view contrib/packaging/build-linux-wheels.sh @ 41033:b373477948df

revlog: limit base to rev size ratio to 500 instead of 50

While a value of 50 provided large speedup in some case (400s → 7s) it also
creates a slow down for a whole class of revision we are seeing in a private
repository (0.1s → 3s). A value of 500 makes them disappear improving the
total runtime (the slower revision still improve significantly (400s → 21s)).

We need to run a wider array of tests on various repositories to see the
effect on speed and size of different values for this acceptable constant.
However, in the meantime, it seems safer to move back to a less impactful
value.

#!/bin/bash
# This file is directly inspired by
# https://github.com/pypa/python-manylinux-demo/blob/master/travis/build-wheels.sh
set -e -x

PYTHON_TARGETS=$(ls -d /opt/python/cp27*/bin)

# Create an user for the tests
useradd hgbuilder

# Bypass uid/gid problems
cp -R /src /io && chown -R hgbuilder:hgbuilder /io

# Compile wheels for Python 2.X
for PYBIN in $PYTHON_TARGETS; do
    "${PYBIN}/pip" wheel /io/ -w wheelhouse/
done

# Bundle external shared libraries into the wheels with
# auditwheel (https://github.com/pypa/auditwheel) repair.
# It also fix the ABI tag on the wheel making it pip installable.
for whl in wheelhouse/*.whl; do
    auditwheel repair "$whl" -w /src/wheelhouse/
done

# Install packages and run the tests for all Python versions
cd /io/tests/

for PYBIN in $PYTHON_TARGETS; do
    # Install mercurial wheel as root
    "${PYBIN}/pip" install mercurial --no-index -f /src/wheelhouse
    # But run tests as hgbuilder user (non-root)
    su hgbuilder -c "\"${PYBIN}/python\" /io/tests/run-tests.py --with-hg=\"${PYBIN}/hg\" --blacklist=/io/contrib/packaging/linux-wheel-centos5-blacklist"
done