jupytext

Improvements

• Jupytext command line has more arguments: --paired-paths to list the paths for the paired representations of the notebook, and --sync to synchronise the content of all paired paths based on the most recent file (#146). In addition, the --from argument is optional even when the notebook is read from stdin (#148).
• The pairing information, and more generally the notebook metadata can be edited with the CLL, see the --set-formats and the --update-metadata arguments (#141).
• Jupytext can --pipe the text representation of a notebook to external programs like black or flake8 (#154, #142)
• The Python representation of notebooks containing PEP8 cells is now expected to be PEP8 compliant (#154).
• Format specification allow prefix and suffix for path and file name (#138, #142). Use ipynb,prefix/suffix.py:percent to pair the current notebook named notebook.ipynb to a script named prefixnotebooksuffix.py. Suffix and prefix can also be configured on the ipynb file, with the same syntax.
• Introducing a new hydrogen format for scripts, which derives from percent. In that format Jupyter magic commands are not commented (#59, #126, #132).
• Introducing a new bare format for scripts, which derives from light. That format has no cell marker. Use a notebook metadata filter {"jupytext": {"notebook_metadata_filter":"-all"}} if you want no YAML header (#152).
• The default format for R script is now light, as for the other languages.
• Added support for q/kdb+ notebooks (#161)
• Python scripts or Markdown documents that have no Jupyter metadata receive a metadata filter that ensures that metadata is not exported back to the text representation (#124).
• Metadata filters are represented as strings rather than dictionaries to make YAML headers shorter. Previous syntax from #105 is still supported. They were also renamed to notebook_metadata_filter and cell_metadata_filter.
• Markdown and RMarkdown formats have a new option split_at_heading to split Markdown cells at heading (#130)

BugFixes

• Main language of scripts is inferred from script extension. Fixes a round trip conversion issue for Python notebooks with a Javascript cell.
• Non-Python scripts opened as notebooks in Jupyter are now correctly saved even when no matching kernel is found.
• Jupyter magic commands like ls are commented in the light and R markdown format (#149).

vs 1.0.0-rc0

• Fixed support for notebooks with jupytext metadata from previous versions of Jupytext (#163)
• Fixed round-trip conversion (#164 , #165)

vs 1.0.0-rc1

• Added new line when writing ipynb, just like nbformat (#165)
• Improved matching of preferred format when using prefixes/suffixes (#170)
• Piping a notebook (e.g. in percent format) preserves the pairing information (e.g. light format)
• Writing a notebook does not change the notebook object

vs 1.0.0-rc2

• Better support for the auto extension, used by our two notebook extensions https://github.com/mwouts/jupyterlab-jupytext and https://github.com/mwouts/jupytext_nbextension
• Fix incorrect handling of user metadata filters (#139)

rc3 vs 1.0.0-rc2

• Sorry, there was no rc3...

rc4 vs 1.0.0-rc2

• Pairing of notebooks to a different folder even when notebook is at the root #174
• Jupyter notebook and lab extensions are distributed with Jupytext #86

rc5 vs 1.0.0-rc4

• Pairing folder is created when missing #176
• Jupyter notebook and lab extensions are distributed with Jupytext #176
• Html and Latex cell magics are supported #179

Improvements

• The language_info section is not part of the default header any more. Language information is now taken from metadata kernelspec.language. (#105).
• When opening a paired notebook, the active file is now the file that was originally opened (#118). When saving a notebook, timestamps of all the alternative representations are tested to ensure that Jupyter's autosave does not override manual modifications.
• Jupyter magic commands are now commented per default in the percent format (#126, #132). Version for the percent format increases from '1.1' to '1.2'. Set an option comment_magics to false either per notebook, or globally on Jupytext's contents manager, or on jupytext's command line, if you prefer not to comment Jupyter magics.
• Jupytext command line has a pre-commit mode (#121).

Improvements

• The language_info section is not part of the default header any more. Language information is now taken from metadata kernelspec.language. (#105).
• When opening a paired notebook, the active file is now the file that was originally opened (#118). When saving a notebook, timestamps of all the alternative representations are tested to ensure that Jupyter's autosave does not override manual modifications.
• Jupyter magic commands are now commented per default in the percent format (#126, #132). Version for the percent format increases from '1.1' to '1.2'. Set an option comment_magics to false either per notebook, or globally on Jupytext's contents manager, or on jupytext's command line, if you prefer not to comment Jupyter magics.
• Jupytext command line has a pre-commit mode (#121).

• The language_info section is not part of the default header any more. Language information is now taken from metadata kernelspec.language. (#105).
• When opening a paired notebook, the active file is now the file that was originally opened (#118). When saving a notebook, timestamps of all the alternative representations are tested to ensure that Jupyter's autosave does not override manual modifications.
• Jupyter magic commands are now commented per default in the percent format (#126, #132). Version for the percent format increases from '1.1' to '1.2'. Set an option comment_magics to false either per notebook, or globally on Jupytext's contents manager, or on jupytext's command line, if you prefer not to comment Jupyter magics.

Improvements

• bash scripts as notebooks (#127)
• R scripts with .r extension are supported (#122)
• Jupytext selects the first kernel that matches the language (#120)

Improvements

• Notebook metadata is filtered - only the most common metadata are stored in the text representation (#105)
• New config option freeze_metadata on the content manager and on the command line interface (defaults to False). Use this option to avoid creating a YAML header or cell metadata if there was none initially. (#110)
• Language magic arguments are preserved in R Markdown, and also supported in light and percent scripts (#111, #114, #115)
• First markdown cell exported as a docstring when using the Sphinx format (#107)

BugFixes

• percent format has a fail-safe for code cells with square brackets in the title (#116)

Improvements

• Notebook metadata is filtered - only the most common metadata are stored in the text representation (#105)
• New config option additional_metadata_on_text_files on the content manager. Defaults to True. Change its value to False to avoid creating a YAML header or cell metadata if there was none initially (#110)
• Language magic arguments are preserved in R Markdown, and also supported in light and percent scripts (#111, #114)
• First markdown cell exported as a docstring when using the Sphinx format (#107)

Improvements

• Notebook metadata is filtered - only the most common metadata are stored in the text representation (#105)
• New config option additional_metadata_on_text_files on the content manager. Defaults to True. Change its value to False to avoid creating a YAML header or cell metadata if there was none initially (#110)
• Language magic arguments are preserved (#111)
• First markdown cell exported as a docstring when using the Sphinx format (#107)

Improvements

• Frozen cells are supported in R Markdown, light and percent scripts (#101)
• Inactive cells extended to percent scripts (#108)
• jupytext gains a --version argument (#103)
• "ExecuteTime" cell metadata is not included in the text representation anymore (#106)

Improvements

• Round trip conversion testing with jupytext --test was improved (#99)
• Round trip conversion tested on Jake Vanderplas' Python for Data Science Handbook.

BugFixes

• Nested lists and dictionaries are now supported in notebook metadata
• Final empty code cell supported in Sphinx representation

BugFixes

• Sphinx format tested on World population notebook (#97)
• Mirror test made stronger on this occasion!
• Markdown representation recognize Julia, Scheme and C++ code cells as such
• Light representation of Scheme and C++ notebooks fixed (#61)

Improvements

• All jupytext related metadata goes to a jupytext section (#91). Please make sure your collaborators use the same version of Jupytext, as the new version can read previous metadata, but not the opposite.
• Notebooks extensions can be prefixed with any prefix of at most three chars (#87).
• Export of the same notebook to multiple formats is now supported. To export to all python formats, plus .ipynb and .md, use "jupytext": {"formats": "ipynb,pct.py:percent,lgt.py:light,spx.py:sphinx,md"},.
• README includes a short section on how to extend light and percent formats to more languages (#61).
• Jupytext's contents manager accepts the auto extension in default_jupytext_formats (#93).
• All Jupyter magics are escaped in light scripts and R markdown documents. Escape magics in other formats with a comment_magics metadata (true or false), or with the contents manager comment_magics global flag (#94).

BugFixes

• Trusting notebooks made functional again.
• Command line jupytext returns a meaningful error when no argument is given.
• Fixed global pairing configuration (#95).

Improvements

• light and percent formats made available for scheme and cpp notebooks. Adding more formats is straightforward - just add a new entry to _SCRIPT_EXTENSIONS in languages.py, a sample notebook and a mirror test (#61)
• Format name is automatically appended to extension in jupytext_formats when notebook is loaded/saved.

BugFixes

• Notebooks extensions can only be prefixed with .nb (#87)

BugFixes

• Markdown cells header in sphinx gallery format may have a space between first # and following.

Improvements

• Jupytext can read and write notebooks as Hydrogen/VScode/Spyder/PyCharm compatible scripts (cells starting with # %%) (#59)
• Jupytext can read and write notebooks as Sphinx-gallery compatible scripts (#80)
• Metadata are supported for all cell types in light python and percent formats (#66). Due to this, light python format version is now 1.3. Light python notebooks in versions 1.1 and 1.2 are still readable.
• Command line jupytext has a from argument, and now accepts notebook from the standard input.

BugFixes

• Fix merging of input and output notebooks (#83)
• Removed extra new line on stdout in command line jupytext (#84)

Improvements

• Jupytext can read and write notebooks as Hydrogen/VScode/Spyder/PyCharm compatible scripts (cells starting with # %%) (#59)
• Jupytext can read and write notebooks as Sphinx-gallery compatible scripts (#80)
• Metadata are supported for all cell types in light python and percent formats (#66). Due to this, light python format version is now 1.3. Light python notebooks in versions 1.1 and 1.2 are still readable.
• Command line jupytext has a from argument, and now accepts notebook from the standard input.

BugFixes

• Fix merging of input and output notebooks (#83)
• Removed extra new line on stdout in command line jupytext (#84)

Improvements

• Code lines that start with a quotation mark in Jupyter are commented in the corresponding Python and Julia scripts (#73)
• Update pypy, add flake8 tests on Travis CI (#74)

BugFixes

• Import notebook.transutils before notebook.services.contents.filemanager (#75)

Improvements

• Jupytext will not load paired notebook when text representation is out of date (#63)
• Package tested against Python 3.7 (#68)

BugFixes

• Allow unicode characters in notebook path (#70)
• Read README.md as unicode in setup.py (#71)

Improvements

• Lighter cell markers for Python and Julia scripts (#57). Corresponding file format version at 1.2. Scripts in previous version 1.1 can still be opened.
• New screenshots for the README.

BugFixes

• Command line conversion tool jupytext fixed on Python 2.7 (#46)

Improvements

• Initial support for Jupyter notebooks as Julia scripts (#56)
• Command line conversion tool jupytext has explicit to and output options (#46)
• Round trip test with jupytext --test improved (#54)
• Improved README (#51)

BugFixes

• testfixtures now in requirements (#55)
• Empty code cells are now preserved (#53)

Improvements

• Package and conversion script renamed from nbrmd to jupytext.
• Cell parsing and exporting done in two specialized classes. This is way easier to read. Pylint score at 9.9 !
• Python file format updated to 1.1: default end of cell for python scripts is one blank space. Two blank spaces are allowed as well. Now you can reformat code in Python IDE without breaking notebook cells (#38).
• Added support for plain markdown files (#40, #44).
• Demonstration notebooks more user friendly (#45).
• Command line tool simpler to use (#46).
• Start code patterns present in Jupyter cells are escaped.
• Default nbrmd_format is empty (mwouts/nbsrc/#5): no Jupyter notebook is created on disk when the user opens a Python or R file and saves it from Jupyter, unless the users asks for it by setting nbrmd_format.

BugFixes

• Fixed message in the nbsrc script (#43)
• Technical metadata don't appear any more in scripts unless required (#42)
• Code cells that are fully commented remain code cells after round trip (#41)