Avoid that tripper.Namespace crashes if the cache directory cannot be accessed

.pre-commit-config.yaml

@@ -44,6 +44,7 @@ repos:
     rev: 24.8.0
     hooks:
     - id: black
+      line-length: 79
 
   - repo: https://github.com/PyCQA/bandit
     rev: 1.7.9

README.md

@@ -9,6 +9,13 @@ Tripper
 [![DOI](https://zenodo.org/badge/547162834.svg)](https://zenodo.org/badge/latestdoi/547162834)
 
 
+Getting started
+---------------
+* [Tutorial]
+* [Discovery of custom backends]
+* [Reference manual]
+* [Known issues]
+
 
 Basic concepts
 --------------
@@ -39,13 +46,6 @@ The submodules `mappings` and `convert` provide additional functionality beyond
 - **tripper.convert**: convert between RDF and other data representations.
 
 
-Documentation
--------------
-* Getting started: See the [tutorial](docs/tutorial.md)
-* [Discovery of custom backends](docs/backend_discovery.md)
-* [Reference manual]
-
-
 Available backends
 ------------------
 The following backends are currently available, either in Tripper or other packages.
@@ -104,11 +104,14 @@ We gratefully acknowledge the following projects for supporting the development
 
 
 
+[Tutorial]: docs/tutorial.md
+[Discovery of custom backends]: docs/backend_discovery.md
+[Reference manual]: https://emmc-asbl.github.io/tripper/latest/api_reference/triplestore/
+[Known issues]: https://emmc-asbl.github.io/tripper/latest/known-issues.md
 [tripper]: https://emmc-asbl.github.io/tripper
 [rdflib]: https://rdflib.readthedocs.io/en/stable/
 [PyPI]: https://pypi.org/project/tripper
 [PyBackTrip]: https://github.com/EMMC-ASBL/PyBackTrip/
-[Reference manual]: https://emmc-asbl.github.io/tripper/latest/api_reference/triplestore/
 [Literal]: https://emmc-asbl.github.io/tripper/latest/api_reference/triplestore/#tripper.triplestore.Literal
 [Namespace]: https://emmc-asbl.github.io/tripper/latest/api_reference/triplestore/#tripper.triplestore.Namespace
 [Triplestore]: https://emmc-asbl.github.io/tripper/latest/api_reference/triplestore/#tripper.triplestore.Triplestore

docs/index.md

@@ -9,6 +9,13 @@ Tripper
 [![DOI](https://zenodo.org/badge/547162834.svg)](https://zenodo.org/badge/latestdoi/547162834)
 
 
+Getting started
+---------------
+* [Tutorial]
+* [Discovery of custom backends]
+* [Reference manual]
+* [Known issues]
+
 
 Basic concepts
 --------------
@@ -39,13 +46,6 @@ The submodules `mappings` and `convert` provide additional functionality beyond
 - **tripper.convert**: convert between RDF and other data representations.
 
 
-Documentation
--------------
-* Getting started: See the [tutorial](tutorial.md)
-* [Discovery of custom backends](backend_discovery.md)
-* [Reference manual]
-
-
 Available backends
 ------------------
 The following backends are currently available, either in Tripper or other packages.
@@ -104,11 +104,14 @@ We gratefully acknowledge the following projects for supporting the development
 
 
 
+[Tutorial]: tutorial.md
+[Discovery of custom backends]: backend_discovery.md
+[Reference manual]: https://emmc-asbl.github.io/tripper/latest/api_reference/triplestore/
+[Known issues]: https://emmc-asbl.github.io/tripper/latest/known-issues.md
 [tripper]: https://emmc-asbl.github.io/tripper
 [rdflib]: https://rdflib.readthedocs.io/en/stable/
 [PyPI]: https://pypi.org/project/tripper
 [PyBackTrip]: https://github.com/EMMC-ASBL/PyBackTrip/
-[Reference manual]: https://emmc-asbl.github.io/tripper/latest/api_reference/triplestore/
 [Literal]: https://emmc-asbl.github.io/tripper/latest/api_reference/triplestore/#tripper.triplestore.Literal
 [Namespace]: https://emmc-asbl.github.io/tripper/latest/api_reference/triplestore/#tripper.triplestore.Namespace
 [Triplestore]: https://emmc-asbl.github.io/tripper/latest/api_reference/triplestore/#tripper.triplestore.Triplestore

docs/known-issues.md

@@ -0,0 +1,10 @@
+Known issues
+============
+
+* If you use the rdflib backend and don't have write permissions to
+  the cache directory (which e.g. can happen if you run Python in
+  docker as a non-root user), you may get a `urllib.error.HTTPError`
+  error when accessing an online rdf resource.
+
+  Setting the environment variable `XDG_CACHE_HOME` to a directory
+  that you have write access to will solve the problem.

tripper/backends/rdflib.py

@@ -2,6 +2,12 @@
 
 For developers: The usage of `s`, `p`, and `o` represent the different parts of
 an RDF Triple: subject, predicate, and object.
+
+There is a issue with rdflib raising an `urllib.error.HTTPError`
+exception if you don't have write permissions to the cache directory.
+See [Known issues](https://emmc-asbl.github.io/tripper/latest/known-issues.md)
+for more details.
+
 """
 
 # pylint: disable=line-too-long

tripper/namespace.py

@@ -152,7 +152,7 @@
             if s.startswith(iri)
         )
 
-    def _get_cachefile(self) -> "Path":
+    def _get_cachefile(self) -> Path:
         """Return path to cache file for this namespace."""
         # pylint: disable=too-many-function-args
         name = self._iri.rstrip("#/").rsplit("/", 1)[-1]
@@ -162,34 +162,59 @@
     def _save_cache(self):
         """Save current cache."""
         # pylint: disable=invalid-name
-        cachefile = self._get_cachefile()
-        if self._iris and not sys.is_finalizing():
-            with open(cachefile, "wb") as f:
-                pickle.dump(self._iris, f)
+        try:
+            cachefile = self._get_cachefile()
+            if self._iris and not sys.is_finalizing():
+                with open(cachefile, "wb") as f:
+                    pickle.dump(self._iris, f)
+        except OSError as exc:
+            warnings.warn(
+                f"Cannot access cache file: {exc}\n\n"
+                "You can select cache directory with the XDG_CACHE_HOME "
+                "environment variable."
+            )
 
     def _load_cache(self) -> bool:
         """Update cache with cache file.
 
         Returns true if there exists a cache file to load from.
         """
         # pylint: disable=invalid-name
-        cachefile = self._get_cachefile()
-        if self._iris is None:
-            self._iris = {}
-        if cachefile.exists():
-            with open(cachefile, "rb") as f:
-                self._iris.update(pickle.load(f))  # nosec
-            return True
-        return False
+        try:
+            cachefile = self._get_cachefile()
+            if self._iris is None:
+                self._iris = {}
+            if cachefile.exists():
+                with open(cachefile, "rb") as f:
+                    self._iris.update(pickle.load(f))  # nosec
+                return True
+            return False
+        except OSError as exc:
+            warnings.warn(
+                f"Cannot create cache directory: {exc}\n\n"
+                "You can select cache directory with the XDG_CACHE_HOME "
+                "environment variable."
+            )
+            return False
 
     def __getattr__(self, name):
         if self._iris and name in self._iris:
             return self._iris[name]
         if self._check:
             msg = ""
-            cachefile = self._get_cachefile()
-            if cachefile.exists():
-                msg = f"\nMaybe you have to remove the cache file: {cachefile}"
+            try:
+                cachefile = self._get_cachefile()
+                if cachefile.exists():
+                    msg = (
+                        "\nMaybe you have to remove the cache file: "
+                        f"{cachefile}"
+                    )
+            except OSError as exc:
+                warnings.warn(
+                    f"Cannot access cache file: {exc}\n\n"
+                    "You can select cache directory with the XDG_CACHE_HOME "
+                    "environment variable."
+                )
             raise NoSuchIRIError(self._iri + name + msg)
         return self._iri + name