Skip to content

TrackDown - Issue Tracking with plain Markdown. If you are missing the "git clone" for your tickets from github.com, codeberg.org, or gitlab.com, then this is for you. A lightweight Ticketing System for distributed and unconnected small Teams.

License

Notifications You must be signed in to change notification settings

mgoellnitz/trackdown

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

TrackDown

Latest Release License Build Download

Issue Tracking with plain Markdown for GIT and Mercurial.

In short: You are missing the git clone or hg clone respectively for your tickets from Codeberg, GitLab, GitHub, or some other services (see below) where we already have this for code and wiki?

You need issue tracking which works for distributed and potentially disconnected situations together with your distributed version control GIT or Mercurial and e.g. also your distributed wiki editing through GIT or Mercurial as well?

Then this here is for you!

Support this project: ko-fi

It is not intended for large, permanently online or connected teams and heavy flows of tickets though, since you will be having only one file with plain Markdown with your issues - and optionally other stuff - collected in it.

The currently open issues of TrackDown itself can be found here, here, and here.

The corresponding roadmap is placed here, here, and here.

Feedback

The primary home of this project for contribution and feedback lives at Codeberg with additional mirrors at GitLab and GitHub.

Design

While TrackDown does not define an issue related workflow, it has some intended workflow elements which are supported:

The issues are defined and maintained in a single Markdown file following the format given here.

The GIT post-commit hook or Mercurial commit hook of TrackDown reads the commit messages and modifies that issue collection if your commit messages relate to some of the issues.

Additionally, a roadmap file is automatically maintained for your tickets. This roadmap file groups the issue's headlines in groups according to their version label and illustrated progress counting issues in progress and resolved issues.

The issue collection this way is held local on your machine and not remote in the database of a tracking system. (Which is something also Fossil supports.) Like with the source code, it is pushed to remote repositories if needed (or possible). The simple Markdown format and the usage of GIT or Mercurial as a backend support distributed, shared editing and later merging of the issues and the related notes in the issue collection. (This is where the parallel with Fossil ends).

The Format

While sticking to only partly structured Markdown the following elements should be maintainable with TrackDown:

  • ID
  • Title
  • Status
  • Commits
  • Target Version
  • Severity
  • Affected Versions
  • Description
  • Comments

These fields are mapped to the following source structure

  ## ID Title (status)

  *Target Version (optional)* - Currently assigned to: `me` (optional)

  ### severity (optional) priority (optional)

  affected versions: 1.0, 1.1 (optional)

  ### Description (optional)

  description

  ### Comments (optional)

  comments (structured)

  ### Commits (auto generated)

  The headline commits at level three is optional. The commit messages are
  inserted just as the last part of the issue's level two text area.

The really fixed non-optional parts of this are

  ## ID Title (status)

  (Commit messages inserted here before the next ticket)

Field Values

ID

Any combination of (English) upper- and lower-case letters and digits.

Title

Any expressible in Markdown.

Status

Anything expressible in Markdown. Automatically set values are "in progress" if you start committing for a certain ID and "resolved", if you are using a prefix of "fixes ID" or "resolves ID".

Other intended values include "new", where the issue is just files, and "closed" when the solution is brought into production.

Target Version

Only digits, letters and dots. No spaces allowed.

Description

Anything expressible in Markdown.

Affected Versions

Anything expressible in Markdown. Is expected to describe which version are affected by the issue (if this is possible to say).

Comments

Anything expressible in Markdown.

Severity

Anything expressible in Markdown.

Setup

There are two ways to set up TrackDown: Have the issues file integrated in your source code repository, or place it in an arbitrary place of your choosing.

The first - default - way is to use it in a separate branch of your source code repository. It is kept visible and editable through a symbolic link at the root level of the source code repository. Of course, this file is touched automatically via commits to your source code through the (post-)commit hook of TrackDown.

The second way is to use the file at a different location - e.g. in the wiki of the project instead of the source code repository, which is described later.

In both cases, the automatically maintained roadmap file resides next to the issue collection file.

Initialize the Repository

If you want to track the issues in a TrackDown branch of your source code repository and not in any other location of your choosing, you need to modify the GIT or Mercurial repository accordingly. Your source code repository must contain at least one commit for this to work.

To initialize your source code repository this way, call the script

trackdown.sh init

This creates the TrackDown branch for the issue tracking. For GIT repositories, you have to manually propagate this thread to your upstream repositories.

git push origin trackdown

TrackDown does not interfere with your remote workflow for any version control system: Also for Mercurial the trackdown branch will only show up in the remote repositories if you push it.

hg push

Initialization must only be executed once for a repository including all of its forks and clones.

If you want to use the issue collection file from a different location than the special TrackDown branch, leave out this step.

Repository Integration

Regardless of the location of the issue collection file, for each clone of the repository you have to set up the TrackDown tooling to be able to use it integrated with your source code commits.

To start using TrackDown for the respective clone, you have to issue

trackdown.sh use

when using the TrackDown branch in the source code repository or

trackdown.sh use <path/to/issues.md>

like in

trackdown.sh use ../wiki/issues.md

when using TrackDown with the issue collection file at a different location. Automatic commit and push (see below) will be switched off in the latter case.

This creates (git or hg ignored) links issues.md and roadmap.md in the root directory of your project pointing to the issue collection file and the roadmap. Additionally, it will configure a post-commit hook for GIT or a commit hook for Mercurial respectively.

After this step, you can edit the issue collection file following the format mentioned above.

Commands in the Commit Messages

TrackDown is supposed to read the commit messages when not used as a plain mirror and interpret the contents as potential commands for the modification of alongside you work.

When using GIT, TrackDown relies on an implementation, which is capable of executing the script hooks, which is - as opposed to Mercurial - not the case for all implementations.

Right now, TrackDown understands only two commands in the commit messages.

refs #id[,id...]

Reference the commit in the list of commits at the end of the issue text.

git commit -m "refs #MYID - comment" files...

This command changes the state to "in progress" from anything like new, nothing, or even resolved. If the commit relates to more than one issue, the issues can be separated by commas.

git commit -m "fixes #ONEID,ANOTHERID - comment" files...
(Future work: lifts the issue up to the top of the list)

resolves|resolve|fixes #id[,id...]

Reference the commit in the list of commits at the end of the issue text.

git commit -m "fixes #MYID - comment" files...

This command changes the state to "resolved" from anything like new, nothing, or in progress. If the commit relates to more than one issue, the issues can be separated by commas.

git commit -m "fixes #ONEID,ANOTHERID - comment" files...
(Future work: moves the issue to the top the part of the list where the resolved issues reside)

Command Line Tools

In addition to the init and integration tools, the following commands are available:

Roadmap

Provided that the issues in the issue collection file are marked with version labels like suggested, the command

trackdown.sh roadmap

prints out a complete roadmap of the project sorted by "target versions" in Markdown format.

The term "target version" could also be read as "release" or "sprint" or anything which describes your development process best.

List

The command ls is used to show all issues marked for a given "target version" like in

trackdown.sh ls 1.1

where all issues intended to be completed in "target version" 1.1 are listed.

The term "target version" could also be read as "release" or "sprint" or anything which describes your development process best.

My Tickets

The command

trackdown.sh mine

lists all issues in the issue collection, which are marked with a

*Version 1.0* - Currently assigned to: `me`

The me placeholder in the case is taken - in that order - from

  • the first parameter on the command line
  • The me entry in the .trackdown/config file
  • The local user name from the environment variable $USER

Optionally, you can add a path to an issue collection file as an additional parameter like in

trackdown.sh mine ../wiki/issues.md

or

trackdown.sh mine UserName ../wiki/issues.md

Show Issue Collection Changes Status

To show the current state of the local editing of the issue collection and roadmap file, even if they reside in a special TrackDown branch and are only used as symbolic links in the source code repository, a shortcut command is available, giving a brief summary of the Mercurial or GIT state if the issue collection and roadmap file.

trackdown.sh status

Quick Sync of Issue Collection and Roadmap

Most times the editing changelog of the issue collection file and roadmap file don't present too much additional information for which is already held in the commit messages of the source code and the issue collection file itself.

In such situations, you can use the shortcut command sync to bring the issue collection and roadmap file on your machine and the remote repository in sync.

trackdown.sh sync

Copy Milestone/Release Contents

The command copy is used to extract the issues related to a given milestone, release, version, or whatever your terminology might be to a separate file named after the given parameter. So

trackdown.sh copy 1.1

copies all notes for the issues marked with "1.1" as a version marker to a separate file 1.1.md to obtain release notes and get the resolved issues from the base issue collection file for your current work.

Issues

The command

trackdown.sh issues

lists all potential issues in the issue collection. Potential means in this case, that there may be some false positives if there are additional elements in your issue collection file, which might be interpreted as issues.

Optionally, you can add a path to an issue collection file as a parameter, like in

trackdown.sh issues ../wiki/issues.md

Configuration

The source repository contains a directory named .trackdown/.

This directory contains a file named config. There are some options in this file, which you might want to change.

Example config file for TrackDown:

  autocommit=true
  autopush=false
  location=../wiki/issues.md
  prefix=https://github.com/user/project/commit/
  me=My Name

Auto Commit all Issue Collection Changes

Automatically commits the new change to the TrackDown branch. If you didn't change the default location where your source code repository contains the a TrackDown branch, you will want to leave the unchanged with the default value true.

In other scenarios, you may switch it to false.

Auto Push all Issue Collection Commits

Automatically pushes after each cs to the upstream repository. If you didn't change the default locations, where your source code repository actually is the upstream repository of your issue collection, you will want to leave this unchanged with the default value of true.

In other scenarios, you may switch it to false. E.g. if the issue collection is part of your project wiki, then automatic pushing might lead to remote operations, which is not desirable.

Online commit summary prefix

With some GIT backends, it is possible to obtain a summary with changes and commit message online for every commit. To use this facility, place a prefix in the config file, where the hash of a commit can be appended, to obtain a valid link for that very commit.

If TrackDown discovers common GIT services, it tries to automatically discover the correct prefix for URLs pointing to single commits.

Username for assignments

To be able to assign tickets to users, and thus to know who you are, the user name as used in the issue collection file can be added here. This makes it possible e.g. to list issues assigned to the current user.

Tickets won't be automatically be assigned when adding commits for an issue. Just the progress flag will be set.

Installation

When building TrackDown locally, preliminary DEB and RPM packages will be created. But when downloading one of the releases or a snapshot, copy the files from bin/ to a place on your $PATH for now. For some functions - especially in the area of issue tracker mirror - jq needs to be installed.

Of course, this way the remaining Windows users are locked out.

A symbolic link td to the trackdown.sh script is recommended for easier use.

Compatibility

TrackDown is tested to work with Debian 11 and newer. It is expected to work on similar Linux systems and MacOS systems. Also [NetBeans][netbeans] internal GIT implementation should work.

There are no plans to support Windows systems directly.

Prerequisites

TrackDown relies on a GIT or Mercurial installation available on the path when used with distributed version control as the backend. The mirror feature in turn heavily relies on the installation of jq available through your path.

Related Projects

I only came across related projects which have certain limitations or are unmaintained. In each case, the limitations have an extent that kept me from using these systems except for very small or test projects.

Fossil SCM

What I liked about fossil is, that it brings the three core elements of development

  • Source Code
  • Documentation or Notes (Wiki)
  • Issues

local to my machine for distributed development or disconnected situations.

You don't have to maintain backups since the remote instances are your backups of the source code, wiki, and ticketing state.

It does not have a wiki capable of shared editing with later merging, like the GIT based wikis of GitLab, GitHub, Forgejo, or Bitbucket.

Also, it is not possible to the contents of the wiki outside the Fossil context e.g. for a documentation website, since you cannot export the wikis raw data. (Yes, Fossil provides means to use the wiki directly as a documentation site system, which is similar but not exactly the same.)

The drawback is, that it does all these things by creating a nearly closed shop system not open to re-use of these elements and not open to external tooling outside the Fossil scripting facility.

Additionally, I have to keep the Fossil internal web server running for each repository I am using, to be able to read the notes and issues for a project.

Also, there is only poor IDE support for Fossil right now, with the exception of support for Idea and my own small plug-in for NetBeans mirrored here.

Forgejo (Gitea and Gogs)

Originally coming from the root of Gitea and Gogs, Forgejo has been forked and extended. Intended for on premises use as a GIT based solution for Code, Wiki and CI together with an issue tracking section, it is also available in the form of the public online incarnation Codeberg.

Of course Forgejo can be used as a TrackDown storage backend or mirroring source.

We also expect the related Gitea and Gogs system to be still fully usable in the same way.

GitHub

GitHub is the obvious solution used in so many GIT powered projects together with a GIT based wiki (as opposed to Bitbucket and GitLab the Wiki is a flat folder - be warned) and many other useful details.

The only thing I'm missing is the distributed offline work for ticketing.

So in this case it is possible to leave out the ticketing of GitHub and use TrackDown with GitHub as the GIT based storage backend. And this is exactly what TrackDown was designed for.

As an alternative, you can at least mirror the issues from GitHub to have your notes with you and at least know the issue IDs for offline code commits. Or you can use the mirroring steps for migration purposes.

GitLab

GitLab not only is a good online solution but also is a piece of on premises software (like Bitbucket for the renamed git-Part - not hg.). It's wiki is also GIT based wiki, and it comes with a wealth of other integrations and useful tools and details.

The only thing I'm missing is the distributed offline work for ticketing.

So in this case it is possible to leave out the ticketing of GitLab and use TrackDown with GitLab as the GIT based storage backend. And this is exactly what TrackDown was designed for.

As an alternative, you can at least mirror the issues from GitLab to have the notes with you and now the issue IDs for offline code commits.

Bitbucket

Bitbucket.org is a decent tool for open source or small projects. It has decent VCS solutions, a wiki which can be used distributed through GIT. In the past they were a brilliant backend for TrackDown with both VCS solutions, since they also provided support for Mercurial, which was abandoned in mid 2020.

The only thing I'm missing is the distributed offline work for ticketing.

So in this case it is possible to leave out the ticketing of Bitbucket and use TrackDown with Bitbucket as the GIT based storage backend. And this is exactly what TrackDown was designed for. For migration purposes, or if the limited issue tracking within bitbucket.org is sufficient, the mirroring feature might come in handy.

Atlassian themselves recommends using Jira.

Trac

A few years ago, a colleague stated that he is running a local VM for each project, he is involved with, to take notes, track issues, and maintain source code.

Of course, this does not imply shared use of the Trac service or disconnected use.

Also, while Trac is a brilliant tool, this leaves me with the necessity to maintain the locally running instances and take backups of them, in addition to the project VCS and source code repositories. This is not the case for the GIT based solutions in this list, which have a remote repository as a backup wiki and source code.

Other VCS Services

A small list of tested backends for TrackDown which don't support any kind of issue tracking but Code and Wiki with the VCS:

MDWiki

Unlike Blogdown where you again start a server - but this time on localhost, MDWiki just runs in your browser to view Markdown files nicely formatted locally.

file:///home/me/somewhere/thats/green/repo/wiki.html#!issues.md
file:///home/me/somewhere/thats/green/repo/wiki.html#!roadmap.md

The output of TrackDown looks pretty usable in this setup and gives a good overview of the issues as the roadmap.

When you also use GitLab, GitHub, or Bitbucket Wikis, MDWiki has a different understanding, how links should be interpreted. To get a fully compatible local and remote viewing setup for these cases, a patched version of MDWiki exists on GitHub.

Unmaintained related Projects

These seem to address similar issues, but are not under active development

Migration and Offline Mirroring

To facilitate the use of TrackDown, the option of migrating an existing base of tickets is helpful, of course. The choice, which systems are taken as a data source for such a migration is driven by personal needs.

GitHub Offline Mirror

For disconnected situations which TrackDown is supposed to support, it is possible to connect a workspace to its GitHub issue tracker and mirror tickets for offline use.

The mirror - of course - is not intended for changing the issues in the issue collection file. State changes will most likely be triggered on GitHub by your commit messages or manually, after which a call of the mirroring script can be helpful.

Instead of trackdown.sh use issue trackdown.sh github to set up the mirror connection.

trackdown.sh github <projectname> <owner> <apitoken>

Afterwards, anytime you can connect to the GitHub system, collect the current mirror state to your local issue collection file and the roadmap.

trackdown.sh mirror

Additionally - since you now are on your command line and perhaps don't want to switch windows every second - there is a remote command to issue commands on the remote mirroring source system.

You have to provide the issue-id and the id of the user, which is also always exported to the issue collection file to facilitate this.

The commands available are

  • assign to assign issues to users
trackdown.sh remote assign 68 XYZ
Assigning 68 to user XYZ
  • comment to comment issues
trackdown.sh remote comment 68 "Just a comment"
Adding comment "Just a comment" to issue 68

GitLab Offline Mirror

For disconnected situations which TrackDown is supposed to support, it is possible to connect a workspace to its GitLab issue tracker and mirror tickets for offline use.

The mirror - of course - is not intended for changing the issues in the issue collection file. State changes will most likely be triggered on GitLab by your commit messages or manually, after which a call of the mirroring script can be helpful.

Instead of trackdown.sh use issue trackdown.sh gitlab to set up the mirror connection.

trackdown.sh gitlab <apitoken> <projectname> [https://<gitlab.host>]

If you omit the URL prefix, https://gitlab.com is used. The project name must be given without any group or user addition.

Afterwards, anytime you can connect to the GitLab system, collect the current mirror state to your local issue collection file and the roadmap.

trackdown.sh mirror

Additionally, since you now are on your command line and perhaps don't want to switch windows every second, there is a remote command to issue commands on the remote mirroring source system.

trackdown.sh remote assign 68 XYZ
Assigning 68 to user XYZ

You have to provide the issue-id and the id of the user, which is also always exported to the issue collection file to facilitate this.

The commands available are

  • assign to assign issues to users
trackdown.sh remote assign 68 XYZ
Assigning 68 to user XYZ
  • comment to comment issues
trackdown.sh remote comment 68 "Just a comment"
Adding comment "Just a comment" to issue 68
  • assign to assign issues to users

  • milestone to put issues into milestones

Gitea Offline Mirror

For disconnected situations which TrackDown is supposed to support, it is possible to connect a workspace to its Gitea issue tracker and mirror tickets for offline use.

Setup parameters default to values from the Git repository your current local directory points to.

The mirror - of course - is not intended for changing the issues in the issue collection file. State changes will most likely be triggered on the Gitea instance in use by your commit messages or manually, after which a call of the mirroring script can be helpful.

Instead of trackdown.sh use issue trackdown.sh gitea to set up the mirror connection.

trackdown.sh gitea <apitoken> <projectname> [https://<gitea.host>]

If you omit the URL prefix and no values can be derived from your current working directory, https://codeberg.org is used.

Afterwards, anytime you can connect to the Gitea system, collect the current mirror state to your local issue collection file and the roadmap.

trackdown.sh mirror

Additionally - since you now are on your command line and perhaps don't want to switch windows every second - there is a remote command to issue commands on the remote mirroring source system.

You have to provide the issue id and the id of the user, which is also always exported to the issue collection file to facilitate this.

The commands available are

  • assign to assign issues to users
trackdown.sh remote assign 68 XYZ
Assigning 68 to user XYZ
  • comment to comment issues
trackdown.sh remote comment 68 "Just a comment"
Adding comment "Just a comment" to issue 68

It is expected that this also works for Gogs backends as well.

Bitbucket.org Offline Mirror

For disconnected situations which TrackDown is supposed to support, it is possible to connect a workspace to its Bitbucket.org issue tracker and mirror tickets for offline use.

Some of my stalled projects reside there, and I already did an export of the issue tracker contents, which is what Bitbucket.org supports.

Offline mirror capabilities are now added to this tool for smother migration away from the proprietary issue tracker.

The mirror again is not intended for changing the issues in the issue collection file. State changes will most likely be triggered on Bitbucket.org by your commit messages or manually, after which a call of the mirroring script can be helpful.

Instead of trackdown.sh use issue trackdown.sh github to set up the mirror connection.

trackdown.sh bitbucket <projectname> <owner> <app-password>

Afterwards, anytime you can connect to the Bitbucket.org system, collect the current mirror state to your local issue collection file and the roadmap.

trackdown.sh mirror

In the case of Bitbucket.org, the mirror script has to ask for you password on Bitbucket.org, if you leave out the app password. App passwords can be generated in the personal Bitbucket.org settings.

Additionally - since you now are on your command line and perhaps don't want to switch windows every second - there is a remote command to issue commands on the remote mirroring source system.

The commands available are

  • assign to assign issues to users
trackdown.sh remote assign 68 XYZ
Assigning 68 to user XYZ
  • comment to comment issues
trackdown.sh remote comment 68 "Just a comment"
Adding comment "Just a comment" to issue 68

Redmine

For historical reasons, my Tangram project used Redmine some time ago and customers also used Redmine. So there are two scenarios where some interfacing would be helpful.

In addition, the roadmap outline of TrackDown is very much inspired by the Redmine roadmap page.

Offline mirror

Since I'm - sad enough - not in the position to tell my enterprise scale customers which ticketing systems to use, there is still the need to have the issue descriptions, ticket ID, target versions, affected versions and even the roadmap available offline.

For an offline mirror without the capability to change the status of tickets, the following setup workflow is used instead of the steps given above:

Instead of trackdown.sh use issue trackdown.sh redmine to set up the mirror connection.

trackdown.sh redmine <apikey> <projectname>[,<projectname>...] https://<redmine.host>/

Afterwards, anytime you can connect to the Redmine system, collect the current mirror state to your local issue collection file and the roadmap.

trackdown.sh mirror

Additionally, since you now are on your command line and perhaps don't want to switch windows every second, there is a remote command to issue commands on the remote mirroring source system.

trackdown.sh remote comment XYZ "Hi there."
Adding comment "Hi there." to XYZ
trackdown.sh remote assign 68 XYZ
Assigning 68 to user XYZ

You have to provide the id of the user - not its name, which is also always exported to the issue collection file to facility this.

Migration

When you think the information mirrored right now is sufficient to cut the ties, you can set up the created issue collection and roadmap as the repository and do a trackdown.sh use.

The full migration is not covered by a command yet, and setting up the mirrored data in the special TrackDown branch or any other location of your choosing must be accomplished manually. The needed steps include:

Latest Mirror

Get the latest mirrored data.

trackdown.sh mirror

Remove Mirror Configuration

rm -rf .trackdown

Initialize TrackDown

Special Branch Flavour:

trackdown.sh init
trackdown.sh use
mv old-issues.md .git/trackdown/issues.md # or .hg
roadmap.md .git/trackdown/roadmap.md # or .hg

Custom location - e.g. wiki

mv old-issues.md ../wiki/issues.md
roadmap.md ../wiki/roadmap.md
trackdown.sh use ../wiki/issues.md
(cd wiki; git add issues.md roadmap.md)  # or hg

Clean Up

Since the mirroring collects as much data as possible, it might be a good idea to separate already closed releases or milestones from the currently relevant issues in the issue collection. Use the copy command to do so:

trackdown.sh copy Milestone1

And now use the remaining issues as the new collection and add the separated issues as a changes/changelog part to your documentation.

mv ../wiki/Milestone1-issues.md ../wiki/issues.md
(cd ../wiki ; git add Milestone1.md) # or hg

Of course, this cannot only be done for mirror issue collections and is e.g. used for TrackDown itself, like for release 1.0 in this and this file.

About

TrackDown - Issue Tracking with plain Markdown. If you are missing the "git clone" for your tickets from github.com, codeberg.org, or gitlab.com, then this is for you. A lightweight Ticketing System for distributed and unconnected small Teams.

Topics

Resources

License

Stars

Watchers

Forks

Sponsor this project

Packages

No packages published

Languages