git – A Geeky Gulati

Updates to twit-sort

I am currently on vacation! In Florida! We drove!

I took the opportunity (in the early mornings / late nights when everybody else is asleep) to work on some code that I wanted to update – I don’t normally get time to do this. Its my twitter-reading app. Which only I use. But hey, that’s fine, I haven’t tried to market it.

Changes

Change #1: I kept seeing “…” with truncated tweets. A little research led to some interesting stuff — basically following retweets to get down to the original tweet to get the text from there.

Along the way I tried to build up a “who quoted who” breadcrumb:

The code is a bit wonky:

However, its confusing – many people (or clients) retweet without actually having a retweet link. There’s “quoting” which is different .. eh. whatever. If I’m given it via the API, passing it through, figure it out later.

Change #2: I surfaced the like count and favorite count, as well as a link per tweet to open the tweet in its own window (as hosted by twitter).

Complications

I have the code in github, which is public, and I don’t want my access keys and tokens and stuff checked in there. So, I had to fudge around a bit – this was my solution:

I created a branch with all the passwords and stuff, and worked from there.
Once it was done, I either did a rebase (with cherry picking) or a direct cherry pick to move commits over to master, which I pushed up.

Granted I could use my azure visual studio hosted git, but I wanted the code to be visible / usable. So.. if anybody has other ideas of how to do this better, please let me know.

What Next (with this project)

Not much else I can do with this in its current codebase. I could certainly make it prettier, but .. eh, that’s not me.

If I had unlimited time, I’d rewrite it – make it so it did all its fetching and filtering and sorting locally (in javascript). I could get a LOT fancier then – things like pulling out hashtags into groups, etc. Maybe adding some sentiment analysis things.

However, I have so many other projects on the burner.. this one won’t make it for a while. My need has been solved, so there’s very itch here to scratch.

The url of the site: http://twit-sort.azurewebsites.net

The code: https://github.com/sunnywiz/twit-sort

git rebase

I’ve never done a rebase before. I figured today would be a good day to try it out. It’s a simple case —

From the client’s point of view, origin/master is ahead by one commit. It happens to be a conflicting change.

In Tortoise (I’m starting out easy), I pick “Rebase”, and I’m saying I want to change my branch SG-EndToEndTest to be “re-based on” remote master.

Its showing me the list of commits that I’ve had, since I started my branch from wherever
Everything is picked. Could choose to not-pick some stuff, leave it behind, like config file changes for local debugging.

I click “Start Rebase”. It starts from the bottom of the list (ID=1) above and triest to re-apply the commits to the new root. It runs into conflicts –

I right click on the conflict, edit it .. turns out TortoiseDiff doesn’t think it’s a conflict, I can easily mark it as resolved. I have to do this twice, both in .csproj files.

When its done, its all still local – server’s not any different – but local shows:

The old commits are still there.
But they’ve been cloned and re-grafted onto a new source node, and the label has been moved.
If I had to, I could undo everything by force-moving SG-EndToEndTest to be based on ad1c.

If I now push up to the server – I get an error:

Instead, I have to push with –force (“may discard unknown changes”) (ie, “tell the server to do stuff where it discards changes that I, the client, know nothing about – JUST MOVE THE REF”)

Now the server looks like this:

That was my first, very simple, rebase (without squashing).

The pull request that I had open against that branch survived as well, and changed from “Cannot merge – conflicts” to “Can merge automatically”:

Enjoying Git

My current client uses an (onsite?) enterprise github thingy. We do basically branch-per-feature, which means a lot of merging. And surprisingly few conflicts. I find myself constantly going back to this graph:

It’s not available on the free public github, AFAIK. You can see where people pulled from Refactor back into their feature branch, and how much context switching there is (we have two folks working on the project).

A video that really explained git to me was this:

[youtube=http://www.youtube.com/watch?v=1ffBJ4sVUb4&w=448&h=252&hd=1]

Awesome Git Video!

It highlighted the directed graph structure, and the role that nodes and labels play. After watching it, I understood Fetch, Pull, Push, Add, etc – it all made sense!

However, I’ll confess – I’m using tortoiseGit, not the git command line.