Merge nbconvert into IPython #3500

minrk · 2013-07-01T01:17:14Z

Supersedes PR #3497. This PR merges the actual repo, and is made from the IPython org so all core devs can finish it up.

This pull request takes the nbconvert package and inserts into the main IPython project as IPython.nbconvert. While nbconvert is not done, we need to get it merged ASAP so continue the work. In addition to the code dump I have made some basic changes:

Update imports to begin to develop the official API that people will see when importing various things.
Fixed the installation logic to include nbconvert.
Created an nbconvert sub-command of ipython so users can do ipython nbconvert.

I will try to get the Travis tests to pass (mostly import issues), but this code does not have tests. These will be added in a future PR that @jdfreder has already started on. One question we should figure out before merging:

Should nbconvert be a subcommand or its own command?
If it is a subcommand, should we call it nbconvert or the simpler convert?

allow nonexistent output

filters. Filter is now dependent on env. type.

(see template.py)

of the template.

Added more comments and header comments.

Instead of the horizantal rule following the target via a minipage, the needspace package is used. This frees the need to preprocess the target block using the sphinxtransformer. This allows us to rely on Jinja more, as intended by nbconvert. Fixed some macro names to stay PEP8 consistent. Removed an unnecessary comment. Unified the output type blocks into the render_output macro.

Centeralized some code into methods Added more comments

This reverts part of commit 994bca5.

Small images were getting streched to larger scales.

minrk · 2013-07-01T01:19:22Z

Re-posting my own note from #3497:

Should nbconvert be a subcommand or its own command?

I think it should be a subcommand. I would like to eventually get rid of all top-level commands (ipengine, ipcontroller, etc.), so I don't think we should ever add new top-level scripts. We haven't had an official conversation on this matter, though.

minrk · 2013-07-01T01:20:39Z

Also: woo!

ellisonbg · 2013-07-01T03:49:24Z

Ok I see how you did this, thanks for doing it the right way.

ellisonbg · 2013-07-01T03:51:37Z

Looks like you missed by last two commits - the ones that fixed the Travis run.

minrk · 2013-07-01T03:56:32Z

ok, they weren't there when I merged. Will cherry pick.

minrk · 2013-07-01T03:59:38Z

Scratch that, they were just after the first comments. Should be in now.

ellisonbg · 2013-07-01T04:12:28Z

OK our Travis build is passing again. Only one more thing to figure out: do we want to ipython nbconvert or ipython convert as the subcommand? Other than that, I would like to merge so @jdfreder can continue his work on adding tests in the morning.

minrk · 2013-07-01T04:29:28Z

I don't have strong feelings on nbconvert vs convert - maybe 51% in favor of just convert?

fperez · 2013-07-01T05:29:31Z

I sort of preferred nbconvert for being more explicit, but when I really think about it, it's hard to imagine ipython convert foo.ipynb being in any way confusing. So I'm starting to warm up to plain convert myself...

ellisonbg · 2013-07-01T05:31:17Z

I think the main question if if convert alone is ambiguous in any way. I can't think of anything else we would want to "convert" at the command line.

fperez · 2013-07-01T05:38:05Z

That's what I was thinking, and honestly from that perspective, I don't see a problem with plain convert.

fperez · 2013-07-01T05:39:00Z

Though if we really want to remain future-proof, the nb prefix isn't that terrible either. Honestly I can be persuaded either way, I don't really have a very strong preference.

ellisonbg · 2013-07-01T05:53:15Z

I too am just about 50/50...makes it a bit tough to decide. Another reason
to use nbconvert is that is what we have been calling it so far.

On Sun, Jun 30, 2013 at 10:39 PM, Fernando Perez
notifications@github.comwrote:

Though if we really want to remain future-proof, the nb prefix isn't that
terrible either. Honestly I can be persuaded either way, I don't really
have a very strong preference.

—
Reply to this email directly or view it on GitHubhttps://github.com//pull/3500#issuecomment-20264196
.

Brian E. Granger
Cal Poly State University, San Luis Obispo
bgranger@calpoly.edu and ellisonbg@gmail.com

jdfreder · 2013-07-01T15:22:03Z

@ellisonbg don't worry about me, I can continue to work on the tests on my machine even though this isn't merged yet. The tests aren't really coupled in a way that would make them break significantly when moved to ipy (just references, easy fix).

ellisonbg · 2013-07-01T17:33:14Z

@jdfreder and I talked about the name of the subcommand. Both of us feel that using nbconvert is slightly better as it is consistent with the code base. This consistency is important because users will interact with the code base in config files. Merging - great work everyone!

Merge nbconvert into IPython

Carreau · 2013-07-01T17:38:34Z

Thanks a lot everyone, I would really like to be able to participate more, but those day I really can't.
Will try to catch up for next hangout.

damianavila · 2013-07-01T20:22:49Z

Great! I just landed a few hours ago and it was already merged... great again! Good job!
And thanks for preserving the history of the commits... now, I am an official IPython contributor! ;-)

Cheers.
Damián.

Merge nbconvert into IPython

minrk and others added 30 commits March 17, 2013 10:24

Merge pull request #109 from Carreau/nodtype

9953d9a

allow nonexistent output

Conditional syntax highlighting, no longer to seperate

9eead17

filters. Filter is now dependent on env. type.

Cleaned up get_lines filter

53a1951

filterOutLatex replaced with built in 'escape_tex' filter

3a4d59b

(see template.py)

Added explicit filters.

45bf487

Sphinx specific metadata now stored in _draft namespace.

ebb2541

Added a nice comment to explain the line splitting.

5d57ab1

Re-arrange imports, added some comments, general cleaning.

34544aa

Added region headers, to help visually separate elements

2e01a53

of the template.

Centeralized all "nbconvert sphinx block" logic (macros)

8f6f645

Added more comments and header comments.

Make sure we dont stomp all over an existing _draft key, value.

0f20af2

Fixed, wrong Sphinx block specifiers used.

3d2e65d

Localized code into jinja macros.

90d97d8

Fixed, backwards defaults.

36d83a4

Fixed, global var interferring with locals vars

43da2d9

Updated comments

0a2f9bb

Centeralized some code into methods Added more comments

Changed maximum line length from 87 to 85 char.

6d48cb3

Fixed string conatation with int value in Jinja.

7663ea6

Fixed single backslash breaking Jinja string output.

4671b25

Removed unnecessary wrapping by listings package.

79c0ec3

Removed indent that showed in verbatim output.

5c4a160

Revert "fix write binary and non math $"

021cba3

This reverts part of commit 994bca5.

PEP8 conformance

766d750

Moved sphinx template related resources into Sphinx namespace

e3e755a

Generalized dictionary based menu input prompting.

1f25d17

Added notebook style input highlighting.

d19125a

Added notebook output formatting style

7403f31

Disable TOC for howto.

8801fcf

Fixed image sizing problem

a11f301

Small images were getting streched to larger scales.

ellisonbg and others added 5 commits June 30, 2013 18:04

Fixing import for nbconvert.

45d39d2

Fixing install logic for nbconvert.

479fc2b

Adding nbconvert subcommand.

4925542

Rename nbconvert entry point to the launch_new_instance.

e8cfe6f

remove nbconvert.py

7457488

minrk mentioned this pull request Jul 1, 2013

Merge nbconvert into IPython #3497

Closed

5 tasks

ellisonbg added 2 commits June 30, 2013 20:59

Fixing imports in setupbase.py.

da3c51a

Fixing test logic for nbconvert to get tests to pass.

acd2e7d

ellisonbg added a commit that referenced this pull request Jul 1, 2013

Merge pull request #3500 from ipython/nbconvert

fba37df

Merge nbconvert into IPython

ellisonbg merged commit fba37df into master Jul 1, 2013

damianavila mentioned this pull request Jul 4, 2013

Add reveal.js library (version 2.4.0) inside IPython #3499

Closed

minrk deleted the nbconvert branch June 20, 2014 18:17

mattvonrocketstein pushed a commit to mattvonrocketstein/ipython that referenced this pull request Nov 3, 2014

Merge pull request ipython#3500 from ipython/nbconvert

e58338c

Merge nbconvert into IPython

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Merge nbconvert into IPython #3500

Merge nbconvert into IPython #3500

minrk commented Jul 1, 2013

minrk commented Jul 1, 2013

minrk commented Jul 1, 2013

ellisonbg commented Jul 1, 2013

ellisonbg commented Jul 1, 2013

minrk commented Jul 1, 2013

minrk commented Jul 1, 2013

ellisonbg commented Jul 1, 2013

minrk commented Jul 1, 2013

fperez commented Jul 1, 2013

ellisonbg commented Jul 1, 2013

fperez commented Jul 1, 2013

fperez commented Jul 1, 2013

ellisonbg commented Jul 1, 2013

jdfreder commented Jul 1, 2013

ellisonbg commented Jul 1, 2013

Carreau commented Jul 1, 2013

damianavila commented Jul 1, 2013

Merge nbconvert into IPython #3500

Merge nbconvert into IPython #3500

Conversation

minrk commented Jul 1, 2013

minrk commented Jul 1, 2013

minrk commented Jul 1, 2013

ellisonbg commented Jul 1, 2013

ellisonbg commented Jul 1, 2013

minrk commented Jul 1, 2013

minrk commented Jul 1, 2013

ellisonbg commented Jul 1, 2013

minrk commented Jul 1, 2013

fperez commented Jul 1, 2013

ellisonbg commented Jul 1, 2013

fperez commented Jul 1, 2013

fperez commented Jul 1, 2013

ellisonbg commented Jul 1, 2013

jdfreder commented Jul 1, 2013

ellisonbg commented Jul 1, 2013

Carreau commented Jul 1, 2013

damianavila commented Jul 1, 2013