[pawank925]

Split writer/reader parts to support libzim built without writer (e.g. Android/reader-only builds).

Fixes #287

[codecov]

Codecov Report

❌ Patch coverage is 95.51282% with 7 lines in your changes missing coverage. Please review.
✅ Project coverage is 95.28%. Comparing base (f3ed793) to head (1f37944).

Files with missing lines	Patch %	Lines
libzim/libzim.pyx	95.51%	7 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #292      +/-   ##
==========================================
+ Coverage   93.70%   95.28%   +1.58%     
==========================================
  Files           1        1              
  Lines         635      636       +1     
==========================================
+ Hits          595      606      +11     
+ Misses         40       30      -10     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[pawank925]

Hey @benoit74 — a design question on the approach for #287 before this goes further.

#287 floated a pip writer extra: reader-only core by default, writer bits via libzim[writer], plus a way to mark the extra unavailable on some platforms. This PR instead does compile-time detection:

• setup.py detects at build time whether the linked libzim was built without writer support (LIBZIM_WITHOUT_WRITER in zim_config.h, falling back to the presence of zim/writer/creator.h).
• When detected, it passes LIBZIM_WITHOUT_WRITER to Cython (compile_time_env) and to the C++ wrappers (-D), so writer-only wrappers/callbacks are not compiled at all.
• The libzim.writer submodule is only registered when writer support is present, so import libzim.writer raises ImportError on reader-only builds (tests skip via a HAS_WRITER probe).

Rationale: one wheel keeps working against both kinds of libzim, and there is no extra whose availability has to be modeled per platform. Does this satisfy what you need for #283, or would you prefer the writer extra packaging route from #287?