Updated readme with modern guidelines on how to build docs (#51)

Author: abuts

README.md

@@ -1,24 +1,48 @@
 Documentation
 =============
-This is the documentation for the [MATLAB](https://github.com/RascalSoftware/RAT) and [Python](https://github.com/RascalSoftware/python-RAT) versions of the RAT project. 
+This is the documentation for the [MATLAB](https://github.com/RascalSoftware/RAT) and [Python](https://github.com/RascalSoftware/python-RAT) versions of the RAT project. The installation instruction provided here, assumes [conda](https://docs.conda.io/) is used to manage python packages, please adapt as needed for a different package manager.
 
 Build docs
 ----------
 The documentation should be built using the provided Sphinx make file. The reStructuredText source is in the source 
 folder while the build will be placed in a build folder. The build requires a python executable and the python packages 
-in the requirements.txt. 
+in the requirements.txt. You need MATLAB version and python version or RAT software installed in your system.
 
-    conda create -n RAT python=3.9
-    conda activate RAT
-    pip install -r requirements.txt
+```bash
+   conda create -n RAT python=3.9
+   conda activate RAT
+   pip install -r requirements.txt
+```
 
-Download the appropriate version of RAT from the GitHub [release](https://github.com/RascalSoftware/RAT/releases) page, and unzip the contents into a folder called API (This folder should be in the root directory). For example on a Linux machine, the nightly can be downloaded as shown 
+You also must have `pandoc` installed to build the Python example Jupiter notebooks. See the installation instructions [here](https://pandoc.org/installing.html). If not previously installed system-wide, install [pandoc](https://pandoc.org/) using [conda](https://docs.conda.io/) as described in their [installation manual](https://pandoc.org/installing.html#conda-forge)
+depending on [conda](https://docs.conda.io/) flavour you are using.
 
+If you do not have RAT code on your machine, download the appropriate version of RAT from the GitHub [release](https://github.com/RascalSoftware/RAT/releases) page, and unzip the contents into a folder called API (This folder should be located within RAT-documentation directory, alongside the **source** folder and main `make.bat` file (see below)). For example on a Linux machine, the nightly can be downloaded as shown:
+
+```bash
+    cd path_to_rat_docs_folder
     wget https://github.com/RascalSoftware/RAT/releases/download/nightly/Linux.zip
     unzip Linux.zip -d API/
+```
+
+Where `path_to_rat_docs_folder` is the location where documentation is cloned or unpacked.
+If RAT is already present, you may create symbolic link to the existing` RAT` directory, e.g.:
+
+for Windows:
+
+    cd path_to_rat_docs_folder  
+    mklink /j API path_to_matlab_rat_folder
+    
+for linux/macOS:
 
-You also must have `pandoc` installed to build the Python example Jupyter notebooks. See the installation instructions [here](https://pandoc.org/installing.html).
+```bash
+   cd path_to_rat_docs_folder
+   ln -s path_to_matlab_rat_folder  API
+```
 
+Where `path_to_rat_docs_folder` is the path to the RAT documentation repository and `path_to_matlab_rat_folder` is the path to the downloaded MATLAB RAT release
+    
+[Python RAT API](https://github.com/RascalSoftware/python-RAT)  should be installed on the same python virtual environment created earlier. Build process adds modules necessary for generating python documentation to the python modules search path. Look at [Python RAT repository](https://github.com/RascalSoftware/python-RAT/blob/main/README.md) for more information on how to build python API.
 
 To build the HTML docs, type the following into a terminal with access to the Python executable: