Merge branch 'master' into refactor-matching

.github/workflows/tests.yml

@@ -42,7 +42,7 @@ jobs:
       - name: Set up Python
         uses: actions/setup-python@v2
         with:
-          python-version: "3.8"
+          python-version: "3.10"
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip
@@ -58,7 +58,7 @@ jobs:
       - name: Set up Python
         uses: actions/setup-python@v2
         with:
-          python-version: "3.8"
+          python-version: "3.10"
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip

CHANGELOG.md

@@ -1,103 +1,131 @@
 # pygdbmi release history
 
-## 0.10.0.3
-* Fixed a bug where notifications without a payload were not recognized as such
-* [dev] Refactored the code to parse MI records to decrease the number of regex matches to perform
+## dev
+
+- Fixed a bug where notifications without a payload were not recognized as such
+- Invalid octal sequences produced by GDB are left unchanged instead of causing a `UnicodeDecodeError` (#64)
+
+Internal changes
+
+- Update and freeze dependencies for documentation generation
+- Refactored the code to parse MI records to decrease the number of regex matches to perform
+
 
 ## 0.10.0.2
-* Strings containing escapes are now unescaped, both for messages in error records, which were previously mangled (#57), and textual records, which were previously left escaped (#58)
-* Dropped support for Python 3.6 and added explicit support for Python 3.9 and 3.10.
+
+- Strings containing escapes are now unescaped, both for messages in error records, which were previously mangled (#57), and textual records, which were previously left escaped (#58)
+- Dropped support for Python 3.6 and added explicit support for Python 3.9 and 3.10.
 
 ## 0.10.0.1
-* Fix bug with `time_to_check_for_additional_output_sec`, as it was not being used when passed to `GdbController`
+
+- Fix bug with `time_to_check_for_additional_output_sec`, as it was not being used when passed to `GdbController`
 
 ## 0.10.0.0
 
- **Breaking Changes**
+**Breaking Changes**
 
-* Drop support for Python 3.5
-* Update `GdbController()` API. New API is `GdbController(command: Optional[List[str]], time_to_check_for_additional_output_sec: Optional[int])`.
-* `GdbController.verify_valid_gdb_subprocess()` was removed
-* Remove `NoGdbProcessError` error
+- Drop support for Python 3.5
+- Update `GdbController()` API. New API is `GdbController(command: Optional[List[str]], time_to_check_for_additional_output_sec: Optional[int])`.
+- `GdbController.verify_valid_gdb_subprocess()` was removed
+- Remove `NoGdbProcessError` error
 
 Other Changes
 
-* Add new `IoManager` class to handle more generic use-cases
-* [dev] use pytest for testing
-* gdb mi parsing remains unchanged
+- Add new `IoManager` class to handle more generic use-cases
+- [dev] use pytest for testing
+- gdb mi parsing remains unchanged
 
 ## 0.9.0.3
 
-* Drop support for 2.7, 3.4
-* Add support for 3.7, 3.8
-* Add `py.typed` file so mypy can enforce type hints on `pygdbmi`
-* Do not log in StringStream (#36)
-* Updates to build and CI tests (use nox)
-* Use mkdocs and mkdocstrings
-* Doc updates
+- Drop support for 2.7, 3.4
+- Add support for 3.7, 3.8
+- Add `py.typed` file so mypy can enforce type hints on `pygdbmi`
+- Do not log in StringStream (#36)
+- Updates to build and CI tests (use nox)
+- Use mkdocs and mkdocstrings
+- Doc updates
 
 ## 0.9.0.2
-* More doc updates
+
+- More doc updates
 
 ## 0.9.0.1
-* Update docs
+
+- Update docs
 
 ## 0.9.0.0
-* Stop buffering output
-* Use logger in GdbController; modify `verbose` arguments.
-* Remove support for Python 3.3
+
+- Stop buffering output
+- Use logger in GdbController; modify `verbose` arguments.
+- Remove support for Python 3.3
 
 ## 0.8.4.0
-* Add method `get_subprocess_cmd` to view the gdb command run in the shell
+
+- Add method `get_subprocess_cmd` to view the gdb command run in the shell
 
 ## 0.8.3.0
-* Improve reading gdb responses on unix (performance, bugfix) (@mouuff)
+
+- Improve reading gdb responses on unix (performance, bugfix) (@mouuff)
 
 ## 0.8.2.0
-* Add support for [record and replay (rr) gdb supplement](http://rr-project.org/)
+
+- Add support for [record and replay (rr) gdb supplement](http://rr-project.org/)
 
 ## 0.8.1.1
-* Discard unexpected text from gdb
+
+- Discard unexpected text from gdb
 
 ## 0.8.1.0
-* Add native Windows support
+
+- Add native Windows support
 
 ## 0.8.0.0
-* Make parsing more efficient when gdb outputs large strings
-* Add new methods to GdbController class: `spawn_new_gdb_subprocess`, `send_signal_to_gdb`, and `interrupt_gdb`
+
+- Make parsing more efficient when gdb outputs large strings
+- Add new methods to GdbController class: `spawn_new_gdb_subprocess`, `send_signal_to_gdb`, and `interrupt_gdb`
 
 ## 0.7.4.5
-* Update setup.py
+
+- Update setup.py
 
 ## 0.7.4.4
-* Fix windows ctypes import (#23, @rudolfwalter)
+
+- Fix windows ctypes import (#23, @rudolfwalter)
 
 ## 0.7.4.3
-* Workaround gdb bug with repeated dictionary keys
+
+- Workaround gdb bug with repeated dictionary keys
 
 ## 0.7.4.2
-* Improved buffering of incomplete gdb mi output (@trapito)
-* Remove support of Python 3.2
+
+- Improved buffering of incomplete gdb mi output (@trapito)
+- Remove support of Python 3.2
 
 ## 0.7.4.1
-* Preserve leading and trailing spaces in gdb/mi output (plus unit tests)
-* Add unit test for buffering of gdb/mi output
-* Documentation updates
-* Refactoring
+
+- Preserve leading and trailing spaces in gdb/mi output (plus unit tests)
+- Add unit test for buffering of gdb/mi output
+- Documentation updates
+- Refactoring
 
 ## 0.7.4.0
-* Add more exception types (`NoGdbProcessError`, `GdbTimeoutError`)
-* Add logic fixes for Windows (@johncf)
-* Use codecs.open() to open the readme.rst, to prevent locale related bugs (@mariusmue)
+
+- Add more exception types (`NoGdbProcessError`, `GdbTimeoutError`)
+- Add logic fixes for Windows (@johncf)
+- Use codecs.open() to open the readme.rst, to prevent locale related bugs (@mariusmue)
 
 ## 0.7.3.3
-* Add alternate pipe implementation for Windows
+
+- Add alternate pipe implementation for Windows
 
 ## 0.7.3.2
-* Replace `epoll` with `select` for osx compatibility (@felipesere)
+
+- Replace `epoll` with `select` for osx compatibility (@felipesere)
 
 ## 0.7.3.1
-* Fix README
+
+- Fix README
 
 ## 0.7.3.0
-* Add support for gdb/mi (optional) tokens (@mariusmue)
+
+- Add support for gdb/mi (optional) tokens (@mariusmue)

MANIFEST.in

@@ -7,6 +7,8 @@ exclude example.py
 exclude .flake8
 exclude noxfile.py
 exclude mkdocs.yml
+exclude *.in
+exclude *.txt
 
 prune tests
 prune docs

mkdoc_requirements.in

@@ -0,0 +1,5 @@
+.
+mkdocstrings[python]
+mkdocs
+mkdocs-material
+pygments

mkdoc_requirements.txt

@@ -0,0 +1,78 @@
+#
+# This file is autogenerated by pip-compile with python 3.10
+# To update, run:
+#
+#    pip-compile mkdoc_requirements.in
+#
+click==8.1.3
+    # via mkdocs
+ghp-import==2.1.0
+    # via mkdocs
+griffe==0.22.0
+    # via mkdocstrings-python
+importlib-metadata==4.12.0
+    # via mkdocs
+jinja2==3.1.2
+    # via
+    #   mkdocs
+    #   mkdocs-material
+    #   mkdocstrings
+markdown==3.3.7
+    # via
+    #   mkdocs
+    #   mkdocs-autorefs
+    #   mkdocs-material
+    #   mkdocstrings
+    #   pymdown-extensions
+markupsafe==2.1.1
+    # via
+    #   jinja2
+    #   mkdocstrings
+mergedeep==1.3.4
+    # via mkdocs
+mkdocs==1.3.1
+    # via
+    #   -r mkdoc_requirements.in
+    #   mkdocs-autorefs
+    #   mkdocs-material
+    #   mkdocstrings
+mkdocs-autorefs==0.4.1
+    # via mkdocstrings
+mkdocs-material==8.3.9
+    # via -r mkdoc_requirements.in
+mkdocs-material-extensions==1.0.3
+    # via mkdocs-material
+mkdocstrings[python]==0.19.0
+    # via
+    #   -r mkdoc_requirements.in
+    #   mkdocstrings-python
+mkdocstrings-python==0.7.1
+    # via mkdocstrings
+packaging==21.3
+    # via mkdocs
+    .
+    # via -r mkdoc_requirements.in
+pygments==2.12.0
+    # via
+    #   -r mkdoc_requirements.in
+    #   mkdocs-material
+pymdown-extensions==9.5
+    # via
+    #   mkdocs-material
+    #   mkdocstrings
+pyparsing==3.0.9
+    # via packaging
+python-dateutil==2.8.2
+    # via ghp-import
+pyyaml==6.0
+    # via
+    #   mkdocs
+    #   pyyaml-env-tag
+pyyaml-env-tag==0.1
+    # via mkdocs
+six==1.16.0
+    # via python-dateutil
+watchdog==2.1.9
+    # via mkdocs
+zipp==3.8.1
+    # via importlib-metadata

noxfile.py

@@ -26,30 +26,25 @@ def lint(session):
     session.run("python", "setup.py", "check", "--metadata", "--strict")
 
 
-doc_dependencies = [
-    ".",
-    "mkdocstrings",
-    "mkdocs",
-    "mkdocs-material",
-    "pygments",
-]
+def install_mkdoc_dependencies(session):
+    session.install("-r", "mkdoc_requirements.txt")
 
 
-@nox.session(python="3.7")
+@nox.session
 def docs(session):
-    session.install(*doc_dependencies)
+    install_mkdoc_dependencies(session)
     session.run("mkdocs", "build")
 
 
-@nox.session(python="3.7")
+@nox.session
 def serve_docs(session):
-    session.install(*doc_dependencies)
+    install_mkdoc_dependencies(session)
     session.run("mkdocs", "serve")
 
 
-@nox.session(python="3.7")
+@nox.session
 def publish_docs(session):
-    session.install(*doc_dependencies)
+    install_mkdoc_dependencies(session)
     session.run("mkdocs", "gh-deploy")
 
 

pygdbmi/gdbescapes.py

@@ -166,7 +166,13 @@ def _unescape_internal(
                     raise ValueError(
                         f"Invalid octal number {octal_number!r} in {escaped_str!r}"
                     ) from exc
-            replaced = octal_sequence_bytes.decode("utf-8")
+            try:
+                replaced = octal_sequence_bytes.decode("utf-8")
+            except UnicodeDecodeError:
+                # GDB should never generate invalid sequences but, according to #64,
+                # it can do that on Windows. In this case we just keep the sequence
+                # unchanged.
+                replaced = f"\\{escaped_octal}"
 
         elif escaped_char is not None:
             # We found a single escaped character.

tests/test_pygdbmi.py

@@ -502,6 +502,9 @@ def test_unescape(self) -> None:
             with self.assertRaisesRegex(ValueError, "Invalid escape character"):
                 unescape(bad)
 
+        # An octal sequence that is not valid UTF-8 doesn't get changes, see #64.
+        assert_match(unescape(r"254 '\376'"), r"254 '\376'")
+
     def test_advance_past_string_with_gdb_escapes(self) -> None:
         """Test the advance_past_string_with_gdb_escapes function"""