Links to .md files require # on Sphinx

Fix #255

docs/formats.md

@@ -15,7 +15,7 @@ Jupytext's implementation for Jupyter notebooks as [Markdown](https://daringfire
 - Code and raw cells are encoded as Markdown code blocks with triple backticks. In a Python notebook, a code cell starts with ` ```python` and ends with ` ``` `. Cell metadata are found after the language information, with a `key=value` syntax, where `value` is encoded in JSON format (Markdown) or R format (R Markdown). R Markdown [code cell options](https://yihui.name/knitr/options/) are mapped to the corresponding Jupyter cell metadata options, when available.
 - Markdown cells are inserted verbatim and separated with two blank lines. When required (cells with metadata, cells that contain two blank lines or code blocks), Jupytext protects the cell boundary with HTML comments: `<!-- #region -->` and `<!-- #endregion -->`. Cells with explicit boundaries are [foldable](https://code.visualstudio.com/docs/editor/codebasics#_folding) in vscode, and can accept both a title and/or metadata in JSON format: `<!-- #region This is the title for my protected cell {"key": "value"}-->`.
 
-See how our `World population.ipynb` notebook in the [demo folder](https://github.com/mwouts/jupytext/tree/master/demo) is represented in [Markdown](https://github.com/mwouts/jupytext/blob/master/demo/World%20population.md) or [R Markdown](https://github.com/mwouts/jupytext/blob/master/demo/World%20population.Rmd).
+See how our `World population.ipynb` notebook in the [demo folder](https://github.com/mwouts/jupytext/tree/master/demo) is represented in [Markdown](https://github.com/mwouts/jupytext/blob/master/demo/World%20population.md#) or [R Markdown](https://github.com/mwouts/jupytext/blob/master/demo/World%20population.Rmd).
 
 When you open a plain Markdown file in Jupytext, the Markdown text is rendered in Markdown cells. Cells breaks occur when the text contains two consecutive blank lines (or code cells). If you want to also split cells on Markdown headers, so that headers prefixed by one blank line appear at the top of a new cell, use the `split_at_heading` option. Set the option either on the command line, or by adding `"split_at_heading": true` to the jupytext section in the notebook metadata, or on Jupytext's content manager:
 
@@ -32,7 +32,7 @@ Pandoc's Markdown format is available in Jupytext as `md:pandoc`. This requires
 conda install pandoc -c conda-forge
 ```
 
-Pandoc's format uses Pandoc divs (`:::`) as explicit cell markers. See how our `World population.ipynb` notebook is [represented](https://github.com/mwouts/jupytext/blob/master/demo/World%20population.pandoc.md#a-quick-insight-at-world-population) in that format. Please also note that `pandoc`, while preserving the HTML rendering, may reformat the text in some of the Markdown cells. If that is an issue for you, please wait until [jgm/pandoc#5408](https://github.com/jgm/pandoc/issues/5408) gets implemented.
+Pandoc's format uses Pandoc divs (`:::`) as explicit cell markers. See how our `World population.ipynb` notebook is [represented](https://github.com/mwouts/jupytext/blob/master/demo/World%20population.pandoc.md#) in that format. Please also note that `pandoc`, while preserving the HTML rendering, may reformat the text in some of the Markdown cells. If that is an issue for you, please wait until [jgm/pandoc#5408](https://github.com/jgm/pandoc/issues/5408) gets implemented.
 
 Jupytext currently strips the output cells before calling `pandoc`. As for the other formats, outputs cells can be preserved in paired notebooks. In the case of the pandoc format, paired notebooks are available with the metadata `"jupytext": {"formats": "md:pandoc,ipynb"}`.
 

docs/introduction.md

@@ -10,7 +10,7 @@ Jupytext can save Jupyter notebooks as
 - Markdown and R Markdown documents,
 - Scripts in many languages.
 
-The languages that are currently supported by Jupytext are: Julia, Python, R, Bash, Scheme, Clojure, Matlab, Octave, C++, q/kdb+, IDL, TypeScript and Javascript. Extending Jupytext to more languages should be easy - read more at [CONTRIBUTING.md](https://github.com/mwouts/jupytext/blob/master/CONTRIBUTING.md). In addition, jupytext users can choose between two formats for notebooks as scripts:
+The languages that are currently supported by Jupytext are: Julia, Python, R, Bash, Scheme, Clojure, Matlab, Octave, C++, q/kdb+, IDL, TypeScript and Javascript. Extending Jupytext to more languages should be easy - read more at [CONTRIBUTING.md](https://github.com/mwouts/jupytext/blob/master/CONTRIBUTING.md#). In addition, jupytext users can choose between two formats for notebooks as scripts:
 - The `percent` format, compatible with several IDEs, including Spyder, Hydrogen, VScode and PyCharm. In that format, cells are delimited with a commented `%%`.
 - The `light` format, designed for this project. Use that format to open standard scripts as notebooks, or to save notebooks as scripts with few cell markers - none when possible.
 
@@ -42,7 +42,7 @@ In short: the Markdown representation of notebooks fits well the notebooks that
 
 Looking for a demo?
 - Read the original [announcement](https://towardsdatascience.com/introducing-jupytext-9234fdff6c57) in Towards Data Science,
-- Watch the [PyParis talk](https://github.com/mwouts/jupytext_pyparis_2018/blob/master/README.md),
+- Watch the [PyParis talk](https://github.com/mwouts/jupytext_pyparis_2018/blob/master/README.md#),
 - or, try Jupytext online with [binder](https://mybinder.org/v2/gh/mwouts/jupytext/master?urlpath=lab/tree/demo/get_started.ipynb)!
 
 ## Want to contribute?