FEAT - All Docker packaging

- added new `docker-package-pypi.sh` script for Docker based `twine` upload;
- changes to `package-pypi.sh` for Docker use:
  - needs to run in script dir for Docker;
  - use explicit config file for credentials;
  - added `twine` installation via `pip`;
- `docker-package-win.sh` now uses `cdrx/pyinstaller-windows` Docker image for Wine and Windows Python packaging;
- added new `jpylyzer.spec` suitable for win32 and win64 builds, these are now controlled via Docker image choice;
- added `requirements.txt` for automatic installation of `six` when building windows packages;
- updated packaging documentation;
- enabled long description in `setup.py`;
- added `realpath` installation to `.travis.yml`;
- removed unneeded `Dockerfile-Wine.build`, `buildwin.sh`, `jpylyzer_win32.spec`, `jpylyzer_win64.spec`;
- small `README.md` tweaks; and
- tidied `.travis.yml`.

.travis.yml

@@ -8,13 +8,9 @@ services:
   - docker
 
 before_install:
+  - sudo apt-get install -y realpath
   - python --version
-  - pip install -U pip
-  - pip install -U pytest
-  - pip install -U pyinstaller
-  - pip install -U pep257
-  - pip install -U twine
-  - pip install -U codacy-coverage
+  - pip install -U pip pytest pyinstaller pep257 twine codacy-coverage
   # Make the Jpylyzer test files available by cloning the repo as a sibling directory
   - git clone https://github.com/openpreserve/jpylyzer-test-files.git ../jpylyzer-test-files
 
@@ -49,9 +45,7 @@ script:
   - ./setup.py sdist
   - twine check dist/*
   - ./docker-package.sh debian:stretch
-  - ./docker-package-win.sh debian:stretch
-# DISABLED - to make way for debian packaging above
-#  - ./build-with-pyinstaller.sh
+  - ./docker-package-win.sh
 
 after_success:
   - ls -alh ./dist

BUILD_HOWTO_WINDOWS.md

@@ -3,33 +3,28 @@
 
 ## Prerequisites
 
-Windows binaries are now built under Linux using [Wine](https://www.winehq.org/). Make sure you have a recent version of Wine installed. Also check if (a recent version of) the *winbind* package is installed. If not, install it using:
-
-    sudo apt-get install winbind
-
+The windows packaging is now Docker based. In order to run Docker as a non-root user, it is helpful to create a Unix *docker* group. Steps (see also the more detailed discussion [here](https://docs.docker.com/install/linux/linux-postinstall/)):
+
+1. Create the *docker* group:
+    ```
+    sudo groupadd docker
+    ```
+1. Add the current user to the group:
+    ```
+    sudo usermod -aG docker $USER
+    ```
+1. Then log out and then log back in for the changes to take effect, or run the following command:
+    ```
+    newgrp docker
+    ```
+1. Verify that everything works by running the following test:
+    ```
+    docker run hello-world
+    ```
 ## Building the binaries
 
 In your console window, go to the root of the jpylyzer directory. Then run:
-
-    ./buildwin.sh
-
-The script first checks for the presence of (portable) 64 and 32 versions of Python 2.7 under *Wine*, and installs them if they are not found. Note that the installers need some manual input. Most importantly, make sure you enter the following installation paths:
-
-* `C:\Python27_64` for the 64-bit version;
-
-* `C:\Python27_32` for the 32-bit version.
-
-The script also automatically installs PyInstaller ifit is not there already.
-
-Once the above dependencies are installed, 64 and 32 bit binaries are built automatically. The (zipped) binaries can be found in the *dist* directory.
-
-## Troubleshooting
-
-If the output of the build script includes this error:
-
-    err:winediag:SECUR32_initNTLMSP ntlm_auth was not found or is outdated. Make sure that ntlm_auth >= 3.0.25 is in your path. Usually, you can find it in the winbind package of your distribution.
-
-Fix this by installing *winbind* (see the top of this page).
-
-
-
+    ```
+    ./docker-package-win.sh
+    ```
+The (zipped) binaries can be found in the `dist` directory.

README.md

@@ -10,12 +10,11 @@ Please visit the jpylyzer homepage for links to the most recent package download
 <http://jpylyzer.openpreservation.org/>
 
 
-## Cd Status
+## CI Status
 
 - [![Build Status](https://travis-ci.org/openpreserve/jpylyzer.svg?branch=master)](https://travis-ci.org/openpreserve/jpylyzer "Jpylyzer Travis-CI integration build") Travis-CI
 
-- [![Build Status](http://jenkins.opf-labs.org/buildStatus/icon?job=jpylyser)](http://jenkins.opf-labs.org/job/jpylyser/) OPF Jenkins
-
+- [![Build Status](https://jenkins.openpreservation.org/buildStatus/icon?job=jpylyzer%2Fjpylyzer)](https://jenkins.openpreservation.org/job/jpylyzer/job/jpylyzer/) OPF Jenkins
 <!-- Start of text to be copied to usage.md of jpylyzer website -->
 
 ## Using jpylyzer from the command line
@@ -92,7 +91,7 @@ with *pip*. Then import *jpylyzer* into your code by adding:
 from jpylyzer import jpylyzer
 ```
 Subsequently you can call any function that is defined in *jpylyzer.py*.
-In practice you will most likely only need the *checkOneFile* function. 
+In practice you will most likely only need the *checkOneFile* function.
 The following minimal script shows how this works:
 
 ```python
@@ -111,7 +110,7 @@ imageHeight = myResult.findtext('./properties/jp2HeaderBox/imageHeaderBox/height
 print(imageHeight)
 ```
 
-Here, *myResult* is an *Element* object that can either be used directly, 
+Here, *myResult* is an *Element* object that can either be used directly,
 or converted to XML using the *ElementTree* module[^3]. The structure of the
 element object follows the XML output that described in [Chapter 5](#output-format).
 

docker-package-pypi.sh

@@ -0,0 +1,12 @@
+#! /usr/bin/env bash
+#
+# Build Debian package in a Docker container
+#
+
+set -e
+
+# Get build platform as 1st argument, and collect project metadata
+pypi_name="$(./setup.py --name)"
+
+# Build in Docker container, save results, and show package info
+docker run -v "$PWD":/src -v "$HOME":/pypirc --entrypoint "/src/package-pypi.sh" --rm python:3.6-stretch