gophr-pm/gophr

Native package management for Go.

Gophr - golang package management

An end-to-end package management solution for Go. No manifest or lock file and a fully versioned dependency graph. Simply place the url in your import path and it's automatically fully versioned.

gophr.pm/author/repo@(semver or SHA)

Native golang dependency management

go get can only retrieve the current master branch. If you ever need to re-download your dependency it could be totally different each time.

  import (
      // Un-versioned github link
      "github.com/a/b"
  )

Gophr dependency management and versioning

Gophr allows you to version lock your dependencies by semver or SHA.

  import (
      // Version current master branch
      "gophr.pm/a/b"
      // Version by semver
      "gophr.pm/a/[email protected]"
      // Version by semver logic
      "gophr.pm/a/b@>1.0.0"
      "gophr.pm/a/b@<1.3.2"
      // Version by partial or full SHA (Anything between 6 - 40 Characters)
      "gophr.pm/a/b@24638c"
      "gophr.pm/a/b@24638c6d1aaa1"
      "gophr.pm/a/b@24638c6d1aaa1a39c14c704918e354fd3949b93c"
  )

The problem with native Golang dependency management

Golang has no ability to version a specific SHA or tag for a repo. Anytime you pull down an import it grabs the current master branch. This not only bad practice but it could potentially silently break your code without you ever knowing why.

There are plenty of Golang versioning tools. What makes Gophr special?

Gophr doesn't require you to install any tooling to use. Simply place the versioned url gophr.pm/author/repo@(semver or SHA) in your import path and you're done.

We give you the power of semver to reference tags and create logical equivalence operations just like in gem and npm.

"gophr.pm/a/b@>1.0.4"
"gophr.pm/a/b@<1.0.0"

We also fully version the entire dependency graph. Meaning, we version lock every sub-dependency as well, so it's perfectly preserved, everytime. Something no one else does.

Gophr Resources

• Home
• Gophr CLI
• Gophr Web
• How do I contribute?
• How does Gophr work?
• Our Microservices
• Setting up dev environment

Questions, comments, concerns? Feel free to open an issue or reach out on slack @shikkic, @skeswa or @ZacharyThomas

gophr's Issues

Make the indexer work with `common/model_package.go`

We need to use the batch insert function defined in the package model when we index go packages.

[gophr-cli] basic lock latest versions

Create basic lock command that will version all github dependencies with the latest ref or semvar

[Indexer] Use github api instead

Unfortunately if we want to invest in other repo attributes that are important to us we will have to make requests to a github endpoint ( that will be throttled) so we'll need to figure out how to pace the number of requests we make through proper channels

[gophr-cli] Initial Refactor of Search

Want to start cleaning up and creating a better codebase for the cli needed to make some big changes to the search command.

Deploy a copy of dev to the cloud and address it to gophr.pm

Create basic gophr-cli search command

Need the basic boiler plate and pretty print the search results

Indexer pull versions for packages

For each package on github pull in versions

Start an initial design skeleton for the gophr.pm web experience

Such material. Very gophr. Wow.

Build a skeleton for the API

Create some stubbed-out routes to demonstrate how the API will be written.

Figure out Cassandra

We need to figure out how to make it work for our use case.

Normalize public errors in router and common

To make the router api fire back more descriptive errors that don't reveal too much internal logic.

Get test coverage for the router back up to 90+%

Pretty straightforward - just tedious.

[API] Install route 100% WORKS

Write an API endpoint for getting the latest version (or ref) of a package.

It should be an HTTP GET to /versions/latest.

Created tests for Indexer

Getting Yaw onboard

...

[Gophr-cli] Install command routes to BE

Write an API endpoint for getting the known versions of a specific package

[Backend] Recursive date-based sub package versioning

Write a standardized environment config utility

We need a shared bit of code that gets environment information and throws it in a struct. e.g. host domain, db domain, port, api keys etc.

Create Cassandra + Lucene Index Docker Container

We'll need to create a docker container for elastisearch and get the config scripts setup

Needs moar java

[Indexer] Create clean up cron job

Write an API endpoint for getting information about a specific package

[gophr-cli] Bash completion

Create a "common" sub-package for shared functionality

So that we can share logic between the router and indexer in the short term. In the long term, we'll have database logic in here.

Create search api

Create Cassandra DB Schema

Self-explanatory

[gophr-cli] Deps command refactor

Want to start cleaning up and creating a better codebase for the cli needed to make some big changes to the deps command.

Incorporate version-segmented endpoints in the API

Just like github has /api/v1/... we would have the same deal.

Determine package download numbers

Popular
Trending
(New, eventually?)

[Router] Support fo sub-packages

We need to be able to handle requests for sub packages

Router should fail when a non-existent version in queried despite there being no candidates

The reproduction is the go-get request on http://gophr.dev/shikkic/gophr-cli@15

Central logging component

https://www.goinggo.net/2013/11/using-log-package-in-go.html

Kubernetes & Deployment on Google Cloud

• Add gophrctl configuration flag support for prod
• Add gophrctl configuration flag support for staging
• Write kubernetes configuration for api, router, indexer, turnstile, db, etc.
• Finish replication controllers for each of the application images
• Mod the ws image for kubernetes
• Mod the cassandra image for kubernetes based on the official example
• Deploy all of the docker images to GCR by adding gophrctl push
• Setup GPG encryption for sensitive docker image files (like SSL certs)

Create frontend web boilerplate

Create web frontend boilerplate using

• React
• Redux
• Babel
• PostCSS
• Mocha

Write an API endpoint for getting the most depended-on packages

Get test coverage for the router back up to 90+%

The new changes obliterated test coverage.

Add github repo stars fetching

Use the API to get a list of every golang repository, and preserve the star count.

Add SemverCandidate.String()

So that we can use it in the indexer.

Write an API endpoint for getting the "up and coming packages"

Indexer integration with Cassandra and Elasticsearch

Basically get data into Cassandra

Get SSL certs for gophr.pm

Refactor router to support git refs

Right now we support:

gophr.pm/a/b@~1.1

But what if github.com/a/b has no version tags? That means that gophr just can't help you, unless we support specific refs in the URL:

gophr.pm/a/b@c7a7db4761eab9d2b803a1c35e2ce905ea759854

This makes a gophr a universal dependency management solution. This issue is basically me refactoring the router to support that.

[gophr-cli] create env flag for gophr urls

Getting more done in GitHub with ZenHub

Hola! @skeswa has created a ZenHub account for the skeswa organization. ZenHub is the leading team collaboration and project management solution built for GitHub.

---

How do I use ZenHub?

To get set up with ZenHub, all you have to do is download the browser extension and log in with your GitHub account. Once you do, you’ll get access to ZenHub’s complete feature-set immediately.

What can ZenHub do?

ZenHub adds a series of enhancements directly inside the GitHub UI:

• Real-time, customizable task boards for GitHub issues;
• Burndown charts, estimates, and velocity tracking based on GitHub Milestones;
• Personal to-do lists and task prioritization;
• “+1” button for GitHub issues and comments;
• Drag-and-drop file sharing;
• Time-saving shortcuts like a quick repo switcher.

Add ZenHub to GitHub

Still curious? See more ZenHub features or read user reviews. This issue was written by your friendly ZenHub bot, posted by request from @skeswa.

Write an API endpoint for getting the most popular packages

Separate common/ into sub packages

Like, for instance:

• common/errors
• common/model

Write an API endpoint for fuzzy searching packages.

This endpoint will be used for autocomplete and other similar features.

Persistant shared volume for DB container

Kinda sucks to have to run the indexer every time I spin down and up