Release v0.4.0 · noseglasses/elf_diff

Badges at the top of README.md file
Documents Weasyprint usage in README.md
Documents PyPI package name inconsistency
Mangling files for compilers with exotic name mangling that do not provide binutils
Similar symbols detection and output can now explicitly disabled for performance reasons
CI tests on Windows
CI code coverage testing (line & branch coverage)
CI codacy code quality testing
Github CI now deploys examples to gh_pages
CONTRIBUTING.md file
CODE_OF_CONDUCT.md file
Github issue templates
A second simple example for testing and documentation (also deployed on gh_pages)
Balloons with additional information are displayed when hovering over HTML table headers
Statistics now comprise the total number and the number of selected and excluded symbols
of each binary in addition to the selection and exclusion regex
Added means to gess whether symbol demangling is available
Plugin system added
Report generation now taken over by plugins
Added YAML, JSON and TXT reports
Between the extraction of symbols from the binaries and the report generation a
new central data structure, the document tree is established
Jinja2 templates of the html export plugin now operate directly on the document tree
Custom plugins can be loaded from the command line
Command line flag added to write the document tree to stdout
HTML reports now additionally output tree formatted information to aid changing the Jinja templates or writing custom versions

Initial example in README.md improved
README.md restructured
Added and improved tests
Tests write their output to a directory structure that matches the fixture and test names
of the Python test definition
Python requirements mostly tied to exact versions
Mass reports deprecated
Mass reports Jinja template simplified
Symbol type 'new' renamed to 'appeared in order not to conflict with the distinction old/new
'Symbol similarity' renamed 'signature similarity'

Provide feedback