[bug#68391] [PATCH 1/3] gnu: Add python-typeguard-4.

[bug#68391] [PATCH 1/3] gnu: Add python-typeguard-4.

[Troy Figiel]

* gnu/packages/python-xyz.scm (python-typeguard-4): New variable.
---
 gnu/packages/python-xyz.scm | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/gnu/packages/python-xyz.scm b/gnu/packages/python-xyz.scm
index 46a5e81cb4..d7bbdbdb13 100644
--- a/gnu/packages/python-xyz.scm
+++ b/gnu/packages/python-xyz.scm
@@ -24804,6 +24804,24 @@ (define-public python-typeguard
 with PEP 484 argument (and return) type annotations.")
     (license license:expat)))
 
+(define-public python-typeguard-4
+  (package
+    (inherit python-typeguard)
+    (name "python-typeguard")
+    ;; This is the latest version we can use, becaue python-typeguard >= 4.0.1
+    ;; requires python-typing-extensions >= 4.7.0.
+    (version "4.0.0")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (pypi-uri "typeguard" version))
+       (sha256
+        (base32 "11yrc7pv2fajmicwiyc92sb5gphlw3zbxww8f2prrsh6rgdv6kqr"))))
+    (build-system pyproject-build-system)
+    (arguments
+     '())
+    (propagated-inputs (list python-typing-extensions))))
+
 (define-public bpython
   (package
     (name "bpython")
-- 
2.42.0

[bug#68391] [PATCH 2/3] gnu: Add python-pandas-stubs.

[Troy Figiel]

* gnu/packages/python-science.scm (python-pandas-stubs): New variable.
---
 gnu/packages/python-science.scm | 58 +++++++++++++++++++++++++++++++++
 1 file changed, 58 insertions(+)

diff --git a/gnu/packages/python-science.scm b/gnu/packages/python-science.scm
index 8065de0e99..d8e0b343fb 100644
--- a/gnu/packages/python-science.scm
+++ b/gnu/packages/python-science.scm
@@ -561,6 +561,64 @@ (define-public python-pandas
 doing practical, real world data analysis in Python.")
     (license license:bsd-3)))
 
+(define-public python-pandas-stubs
+  (package
+    (name "python-pandas-stubs")
+    ;; The versioning follows that of Pandas and uses the date of the
+    ;; python-pandas-stubs release. This is the latest version of
+    ;; python-pandas-stubs for python-pandas 1.4.4.
+    (version "1.4.4.220919")
+    (source
+     (origin
+       ;; No tests in the PyPI tarball.
+       (method git-fetch)
+       (uri (git-reference
+             (url "https://github.com/pandas-dev/pandas-stubs")
+             (commit (string-append "v" version))))
+       (file-name (git-file-name name version))
+       (sha256
+        (base32 "14fhj1y71akwl41ws7cpazsbq5b8wf4rwaydqq2h39q7gylpcp99"))))
+    (build-system pyproject-build-system)
+    (arguments
+     '(#:phases (modify-phases %standard-phases
+                  (add-before 'check 'prepare-x
+                    (lambda _
+                      (system "Xvfb &")
+                      (setenv "DISPLAY" ":0")
+                      ;; xsel needs to write a log file.
+                      (setenv "HOME"
+                              (getcwd)))))))
+    (propagated-inputs (list python-types-pytz))
+    ;; Add python-fastparquet to native inputs once it has been packaged. Its
+    ;; tests will be skipped for now.
+    (native-inputs (list python-lxml
+                         python-matplotlib
+                         python-pandas
+                         python-poetry-core
+                         python-pyarrow
+                         python-pyreadstat
+                         python-pytest
+                         python-scipy
+                         python-sqlalchemy
+                         python-tables
+                         python-tabulate
+                         python-xarray
+                         ;; Needed to test clipboard support.
+                         which
+                         xclip
+                         xorg-server-for-tests
+                         xsel))
+    (home-page "https://pandas.pydata.org")
+    (synopsis "Type annotations for pandas")
+    (description
+     "This package contains public type stubs for @code{python-pandas},
+following the convention of providing stubs in a separate package, as
+specified in @acronym{PEP, Python Enhancement Proposal} 561.  The stubs cover
+the most typical use cases of @code{python-pandas}.  In general, these stubs
+are narrower than what is possibly allowed by pandas, but follow a convention
+of suggesting best recommended practices for using @code{python-pandas}.")
+    (license license:bsd-3)))
+
 (define-public python-pythran
   (package
     (name "python-pythran")
-- 
2.42.0

[bug#68391] [PATCH 3/3] gnu: Add python-pandera.

[Troy Figiel]

* gnu/packages/python-science.scm (python-pandera): New variable.
---
 gnu/packages/python-science.scm | 75 +++++++++++++++++++++++++++++++++
 1 file changed, 75 insertions(+)

diff --git a/gnu/packages/python-science.scm b/gnu/packages/python-science.scm
index d8e0b343fb..f82cbdb79c 100644
--- a/gnu/packages/python-science.scm
+++ b/gnu/packages/python-science.scm
@@ -619,6 +619,81 @@ (define-public python-pandas-stubs
 of suggesting best recommended practices for using @code{python-pandas}.")
     (license license:bsd-3)))
 
+(define-public python-pandera
+  (package
+    (name "python-pandera")
+    (version "0.17.2")
+    (source
+     (origin
+       ;; No tests in the PyPI tarball.
+       (method git-fetch)
+       (uri (git-reference
+             (url "https://github.com/unionai-oss/pandera")
+             (commit (string-append "v" version))))
+       (file-name (git-file-name name version))
+       (sha256
+        (base32 "1mnqk583z90k1n0z3lfa4rd0ng40v7hqfk7phz5gjmxlzfjbxa1x"))
+       (modules '((guix build utils)))
+       ;; These tests require PySpark. We need to remove the entire directory,
+       ;; since the conftest.py in this directory contains a PySpark import.
+       ;; (See: https://github.com/pytest-dev/pytest/issues/7452)
+       (snippet '(delete-file-recursively "tests/pyspark"))))
+    (build-system pyproject-build-system)
+    (arguments
+     (list
+      #:test-flags '(list "-k"
+                          (string-append
+                           ;; Needs python-pandas >= 1.5
+                           "not test_python_std_list_dict_generics"
+                           " and not test_python_std_list_dict_empty_and_none"
+                           " and not test_pandas_modules_importable"))))
+    ;; Pandera comes with a lot of extras. We test as many as possible, but do
+    ;; not include all of them in the propagated-inputs. Currently, we have to
+    ;; skip the pyspark and io tests due to missing packages python-pyspark
+    ;; and python-frictionless.
+    (propagated-inputs (list python-hypothesis ;strategies extra
+                             python-multimethod
+                             python-numpy
+                             python-packaging
+                             python-pandas
+                             python-pandas-stubs ;mypy extra
+                             python-pydantic
+                             python-scipy ;hypotheses extra
+                             python-typeguard-4
+                             python-typing-inspect
+                             python-wrapt))
+    (native-inputs (list python-dask ;dask extra
+                         python-fastapi ;fastapi extra
+                         python-geopandas ;geopandas extra
+                         python-modin ;modin extra
+                         python-pyarrow ;needed to run fastapi tests
+                         python-pytest
+                         python-pytest-asyncio
+                         python-sphinx
+                         python-uvicorn)) ;needed to run fastapi tests
+    (home-page "https://github.com/unionai-oss/pandera")
+    (synopsis "Perform data validation on dataframe-like objects")
+    (description
+     "@code{python-pandera} provides a flexible and expressive API for
+performing data validation on dataframe-like objects to make data processing
+pipelines more readable and robust.  Dataframes contain information that
+@code{python-pandera} explicitly validates at runtime.  This is useful in
+production-critical data pipelines or reproducible research settings.  With
+@code{python-pandera}, you can:
+
+@itemize
+@item Define a schema once and use it to validate different dataframe types.
+@item Check the types and properties of columns.
+@item Perform more complex statistical validation like hypothesis testing.
+@item Seamlessly integrate with existing data pipelines via function decorators.
+@item Define dataframe models with the class-based API with pydantic-style syntax.
+@item Synthesize data from schema objects for property-based testing.
+@item Lazily validate dataframes so that all validation rules are executed.
+@item Integrate with a rich ecosystem of tools like @code{python-pydantic},
+@code{python-fastapi} and @code{python-mypy}.
+@end itemize")
+    (license license:expat)))
+
 (define-public python-pythran
   (package
     (name "python-pythran")
-- 
2.42.0

[bug#68391] [PATCH 0/3] Add python-pandera

[Troy Figiel]

This patch series adds python-pandera and its dependencies.

Unfortunately, directly upgrading python-typeguard to a more recent
version breaks a variety of packages (it has 300+ dependents). For that
reason, python-typeguard-4 has been introduced.

Troy Figiel (3):
  gnu: Add python-typeguard-4.
  gnu: Add python-pandas-stubs.
  gnu: Add python-pandera.

 gnu/packages/python-science.scm | 133 ++++++++++++++++++++++++++++++++
 gnu/packages/python-xyz.scm     |  18 +++++
 2 files changed, 151 insertions(+)


base-commit: 637b72e2b83a6332849218ef1f193124fa8239eb
-- 
2.42.0

[bug#68391] Small description adjustment

[Troy Figiel]

I just noticed there is a single `pandas' I did not rewrite as
`@code{python-pandas}' in the description of `python-pandas-stubs'. If
the patches look good as they are, I kindly ask a committer to make that
small adjustment before applying them, as to keep the look of the
description uniform.

Thanks!

Troy

bug#68391: Small description adjustment

[Mathieu Othacehe]

Hey,

> I just noticed there is a single `pandas' I did not rewrite as
> `@code{python-pandas}' in the description of `python-pandas-stubs'. If
> the patches look good as they are, I kindly ask a committer to make that
> small adjustment before applying them, as to keep the look of the
> description uniform.

Fixed that and a typo in the first patch before applying,

Thanks,

Mathieu