I've worked with websockets but I've never worked with MigratoryData. Writing a basic websockets client from scratch probably would not be that hard with access to the spec. You could probably literally do it in almost any language that supports network programming, which is almost anything.

The trick here is going to be... not having access to the specs. Your next best bet is going to be spying on one of their connections, and it'll be ideal if you know the data that's going across the wire. Many years ago I reverse engineered the network protocols of poker clients, which I could do because I was playing the hands I was recording so I could "follow" the data stream.

Still I always had problems - edge cases I hadn't forseen (because without a spec you're a bit in the dark as to what 'should' happen in any case you haven't specifically seen.

uh I'd start by seeing if you can see usable data in the ws part of chrome tools (frames), if so you can probably set up a server to grab that.

Yeah. I have access to the javascript client API, but it's minified. I also have, unminified and unobfuscated, the javascript from the website that invokes the client API. I was thinking about just running the javascript using Java's JS engine, but I'm not sure if it uses any unsupported stuff (as Java's JS engine doesn't support browser-centric stuff like DOM manipulation). It would be nicer to do it natively. I was hoping to figure out something clearer about what direction to go but I guess I'm just going to have to jump in and try things out.

I do lots of websocket stuff but wouldn't have the first clue on how to do what you're looking for without a lot of almost certainly frustrating trial and error dev.

oh man i pissed one of my colleagues off today really bad and managed to make myself look like a total donkey.

tbf it wasn't really my fault - everyone has stuff they're supposed to be working on, i didn't have a lot to do today (did a lot the last few work days). So, a new request by the product owner comes in for a release we were gonna have today, he wants some copyright info prepended to like 150 source files. Cool, I can do that. I assign myself the task on github, made a comment, tagged my boss in it. Pretty clear. He even came over to my desk and said yes it's ok to do that, I told him I could write a script in about 10 mins, bingo bango bongo.

Had a slight issue with sed being weird on my linux VM so I worked around that and it took me about 45 mins to make sure everything looked good, ok cool, create a branch for it, push my commits to the branch, then request a merge to the main branch.

Girl comes over to my desk SUPER pissed about 5 mins later saying I pushed a bunch of stuff to her branch. I was super confused like what??? And apparently she said she'd been working on that issue all morning (what the **** like it's seriously a 2 liner script and don't you have other stuff to do??) and created a branch for it named the same identical thing as I did so we were pushing onto each other's ****.

Well, it didn't really **** anything up except a few files got an extra ending comment tag. But she said I should've checked for an existing branch or assigned myself the project on git - and the thing is I'm like 99% sure I created the branch first because "$ git -b checkout ______" didnt throw anything weird at me, but I'm not sure if it's supposed to and I don't even know how to check if a branch exists already and I'm not even really sure I should have because why would anyone else be working on that issue, #1, and #2, why would they name it the same branch i did??? it was a dumb accident. I tried to explain that I assigned myself the task and commented on the issue like an hour before hand but she was still pissed and I'm like 99% sure she was complaining about it to other people.

Like come on. I was too bewildered to really say anything to her, I made a comment on the post that said "Sorry, I assigned myself this task about an hour ago, there was a mixup."

She's a petty idiot, pay her no mind.

Anyone who agrees with her is no friend to progress and is also a friend of drama and pettiness, pay them no mind either.

This is one of the reasons I like the fork and pull request workflow. You're always working on a branch on your own remote. Any interaction with the upstream (shared) branch happens via pull request

Can you explain that to me like I’m a moron?

At first i thought I did something really dumb because I’m honestly still learning command line git stuff but on further research it seems like it’s actually sort of a pain to check if a branch already exists.

Regardless it’s kind of a free for all here obviously so I’d like to not repeat something like that again.

We use branch names like Matthew/foo-work and it helps with that issue

Most stuff you do in git just affects your local git repo until you do a push. So git checkout -b mybranch makes a local branch and until you push that branch to origin no one else can see it.

I don't think her story really checks out, although I'd want to try it to be sure. Pretty sure whoever pushed to origin last would get an error message and at the very least have to pull/merge first. (That is, her story doesn't check out unless you had to pull/merge before you could push your branch...)

Well, kind of worst case scenario, there are headless browser-based javascript environments, that's one of the ways that front-end automated testing works, and it's how google scrapes JS websites.

https://developers.google.com/web/up...eadless-chrome

ok, this is kind of what I thought anyway. Something was weird about it.

Anyway, work's been going well, it changes day to day with being extremely busy and extremely bored but I guess that's the nature of the business.

Not to get too bloggy but I did a good thing today, one of our interns is really inquisitive and let's put it in a nice way and say he's VERY outgoing about questions, I asked my rockstar if he was being annoyed by the intern and if he'd like me to ask him to schedule a set time for questions, and he replied in all caps YES PLEASE

so, diplomatically I gotta deal with that, but that's about the extent of my PM duties for the rest of this week, other than 1 meeting. I think I could maybe juggle these two jobs for a while but I understand what people are saying that at one point I need to go one way or the other.

I wrote a smoke test for that ancient issue that I pointed our team's direction in a few weeks ago, and it got a lot of comments on github and sparked a lot of discussion, so I think I did a good thing there too. Plus the cranky/toxic guy approached me about it and wanted to discuss it. So that was a big deal.

I don't know how long I'm gonna last here, lol.

This was a common workflow when working on open source projects. Not sure if that is something that is enabled as part of github enterprise though.

But the whole story doesn't make a lot of sense for two reasons:

1. When naming your own branch, I find it rare to ever name it the same thing.

2. If jmakin was able to push commits on her branch that would mean she had no commits pushed on top of master. So what is the big deal? No one was overriding each other's work. She should have assigned the ticket to herself before she started it!

Anyways, I find the whole thing silly to get pissed off about and wouldn't worry too much about.

Hah that's awesome. We have an intern and new hire on our team. They are both quiet and are afraid to ask questions. Sometimes I get neck deep in my work and often feel bad when I neglect to check up on them. So I'm working on being more proactive and checking up more often on them.

jmakin's story doesn't make sense (like most of his stories lol). Yes, `git checkout -b` will not throw an error unless that branch exists locally, even if a branch with the same name exists remotely. However, without `--force` or `--force-with-least, `push` will get rejected unless you are only pushing new commits at the tip of the branch. So assuming you successfully pushed without `--force` or `--force-with-lease`, you didn't overwrite her work.

She's probably pissed off because she has unpushed commits and she was the one getting rejected. If that's true, all she needed to do was `git fetch` and `git rebase origin/branch`. This shouldn't generate any conflicts because you just added lines. Or even simpler, she should just push her commits with `--force-with-lease`, and have your run your simple sed command once more.

She's in the wrong here, not you. People make mistakes, it's absolutely fine and to be expected. Just work together in a courteous and professional way to solve the problem.

Nobody can be expected to prevent first time mistakes, as long as you learn from them. What you should expect from your coworkers however is for them to control their ****ty attitude. She needs to work on her people skills.

Grimreaper shut the **** up seriously. Your shtick is annoying and old.

Put me on ignore if you have to, but i’m going to make it my personal mission to annoy you until you do.

jmakin,

Everything he said is correct. You are the one who seems to be annoyed.

On the topic of git my favorite feature is git stash. I use it frequently, it's especially nice when you want to show a coworker something that you are fixing that is a work in progress.

Stash your current work, go back to the existing/previous state, and then go right back to where you left off instantly.

It’s the tone and unnecessary haughtiness of his posts. He’s ignored now.

The other two interns we have are like this - I have to check on them at least once a day to make sure they're not stuck, and they usually are. And a lot of the time I'm unable to help them so I have to find someone who can.

But this other kid is super disruptive, he rarely sits at his own desk and prefers spending time in other people's offices asking questions whenever one pops into his head. My colleague said "i wouldnt mind if it was even once an hour but it's just constant."

My colleague is super productive and does like 90% of our work so i don't want him getting interrupted like that, lol. But I also don't want to stifle this kid's enthusiasm obviously, it's a good thing. I'm just going to send an email out to all the interns telling them to please don't be afraid to ask any questions, but if you do have questions, please email them (if they're small) so the person can respond at their leisure, or compile their questions all at once and then schedule a time to meet with someone who can help with the problem.

This is where Slack is useful. You throw your question in the team channel and let anyone on the team including the other interns a chance to answer the question. It also allows the developers to answer the question on their own schedule.

lol I am a proponent of stash too. my coworkers are probably tired of hearing it. its great for moving around branches.

And Slack goes does down globally lol

I had a very senior backend dev drive on my machine and he stashed/checkout -b/pull from his remote/build/test/compare and then afterwards went back to my original branch and applied the code.

I was like, ok yes that was awesome I'm immediately learning how to do that.

this might stifle the other interns even more. they may be even less inclined to seek out face to face help (I mean, you are pretty much telling them not to).

and the annoying guy will proly just plow ahead like nothings changed.

not that I have a good solution.

Can't you just have a face to face with him and be 100% honest. You like his enthusiasm, he just needs to modulate it to a reasonable volume/frequency.