⬆️ Propose v/1.4.0 | ⚡ Few Improvements | ➖ Drop dependencies to reduce the footprint

[Ousret]

Changes :

• Dependency: ➖ Using standard logging instead of using the package loguru.
• Dependency: ➖ Dropping nose test framework in favor of the maintained pytest.
• Dependency: ➖ Choose to not use dragonmapper package to help with gibberish Chinese/CJK text. Prefer using characters occurrences instead.
• Dependency: 🔧 ➖ Require cached_property only for Python 3.5 due to constraint. Dropping for every other interpreter version.
• Bugfix: 🐛 BOM marker in a CharsetNormalizerMatch instance could be False in rare case even if obviously present. Due to the sub-match factoring process.
• Improvement: ❇️ Return ASCII if given sequences fit.
• Performance: ⚡ Huge improvement over large payload.
• Change: 🔥 Stop support for UTF-7. Could be reverted

[codecov-commenter]

Codecov Report

Merging #41 (716f9b0) into master (65dea98) will decrease coverage by 0.80%.
The diff coverage is 88.41%.

@@            Coverage Diff             @@
##           master      #41      +/-   ##
==========================================
- Coverage   84.56%   83.76%   -0.81%     
==========================================
  Files          13       14       +1     
  Lines         920      967      +47     
==========================================
+ Hits          778      810      +32     
- Misses        142      157      +15     

Impacted Files	Coverage Δ
charset_normalizer/probe_inherent_sign.py	88.88% <ø> (+38.88%)	⬆️
charset_normalizer/cached_property/__init__.py	77.08% <77.08%> (ø)
charset_normalizer/probe_chaos.py	80.83% <84.00%> (-1.35%)	⬇️
charset_normalizer/normalizer.py	82.68% <85.71%> (-2.93%)	⬇️
charset_normalizer/unicode.py	93.60% <94.11%> (-0.57%)	⬇️
charset_normalizer/__init__.py	100.00% <100.00%> (ø)
charset_normalizer/probe_coherence.py	94.89% <100.00%> (-0.04%)	⬇️
charset_normalizer/probe_words.py	88.88% <100.00%> (+2.77%)	⬆️
charset_normalizer/version.py	100.00% <100.00%> (ø)
... and 2 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 65dea98...716f9b0. Read the comment docs.

[Ousret]

For now, we observe a slight decrease in Chinese-related CP detection. Should be fixed soon. Thinking of ways to resolve this.

[Ousret]

Otherwise,

• Dependencies tree has been reduced to the minimum
• ASCII resolution when content fits to it
• Fix UTF-16 nonsense detection when content contain only numbers (~1KB)
• Performance on large file to be resolved
• Chinese detection to be improved, need reach previous detection threshold/success