meta_repo_links: Start hook.

conservancy_beancount/data.py

@@ -37,6 +37,16 @@ from .beancount_types import (
     Transaction,
 )
 
+LINK_METADATA = frozenset([
+    'approval',
+    'check',
+    'contract',
+    'invoice',
+    'purchase-order',
+    'receipt',
+    'statement',
+])
+
 class Account(str):
     """Account name string
 

conservancy_beancount/errors.py

@@ -38,6 +38,14 @@ class Error(Exception):
 
 Iter = Iterable[Error]
 
+class BrokenLinkError(Error):
+    def __init__(self, txn, key, link, source=None):
+        super().__init__(
+            "{} not found in repository: {}".format(key, link),
+            txn,
+            source,
+        )
+
 class ConfigurationError(Error):
     def __init__(self, message, entry=None, source=None):
         if source is None:

conservancy_beancount/plugin/meta_repo_links.py

@@ -0,0 +1,59 @@
+"""meta_repo_links - Check that repository links are valid"""
+# Copyright © 2020  Brett Smith
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <https://www.gnu.org/licenses/>.
+
+import re
+
+from . import core
+from .. import config as configmod
+from .. import data
+from .. import errors as errormod
+from ..beancount_types import (
+    MetaKey,
+    MetaValue,
+    Transaction,
+)
+
+from typing import (
+    Mapping,
+)
+
+class MetaRepoLinks(core.TransactionHook):
+    HOOK_GROUPS = frozenset(['linkcheck'])
+    PATH_PUNCT_RE = re.compile(r'[:/]')
+
+    def __init__(self, config: configmod.Config) -> None:
+        repo_path = config.repository_path()
+        if repo_path is None:
+            raise errormod.ConfigurationError("no repository configured")
+        self.repo_path = repo_path
+
+    def _check_links(self,
+                     txn: Transaction,
+                     meta: Mapping[MetaKey, MetaValue],
+    ) -> errormod.Iter:
+        for key in data.LINK_METADATA.intersection(meta):
+            for link in str(meta[key]).split():
+                match = self.PATH_PUNCT_RE.search(link)
+                if match and match.group(0) == ':':
+                    pass
+                elif not (self.repo_path / link).exists():
+                    yield errormod.BrokenLinkError(txn, key, link)
+
+    def run(self, txn: Transaction) -> errormod.Iter:
+        yield from self._check_links(txn, txn.meta)
+        for post in txn.postings:
+            if post.meta is not None:
+                yield from self._check_links(txn, post.meta)

tests/test_meta_repo_links.py

@@ -0,0 +1,135 @@
+"""Test link checker for repository documents"""
+# Copyright © 2020  Brett Smith
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <https://www.gnu.org/licenses/>.
+
+import itertools
+
+from pathlib import Path
+
+import pytest
+
+from . import testutil
+
+from conservancy_beancount import errors as errormod
+from conservancy_beancount.plugin import meta_repo_links
+
+METADATA_KEYS = [
+    'approval',
+    'check',
+    'contract',
+    'invoice',
+    'purchase-order',
+    'receipt',
+    'statement',
+]
+
+GOOD_LINKS = [Path(s) for s in [
+    'Projects/project-data.yml',
+    'Projects/project-list.yml',
+]]
+
+BAD_LINKS = [Path(s) for s in [
+    'NonexistentDirectory/NonexistentFile1.txt',
+    'NonexistentDirectory/NonexistentFile2.txt',
+]]
+
+NOT_FOUND_MSG = '{} not found in repository: {}'.format
+
+def build_meta(keys=None, *sources):
+    if keys is None:
+        keys = iter(METADATA_KEYS)
+    sources = (itertools.cycle(src) for src in sources)
+    return {key: ' '.join(str(x) for x in rest)
+            for key, *rest in zip(keys, *sources)}
+
+@pytest.fixture(scope='module')
+def hook():
+    config = testutil.TestConfig(repo_path='repository')
+    return meta_repo_links.MetaRepoLinks(config)
+
+def test_error_with_no_repository():
+    config = testutil.TestConfig(repo_path=None)
+    with pytest.raises(errormod.ConfigurationError):
+        meta_repo_links.MetaRepoLinks(config)
+
+def test_good_txn_links(hook):
+    meta = build_meta(None, GOOD_LINKS)
+    txn = testutil.Transaction(**meta, postings=[
+        ('Income:Donations', -5),
+        ('Assets:Cash', 5),
+    ])
+    assert not list(hook.run(txn))
+
+def test_good_post_links(hook):
+    meta = build_meta(None, GOOD_LINKS)
+    txn = testutil.Transaction(postings=[
+        ('Income:Donations', -5, meta),
+        ('Assets:Cash', 5),
+    ])
+    assert not list(hook.run(txn))
+
+def test_bad_txn_links(hook):
+    meta = build_meta(None, BAD_LINKS)
+    txn = testutil.Transaction(**meta, postings=[
+        ('Income:Donations', -5),
+        ('Assets:Cash', 5),
+    ])
+    expected = {NOT_FOUND_MSG(key, value) for key, value in meta.items()}
+    actual = {error.message for error in hook.run(txn)}
+    assert expected == actual
+
+def test_bad_post_links(hook):
+    meta = build_meta(None, BAD_LINKS)
+    txn = testutil.Transaction(postings=[
+        ('Income:Donations', -5, meta.copy()),
+        ('Assets:Cash', 5),
+    ])
+    expected = {NOT_FOUND_MSG(key, value) for key, value in meta.items()}
+    actual = {error.message for error in hook.run(txn)}
+    assert expected == actual
+
+@pytest.mark.parametrize('ext_doc', [
+    'rt:123',
+    'rt:456/789',
+    'rt://ticket/23',
+    'rt://ticket/34/attachments/567890',
+])
+def test_docs_outside_repository_not_checked(hook, ext_doc):
+    txn = testutil.Transaction(
+        receipt='{} {} {}'.format(GOOD_LINKS[0], ext_doc, BAD_LINKS[1]),
+        postings=[
+            ('Income:Donations', -5),
+            ('Assets:Cash', 5),
+        ])
+    expected = {NOT_FOUND_MSG('receipt', BAD_LINKS[1])}
+    actual = {error.message for error in hook.run(txn)}
+    assert expected == actual
+
+def test_mixed_results(hook):
+    txn = testutil.Transaction(
+        approval='{} {}'.format(*GOOD_LINKS),
+        contract='{} {}'.format(BAD_LINKS[0], GOOD_LINKS[1]),
+        postings=[
+            ('Income:Donations', -5, {'invoice': '{} {}'.format(*BAD_LINKS)}),
+            ('Assets:Cash', 5, {'statement': '{} {}'.format(GOOD_LINKS[0], BAD_LINKS[1])}),
+        ])
+    expected = {
+        NOT_FOUND_MSG('contract', BAD_LINKS[0]),
+        NOT_FOUND_MSG('invoice', BAD_LINKS[0]),
+        NOT_FOUND_MSG('invoice', BAD_LINKS[1]),
+        NOT_FOUND_MSG('statement', BAD_LINKS[1]),
+    }
+    actual = {error.message for error in hook.run(txn)}
+    assert expected == actual