nbconverted. #3576
Conversation
| @@ -61,7 +61,7 @@ $$ | |||
|
|
|||
| {% block headingcell scoped %} | |||
|
|
|||
| {{ '#' * cell.level }} {{ cell.source}} | |||
| {{ '#' * cell.level }} {{ cell.source | wrap(80)}} | |||
There was a problem hiding this comment.
how does the wrap 80 behave on more than 80 char title ?
There was a problem hiding this comment.
The same way than with other text... it is probably redundant because it is no sense to have a title with more than 80 char at least you are working in the academia... hehe... joke aside, I think we need to wrap the tile for deal with longer ones...
There was a problem hiding this comment.
my problem is :
# if this is a really long long title in markdown
this will probably not be...
# if this is a really long
long title in markdown
There was a problem hiding this comment.
Yeah, you are right about that... maybe is better not to call wrap in this case...
I prefer to have consistency in the title... I will delete this call of the wrap function.
Thanks.
There was a problem hiding this comment.
jinja2 has a built-in wordwrap filter which support the definition of wrapstrings. If I understand the docs correctly you might be able to apply this filter here, like {{ cell.source | wordwrap(80, False, '\n'+'#' * cell.level)}}
Ok, it is some rude workaround, but should be possible.
There was a problem hiding this comment.
Interesting, let me try it...
There was a problem hiding this comment.
This could even replace our wrap filter totally I guess !
There was a problem hiding this comment.
OK, it works great... but I added a whitespace to the function to be rendered correctly, if not present we have the text immediately next to # and it will be not rendered OK by a lot of markdown parsers:
{{ cell.source | wordwrap(80, False, '\n'+'#' * cell.level+' ')}}
|
+2 (+1, +1 because I like nbconverting (verb): IPython conversion process. Example "He was a matlab guy, but he as recently been nbconverted to IPython." |
|
HeHe... From your definition, "nbconverted to IPython" is redundant... Example "He was a matlab guy, but he as recently been nbconverted." That's all... HeHe As you with iPtyhon vs. IPython campaign... I have to start my own campaign with the |
…p the titles in the correct way.
| @@ -61,7 +61,7 @@ $$ | |||
|
|
|||
| {% block headingcell scoped %} | |||
|
|
|||
| {{ '#' * cell.level }} {{ cell.source}} | |||
| {{ '#' * cell.level }} {{ cell.source | wordwrap(80, False, '\n'+'#' * cell.level+' ')}} | |||
There was a problem hiding this comment.
and... why? if we have a very long title it would be nice to have it wraped.
There was a problem hiding this comment.
Tools that auto hard-wrap without user input bug me, but you can leave this if you want.
There was a problem hiding this comment.
I understand you, but if we are already wrapping inside markdown cells, I think we have to be consistent and wrap also inside the headers cells... or not wrap anything... If I have to choose, I prefer to wrap it all.
There was a problem hiding this comment.
consistency argument wins, original comment withdrawn.
There was a problem hiding this comment.
Real question is markdown interpretation is the following one or more title
## Am I a nice
## multiline title or
## 4 titles without
## any text ?
There was a problem hiding this comment.
(definitively 4 titles if I past that in a MD cell in notebook.)
There was a problem hiding this comment.
Okay then, original comment reinstated, do not wrap this.
There was a problem hiding this comment.
OK, you convinced me... I will not wrap the headers.
Updating the PR...
…rpretation of multiline headers.
|
+1 then :-) |
Added support for markdown in heading cells when they are nbconverted.
|
Thanks! |
Added support for markdown in heading cells when they are nbconverted.
@minrk as posted a PR where provides the possibility to write markdown in heading cells: #3531
@Carreau has pointed out the need of a counterpart in the nbconvert machinery.
@jakobgager and @jdfreder have pointed out that the latex exporter already supported markdown in heading.
So, I have made some little changes for the remaining basic templates: basichtml, rst, markdown (python tpl don't need changes) to cover this issue from the nbconvert machinery point of view.
Damián.