Signed-off-by: Wissam Abu Ahmad <wabuahma@redhat.com>

With the release of PEP-639 the best practice for specifying the license
is now to use a license expression in the license field and specify any
license files in license-files rather than the table-based approach from
PEP-621. Including the license in the classifiers is also no longer
allowed when using PEP-639 and has been removed.

Signed-off-by: Chris Marchbanks <csmarchbanks@gmail.com>

Signed-off-by: Chris Marchbanks <csmarchbanks@gmail.com>

Part of #1013

Signed-off-by: Owen Williams <owen.williams@grafana.com>

UTF-8 Content Negotiation

Readd parts of `MANIFEST.in` responsible for including the test data
in the source distribution.  Without that, setuptools includes only
`.py` files from the test tree, leading to test failures.

Fixes #1112

Signed-off-by: Michał Górny <mgorny@gentoo.org>

Signed-off-by: Chris Marchbanks <csmarchbanks@gmail.com>

This saves ~10% in the benchmark.

Signed-off-by: Chris Marchbanks <csmarchbanks@gmail.com>

Enumerating rather than using a while loop saves significant CPU when
looking for an unquoted character. This ends up improving the benchmark
~20% on its own.

Signed-off-by: Chris Marchbanks <csmarchbanks@gmail.com>

Split the term into the label name and label value portions in one swoop
rather than starting from the beginning to find an = character after
already going through the full term. This saves ~5% on the benchmark.

Signed-off-by: Chris Marchbanks <csmarchbanks@gmail.com>

Signed-off-by: Chris Marchbanks <csmarchbanks@gmail.com>

Optimize parser

* Add support to write_to_textfile for custom tmpdir

While the try/except block does prevent most of the temp files from
persisting, if there is a non catchable exception, those temp files
continue to pollute the directory. Optionally set the temp
directory would let us write to something like /tmp, so the target
directory isn't polluted

Signed-off-by: Aaditya Dhruv <aadityadhruv@mailbox.org>

* Modify write_to_textfile to ensure tmpdir is on same filesystem

The tmpdir must be on the same filesystem to ensure an atomic operation
takes place. If this is not enforced, there could be partial writes
which can lead to partial/incorrect metrics being exported

Signed-off-by: Aaditya Dhruv <aadityadhruv@mailbox.org>

---------

Signed-off-by: Aaditya Dhruv <aadityadhruv@mailbox.org>

* Start implement OM text exposition for nh, add first no obs test
* Correct template for nh sample spans, add test
* Correct templating and appending for deltas, add longer spans test
* Add tests for nh with labels, remove labels sorting
* Break down logic classic vs nh samples, add tests for classic-native histograms cohabitation
* Move classic sample logic back to where it belongs
* Assign nh to value, correct nil values in tests, clean up white spaces
* Add logic for exposing nh exemplars
* Please linters
* Assign nh_exemplars to exemplarstr
* Add Any type to metric_family in OM exposition test
* Change printing order of  nh spans and deltas according to OM 2.0 proposal
* Shorten name of spans and deltas as per OM 2.0 proposal
* Adapt nh with UTF-8 tests to new testing framework
* Update prometheus_client/openmetrics/exposition.py
* Update prometheus_client/openmetrics/exposition.py
* Eliminate erroneous abbreviation for spans and deltas

---------

Signed-off-by: Arianna Vespri <arianna.vespri@yahoo.it>
Signed-off-by: Arianna Vespri <36129782+vesari@users.noreply.github.com>
Co-authored-by: Chris Marchbanks <csmarchbanks@gmail.com>