v4.0

.github/workflows/python-package.yml

@@ -12,14 +12,14 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: [3.9, "3.10", 3.11, 3.12]
+        python-version: ["3.10", 3.11, 3.12, 3.13, 3.14]
 
     steps:
-    - uses: actions/checkout@v4
+    - uses: actions/checkout@v6
       with:
         submodules: recursive
     - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v5
+      uses: actions/setup-python@v6
       with:
         python-version: ${{ matrix.python-version }}
     - name: Install dependencies

CHANGES

@@ -1,6 +1,19 @@
 Changelog
 =========
 
+Unreleased
+----------
+
+- removed dependency on `attrs`
+
+Backwards incompatibility
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+- Since formerly `@attr.s` decorated classes are now dataclasses, any class inheriting from these
+  will be broken.
+- Some functions have been moved to different modules, so imports may be broken.
+
+
 Version 3.7.0
 -------------
 

README.md

@@ -9,6 +9,10 @@ This package provides
 - a Python API to read and write relational, tabular data according to the [CSV on the Web](https://csvw.org/) specification and 
 - commandline tools for reading and validating CSVW data.
 
+> [!IMPORTANT]  
+> The Python API provided by `csvw` 4.x is not fully backwards compatible with `csvw` < 4.
+> See [CHANGES](CHANGES) for more information.
+
 
 ## Links
 
@@ -19,7 +23,7 @@ This package provides
 
 ## Installation
 
-This package runs under Python >=3.8, use pip to install:
+This package runs under Python >=3.9, use pip to install:
 
 ```bash
 $ pip install csvw

RELEASING.md

@@ -7,11 +7,24 @@ Releasing csvw
 tox -r
 ```
 
+- Run the integration test from the README:
+```python
+import json
+from csvw import CSVW
+data = CSVW('https://raw.githubusercontent.com/cldf/csvw/master/tests/fixtures/test.tsv')
+print(json.dumps(data.to_json(minimal=True), indent=4))
+```
+
 - Make sure flake8 passes:
 ```shell
 flake8 src
 ```
 
+- Make sure pylint passes with a score of 10:
+```shell
+pylint src/csvw
+```
+
 - Make sure docs can be created:
 ```shell
 cd docs

setup.cfg

@@ -23,12 +23,12 @@ classifiers =
     Natural Language :: English
     Operating System :: OS Independent
     Programming Language :: Python :: 3
-    Programming Language :: Python :: 3.8
     Programming Language :: Python :: 3.9
     Programming Language :: Python :: 3.10
     Programming Language :: Python :: 3.11
     Programming Language :: Python :: 3.12
     Programming Language :: Python :: 3.13
+    Programming Language :: Python :: 3.14
     Programming Language :: Python :: Implementation :: CPython
     Programming Language :: Python :: Implementation :: PyPy
     License :: OSI Approved :: Apache Software License
@@ -38,16 +38,14 @@ zip_safe = False
 packages = find:
 package_dir =
     = src
-python_requires = >=3.8
+python_requires = >=3.9
 install_requires =
-    attrs>=18.1
     isodate
     python-dateutil
     # Pin until fix for 2.0.0 is released (https://pypi.org/project/rfc3986/#history):
     rfc3986<2
     uritemplate>=3.0.0
     babel
-    requests
     language-tags
     rdflib
     termcolor
@@ -76,7 +74,6 @@ test =
     frictionless
     pytest>=5
     pytest-mock
-    requests-mock
     pytest-cov
 docs =
     sphinx<7
@@ -112,7 +109,7 @@ show_missing = true
 skip_covered = true
 
 [tox:tox]
-envlist = py38, py39, py310, py311, py312, py313
+envlist = py39, py310, py311, py312, py313, py314
 isolated_build = true
 skip_missing_interpreter = true
 

src/csvw/__init__.py

@@ -1,4 +1,6 @@
-# csvw - https://w3c.github.io/csvw/primer/
+"""
+csvw - https://w3c.github.io/csvw/primer/
+"""
 
 from .metadata import (
     TableGroup, Table, Column, ForeignKey, Link, NaturalLanguage, Datatype, URITemplate, CSVW,

src/csvw/__main__.py

@@ -1,3 +1,6 @@
+"""
+CLI for the csvw package.
+"""
 import sys
 import json
 import shutil
@@ -9,10 +12,11 @@
 
 from csvw import CSVW, TableGroup
 from csvw.db import Database
-from csvw.utils import metadata2markdown
+from csvw.metadata_utils import metadata2markdown
 
 
 def parsed_args(desc, args, *argspecs):
+    """Add custom arguments to the parser and parse."""
     if args is None:  # pragma: no cover
         parser = argparse.ArgumentParser(description=desc)
         for kw, kwargs in argspecs:
@@ -21,23 +25,25 @@ def parsed_args(desc, args, *argspecs):
     return args
 
 
-def exit(ret, test=False):
+def exit(ret, test=False):  # pylint: disable=redefined-builtin
+    """We don't want to exit the test suite"""
     if test:
         return ret
     sys.exit(ret)  # pragma: no cover
 
 
 def csvwdescribe(args=None, test=False):
+    """Describe a (set of) CSV file(s) with basic CSVW metadata."""
     frictionless = shutil.which('frictionless')
     if not frictionless:  # pragma: no cover
         raise ValueError('The frictionless command must be installed for this functionality!\n'
                          'Run `pip install frictionless` and try again.')
 
     args = parsed_args(
-        "Describe a (set of) CSV file(s) with basic CSVW metadata.",
+        csvwdescribe.__doc__,
         args,
-        (['--delimiter'], dict(default=None)),
-        (['csv'], dict(nargs='+', help="CSV files to describe as CSVW TableGroup")),
+        (['--delimiter'], {'default': None}),
+        (['csv'], {'nargs': '+', 'help': "CSV files to describe as CSVW TableGroup"}),
     )
     fargs = ['describe', '--json']
     if args.delimiter:
@@ -53,19 +59,20 @@ def csvwdescribe(args=None, test=False):
 
     dp = json.loads(subprocess.check_output([frictionless] + fargs + args.csv))
     if onefile:
-        dp = dict(resources=[dp], profile='data-package')
+        dp = {'resources': [dp], 'profile': 'data-package'}
 
     tg = TableGroup.from_frictionless_datapackage(dp)
     print(json.dumps(tg.asdict(), indent=4))
-    return exit(0, test=test)
+    return exit(0, test=test)  # pylint: disable=R1722
 
 
 def csvwvalidate(args=None, test=False):
+    """Validate a (set of) CSV file(s) described by CSVW metadata."""
     args = parsed_args(
-        "Validate a (set of) CSV file(s) described by CSVW metadata.",
+        csvwvalidate.__doc__,
         args,
-        (['url'], dict(help='URL or local path to CSV or JSON metadata file.')),
-        (['-v', '--verbose'], dict(action='store_true', default=False)),
+        (['url'], {'help': 'URL or local path to CSV or JSON metadata file.'}),
+        (['-v', '--verbose'], {'action': 'store_true', 'default': False}),
     )
     ret = 0
     try:
@@ -83,80 +90,85 @@ def csvwvalidate(args=None, test=False):
         print(colored('FAIL', 'red', attrs=['bold']))
         if args.verbose:
             print(colored(str(e), 'blue'))
-    return exit(ret, test=test)
+    return exit(ret, test=test)  # pylint: disable=R1722
 
 
 def csvw2datasette(args=None, test=False):
+    """Convert CSVW to data for datasette (https://datasette.io/)."""
     args = parsed_args(
-        "Convert CSVW to data for datasette (https://datasette.io/).",
+        csvw2datasette.__doc__,
         args,
-        (['url'], dict(help='URL or local path to CSV or JSON metadata file.')),
-        (['-o', '--outdir'], dict(type=pathlib.Path, default=pathlib.Path('.'))),
+        (['url'], {'help': 'URL or local path to CSV or JSON metadata file.'}),
+        (['-o', '--outdir'], {'type': pathlib.Path, 'default': pathlib.Path('.')}),
     )
     dbname, mdname = 'datasette.db', 'datasette-metadata.json'
     csvw = CSVW(args.url)
     db = Database(csvw.tablegroup, fname=args.outdir / dbname)
     db.write_from_tg()
     md = {}
     for k in ['title', 'description', 'license']:
-        if 'dc:{}'.format(k) in csvw.common_props:
-            md[k] = csvw.common_props['dc:{}'.format(k)]
-    # FIXME: flesh out, see https://docs.datasette.io/en/stable/metadata.html
+        if f'dc:{k}' in csvw.common_props:
+            md[k] = csvw.common_props[f'dc:{k}']
     args.outdir.joinpath(mdname).write_text(json.dumps(md, indent=4))
-    print("""Run
-    datasette {} --metadata {}
-and open your browser at
-    http://localhost:8001/
-to browse the data.
-""".format(args.outdir / dbname, args.outdir / mdname))
-    return exit(0, test=test)
+    for line in [
+        "Run",
+        f"    datasette {args.outdir / dbname} --metadata {args.outdir / mdname}",
+        "and open your browser at",
+        "    http://localhost:8001/",
+        "to browse the data.",
+    ]:
+        print(line)
+    return exit(0, test=test)  # pylint: disable=R1722
 
 
 def csvw2json(args=None, test=False):
+    """Convert CSVW to JSON, see https://w3c.github.io/csvw/csv2json/"""
     args = parsed_args(
-        "Convert CSVW to JSON, see https://w3c.github.io/csvw/csv2json/",
+        csvw2json.__doc__,
         args,
-        (['url'], dict(help='URL or local path to CSV or JSON metadata file.')),
+        (['url'], {'help': 'URL or local path to CSV or JSON metadata file.'}),
     )
     csvw = CSVW(args.url)
     print(json.dumps(csvw.to_json(), indent=4))
-    return exit(0, test=test)
+    return exit(0, test=test)  # pylint: disable=R1722
 
 
-def csvw2sqlite(args=None, test=False):  # pragma: no cover
+def csvw2sqlite(args=None, test=False):
+    """Convert CSVW to SQLite"""
     args = parsed_args(
-        "Convert CSVW to JSON, see https://w3c.github.io/csvw/csv2json/",
+        csvw2sqlite.__doc__,
         args,
         (
             ['url'],
-            dict(help='URL or local path to CSVW metadata file describing a TableGroup.\n\n'
-                      'Note that not all valid CSVW datasets can be converted to SQLite. One '
-                      'limitation is that all tables which are referenced by foreign keys must '
-                      'have a primary key.')),
+            {'help': 'URL or local path to CSVW metadata file describing a TableGroup.\n\n'
+                     'Note that not all valid CSVW datasets can be converted to SQLite. One '
+                     'limitation is that all tables which are referenced by foreign keys must '
+                     'have a primary key.'}),
         (
             ['output'],
-            dict(help='Path for the generated SQLite database file.')),
+            {'help': 'Path for the generated SQLite database file.'}),
     )
     tg = TableGroup.from_file(args.url)
     db = Database(tg, args.output)
     db.write_from_tg(_force=True)
-    return exit(0, test=test)
+    return exit(0, test=test)  # pylint: disable=R1722
 
 
 def csvw2markdown(args=None, test=False):
+    """Create a Markdown document containing the CSVW metadata in human readable form."""
     args = parsed_args(
         "Convert CSVW to JSON, see https://w3c.github.io/csvw/csv2json/",
         args,
         (
             ['url'],
-            dict(help='URL or local path to CSVW metadata file describing a TableGroup.\n\n'
-                      'Note that not all valid CSVW datasets can be converted to SQLite. One '
-                      'limitation is that all tables which are referenced by foreign keys must '
-                      'have a primary key.')),
+            {'help': 'URL or local path to CSVW metadata file describing a TableGroup.\n\n'
+                     'Note that not all valid CSVW datasets can be converted to SQLite. One '
+                     'limitation is that all tables which are referenced by foreign keys must '
+                     'have a primary key.'}),
     )
     tg = TableGroup.from_file(args.url)
     print(metadata2markdown(tg, link_files=True))
-    return exit(0, test=test)
+    return exit(0, test=test)  # pylint: disable=R1722
 
 
 if __name__ == '__main__':  # pragma: no cover

src/csvw/_compat.py

@@ -0,0 +1,24 @@
+"""
+Functionality to address python compatibility issues.
+"""
+import re
+import sys
+import datetime
+
+
+if (sys.version_info.major, sys.version_info.minor) >= (3, 11):  # pragma: no cover
+    fromisoformat = datetime.datetime.fromisoformat
+else:
+    def fromisoformat(s: str) -> datetime.datetime:  # pragma: no cover
+        """Somewhat hacky backport of the more full-fledged date parsing support in py3.11."""
+        s = s.replace('Z', '+00:00')
+        ms_p = re.compile(r'(?P<ms>\.[0-9]+)')
+        m = ms_p.search(s)
+        ms = None
+        if m:
+            s = ms_p.sub('', s)
+            ms = float(f'0{ms}')
+        res = datetime.datetime.fromisoformat(s)
+        if ms:
+            res = res.replace(microsecond=(ms * 1000000) % 1000000)
+        return res