kemitix.net

tree-type: Type-Safe Filesystem Paths for Rust

2026-01-21T14:00:00+00:00

I just presented the following Lightning Talk for my tree-type</a> crate.

The Problem: String-Based Paths Are Error-Prone</h2>
When working with filesystem paths in code, we typically use strings:
fn update_project(root: &Path) -> Result<()> { // Easy to make typos let changelog = root.join("CHANGLOG.md"); // Oops! Missing 'E' // Easy to get case wrong let cargo = root.join("cargo.toml"); // Should be "Cargo.toml" // No IDE autocomplete let config = root.join("config").join("settings.toml"); // Errors discovered at runtime (or in production!) let content = fs::read_to_string(&changelog)?; // Runtime error! } </code></pre> Problems:

❌ Typos caught at runtime</li>
❌ No IDE autocomplete</li>
❌ Easy to mix up similar paths</li>
❌ Refactoring is error-prone</li>
❌ No compile-time validation</li> </ul>

The Solution: tree-type</h2>
Define your project structure once, get type safety everywhere:
use tree_type::tree_type; tree_type! { CurrentRoot { changelog("CHANGELOG.md") as ChangeLogFile, cargo("Cargo.toml") as CargoFile, config/ { settings("settings.toml") as SettingsFile, } } } </code></pre> Now use it safely:
fn update_project(root: &CurrentRoot) -> Result<()> { // Type-safe access - impossible to typo! let changelog: ChangeLogFile = root.changelog(); // IDE autocomplete works let cargo: CargoFile = root.cargo(); // Nested navigation is type-safe let settings: SettingsFile = root.config().settings(); // Compiler catches mistakes before code runs let content = changelog.read_to_string()?; } </code></pre> Benefits:

✅ Compile-time path validation</li>
✅ IDE autocomplete for all paths</li>
✅ Impossible to typo filenames</li>
✅ Refactoring is safe</li>
✅ Each path has its own type</li> </ul>

Real Example: Cairn Project</h2>
From the cairn</a> release automation tool:
tree_type! { CurrentRoot { jj/(".jj") as JujutsuRepoDir, git/(".git") as GitRepoDir, changelog("CHANGELOG.md") as ChangeLogFile, cargo("Cargo.toml") as CargoFile, ci_cache(".cairn-ci-cache") as CiCacheFile, } } // Usage throughout the codebase if !self.root.changelog().exists() { return Err("CHANGELOG.md not found"); } let content = self.root.changelog().read_to_string()?; self.root.ci_cache().write(&cache_data)?; </code></pre> Each file has its own type with built-in methods:

`.read_to_string()</code> - Read file contents</li>`
`.write()</code> - Write to file</li>`
`.exists()</code> - Check existence</li>`
`.remove()</code> - Delete file</li>`
`And more…</li> </ul>`
Feature: Type-Safe Navigation with parent()</h2>
The Problem with Generic Paths:
// Generic approach - returns Option let dir = GenericDir::new("/path/to/dir")?; let parent = dir.parent(); // Returns Option<GenericDir> if let Some(p) = parent { // Have to handle None case } </code></pre> tree-type Solution - Exact Parent Types:
tree_type! { ProjectRoot { src/ { lib("lib.rs") as LibFile, main("main.rs") as MainFile, } } } let project = ProjectRoot::new("/project")?; let src = project.src(); let lib_file = src.lib(); // parent() returns exact parent type - no Option! let src_dir: ProjectRootSrc = lib_file.parent(); let project_root: ProjectRoot = src_dir.parent(); </code></pre> Why this matters:

Compiler knows the structure</li>
No runtime checks needed</li>
Can’t accidentally navigate to wrong parent</li>
Type system enforces correct relationships</li> </ul>

Feature: Type-Safe Directory Iteration with children()</h2>
The Problem with read_dir():
// Generic read_dir returns untyped entries for entry in src_dir.read_dir()? { let entry = entry?; // What type is this? File? Directory? Which one? // No type safety, have to check at runtime } </code></pre> tree-type Solution - Typed Children: tree_type! { ProjectRoot { src/ { lib("lib.rs") as LibFile, main("main.rs") as MainFile, utils("utils.rs") as UtilsFile, } } } let src = project.src(); // children() returns an iterator of the correct types! for child in src.children() { match child { ProjectRootSrcChild::Lib(lib_file) => { // lib_file is LibFile type println!("Found lib: {}", lib_file.as_path().display()); } ProjectRootSrcChild::Main(main_file) => { // main_file is MainFile type println!("Found main: {}", main_file.as_path().display()); } ProjectRootSrcChild::Utils(utils_file) => { // utils_file is UtilsFile type println!("Found utils: {}", utils_file.as_path().display()); } } } </code></pre> Benefits: ✅ Each child has its correct type</li> ✅ Exhaustive pattern matching</li> ✅ Compiler ensures you handle all cases</li> ✅ No runtime type checking needed</li> </ul> Feature: Dynamic IDs for Runtime Paths</h2> Sometimes you need paths determined at runtime. Here’s a real example from fireforge: tree_type! { AppDataDir { repos/ as AppReposDir { [owner: RepositoryOwner]/ as RepoOwnerDir { [name: RepositoryName]/ as RepoBaseDir { metadata/ as RepoMetadataDir { file("info.json") as RepoMetadataFile, issues/ as RepoIssuesDir { [issue: IssueId]/ as RepoIssueDir { file("issue.md") as RepoIssueFile, } } } } } } } } // Type-safe navigation with runtime IDs let app_data = AppDataDir::new("/data")?; let owner_dir = app_data.repos().owner("kemitix"); let repo_dir = owner_dir.name("tree-type"); let metadata: RepoMetadataFile = repo_dir.metadata().file(); let issue_dir = repo_dir.metadata().issues().issue(42); let issue_file: RepoIssueFile = issue_dir.file(); // Each level is still type-safe! assert_eq!( issue_file.as_path(), Path::new("/data/repos/kemitix/tree-type/metadata/issues/42/issue.md") ); </code></pre> Benefits: ✅ Runtime flexibility with compile-time safety</li> ✅ Nested dynamic IDs (multiple levels) - unlimited with codegen-v2</code> feature</li> ✅ Type-safe navigation at every level</li> ✅ IDE autocomplete still works</li> </ul> Note: Default code generation supports up to 3 levels of dynamic ID nesting. Enable the codegen-v2</code> feature for unlimited nesting depth. Feature: Extending Generated Types with Custom Methods</h2> Since generated types exist in your crate, you can add your own methods: tree_type! { ProjectRoot { cargo("Cargo.toml") as CargoFile, src/ as SrcDir { lib("lib.rs") as LibFile, } } } // Add custom methods to generated types impl CargoFile { pub fn parse_version(&self) -> Result<String> { let content = self.read_to_string()?; // Parse version from Cargo.toml Ok(extract_version(&content)) } pub fn update_version(&self, new_version: &str) -> Result<()> { let content = self.read_to_string()?; let updated = replace_version(&content, new_version); self.write(&updated) } } impl SrcDir { pub fn count_rust_files(&self) -> Result<usize> { Ok(self.read_dir()?.filter(|e| e.path().extension() == Some("rs")).count()) } } // Use your custom methods let project = ProjectRoot::new("/project")?; let version = project.cargo().parse_version()?; project.cargo().update_version("0.2.0")?; let file_count = project.src().count_rust_files()?; </code></pre> Benefits: ✅ Domain-specific operations on your types</li> ✅ Encapsulate business logic with filesystem structure</li> ✅ Type safety extends to your custom methods</li> ✅ Clean API that combines navigation and operations</li> </ul> Feature: File Operations with Compile-Time Safety</h2> Before (string-based): fn process_files(root: &Path) -> Result<()> { // Can pass wrong file to wrong function let changelog = root.join("CHANGELOG.md"); let cargo = root.join("Cargo.toml"); // Compiler can't help if you mix them up parse_cargo(&changelog)?; // Runtime error! update_changelog(&cargo)?; // Runtime error! } </code></pre> After (tree-type): fn process_files(root: &CurrentRoot) -> Result<()> { let changelog: ChangeLogFile = root.changelog(); let cargo: CargoFile = root.cargo(); // Compiler enforces correct types parse_cargo(&cargo)?; // ✅ Correct update_changelog(&changelog)?; // ✅ Correct // These won't compile: // parse_cargo(&changelog)?; // ❌ Compile error! // update_changelog(&cargo)?; // ❌ Compile error! } fn parse_cargo(file: &CargoFile) -> Result<()> { / ... / } fn update_changelog(file: &ChangeLogFile) -> Result<()> { / ... / } </code></pre> What you get: Different types for different files</li> Can’t accidentally pass wrong file to wrong function</li> Compiler catches mistakes before code runs</li> Refactoring is safe - rename type, compiler finds all uses</li> </ul> Summary: Why tree-type?</h2> Type Safety Benefits: Compile-time validation - Catch path errors before runtime</li> IDE autocomplete - Discover available paths as you type</li> Refactoring safety - Rename types, compiler finds all uses</li> Impossible to typo - Filenames defined once, used safely everywhere</li> Type-safe navigation - parent()</code> returns exact types, no Option needed</li> Dynamic flexibility - Runtime IDs with compile-time structure</li> Function safety - Can’t pass wrong file type to wrong function</li> </ol> When to use: ✅ Projects with fixed directory structures</li> ✅ Build tools and automation</li> ✅ Configuration management</li> ✅ Test fixtures and data directories</li> ✅ Any code that works with known filesystem layouts</li> </ul> When NOT to use: ❌ Completely dynamic/unknown directory structures</li> ❌ User-provided arbitrary paths</li> ❌ Simple scripts with 1-2 paths</li> </ul> Getting Started</h2> [dependencies] tree-type = "0.4" # Optional features # tree-type = { version = "0.4", features = ["serde", "enhanced-errors", "walk", "pattern-validation", "codegen-v2"] } </code></pre> use tree_type::tree_type; tree_type! { MyProject { src/, target/, readme("README.md") } } let project = MyProject::new(project_root)?; project.sync()?; // Creates all directories and required files </code></pre> Optional Features: serde</code> - Serialize/deserialize support</li> enhanced-errors</code> - Better error messages with context</li> walk</code> - Directory traversal methods</li> pattern-validation</code> - Regex validation for dynamic IDs</li> codegen-v2</code> - Unlimited dynamic ID nesting depth</li> </ul> Learn more: Crate: crates.io/crates/tree-type</a></li> Docs: docs.rs/tree-type</a></li> Source: codeberg.org/kemitix/tree-type</a></li> </ul> Q&A - Anticipated Questions</h2> Q: Does this work with existing codebases? A: Yes! You can gradually adopt tree-type. Define types for your structure, then refactor incrementally. Old string-based code continues to work. Q: What about performance? A: Zero runtime overhead. tree-type is a compile-time abstraction - generated code is as efficient as hand-written PathBuf operations. Q: Can I use this with other languages? A: tree-type is Rust-specific (uses procedural macros). The concept could be adapted to other languages with strong type systems (TypeScript, Kotlin, etc.). Q: What if my directory structure changes? A: Update the tree_type! definition, and the compiler will show you everywhere that needs updating. This is a feature - you can’t forget to update code! Q: Does it work on Windows? A: Yes! tree-type uses Rust’s standard Path/PathBuf which handle platform differences automatically. Q: Can I have optional files/directories? A: Yes! Use #[required]</code> attribute for must-exist paths, or check .exists()</code> for optional ones. The sync()</code> method only creates required paths. Q: What about symbolic links? A: Supported with #[symlink(target)]</code> attribute (Unix only). tree-type handles symlink creation and validation. Questions?



Recipe: 30 Minute Vegetarian Puerto Rican Style Picadillo
2026-01-03T20:23:00+01:00
I went vegetarian last year, and I want to try this recipe.</p>
30 Minute Vegetarian Puerto Rican Style Picadillo</h1>
A delicious meat-free version of the classic Puerto Rican dish with bold, savory flavors.</p>
Servings:</strong> 2

Time:</strong> 30 minutes

Difficulty:</strong> Easy</p>
Ingredients</h2>

1 tsp olive oil</li>
1 small yellow onion, diced</li>
1 red bell pepper, diced</li>
3 cloves garlic, minced</li>
2 tbsp tomato purée</li>
1 tsp ground cumin</li>
1 tsp smoked paprika</li>
½ tsp dried oregano</li>
½ tsp ground coriander</li>
¼ tsp black pepper</li>
Pinch of cayenne pepper (optional)</li>
225g plant-based ground (such as Beyond Meat, Impossible, or crumbled extra-firm tofu)</li>
2 bay leaves</li>
200g tinned tomato sauce</li>
1 tbsp capers (optional)</li>
60ml water</li>
Salt to taste</li>
Fresh coriander to garnish</li>
</ul>
Instructions</h2>


Heat olive oil in a pan over medium heat. Add onions and peppers and cook for 3-4 minutes until the onions become translucent.</p>
</li>

Add the garlic, tomato purée, cumin, smoked paprika, oregano, ground coriander, black pepper, and cayenne (if using). Stir and cook for about 1 minute until fragrant.</p>
</li>

Add the plant-based ground or crumbled tofu. If using tofu, break it up with a wooden spoon to create a ground meat texture. Cook for 3-4 minutes until lightly browned.</p>
</li>

Add the bay leaves, tomato sauce, capers (if using), and water. Stir to combine and bring to a boil.</p>
</li>

Reduce heat and simmer for 10 minutes, stirring occasionally, until everything is well combined and heated through. Add salt to taste, remove bay leaves, and garnish with fresh coriander.</p>
</li>
</ol>
Serving Suggestions</h2>
Serve with white rice, sweet plantains, crusty bread, or use as a filling for tacos or burritos.</p>


Development Environment Updates - May 8-25, 2025
2025-05-25T18:25:00+01:00
Git History Summarization Upgrade</h2>

Overhauled my-git-history-summary</code> to use absolute dates instead of hours for clearer time references</li>
Added repository list validation and host-specific configuration support</li>
Enhanced error handling and output formatting with clearer section separation</li>
</ul>
Environment Configuration</h2>

Added memory usage tracking to Starship prompt</li>
Installed JJ version control system integration in Starship</li>
Migrated from Oh My Posh to Starship for terminal theming</li>
</ul>
AI Development Toolkit</h2>

Expanded ai</code> script with devstral</code> and gemma</code> models support</li>
Configured Ollama API endpoints for local AI services</li>
Added model override flags for specialized AI workloads</li>
</ul>
Infrastructure Improvements</h2>

Created systemd</code> services for automated backups</li>
Added new host configuration with port forwarding for my new server</li>
</ul>
Maintenance & Bug Fixes</h2>

Resolved path issues with Kanata keyboard daemon</li>
Patched Ollama API connection settings</li>
Improved cross-platform compatibility checks</li>
</ul>


Project Release: git-next v2025.5.1
2025-05-25T15:16:00+01:00
Today I released version 2025.5.1</code> of git-next</a>, my Trunk-based development manager.
This update fixes an embarrasing bug in the previous version, where it couldn’t update the next</code> branch.
Which is basically git-next</code>’s whole thing.</p>
Key Changes:</h2>
Git Operations Overhaul 🔧</h3>

Fixed remote push synchronization issues</li>
</ul>

While switching to git2</code> in the previous release I got the process wrong and my tests weren’t sufficient to pick up on it.</p>
</blockquote>

Added detailed push conflict detection</li>
</ul>

So, working through the push process I’ve added more checks.</p>
</blockquote>

Enhanced rebase handling with parent commit validation

Improved commit log traversal with duplicate filtering</li>
</ul>
</li>
</ul>
Configuration & Maintenance 🛠</h3>

Removed legacy external repo support</li>
</ul>

This had originally been added as an option, but I’m not aware that it get’s
used and it complicates matters when deciding the name of the remote that we
should be pushing to.
Where as, when we have only internal repos. we always know that they will
called ‘origin’.</p>
</blockquote>

Simplified remote management (now always uses ‘origin’)</li>
Refactored ServerRepoConfig to builder pattern</li>
</ul>

I’m moving away from constructor to use builders more</p>
</blockquote>

Added Docker publishing workflow to Justfile</li>
</ul>

Because I kept forgetting to push the image to codeberg:</p>
</blockquote>
   just</span></span> publish-docker-image v2025.5.1</span>
</span></code></pre>
Diagnostics & Reliability 🩺</h3>

Added more logging:

Added detailed push failure diagnostics</li>
Implemented repository state logging</li>
</ul>
</li>
Fixed shutdown error visibility</li>
</ul>

When using the --ui</code> or --xui</code> any error that caused the server to abort we
lost when the screen reset to normal.
Now it is logged to the screen after the console is reset.</p>
</blockquote>

Removed debug symbols from dev builds</li>
</ul>

An optimisation for development only.</p>
</blockquote>
Documentation Updates 📚</h3>

Corrected Forgejo naming throughout codebase</li>
</ul>

I had been using ForgeJo</code> consistently, but wrongly.</p>
</blockquote>

Removed deprecated gitdir references</li>
</ul>

Now that we only support internal repos, we don’t support the gitdir</code> option
for repos.</p>
</blockquote>
Testing Improvements ✅</h3>

Added comprehensive force push tests</li>
</ul>

Given the mess I made with git push, there are now more tests around this.</p>
</blockquote>

Implemented remote test setup framework</li>
</ul>

Part of that is setting up two git repos on disk and making one the remote of
the other.</p>
</blockquote>

Added 6 new test cases for branch validation</li>
</ul>

Tests to check that we use --force-with-lease</code> only when required, which is
when CI fails and next</code> needs to be reset back to main</code>.</p>
</blockquote>

Enhanced commit history verification</li>
</ul>

Again, more tests to check this is doing what it is supposed to.</p>
</blockquote>
This release focuses on hardening Git operations while improving
maintainability and diagnostics.</p>


PR: Mirror Branch Filter for Forgejo
2025-05-09T07:46:00+01:00
I use a private Forgejo</a> instance to host all my projects, mirroring some to codeberg.org</a>.</p>
However, I’d like to be able to only publish my main</code> branch for each repo to Codeberg.
Forgejo doesn’t support this, so I created a Feature Request</a>.
Now that I’ve fininished the update to git-next</a> that I had been working on, I’ve made a stab at implementing this.</p>
Forgejo is written in Go</a> which I’ve only used very briefly about 5 years ago.</p>
The Pull Request</a>.</p>
It’s not complete, but I think it hits the main points needed for the implementation.</p>
It is missing things, like tests and documentation.</p>
Hopefully I can get some feedback to make sure I’m even approaching the problem correctly.</p>
Mirror Branch Filtering</h2>
A new feature to allow filtering which branches get mirrored. The changes span multiple components:</p>

Added BranchFilter</code> field to mirror settings and API models</li>
Updated database schema with migration</li>
Modified push mirror logic to handle branch filtering</li>
Added UI controls in repository settings</li>
When branch filter is specified, mirrors will only push the listed branches (comma-separated)</li>
</ul>
The mirror branch filtering provides more granular control over repository mirroring operations.</p>


Project Update: git-next
2025-05-08T18:44:00+01:00
Migration Cleanup Completed</h2>
The .dev/TODO.md</code> file has been significantly simplified after completing the migration from gix</code> to git2</code> in the gitlib</code> module.
The detailed migration checklist has been replaced with a fresh TODO template for the next project.</p>
New Docker Publishing Workflow</h2>
A new just</a> recipe has been added to streamline Docker image publishing.
The publish-docker-image</code> command now handles:</p>
• Pulling images from the internal registry
• Retagging for Codeberg.org</a>
• Pushing to the public registry</p>
Example usage:</p>
just</span></span> publish-docker-image v1.2.3</span>
</span></code></pre>


Blog theme: tabi
2025-05-08T08:21:38+01:00
Switched the blog theme to tabi</a>.</p>
Tabi is an accessible Zola theme with features like:</p>

Search functionality</li>
Multi-language support</li>
Optional JavaScript</li>
Perfect Lighthouse score</li>
Comprehensive documentation</li>
</ul>
Tabi is a modern, accessible design with good performance
characteristics. The theme appears to be well-documented and
actively maintained based on its GitHub repository</p>


Project Release: git-next v2025.5.0
2025-05-07T21:44:00+01:00
Today I released version 2025.5.0</code> of git-next</a>, my Trunk-based development manager.
This major update includes significant backend changes and reworked feature selections.</p>
Key Changes:</h2>
Backend Migration:</h3>

Completed migration from gix</a> to git2</a> for Git operations</li>
Implemented git2-based functionality for:

Rebase operations (with new tests)</li>
Remote operations (fetch/push)</li>
Branch switching</li>
File reading</li>
Clone operations</li>
Commit log access</li>
</ul>
</li>
</ul>
Cargo Features:</h3>

Moved email and desktop notifications into two new non-default features:

notify-desktop</code></li>
notify-email</code></li>
</ul>
</li>
Renamed the features for the forges selections to have forge-</code> prefixed:

forgejo</code> -> forge-forgeoj</code></li>
github</code> -> forge-github</code></li>
</ul>
</li>
</ul>
Improvements:</h2>

Enhanced workflow safety and performance in CI</li>
Added OpenSSL build dependencies for Forgejo workflow, this was required to use https</code> with git2</code>.</li>
Improved cargo registry caching in CI</li>
Updated numerous dependencies to latest versions</li>
</ul>
Documentation:</h2>

Added project coding rules</a> and conventions</a></li>
Added TODO</a> file for future work</li>
</ul>
This release represents a significant step forward in stability and functionality. The complete migration to git2 provides a more robust
foundation for future development.</p>
The full changelog</a> and updated documentation can be found in the project repository.</p>


Development Environment Updates - May 6-7, 2025
2025-05-06T21:33:06+01:00
New Git History Summary Tool</h2>
Added a new bash script my-git-history-summary</code> that:</p>

Takes a list of repositories and generates summaries of recent changes</li>
Uses aider</a> with the deepseek</a> model to analyze git diffs</li>
Defaults to showing changes from the last 24 hours</li>
Includes error handling for invalid repos/directories</li>
Comes with a sample repos.list.dorida</code> configuration file for my main dev machine</li>
</ul>
#</span></span>!/usr/bin/env bash</span>
</span></span>
</span>set</span></span> -</span>e</span></span>
</span>
</span>#</span></span> Read the list of repositories from the file</span>
</span></span>repos_file</span>=</span>"</span>$</span>HOME</span></span>/repos.list"</span></span></span>
</span>
</span>if</span> [[</span> $</span>#</span></span> -</span>ge</span> 1 ]]</span></span>;</span> then</span>
</span>    repos_file</span>=</span>"</span>$</span>1</span></span>"</span></span></span>
</span>else</span>
</span>    echo</span></span> "</span>Usage: $</span>0</span></span> <repositories_file> [hours]"</span></span></span>
</span>    exit</span></span> 1</span>
</span>fi</span>
</span>
</span>#</span></span> Parse hours, default to 24 if not provided</span>
</span></span>if</span> [[</span> $</span>#</span></span> -</span>gt</span> 1 ]]</span></span>;</span> then</span>
</span>    hours</span>=</span>"</span>$</span>2</span></span>"</span></span></span>
</span>else</span>
</span>    hours</span>=</span>24</span>
</span>fi</span>
</span>
</span>#</span></span> Format time for git log's --since option (relative date)</span>
</span></span>since_time</span>=</span>"</span>$</span>{</span></span>hours</span></span>}</span></span> hours ago"</span></span></span>
</span>
</span>#</span></span> Check if repos file exists</span>
</span></span>if</span> [</span> !</span> -</span>f</span> "</span>$</span>repos_file</span></span>"</span></span> ]</span></span>;</span> then</span>
</span>    echo</span></span> "</span>Error: Repositories file $</span>repos_file</span></span> not found!"</span></span></span>
</span>    exit</span></span> 1</span>
</span>fi</span>
</span>
</span>#</span></span> Read repository paths into an array</span>
</span></span>mapfile</span></span> -</span>t</span> repos <</span>"</span>$</span>repos_file</span></span>"</span></span></span>
</span>
</span>#</span></span> Process each repository</span>
</span></span>for</span> repo in</span> "</span>$</span>{</span></span>repos[</span>@</span>]</span></span></span>}</span></span>"</span></span></span>;</span> do</span>
</span>    echo</span></span> "</span>$</span>repo</span></span>:"</span></span></span>
</span>    #</span></span> Check if it's a valid Git repository</span>
</span></span>    if</span> [</span> !</span> -</span>d</span> "</span>$</span>HOME</span></span>/$</span>repo</span></span>"</span></span> ]</span></span>;</span> then</span>
</span>        echo</span></span> "</span>Error: $</span>repo</span></span> is not a valid directory."</span></span></span>
</span>        exit</span></span> 1</span>
</span>    fi</span>
</span>    (</span>
</span>        cd</span></span> "</span>$</span>HOME</span></span>/$</span>repo</span></span>"</span></span></span> ||</span> (</span>echo</span></span> "</span>Error: $</span>repo</span></span> is not accessible"</span></span></span> ;</span> exit</span></span> 1</span>)</span>
</span>
</span>
</span>        if</span> !</span> git</span></span> rev-parse --</span>is-inside-work-tree</span> ></span>/dev/null 2</span>>&</span>1</span></span> ;</span> then</span>
</span>            echo</span></span> "</span>Error: $</span>repo</span></span> is not a valid Git repository."</span></span></span>
</span>            exit</span></span> 1</span>
</span>        fi</span>
</span>
</span>        #</span></span> Get the log output for commits within the last x hours</span>
</span></span>        git_diff</span>=</span>$</span>(</span>git</span></span> log --</span>since</span> "</span>$</span>since_time</span></span>"</span></span> --</span>patch</span></span>)</span></span></span>
</span>        if</span> [[</span> -</span>n</span> "</span>$</span>git_diff</span></span>"</span></span> ]]</span></span>;</span> then</span>
</span>            #</span></span> Use aider to generate a summary of changes</span>
</span></span>            echo</span></span> "</span>=== Summary generated by aider ==="</span></span></span>
</span>            aider</span></span> --</span>model</span> deepseek --</span>message</span> "</span>Generate a summary of the changes in the diff: $</span>git_diff</span></span>"</span></span></span>
</span>            echo</span></span> "</span>=== End of summary ==="</span></span></span>
</span>        else</span>
</span>            echo</span></span> "</span>No changes since $</span>since_time</span></span>"</span></span></span>
</span>        fi</span>
</span>    )</span>
</span>done</span>
</span></code></pre>
Mise Configuration Updates</h2>

Added fzf</a> as a required tool in mise</a> configuration</li>
Refactored bashrc’s mise</code> initialization into four separate modules:

mise-activate</code> for core activation</li>
mise-reshim</code> for reshimming</li>
mise-completion</code> for bash completion</li>
mise-tools</code> for tool setup
This provides more granular control over the initialization process.</li>
</ul>
</li>
</ul>
SSH Configuration</h2>
Added Amazon Q</a> SSH integration by including its configuration file at the bottom of .ssh/config</code>.
The integration is set to match all hosts.</p>


New Zola (with Facebook)
2025-01-06T07:33:06+01:00
It’s a new year and I’ve been knocking off the dust from this blog.</p>
I’ve converted it to zola</a>. It had been built with Hugo</a> previously.</p>

EDIT: my facebook archive has been removed and is now hosted on a private site, not on the Internet.</p>
</blockquote>
The site now includes part of my Facebook archive. It’s just part of the data export in HTML, os it’s integrated into the blog.</del></p>
Facebook Archive</del></h3>

photos</del></li>
videos</del></li>
posts</del></li>
more photos</del></li>
check ins</del></li>
</ul>


Mastodon Comments
2023-09-24T19:47:06+01:00
I’ve been looking at tweaking how my blog works, in an effort to encourage me to write more for it.</p>
On the way I found Carl Schwan’s implementation</a> for adding replies to a post on Mastodon to a static site as comments.</p>
So, I’ve added a very slightly modified version here. With tweaks to work with my version of Hugo and the LoveIt theme.</p>
(edit: currently disabled - since switching from hugo to zola</a> this will need to be reimplmented)</em></p>


Blocked From My Own Mailserver
2023-07-29T15:53:08+01:00
Yesterday I saw the announcement the Thunderbird 115 was finally available on flathub.
I thought I would install it and try it out on my laptop.
Normally I use a Roundcube server as my main mail client, with Mail K-9 for my mobile access.</p>
Installed Thunderbird without a hitch.
But then I ran into a problem when I tried to add my first email account.</p>
It didn’t pickup the correct mail server hostnames from autodiscovery, so I have to change those.
Then it wouldn’t authenticate. The program hung.
I didn’t have time to investigate, so I killed the program and went on to do something else.</p>
I later found out that I could no longer access my mail server from mobile or Roundcube.</p>
I could telnet to port 993 from the server with telnet localhost 993</code>, but not from any other
machine on my network. I was also able to access my email from my mobile if I turned off my WiFi and
came in over the Internet.</p>
I was very confused.</p>
My mail server is a self-hosted instance of docker-mailserver</a>.</p>
One of the features docker-mailserver provides is integration with fail2ban</a>.</p>

Fail2ban scans log files (e.g. /var/log/apache/error_log) and bans IPs that show the malicious signs – too many password failures, seeking for exploits, etc.</p>
</blockquote>
It took me a few hours to figure out that fail2ban had added my own IP to the ban address, presumably because
of the misbehaviour of Thunderbird.</p>
The fix was simple:</p>
cd</span></span> mailserver</span>
</span>./setup.sh</span></span> fail2ban unban $</span>MY_IP</span></span></span>
</span></code></pre>
setup.sh</code> is the management script provided by docker-mailserver, and provides easy access to the commands installed in the docker image.</p>
I do plan on trying Thunderbird again, but this time I’ll keep an eye on my mailserver logs to make sure I don’t get locked out again.</p>


Create Custom Garmin Activity Type
2023-07-20T18:50:31+01:00
I use Garmin Forerunner 255 Music as my activity tracker for running, walking and OT class.
For the OT class, we occasionally have a boxing session.
My Garmin has a handy ‘Incident Detection’ feature, or a fall-detector.
This is a safety feature that will send a message to select contacts with my location if it thinks I’ve fallen.
It does give me time to cancel the message, but in the middle of a boxing circuit it’s a bit annoying to stop to fiddle with the watch.
Especially when wearing boxing gloves.</p>
So, to get around this I created a custom activity type for the OT class, where the Incident Detection feature is turned off.</p>
Create a new Activity Type with Incident Detection turned off:</p>

Press the Start button to see a list of activity types to start.</li>
Press the Up button to jump to the bottom of the list. :)</li>
Press the Down button to select the ‘+ Add’.</li>
Press the Up button to select Copy Activity.</li>
Press the Down button, as needed, to select the Activity Type I want to copy.</li>
Press the Start button to copy the Activity Type.</li>
Press the Up/Down/Start buttons to navigate the text editing widget to give the Activity Type a new name.</li>
Select the Green Tick when finished.</li>
Select Done option.</li>
Select as favourite or not.</li>
Press the Up/Down/Start buttons to select where in the menu to place it.</li>
Press the Back button to return to the main screen.</li>
Press and hold the Up key to select settins menu.</li>
Select Safety & Tracking.</li>
Select Incident Detection.</li>
Read the message and press Start.</li>
Press the Up/Down buttons to select the Activity Type.</li>
Press the Start button to toggle Incident Detection on/off.</li>
</ol>


Developer AI: phind.com
2023-05-30T08:56:49+01:00
AI is all the rage. Hyped to the moon.</p>
I’ve generally been avoiding it.</p>
I’ve found one that does interest me though: phind</a></p>
It labels itself a an AI search engine for developers.</p>

Phind is a search engine that simply tells users what the answer is.
Optimized for developers and technical questions, Phind instantly
answers questions with detailed explanations and relevant code snippets
from the web.</p>
</blockquote>

Phind is powered by large AI language models. Unlike some other AI
assistants, Phind pulls information from the internet and is always up
to date. It’s smart enough to generate answers based on information from
multiple sources. With Phind, finding information is as seamless and
informative as talking to a friend.</p>
</blockquote>
It has successfully answered my questions so far.</p>


Fixing missing Ubuntu dependencies installing Python with RTX
2023-05-29T10:40:03+01:00
Fixing the install of python via rtx</a> (now ‘mise’).</p>

ModuleNotFoundError: No module named '_bz2'</code></li>
ModuleNotFoundError: No module named '_curses' </code></li>
ModuleNotFoundError: No module named 'readline'</code></li>
ModuleNotFoundError: No module named '_tkinter'</code></li>
ModuleNotFoundError: No module named 'yaml'</code></li>
</ul>
Most of these modules aren’t really needed by me, but I’d rather not have the errors.</p>
apt-get</span></span> install libbz2-dev libncurses5-dev libreadline-dev libyaml-dev</span>
</span>rtx</span></span> uninstall python</span>
</span>rtx</span></span> install python</span>
</span></code></pre>
I’ve not found how to install the requirements for tkinter yet. Given that rtx is compiling python, installing python-tk</code> kind of defeats that purpose.</p>


Bootstrap Ubuntu
2023-05-10T10:31:20+01:00
My rough notes on bootstrapping a plain Ubuntu instance into my dev environment.</p>
Prequisites</h2>

Ubuntu 22.04 (I use Ubuntu MATE)</li>
My user with sudo</code> permissions</li>
</ul>
Install any updates.</p>
sudo</span></span> apt-update</span>
</span>sudo</span></span> apt-upgrade</span>
</span></code></pre>
Remote Access</h2>

Pin the IP address</li>
Install an SSH server</li>
Add my public key</li>
Verify SSH access</li>
</ul>
Pin the IP address</h3>
Currently my internal network is managed by [eero].</p>
Verify that, if using WiFi, that I’m connected to the correct WiFi network.</p>

Open [eero] app on phone</li>
Settings</li>
Network Settings</li>
Reservations & port forwarding</li>
Add a Resrvation</li>
Select device</li>
Save</li>
</ul>
Install an SSH server</h3>
sudo</span></span> apt install openssh-server molly-guard</span>
</span></code></pre>
Assuming the hostname is anker</code> with an IP address of 192.168.2.22</code>:</p>
Add host details to ~/.ssh/config</code>:</p>
Host anker
</span>    Hostname 192.168.2.22
</span>    User pcampbell
</span></code></pre>
Test that you can ssh to the server:</p>
ssh</span></span> anker</span>
</span></code></pre>
Exit and return to the local machine.</p>
Add my public key</h3>
Add public key to server:</p>
ssh-copy-id</span></span> -</span>i</span> ~</span></span>/.ssh/id_rsa anker</span>
</span>ssh-copy-id</span></span> -</span>i</span> ~</span></span>/.ssh/id_ed25519 anker</span>
</span></code></pre>
Verify SSH access</h3>
Verify that you don’t need to enter your password (assuming your ssh-agent is working):</p>
ssh</span></span> anker</span>
</span></code></pre>
Configure home directory</h2>
I use [yadm] to track my user settings, although I want to have a look at [chezmoi] as a possible alternative.</p>

Install rust, rtx and yadm</li>
Clone yadm dotfiles</li>
Verify submodules</li>
Verify ssh access retained</li>
Verify neovim config loaded</li>
</ul>
Install rust, rtx and yadm</h3>
Run the following commands on the server.</p>
Rust</h4>
sudo</span></span> apt install curl git direnv keychain</span>
</span>curl</span></span> --</span>proto</span> '</span>=https'</span></span> --</span>tlsv1</span>.2 -</span>sSf</span> https://sh.rustup.rs</span> |</span> shcurl</span></span> --</span>proto</span> '</span>=https'</span></span> --</span>tlsv1</span>.2 -</span>sSf</span> https://sh.rustup.rs</span> |</span> sh</span></span>
</span></code></pre>
Quit and restart ssh and verify ~/.cargo/bin/</code> is in your PATH</code>.</p>
RTX</h4>
cargo</span></span> install rtx-cli</span>
</span></code></pre>

Note, when using NuShell, I’m not able to run rtx without specifying the full path, ~/.cargo/bin/rtx</code>.</p>

See: https://github.com/jdxcode/rtx/issues/514</a></li>
</ul>
</blockquote>
Yadm</h4>
rtx</span></span> global yadm@latest</span>
</span>rtx</span></span> install</span>
</span></code></pre>
Now clone the yadm dotfiles repo:</p>
$</span>(</span>rtx</span></span> which yadm</span>)</span></span></span></span> clone --</span>depth</span> 1 --</span>recursive</span> git@git.kemitix.net:kemitix/dotfiles-yadm.git </span>
</span>$</span>(</span>rtx</span></span> which yadm</span>)</span></span></span></span> checkout $</span>HOME</span></span></span>
</span>$</span>(</span>rtx</span></span> which yadm</span>)</span></span></span></span> enter</span>
</span>git</span></span> submodule update --</span>init</span> --</span>recursive</span></span>
</span>cd</span></span> ~</span></span>/.config/nvim/lua</span>
</span>ln</span></span> -</span>s</span> ~</span></span>/.config/nvim-lua-user/ user</span>
</span>exit</span></span>
</span>rtx</span></span> install</span>
</span></code></pre>
Boostrap</h3>
Run the bootstrap:</p>
~</span></span>/bin/bootstrap</span></span> all</span>
</span></code></pre>
Install Desktop software</h2>
If using the machine as a desktop.</p>

Vivaldi</li>
Thunderbird</li>
</ul>
Install Docker</h2>
These are the instructions from docker.com</p>
sudo</span></span> apt-get remove docker docker-engine docker.io containerd runc</span>
</span>sudo</span></span> apt-get update</span>
</span>sudo</span></span> apt-get install ca-certificates curl gnupg</span>
</span>sudo</span></span> install -</span>m</span> 0755 -</span>d</span> /etc/apt/keyrings</span>
</span>curl</span></span> -</span>fsSL</span> https://download.docker.com/linux/ubuntu/gpg</span> |</span> sudo</span></span> gpg --</span>dearmor</span> -</span>o</span> /etc/apt/keyrings/docker.gpg</span>
</span>sudo</span></span> chmod a+r /etc/apt/keyrings/docker.gpg</span>
</span>echo</span></span> \
</span></span></span>  "</span>deb [arch="</span></span>$</span>(</span>dpkg</span></span> --</span>print-architecture</span></span>)</span></span>"</span> signed-by=/etc/apt/keyrings/docker.gpg] https://download.docker.com/linux/ubuntu \
</span></span></span></span>  "</span></span>$</span>(</span>.</span></span> /etc/os-release</span> &&</span> echo</span></span> "</span>$</span>VERSION_CODENAME</span></span>"</span></span></span>)</span></span>"</span> stable"</span></span></span> |</span> \
</span></span>sudo</span></span> tee /etc/apt/sources.list.d/docker.list ></span> /dev/null</span>
</span>sudo</span></span> apt-get update</span>
</span>sudo</span></span> apt-get install docker-ce docker-ce-cli containerd.io docker-buildx-plugin docker-compose-plugin</span>
</span>sudo</span></span> docker run hello-world</span>
</span></code></pre>


Thread Macros in Rust
2023-05-05T11:38:46+01:00
In my current WIP in Rust I needed to spawn a number of different threads, then wait for them to complete.</p>
Niave Thread Spawn and Join</h2>
My first attempt was pretty much what the docs say to do:</p>
let</span> mut</span> threads =</span> vec!</span>[</span>]</span></span>;</span>
</span>threads.</span>push</span>(</span>thread::</span></span>spawn(</span>move</span> ||</span> {</span>
</span></span></span></span>    reader::</span></span>run(</span>sync_sets.</span>clone</span>(</span></span>)</span></span>,</span> cli,</span> tx_reader_to_core</span>)</span></span>
</span></span></span></span></span>}</span></span></span>)</span></span></span>)</span></span>;</span>
</span>threads.</span>push</span>(</span>thread::</span></span>spawn(</span>move</span> ||</span> {</span>
</span></span></span></span>    writer::</span></span>run(</span>sync_sets,</span> rx_core_to_writer,</span> tx_writer_to_core</span>)</span></span>
</span></span></span></span></span>}</span></span></span>)</span></span></span>)</span></span>;</span>
</span>threads.</span>push</span>(</span>thread::</span></span>spawn(</span>move</span> ||</span> {</span>
</span></span></span></span>    store::</span></span>run(</span>store,</span> rx_core_to_store,</span> tx_store_to_core</span>)</span></span>
</span></span></span></span></span>}</span></span></span>)</span></span></span>)</span></span>;</span>
</span>threads.</span>push</span>(</span>thread::</span></span>spawn(</span>move</span> ||</span> {</span>
</span></span></span></span>    core::</span></span>run(</span>
</span></span></span></span></span>        rx_reader_to_core,</span>
</span></span></span></span></span>        rx_store_to_core,</span>
</span></span></span></span></span>        rx_writer_to_core,</span>
</span></span></span></span></span>        tx_core_to_store,</span>
</span></span></span></span></span>        tx_core_to_writer,</span>
</span></span></span></span></span>    </span>)</span></span>
</span></span></span></span></span>}</span></span></span>)</span></span></span>)</span></span>;</span>
</span>
</span>for</span> thread in</span> threads {</span>
</span></span>    thread.</span>join</span>(</span></span>)</span></span>.</span>unwrap_or</span>(</span>(</span></span>)</span></span></span>)</span></span>;</span>
</span></span></span>}</span></span>
</span></code></pre>
There’s a fair bit of repetition there for each thread, and it’s hiding the important details in the middle of it all.</p>
Repetition like this is where Rust’s declaritive macros can help.</p>
Using Macros</h2>
I created two macros, one to handle the spawning and another to wrap that and join all the threads.</p>
macro_rules!</span> spawn</span> {</span></span></span>
</span></span></span>    (</span>$</span>(</span>$x</span>:</span>expr</span>)</span>,*</span></span>)</span></span> =></span> {</span>
</span></span></span></span>        {</span>
</span></span></span></span></span>            let</span> mut</span> threads =</span> vec!</span>[</span>]</span></span>;</span>
</span></span></span></span></span>            $</span>(</span>threads.</span>push</span>(</span>std::</span></span>thread::</span></span>spawn(</span>move</span> ||</span> $x</span></span>)</span></span></span>)</span></span>;</span></span>)</span></span>*</span>
</span></span></span></span></span>            threads
</span></span></span></span></span>        </span>}</span></span>
</span></span></span></span>    }</span></span>;</span>
</span></span></span></span></span>}</span></span></span>
</span>
</span>macro_rules!</span> spawn_and_join</span> {</span></span></span>
</span></span></span>    (</span>$</span>(</span>$x</span>:</span>expr</span>)</span>,*</span></span>)</span></span> =></span> {</span>
</span></span></span></span>        let</span> threads =</span> spawn!</span>(</span>$</span>(</span>$x</span></span>)</span></span>,</span>*</span></span>)</span></span>;</span>
</span></span></span></span>        for</span> thread in</span> threads {</span>
</span></span></span></span></span>            thread.</span>join</span>(</span></span>)</span></span>.</span>unwrap_or</span>(</span>(</span></span>)</span></span></span>)</span></span>;</span>
</span></span></span></span></span>        </span>}</span></span>
</span></span></span></span>    }</span></span>;</span>
</span></span></span></span></span>}</span></span></span>
</span></code></pre>
This allows the code we saw earlier to be condensed to just:</p>
spawn_and_join!</span>(</span>
</span></span>    writer::</span></span>run(</span>writer_syncfile,</span> rx_core_to_writer,</span> tx_writer_to_core</span>)</span></span>,</span>
</span></span>    store::</span></span>run(</span>store,</span> rx_core_to_store,</span> tx_store_to_core</span>)</span></span>,</span>
</span></span>    core::</span></span>run(</span>
</span></span></span>        rx_reader_to_core,</span>
</span></span></span>        rx_store_to_core,</span>
</span></span></span>        rx_writer_to_core,</span>
</span></span></span>        tx_core_to_store,</span>
</span></span></span>        tx_core_to_writer,</span>
</span></span></span>    </span>)</span></span>,</span>
</span></span>    reader::</span></span>run(</span>reader_syncfile,</span> cli,</span> tx_reader_to_core</span>)</span></span>
</span></span></span>)</span></span>;</span>
</span></code></pre>
I then saw that a todo!()</code> in my writer</code> thread was causing the thread to panic, and the log message was that a thread unnamed</code> had paniced.
There was more details that pointed to what I need to do next in this project, but I thought I should name the threads before moving on to that.</p>
Named Threads</h2>
I didn’t want to supply the thread names as a String</code> or &str</code>. I want to keep the noise in the code down to a minimum.
So I use an ident</code> and the stringify!</code> macro to let me do the following:</p>
spawn_and_join!</span>(</span>
</span></span>   writer:</span> writer::</span></span>run(</span>writer_syncfile,</span> rx_core_to_writer,</span> tx_writer_to_core</span>)</span></span>,</span>
</span></span>   store:</span> store::</span></span>run(</span>store,</span> rx_core_to_store,</span> tx_store_to_core</span>)</span></span>,</span>
</span></span>  core:</span> core::</span></span>run(</span>
</span></span></span>       rx_reader_to_core,</span>
</span></span></span>       rx_store_to_core,</span>
</span></span></span>       rx_writer_to_core,</span>
</span></span></span>       tx_core_to_store,</span>
</span></span></span>       tx_core_to_writer,</span>
</span></span></span>    </span>)</span></span>,</span>
</span></span>    reader:</span> reader::</span></span>run(</span>reader_syncfile,</span> cli,</span> tx_reader_to_core</span>)</span></span>
</span></span></span>)</span></span>;</span>
</span></code></pre>
Each thread name is an ident</code> prefixing the expr</code> that is the body of the new thread.</p>
To specify the name of the thread, we can’t just use std::thread::spawn</code>, we need to use std::thread::Builder::new().name(...)</code>.</p>
The updated macros are here:</p>
macro_rules!</span> spawn</span> {</span></span></span>
</span></span></span>    (</span>$</span>(</span>$name</span>:</span>ident</span>: $x</span>:</span>expr</span>)</span>,*</span></span>)</span></span> =></span> {</span>
</span></span></span></span>        {</span>
</span></span></span></span></span>            let</span> mut</span> threads =</span> vec!</span>[</span>]</span></span>;</span>
</span></span></span></span></span>            $</span>(</span>threads.</span>push</span>(</span>{</span>
</span></span></span></span></span></span></span></span>                let</span> thread =</span> std::</span></span>thread::</span></span>Builder::</span></span>new(</span></span>)</span></span>.</span>name</span>(</span>stringify!</span>(</span>$name</span></span>)</span></span>.</span>to_owned</span>(</span></span>)</span></span></span>)</span></span>;</span>
</span></span></span></span></span></span></span></span>                thread.</span>spawn</span>(</span>move</span> ||</span> $x</span></span>)</span></span>?</span>
</span></span></span></span></span></span></span></span>            </span>}</span></span></span>)</span></span>;</span></span>)</span></span>*</span>
</span></span></span></span></span>            threads
</span></span></span></span></span>        </span>}</span></span>
</span></span></span></span>    }</span></span>;</span>
</span></span></span></span></span>}</span></span></span>
</span>
</span>macro_rules!</span> spawn_and_join</span> {</span></span></span>
</span></span></span>    (</span>$</span>(</span>$name</span>:</span>ident</span>: $x</span>:</span>expr</span>)</span>,*</span></span>)</span></span> =></span> {</span>
</span></span></span></span>        let</span> threads =</span> spawn!</span>(</span>$</span>(</span>$name</span>:</span> $x</span></span>)</span></span>,</span>*</span></span>)</span></span>;</span>
</span></span></span></span>        for</span> thread in</span> threads {</span>
</span></span></span></span></span>            thread.</span>join</span>(</span></span>)</span></span>.</span>unwrap_or</span>(</span>(</span></span>)</span></span></span>)</span></span>;</span>
</span></span></span></span></span>        </span>}</span></span>
</span></span></span></span>    }</span></span>;</span>
</span></span></span></span></span>}</span></span></span>
</span></code></pre>


Astronvim Mason Dap Error
2023-04-28T19:41:24+01:00
I’ve been trying to use Neovim</a> as my IDE, with the AstroNvim</a> configuration.
Switching from Emacs key bindings has been a small challenge, but I’m getting the hang of it after only a week.</p>
I did try a few days ago to add the java pack from the astrocommunity.
I wasn’t able to figure out how to configure it, so backed it out.
I’m not needing to do any Java work for a while, it was only a curiosity.</p>
Problem</h2>
Unfortunately, every time I start up nvim and opena file, I get this error appearing in a popup:</p>
Error executing vim.schedule lua callback: ...m-dap.nvim/lua/mason-nvim-dap/mappings/adapters/init.lua:7: module 'mason-nvim-dap.mappings.adapters.js' not found
</span>        no field package.preload['mason-nvim-dap.mappings.adapters.js']                                                                                                                                            
</span>cache_loader: module mason-nvim-dap.mappings.adapters.js not found                                                                                                                                                 
</span>cache_loader_lib: module mason-nvim-dap.mappings.adapters.js not found                                                                                                                                             
</span>        no file './mason-nvim-dap/mappings/adapters/js.lua'                                                                                                                                                        
</span>        no file '/home/runner/work/neovim/neovim/.deps/usr/share/luajit-2.1.0-beta3/mason-nvim-dap/mappings/adapters/js.lua'                                                                                       
</span>        no file '/usr/local/share/lua/5.1/mason-nvim-dap/mappings/adapters/js.lua'                                                                                                                                 
</span>        no file '/usr/local/share/lua/5.1/mason-nvim-dap/mappings/adapters/js/init.lua'                                                                                                                            
</span>        no file '/home/runner/work/neovim/neovim/.deps/usr/share/lua/5.1/mason-nvim-dap/mappings/adapters/js.lua'                                                                                                  
</span>        no file '/home/runner/work/neovim/neovim/.deps/usr/share/lua/5.1/mason-nvim-dap/mappings/adapters/js/init.lua'                                                                                             
</span>        no file './mason-nvim-dap/mappings/adapters/js.so'                                                                                                                                                         
</span>        no file '/usr/local/lib/lua/5.1/mason-nvim-dap/mappings/adapters/js.so'                                                                                                                                    
</span>        no file '/home/runner/work/neovim/neovim/.deps/usr/lib/lua/5.1/mason-nvim-dap/mappings/adapters/js.so'                                                                                                     
</span>        no file '/usr/local/lib/lua/5.1/loadall.so'                                                                                                                                                                
</span>        no file './mason-nvim-dap.so'                                                                                                                                                                              
</span>        no file '/usr/local/lib/lua/5.1/mason-nvim-dap.so'                                                                                                                                                         
</span>        no file '/home/runner/work/neovim/neovim/.deps/usr/lib/lua/5.1/mason-nvim-dap.so'                                                                                                                          
</span>        no file '/usr/local/lib/lua/5.1/loadall.so'                                                                                                                                                                
</span>stack traceback:                                                                                                                                                                                                   
</span>        [C]: in function 'require'                                                                                                                                                                                 
</span>        ...m-dap.nvim/lua/mason-nvim-dap/mappings/adapters/init.lua:7: in function '__index'                                                                                                                       
</span>        ...vim/lazy/mason-nvim-dap.nvim/lua/mason-nvim-dap/init.lua:49: in function 'fn'                                                                                                                           
</span>        ...l/share/nvim/lazy/mason.nvim/lua/mason-core/optional.lua:91: in function 'if_present'                                                                                                                   
</span>        ...vim/lazy/mason-nvim-dap.nvim/lua/mason-nvim-dap/init.lua:41: in function 'fn'                                                                                                                           
</span>        ...l/share/nvim/lazy/mason.nvim/lua/mason-core/optional.lua:91: in function 'if_present'                                                                                                                   
</span>        ...vim/lazy/mason-nvim-dap.nvim/lua/mason-nvim-dap/init.lua:65: in function ''                                                                                                                             
</span>        vim/_editor.lua: in function <vim/_editor.lua:0>  
</span></code></pre>
Investigation</h2>
I’ve tried the following to verify that my configuration is clean:</p>
cd</span></span> ~</span></span>/.config/nvim/</span>
</span>git</span></span> status</span>
</span></code></pre>
HEAD detached at v3.12.1                                                                                                                                                                                           
</span>nothing to commit, working tree clean
</span></code></pre>
cd</span></span> ~</span></span>/.config/nvim/lua/user/</span>
</span>git</span></span> reset --</span>hard</span> upstream/main</span>
</span></code></pre>
So far, no change.</p>
Next step: wipe the ~/.local/state/nvim/</code> then ~/.local/share/nvim/</code> directories.</p>
rm</span></span> -</span>r</span> ~</span></span>/.local/state/nvim/</span>
</span></code></pre>
Nope.</p>
Solution</h2>
rm</span></span> -</span>r</span> ~</span></span>/.local/share/nvim/</span>
</span></code></pre>
That appears to have fixed it.
All the plugins were resintalled as soon as I restarted nvim.</p>
I was then able to successfully restore my ~/.config/nvim/lua/user/</code> configuration without introducing any new problems.</p>


MATE Menu Tray is missing
2023-04-28T00:00:00+00:00
The Empty Desktop</h2>
Trying to login to my work laptop, which has the MATE</a> desktop, and the menu bar is missing.
This is going to make starting new programs interesting.</p>
I just have a blank screen and my mouse pointer.</p>
Now Slack has started.</p>
Investigation</h2>
This is good, I can click on a link in my Google Calendar integration and now I have a browser open.</p>
A quick trip to duckduckgo: mate desktop not loading</code></p>
Fourth answer: Linux Mint Forums</a>:</p>
</p>
There at the end of the snippet: TL, DR - Can get to the terminal by using ctrl+alt+T</strong></p>
That works, and now I have a terminal. (And a second terminal, so I can start typing this up as I go.)</p>
I know that the Welcome app in the MATE Control Center lets you pick your desktop style.
Maybe selecting another style will restart the task bar.</p>
Solution</h2>
Now that I know I can use Ctrl-Alt-T</code> to get a terminal, I can start the Control Center.</p>
mate-control-center</span></span>
</span></code></pre>
Under Administration</em> select Welcome</em>.</p>
</p>
Now Desktop Layout</em>.
The selected layout is Redmond</em>.
(I hadn’t realised I’d gone for that one.)
Let’s pick Familiar</em> and see if anything happens.</p>
</p>
Success</strong>, I now have my panels back.</p>
I’ll leave it on Familiar</em> for now to see if I like it.</p>
In the meantime, I can now get to work on something that isn’t fixing something that is broken.</p>
Oh, I have a new support ticket in for something else that is broken. sigh</em></p>


New site: kemitix.net
2023-04-27T00:00:00+00:00
I’ve never been one for having much of a blog.
I tried a few years ago to start bloging my tech notes.
That was on github pages with Jekyll.
I didn’t update it much past the initial few posts.
Hopfully this time I’ll me more consistent.
Which is why I’m particularly interested in making it as frictionless as possible to add new posts.</p>
I really didn’t like my experiance Jekyll, and when I can to try and revive my site, the Ruby installation process was more than I cared to deal with.
I ran into some errors as soon as I tried to install, and I’m not looking to get into the technical details of how my blog works.
I just want to create a markdown file, commit the file, maybe run a simple command, and that’s it published.</p>
I like static sites.
I’ve a magazine hosted on a S3 bucket: https://www.cossmass.com/</a>.
It is created using Hugo</a>, and I’ve been able to do some interesting things with the templates and loading data from yaml files.
So, after a look around at some other static site generators, I’ve settled on what I know, and recreated my ‘tech-blog’ using Hugo.</p>
Copying over the markdown files from Jekyll required no editing for them to work with Hugo.
I have cleaned up the date to remove the time stamp, and to add some tags, but I didn’t need</em> to.
I also renamed the extension to .md</code> from .markdown</code>, because it just hurt my eyes to look at them.</p>
As for hosting, I did think about creating another S3 bucket, but opted instead to use my existing home server.
It is already using Caddy</a> to proxy over a half dozen other sites, so adding a static site isn’t going to strain it.</p>
Once I eventually remembered that I needed to mount my public</code> directory inside my Caddy docker container, it just worked.</p>
I’ve still to work my way through most of the docs for the theme I’m using, Loveit</a>, so things may change as I tweek settings.</p>
Caddy Configuration</h3>
The revelent section of my Caddyfile</code>:</p>
kemitix.net {
</span>    root * /blog
</span>    file_server
</span>}
</span></code></pre>
Docker</h3>
And my docker-compose.yaml</code>:</p>
services</span></span>:</span>
</span>  caddy</span></span>:</span>
</span>    image</span></span>:</span> caddy:alpine</span>
</span>    restart</span></span>:</span> unless-stopped</span>
</span>    container_name</span></span>:</span> caddy</span>
</span>    volumes</span></span>:</span>
</span>      -</span> ./Caddyfile:/etc/caddy/Caddyfile</span>
</span>      -</span> ./caddy/certs:/certs</span>
</span>      -</span> ./caddy/caddy/config:/config</span>
</span>      -</span> ./caddy/data:/data</span>
</span>      -</span> ./caddy/sites:/srv</span>
</span>      -</span> ../blog/public:/blog</span>
</span>    network_mode</span></span>:</span> "</span>host"</span></span>
</span></code></pre>


ACPI BIOS Error (bug): Could not resolve symbol [ TZ.ETMD], AE_NOT_FOUND (20220313/psargs-330)
2023-04-06T00:00:00+00:00
I started seeing this error on my work laptop, a DELL XPS 17 9710</strong>:</p>
ACPI BIOS Error (bug): Could not resolve symbol [\_TZ.ETMD], AE_NOT_FOUND (20220313/psargs-330)
</span>
</span>No Local Variables are initialized for Method [_OSC]
</span> 
</span>Initialized Arguments for Method [_OSC]:  (4 arguments defined for method invocation)
</span>  Arg0:   00000000fb231197 <Obj>           Buffer(16) 5D A8 3B B2 B7 C8 42 35
</span>  Arg1:   0000000025d9a193 <Obj>           Integer 0000000000000001
</span>  Arg2:   0000000078b55545 <Obj>           Integer 0000000000000002
</span>  Arg3:   00000000bf92f763 <Obj>           Buffer(8) 00 00 00 00 05 00 00 00
</span> 
</span>ACPI Error: Aborting method \_SB.IETM._OSC due to previous error (AE_NOT_FOUND) (20220331/psparse-529)
</span></code></pre>
It would appear when I was attempting to power off or restart, and
would freeze at this point. I would then need to press and hold the
power button to power off the laptop.</p>
When starting Linux, I wouldn’t get the graphical login, but only the
terminal prompt.</p>
My current kernel is Linux 5.19.0-38-generic</code>, but my Grub menu, also
has Linux 5.15.0-25-generic</code> available.</p>
Solution</h2>
I was able to boot into the graphical login by selecting the Linux 5.15.0-25-generic</code> kernel from the grub menu. Once booted with this
kernel, I’m now able to reboot and power off as normal.</p>
Now I just need to set grub to use that kernel by default.</p>
Adding the following lines to /etc/default/grub</code> will cause grub to
save the selection each time, and default to that on the next boot.</p>
GRUB_SAVEDEFAULT=true
</span>GRUB_DEFAULT=saved
</span></code></pre>
Remove or comment out any other lines, such as GRUB_DEFAULT=0</code>.</p>
Run sudo update-grub</code>, then reboot to test it out.</p>


Make grub menu readable on high-resolution screens
2023-04-06T00:00:00+00:00
When grub loads on my work laptop, a DELL XPS 17 9710</strong>, the font is
tiny and practically unreadable.</p>
You can change the font of this screen.</p>
From /etc/default/grub</code>:</p>
# The resolution used on graphical terminal
</span># note that you can use only modes which your graphic card supports via VBE
</span># you can see them in real GRUB with the command `vbeinfo'
</span>#GRUB_GFXMODE=640x480
</span></code></pre>
To find the available resolutions on your machine, boot into the grub
menu and press C</code> to enter console mode. Then type vbeinfo</code>. Look
for the list of available resolutions and choose one.</p>
I went with 640x480</code>, which meant I only had to uncomment the
GRUB_GFXMODE</code> line.</p>
Then run sudo update-grub</code> to save the change. Reboot and you should
be able to read the screen.</p>


Disable Ubuntu's DNS stub resolver
2018-03-02T00:00:00+00:00
I’m starting to experiment with Rancher</a> to manage docker instances. However, when trying to starting the rancher-agent,
the logs were showing that it didn’t like the lookback address being used for DNS resolution. It then promptly refused to go
any further.</p>
The stub resolver is intended to make things easier for users who move between networks and need a more dynamic method to
select their DNS resolution server. Whether that is because they are a laptop user or they need to connect to a VPN
periodically.</p>
For a more static instance, such as my development machine, a fixed DNS resolution address will do me fine.</p>
The following instructions are from Bastian Voigt’s</a> answer on
askUbuntu.com</a>:</p>
How to disable systemd-resolved in Ubuntu?</a>:</p>

Disable the systemd-resolved service and stop it:</p>
sudo</span></span> systemctl disable systemd-resolved.service</span>
</span>sudo</span></span> service systemd-resolved stop</span>
</span></code></pre>
Put the following line in the [main]</code> section of your /etc/NetworkManager/NetworkManager.conf</code>:</p>
dns=default
</span></code></pre>
Delete the symlink /etc/resolv.conf</code></p>
sudo</span></span> rm /etc/resolv.conf</span>
</span></code></pre>
Restart network-manager</p>
sudo</span></span> service network-manager restart</span>
</span></code></pre>
</blockquote>
Then I simply restarted my rancher-agent docker instance and it’s happy to continue.</p>


Building this site in Github Pages
2018-02-24T00:00:00+00:00

Outdated</strong> - this site is not longer hosted on github nor is it using jekyll.</p>
</blockquote>
The following are instructions to build this site. Largely it is an exercise in both verifying the steps for myself and in using the site building tools to create and publish this post.</p>
These instructions are based on building a user or organisation site, rather than a project site.</p>
Create initial site</h2>

Create Repository: kemitix.github.io</code></li>
Browse to the Settings</em> page</li>
Scroll down to GitHub Pages</em></li>
Tick Enforce HTTPS</em></li>
</ul>
Create local Jekyll project</h2>
sudo</span></span> apt install ruby-dev</span>
</span>sudo</span></span> gem install jekyll bundler</span>
</span>jekyll</span></span> new kemitix.github.io</span>
</span>cd</span></span> kemitix.github.io</span>
</span></code></pre>
Configure So Simple theme</h2>
Add plugins to Gemfile</em>:</p>
gem "github-pages", group</span></span>:</span> :jekyll_plugins</span>
</span>
</span>group :jekyll_plugins do</span>
</span>    gem "jekyll-remote-theme"</span>
</span>    gem "jekyll-sitemap"</span>
</span>end</span>
</span></code></pre>
Install plugin:</p>
bundle</span></span> install</span>
</span></code></pre>
Enable plugins and select theme in _config.yml</em>:</p>
plugins</span></span>:</span>
</span>  -</span> jekyll-remote-theme</span>
</span>  -</span> jekyll-sitemap</span>
</span>remote_theme</span></span>:</span> mmistakes/so-simple-theme</span>
</span></code></pre>
Review site locally</h2>
bundle</span></span> exec jekyll serve --</span>incremental</span></span>
</span></code></pre>
Open browser: http://localhost:4000/</code></p>
Publish to github</h2>
git</span></span> init</span>
</span>git</span></span> add .</span>
</span>git</span></span> commit -</span>m</span>"</span>Initial commit"</span></span></span>
</span>git</span></span> push -</span>u</span> origin master</span>
</span></code></pre>

Open browser: https://kemitix.github.io/</code></li>
</ul>
Refs</h2>

http://pages.github.com/</li>
https://jekyllrb.com/docs/installation/</li>
</ul>


gpg: signing failed
2018-02-24T00:00:00+00:00
Trying to setup Jenkins to sign artifacts with the maven-gpg-plugin</code> to allow me to deploy to the Sonatype OSS Nexus. However I was getting the error:</p>
[INFO] --- maven-gpg-plugin:1.6:sign (sign-artifacts) @ parent ---
</span>gpg: signing failed: No such file or directory
</span></code></pre>
Choon-Chern Lim</a> suggests adding --pinentry-mode loopback</code> to fix this.</p>
<</span>configuration</span>></span></span>
</span>    <</span>gpgArguments</span>></span></span>
</span>        <</span>arg</span>></span></span>--pinentry-mode</</span>arg</span>></span></span>
</span>        <</span>arg</span>></span></span>loopback</</span>arg</span>></span></span>
</span>    </</span>gpgArguments</span>></span></span>
</span></</span>configuration</span>></span></span>
</span></code></pre>
Which then brought me to the error:</p>
[INFO] --- maven-gpg-plugin:1.6:sign (sign-artifacts) @ parent ---
</span>gpg: signing failed: Inappropriate ioctl for device
</span></code></pre>
Which led me to actually read the gpg man page</a>.
Where I found the option --batch</code>.</p>

–batch</strong>

–no-batch</strong>

Use batch mode. Never ask, do not allow interactive commands. –no-batch disables this option. Note that even with a filename given on the command line, gpg might still need to read from STDIN (in particular if gpg figures that the input is a detached signature and no data file has been specified). Thus if you do not want to feed data via STDIN, you should connect STDIN to oq/dev/nullcq.</p>
</blockquote>
My next iteration of maven-gpg-plugin</code> config is now:</p>
<</span>plugin</span>></span></span>
</span>    <</span>groupId</span>></span></span>org.apache.maven.plugins</</span>groupId</span>></span></span>
</span>    <</span>artifactId</span>></span></span>maven-gpg-plugin</</span>artifactId</span>></span></span>
</span>    <</span>version</span>></span></span>1.6</</span>version</span>></span></span>
</span>    <</span>executions</span>></span></span>
</span>        <</span>execution</span>></span></span>
</span>            <</span>id</span>></span></span>sign-artifacts</</span>id</span>></span></span>
</span>            <</span>phase</span>></span></span>verify</</span>phase</span>></span></span>
</span>            <</span>goals</span>></span></span>
</span>                <</span>goal</span>></span></span>sign</</span>goal</span>></span></span>
</span>            </</span>goals</span>></span></span>
</span>            <</span>configuration</span>></span></span>
</span>                <</span>gpgArguments</span>></span></span>
</span>                    <</span>arg</span>></span></span>--batch</</span>arg</span>></span></span>
</span>                    <</span>arg</span>></span></span>--pinentry-mode</</span>arg</span>></span></span>
</span>                    <</span>arg</span>></span></span>loopback</</span>arg</span>></span></span>
</span>                </</span>gpgArguments</span>></span></span>
</span>            </</span>configuration</span>></span></span>
</span>        </</span>execution</span>></span></span>
</span>    </</span>executions</span>></span></span>
</span></</span>plugin</span>></span></span>
</span></code></pre>


Template: Jenkinsfile
2018-02-24T00:00:00+00:00

Outdated</strong> - I haven’t used Jenkins for my CI in years.</p>
</blockquote>
Features:</p>

Records all environment variables</li>
Aborts if attempting to build a SNAPSHOT version on either the master</code> branch or a pull request onto the master</code> branch</li>
Perform Static Code Analysis with checkstyle and pmd if there are Java source files - will fail if plugins are not available</li>
Performs a test build/install with Java 9 to validate compatibility</li>
Performs build/install with Java 8 for deploy candidate</li>
Records test results and code coverage in Jenkins if tests were run by surefire/failsafe</li>
Publishes code coverage to Codacy if tests were run by surefire/failsafe - see helpers below</li>
Archives any jar file created</li>
Deploy if on master branch and is attached to a remote git repo (i.e. not a local file:/ repo).</li>
</ul>
final</span> String</span> mvn =</span> "</span>mvn --batch-mode --update-snapshots"</span></span>
</span>
</span>pipeline {</span>
</span></span>    agent any
</span></span>    stages {</span>
</span></span></span>        stage</span>(</span>'</span>Environment'</span></span>)</span></span> {</span>
</span></span></span></span>            steps {</span>
</span></span></span></span></span>                sh '</span>set'</span></span>
</span></span></span></span></span>            }</span></span>
</span></span></span></span>        }</span></span>
</span></span></span>        stage</span>(</span>'</span>no SNAPSHOT in master'</span></span>)</span></span> {</span>
</span></span></span></span>            //</span> checks that the pom version is not a snapshot when the current or target branch is master
</span></span></span></span></span>            when {</span>
</span></span></span></span></span>                expression {</span>
</span></span></span></span></span></span>                    (env.GIT_BRANCH</span> ==</span> '</span>master'</span></span> ||</span> env.CHANGE_TARGET</span> ==</span> '</span>master'</span></span>) &&</span>
</span></span></span></span></span></span>                            (readMavenPom</span>(</span>file</span>:</span> '</span>pom.xml'</span></span>)</span></span>.</span>version).</span>contains</span>(</span>"</span>SNAPSHOT"</span></span>)</span></span>
</span></span></span></span></span></span>                }</span></span>
</span></span></span></span></span>            }</span></span>
</span></span></span></span>            steps {</span>
</span></span></span></span></span>                error</span>(</span>"</span>Build failed because SNAPSHOT version"</span></span>)</span></span>
</span></span></span></span></span>            }</span></span>
</span></span></span></span>        }</span></span>
</span></span></span>        stage</span>(</span>'</span>Static Code Analysis'</span></span>)</span></span> {</span>
</span></span></span></span>            when {</span> expression {</span> findFiles</span>(</span>glob</span>:</span> '</span>**/src/main/java/*.java'</span></span>)</span></span>.</span>length ></span> 0</span> }</span></span> }</span></span>
</span></span></span></span>            steps {</span>
</span></span></span></span></span>                withMaven</span>(</span>maven</span>:</span> '</span>maven 3.5.2'</span></span>,</span> jdk</span>:</span> '</span>JDK 1.8'</span></span>)</span></span> {</span>
</span></span></span></span></span></span>                    sh "</span>${</span>mvn}</span></span> compile checkstyle:checkstyle pmd:pmd"</span></span>
</span></span></span></span></span></span>                }</span></span>
</span></span></span></span></span>                pmd canComputeNew</span>:</span> false</span>, defaultEncoding</span>:</span> '</span>'</span></span>, healthy</span>:</span> '</span>'</span></span>, pattern</span>:</span> '</span>'</span></span>, unHealthy</span>:</span> '</span>'</span></span>
</span></span></span></span></span>            }</span></span>
</span></span></span></span>        }</span></span>
</span></span></span>        stage</span>(</span>'</span>Build Java 9'</span></span>)</span></span> {</span>
</span></span></span></span>            steps {</span>
</span></span></span></span></span>                withMaven</span>(</span>maven</span>:</span> '</span>maven 3.5.2'</span></span>,</span> jdk</span>:</span> '</span>JDK 9'</span></span>)</span></span> {</span>
</span></span></span></span></span></span>                    sh "</span>${</span>mvn}</span></span> clean install"</span></span>
</span></span></span></span></span></span>                }</span></span>
</span></span></span></span></span>            }</span></span>
</span></span></span></span>        }</span></span>
</span></span></span>        stage</span>(</span>'</span>Build Java 8'</span></span>)</span></span> {</span>
</span></span></span></span>            steps {</span>
</span></span></span></span></span>                withMaven</span>(</span>maven</span>:</span> '</span>maven 3.5.2'</span></span>,</span> jdk</span>:</span> '</span>JDK 1.8'</span></span>)</span></span> {</span>
</span></span></span></span></span></span>                    sh "</span>${</span>mvn}</span></span> clean install"</span></span>
</span></span></span></span></span></span>                }</span></span>
</span></span></span></span></span>            }</span></span>
</span></span></span></span>        }</span></span>
</span></span></span>        stage</span>(</span>'</span>Test Results'</span></span>)</span></span> {</span>
</span></span></span></span>            when {</span> expression {</span> findFiles</span>(</span>glob</span>:</span> '</span>**/target/surefire-reports/*.xml'</span></span>)</span></span>.</span>length ></span> 0</span> }</span></span> }</span></span>
</span></span></span></span>            steps {</span>
</span></span></span></span></span>                junit '</span>**/target/surefire-reports/*.xml'</span></span>
</span></span></span></span></span>                jacoco exclusionPattern</span>:</span> '</span>**/*{Test|IT|Main|Application|Immutable}.class'</span></span>
</span></span></span></span></span>                withMaven</span>(</span>maven</span>:</span> '</span>maven 3.5.2'</span></span>,</span> jdk</span>:</span> '</span>JDK 1.8'</span></span>)</span></span> {</span>
</span></span></span></span></span></span>                    sh "</span>${</span>mvn}</span></span> com.gavinmogan:codacy-maven-plugin:coverage "</span></span> +</span>
</span></span></span></span></span></span>                            "</span>-DcoverageReportFile=target/site/jacoco/jacoco.xml "</span></span> +</span>
</span></span></span></span></span></span>                            "</span>-DprojectToken=`$JENKINS_HOME</span>/codacy/token` "</span></span> +</span>
</span></span></span></span></span></span>                            "</span>-DapiToken=`$JENKINS_HOME</span>/codacy/apitoken` "</span></span> +</span>
</span></span></span></span></span></span>                            "</span>-Dcommit=`git rev-parse HEAD`"</span></span>
</span></span></span></span></span></span>                }</span></span>
</span></span></span></span></span>            }</span></span>
</span></span></span></span>        }</span></span>
</span></span></span>        stage</span>(</span>'</span>Archiving'</span></span>)</span></span> {</span>
</span></span></span></span>            when {</span> expression {</span> findFiles</span>(</span>glob</span>:</span> '</span>**/target/*.jar'</span></span>)</span></span>.</span>length ></span> 0</span> }</span></span> }</span></span>
</span></span></span></span>            steps {</span>
</span></span></span></span></span>                archiveArtifacts '</span>**/target/*.jar'</span></span>
</span></span></span></span></span>            }</span></span>
</span></span></span></span>        }</span></span>
</span></span></span>        stage</span>(</span>'</span>Deploy'</span></span>)</span></span> {</span>
</span></span></span></span>            when {</span> expression {</span> (env.GIT_BRANCH</span> ==</span> '</span>master'</span></span> &&</span> env.GIT_URL</span>.</span>startsWith</span>(</span>'</span>https://'</span></span>)</span></span>) }</span></span> }</span></span>
</span></span></span></span>            steps {</span>
</span></span></span></span></span>                withMaven</span>(</span>maven</span>:</span> '</span>maven 3.5.2'</span></span>,</span> jdk</span>:</span> '</span>JDK 1.8'</span></span>)</span></span> {</span>
</span></span></span></span></span></span>                    sh "</span>${</span>mvn}</span></span> deploy --activate-profiles release -DskipTests=true"</span></span>
</span></span></span></span></span></span>                }</span></span>
</span></span></span></span></span>            }</span></span>
</span></span></span></span>        }</span></span>
</span></span></span>    }</span></span>
</span></span>}</span></span>
</span></code></pre>
Codacy</h2>
Publishing to Codacy requires having a secret API Token and a Repo Token. Neither of these should be added to the Jenkinsfile. Instead the two scripts here are interpolated into the maven command line that invokes the codacy-maven-plugin</code> in the Test Results</em> stage.</p>
Requirements: xpath</code></p>

sudo apt install libxml-xpath-perl</code></p>
</blockquote>
token</h3>
#</span></span>!/usr/bin/env bash</span>
</span></span>
</span>TOKENS_FILE</span>=</span></span>`</span>dirname</span></span> $</span>0</span></span></span>`</span></span>/tokens</span>
</span>ARTIFACT_ID</span>=</span></span>`</span>xpath</span></span> -</span>q</span> -</span>e</span> "</span>/project/artifactId/text()"</span></span> <</span> pom.xml</span>`</span></span></span>
</span>
</span>grep</span></span> "</span>^$</span>{</span></span>ARTIFACT_ID</span></span>}</span></span>="</span></span> $</span>{</span></span>TOKENS_FILE</span></span>}</span></span></span>|</span>cut</span></span> -</span>d</span>=</span> -</span>f2</span></span>
</span></code></pre>
tokens</h3>
Provides the repo token by searching for it in the text file tokens</code>. e.g.</p>
repo1=repo1key
</span>repo2=repo2key
</span></code></pre>
These would match against the artifactId</code> of the root module pom.xml</code>.</p>
apitoken</h3>
Simply echos the apikey to STDOUT.</p>
#</span></span>!/usr/bin/env bash</span>
</span></span>
</span>echo</span></span> "</span>xxxXXXxxxXXXxxxXXXxxx"</span></span></span>
</span></code></pre>


Navel Gazing Meetings
2005-04-15T00:00:00+00:00
I have to give up an afternoon’s perfectly good working time, where in I could actually get some good solid work done. The peace and quite that is available to juts get my head down, without the constant interuptions for help and advice from other members of staff. Most of whom take Friday afternoons of. Which usually means I can get some work done.</p>
Not this week. My manager is pulling the whole department into a planning meeting. The third in a series. It’ll take almost all of the afternoon and all we will have decided is how to grade ourselves based on a recent survey that was sent out staff in other departments. The navel gazing as we pour over a similar survey we all had to fill out on how we felt we were doing our job, together with the interrogation of anyone who felt we were anything less than ‘great’ and any area.</p>
Leave me alone and let me do something Productive for Entropy’s sake!</p>


Creative Archive Licence
2005-04-13T00:00:00+00:00
[edit: 2025-01-06 The BBC Creative Archive pilot ended in 2006. more info</a>]</p>
The BBC are leading an initiative that is almost Creative Commons, but has some extra restrictions.
The most significant is that you have to be within the UK, at least most of the time, to use it.
(The Rules in Brief</a>)
It is still in a consultation stage, having only been launched today.
Just now they are looking for feedback on what areas are of interest, so they can prioritise making that content available.</p>
And this taken direct from their FAQ</a>:</p>

Will I be able to get old Dr Who episodes?</strong></p>
Sadly you will not be able to get old Dr Who episodes in the early days of the Creative Archive.
However, it is hoped that the BBC will in time be able to expand its contribution to the
Creative Archive to include programmes from all genres.</p>
</blockquote>
and, rebuting some of the FUD:</p>

I heard that Creative Archive will be using P2P technology. Isn’t that illegal?</strong></p>
Peer to Peer (P2P) distribution technologies are certainly one of the technologies that will make it possible for large
files to be sustainably distributed online and therefore of great interest to the Creative Archive.
The good news is that the technology is not illegal. When used for the distribution of legally licensed material,
it provides an efficient and cost effective means of sharing and distributing content on the internet.</p>
</blockquote>


UK Election: Who Would I Vote For?
2005-04-12T00:00:00+00:00
I heard last week that there is to be a General Election here in the UK. Being somewhat oblivious to most maters of state that aren’t technology related I found this quite interesting. Some years ago I did look through all the various pollicies of all the available parties and weighed their pros and cons, settling on the LibDems as being the closest to my own beliefs in most areas.</p>
In any party political electoral system most people end up having to pick a compromise in selecting the party the most reflects their views. In that there are sometime the odd bitter pill that must be swallowed. For some people a single issue will decide their voting. Generating an almost knee-jerk response amongst the electorate that allows itself to become blinded to all the other issues in their reaction to a single concern. Whether it be Pro-Euro vs Pro-Sterling, or to take an American hot button issue, Pro-Life vs Pro-Choice.</p>
As far as the Euro is concerned, I’m largely unconcerned about whether we join Euroland or not. Money is money. I don’t see it making such a significant difference which currency we have. I’m not going to say I entirely understand all the implications of it, that’s what economists are for. My inclination is more towards it than not, however.</p>
When it comes to the abortion issue, I am very strongly in favour of the mother having a complete free choice in the matter. Don’t get me started on the Pro-Lifers gunning down doctors in the street. I personally am the result of an unplanned pregnancy, and I can see the trouble that it brought my mother having had her life so completely forced in a particular direction. Not that either of us wish it hadn’t happened, but perhaps at a later time in my mother’s life and with a father who was prepared to stand by that responsibility. Not that I would have become the person I am. At a certain level I’d rather not be who I am, but I am and I have to keep working it out from here on in.</p>
Getting back to the UK Election. Having looked at the parties policies some years ago, I don’t follow politics any more. It’s a confused mess of media-hype, compromise and back-room double-crossing and double-dealing. You don’t look at what goes into making a sausage unless you absolutely have to, and the same applies to government. The only person who deserves power is the only person who doesn’t want it. Which leaves us in a pretty pickle.</p>
Below is the results of an interesting tool, which I’m glad to see confirms that the LibDems still prodominantly have the policies that I agree with. So the drift through life that my moral structure had sailed under the last few years, is still firmly within LibDem waters.</p>

Who should I vote for?</p>
Your expected outcome</strong>: Liberal Democrat</p>
Your actual outcome</strong>:</p>
Labour 8
Conservative -21

Liberal Democrat 50
UK Independence Party 8
Green 21</p>
You should vote</strong>: Liberal Democrat</p>
The LibDems</a> take a strong stand against tax cuts and a strong one in favour of public services: they would make long-term residential care for the elderly free across the UK, and scrap university tuition fees. They are in favour of a ban on smoking in public places, but would relax laws on cannabis. They propose to change vehicle taxation to be based on usage rather than ownership.</p>
</blockquote>
Take the test at Who Should You Vote For</a></p>


Cacao Epic
2005-04-11T00:00:00+00:00
I’m trying some of Green & Blacks Organic Hot Chocolate. Nice. Should help me get to sleep tonight. Hot milk with chocolate. It just prompted me to ask Google Suggest where it originally</a> came from. My first impulse was that it entered the Western culture when Grijalva’s expedition arrived at Vera Cruz, and followed a year later in 1519 by Cortes. But Grijalva’s expedition either didn’t notice it, or missed it when they were there.</p>
I was reading recently the tale of the Conquistadors from Bernal Diaz’ first hand account. All the time as I was reading it, I wanted more detail. To read it as a full blown epic novel or watch it as a film. While he went into detail over a lot of what happened, I still felt that it could be turned into a fantastic novel or screenplay. Maybe it already has been done, and I just don’t know about it. Certainly it was a very entertaining read.</p>


nasa lacks confidence
2005-04-10T00:00:00+00:00
from technewsworld</a> (article long since removed):</p>

recent events have underscored the need for a rescue option. NASA officials said last month that they aren’t yet confident in new methods for fixing a shuttle in orbit. Without viable repair methods, the only way to rescue a marooned crew would be through a second shuttle.</p>
</blockquote>
NASA are preparing the shuttle Atlantis to be able to launch within a month of Discovery’s scheduled launch. It would be sent up to rescue the seven crew should something happen to Discovery that made mission control fear another Columbia-like incident. This would this require NASA technicians working round the clock to get Atlantis ready, possibly even putting aside several safety protocols devised after Columbia. While all this was happening Discovery’s crew would have to quietly wait out their time on board the International Space Station. Assuming of course that they can reach it.</p>
The final decision to send Atlantis would fall to the newly appointed NASA Administrator. Was this what he signed up for, I wonder? They would destroy Discovery in the process, ditching in the Pacific. That part I’m slightly confused about. A $3 billion piece of equipment. Admittedly it must have it’s faults, or else there would be no need for a rescue mission. These things can pretty much fly themselves. If there are fears that it might break up over a population center then why no put it into a parking orbit and go fix it on another shuttle mission.</p>
They talk about how the shuttle program would not be maintainable with only the two shuttles that would be left. That the ISS needs a working shuttle program to even finish being built. That the cost of a shuttle mission is in the region of $400 million. Surely a single mission specifically to fix a broken shuttle has to be cost effective against the cost of losing the shuttle altogether.</p>
Because to lose yet another shuttle, would mean that three of the original five shuttles, had been lost. Sadly they have done so taking all hands with them. More importantly for the future of the NASA’s manned missions it would virtually cripple the effort. Congress have lots of ideas of where they could spend all those billions of dollars that would keep the shuttle program alive. No, I don’t think it would be on a new vehicle.</p>
At least the US aren’t the only nation that is pushing men into space. The Chinese</a> are pushing ahead. As are the Japanese</a>. Russia has been doing it for even longer than the U.S. and here’s hoping that the aftermath from the X-Prize</a> and other commercial</a> innovators either get us up there or embarrass the U.S. into at least trying to keep up.</p>
After all, it took competition from the Russian space program to get Kennedy to reach for the Moon in the 60’s. For now we have to wait until Bush has got his head out of the sand. Or replaced with a visionary, which will probably be the sooner of the two options.</p>


Welcome to Nullodium
2005-04-10T00:00:00+00:00
So we have the first post here on Nullodium. Not that anyone will be around to read this first post as it is posted. If you are, gee, you’re weird. But welcome anyway.</p>
What can you expect to find here?</p>
To be honest, I’m not entirely sure yet. This isn’t my first blog. Nor my second. It is however, my first annonymous blog. So I can let it all hang out, and say what I want and mean without fear or risk of embarassment. Some technology, and media posts no doubt, as well as the odd post about current, or perhaps not so current events.</p>
In short, probably not much to write home about. But I hope you’ll stick around anyway.</p>

kemitix.net

tree-type: Type-Safe Filesystem Paths for Rust

The Problem: String-Based Paths Are Error-Prone</h2> When working with filesystem paths in code, we typically use strings:</p>

The Solution: tree-type</h2> Define your project structure once, get type safety everywhere:</p>

Feature: Type-Safe Navigation with parent()</h2> The Problem with Generic Paths:</strong></p>

Recipe: 30 Minute Vegetarian Puerto Rican Style Picadillo

30 Minute Vegetarian Puerto Rican Style Picadillo</h1> A delicious meat-free version of the classic Puerto Rican dish with bold, savory flavors.</p> Servings:</strong> 2 Time:</strong> 30 minutes Difficulty:</strong> Easy</p>

Serving Suggestions</h2> Serve with white rice, sweet plantains, crusty bread, or use as a filling for tacos or burritos.</p>

Development Environment Updates - May 8-25, 2025

Maintenance & Bug Fixes</h2> Resolved path issues with Kanata keyboard daemon</li> Patched Ollama API connection settings</li> Improved cross-platform compatibility checks</li> </ul>

Project Release: git-next v2025.5.1

PR: Mirror Branch Filter for Forgejo

Project Update: git-next

Migration Cleanup Completed</h2> The .dev/TODO.md</code> file has been significantly simplified after completing the migration from gix</code> to git2</code> in the gitlib</code> module. The detailed migration checklist has been replaced with a fresh TODO template for the next project.</p>

Blog theme: tabi

Project Release: git-next v2025.5.0

Key Changes:</h2> Backend Migration:</h3> Completed migration from gix</a> to git2</a> for Git operations</li> Implemented git2-based functionality for: Rebase operations (with new tests)</li> Remote operations (fetch/push)</li> Branch switching</li> File reading</li> Clone operations</li>

Backend Migration:</h3> Completed migration from gix</a> to git2</a> for Git operations</li> Implemented git2-based functionality for: Rebase operations (with new tests)</li> Remote operations (fetch/push)</li> Branch switching</li> File reading</li> Clone operations</li>

Cargo Features:</h3> Moved email and desktop notifications into two new non-default features: notify-desktop</code></li> notify-email</code></li> </ul> </li>

Development Environment Updates - May 6-7, 2025

New Git History Summary Tool</h2> Added a new bash script my-git-history-summary</code> that:</p>

New Zola (with Facebook)

Facebook Archive</del></h3> photos</del></li> videos</del></li> posts</del></li> more photos</del></li> check ins</del></li> </ul>

Mastodon Comments

Blocked From My Own Mailserver

Create Custom Garmin Activity Type

Developer AI: phind.com

Fixing missing Ubuntu dependencies installing Python with RTX

Bootstrap Ubuntu

Prequisites</h2> Ubuntu 22.04 (I use Ubuntu MATE)</li>

Remote Access</h2> Pin the IP address</li> Install an SSH server</li> Add my public key</li>

Install rust, rtx and yadm</h3> Run the following commands on the server.</p>

Thread Macros in Rust

Astronvim Mason Dap Error

MATE Menu Tray is missing

New site: kemitix.net

ACPI BIOS Error (bug): Could not resolve symbol [ TZ.ETMD], AE_NOT_FOUND (20220313/psargs-330)

Make grub menu readable on high-resolution screens

Disable Ubuntu's DNS stub resolver

Building this site in Github Pages

Refs</h2> http://pages.github.com/</li> https://jekyllrb.com/docs/installation/</li> </ul>

gpg: signing failed

Template: Jenkinsfile

apitoken</h3> Simply echos the apikey to STDOUT.</p> #</span></span>!/usr/bin/env bash</span> </span></span> </span>echo</span></span> "</span>xxxXXXxxxXXXxxxXXXxxx"</span></span></span> </span></code></pre>

Navel Gazing Meetings

Creative Archive Licence

UK Election: Who Would I Vote For?

Cacao Epic

nasa lacks confidence

Welcome to Nullodium

30 Minute Vegetarian Puerto Rican Style Picadillo</h1>
A delicious meat-free version of the classic Puerto Rican dish with bold, savory flavors.</p>
Servings:</strong> 2
Time:</strong> 30 minutes
Difficulty:</strong> Easy</p>

Serving Suggestions</h2>
Serve with white rice, sweet plantains, crusty bread, or use as a filling for tacos or burritos.</p>

Maintenance & Bug Fixes</h2>

Resolved path issues with Kanata keyboard daemon</li>
Patched Ollama API connection settings</li>
Improved cross-platform compatibility checks</li> </ul>

Migration Cleanup Completed</h2>
The `.dev/TODO.md</code> file has been significantly simplified after completing the migration from gix</code> to git2</code> in the gitlib</code> module. The detailed migration checklist has been replaced with a fresh TODO template for the next project.</p>`

Facebook Archive</del></h3>

photos</del></li>
videos</del></li>
posts</del></li>
more photos</del></li>
check ins</del></li> </ul>

Install rust, rtx and yadm</h3>
Run the following commands on the server.</p>

Refs</h2>

http://pages.github.com/</li>
https://jekyllrb.com/docs/installation/</li> </ul>

`apitoken</h3> Simply echos the apikey to STDOUT.</p> #</span></span>!/usr/bin/env bash</span> </span></span> </span>echo</span></span> "</span>xxxXXXxxxXXXxxxXXXxxx"</span></span></span> </span></code></pre>`