|
the git book would be better if the first chapter included "here's how we want you to do everything and here are the socialized expectations of contributors to a project"
|
# ? Sep 14, 2021 17:59 |
|
|
# ? Jun 10, 2024 02:15 |
|
Share Bear posted:i'm on team "always merge" so if i can avoid doing rebases (unless the project owner wants me to) i go there oh I thought we were talking about a local branch
|
# ? Sep 14, 2021 18:11 |
|
Share Bear posted:i'm on team "always merge" so if i can avoid doing rebases (unless the project owner wants me to) i go there I've never had an issue with it and you get the added bonus of having to look at your changes one more time before committing.
|
# ? Sep 14, 2021 18:34 |
|
DoomTrainPhD posted:My workflow: that’s just squash-merge with extra steps
|
# ? Sep 14, 2021 19:38 |
|
it's more of a local squash with a bunch of extra steps. i am continually impressed by the amount of effort people put in to avoiding learning how to use git.
|
# ? Sep 14, 2021 21:23 |
|
Plorkyeran posted:it's more of a local squash with a bunch of extra steps. i am continually impressed by the amount of effort people put in to avoiding learning how to use git. i use git, it works, i branch, i merge what you're talking about the expectations of contributing to a project in git
|
# ? Sep 14, 2021 21:31 |
|
Plorkyeran posted:it's more of a local squash with a bunch of extra steps. i am continually impressed by the amount of effort people put in to avoiding learning how to use git. I know how to use rebase, merge, and squash from the CLI. I just find that I like breaking up my commits into smaller, easier to manage multiple commits for code review, and the few extra steps allow me to do so.
|
# ? Sep 14, 2021 21:34 |
|
DoomTrainPhD posted:26 files changed, 4637 insertions(+), 3970 deletions(-) hopefully it was worth that 1 story point
|
# ? Sep 14, 2021 21:43 |
|
Plorkyeran posted:it's more of a local squash with a bunch of extra steps. i am continually impressed by the amount of effort people put in to avoiding learning how to use git. To be fair, I can do multi remote workflows and still don't feel like I know how to git.
|
# ? Sep 14, 2021 22:05 |
|
i don't understand. what is "copy the stock repo to repo-pr"? are you running the command "cp" for this?
|
# ? Sep 14, 2021 22:46 |
|
DoomTrainPhD posted:I know how to use rebase, merge, and squash from the CLI. I just find that I like breaking up my commits into smaller, easier to manage multiple commits for code review, and the few extra steps allow me to do so. you can do so without those extra steps.
|
# ? Sep 14, 2021 22:57 |
|
yes but the git ux is so terrible that many people are inclined to work around it. the whole point of the system is to track all the changes and never lose them, but you can easily lose your changes by loving up in git
|
# ? Sep 14, 2021 23:21 |
DELETE CASCADE posted:yes but the git ux is so terrible that many people are inclined to work around it. the whole point of the system is to track all the changes and never lose them, but you can easily lose your changes by loving up in git it should really be an apache foundation project
|
|
# ? Sep 14, 2021 23:38 |
|
cinci zoo sniper posted:it should really be an apache foundation project
|
# ? Sep 15, 2021 00:01 |
|
it is very difficult to actually lose something you committed to git unless you are in the habit of deleting .git directories or something
|
# ? Sep 15, 2021 00:33 |
|
the only time i've ever lost data in git was when i was turning an existing folder into a new repo, while distracted, and ended up doing a hard reset to the tip of main before i committed anything to it if you are interacting with a steady state git repository it's basically impossible to lose work
|
# ? Sep 15, 2021 00:39 |
|
git reflog
|
# ? Sep 15, 2021 01:10 |
|
ok but keep in mind that "actually lost the work, permanently" is functionally the same as "must go hat-in-hand to the git wizard at the company who knows the magical incantation to fix my repo", at least for the purpose of making the decision to just cp the dir and avoid doing anything in git beyond the 5 commands you understand
|
# ? Sep 15, 2021 04:47 |
|
git reflog Now yer the git wizzard
|
# ? Sep 15, 2021 06:53 |
|
Xarn posted:git reflog yeah, fixing the commit graph can be a giant pain in the rear end but you can put the instructions for retrieving stuff from the reflog on a post-it
|
# ? Sep 15, 2021 08:28 |
|
git re-flog --harder --with-gag
|
# ? Sep 15, 2021 18:03 |
|
DELETE CASCADE posted:ok but keep in mind that "actually lost the work, permanently" is functionally the same as "must go hat-in-hand to the git wizard at the company who knows the magical incantation to fix my repo", at least for the purpose of making the decision to just cp the dir and avoid doing anything in git beyond the 5 commands you understand or you can take the opportunity to learn how to solve your problem yourself and get better at using the tool you use every day. being able to incrementally learn things as needed is one of the most important skills for a software developer, and that applies to languages, libraries, and tools.
|
# ? Sep 15, 2021 18:12 |
|
Plorkyeran posted:or you can take the opportunity to learn how to solve your problem yourself and get better at using the tool you use every day. being able to incrementally learn things as needed is one of the most important skills for a software developer, and that applies to languages, libraries, and tools. Learn? git outta here
|
# ? Sep 15, 2021 18:14 |
|
i agree with him. it's crazy that i did the dumbest thing possible to an empty repository and passed all the --yes-really-gently caress-my-poo poo-up flags that exist and i could still have apparently reverted my blunder with the reflog i am begging you, with tears in my eyes, to please learn how to use your tools
|
# ? Sep 15, 2021 18:23 |
|
TI sucks and their compilers are worse. Why yes, Windows and Linux builds are producing different hex files due to alignment issues and TI's response is
|
# ? Sep 15, 2021 18:43 |
|
DoomTrainPhD posted:TI sucks and their compilers are worse. Why yes, Windows and Linux builds are producing different hex files due to alignment issues and TI's response is all embedded compiler vendors are both dishonest and incompetent in my experience ask me why I don’t do embedded dev anymore
|
# ? Sep 15, 2021 23:05 |
|
our "it transformation project leadership team" asked me to manually fill out a confluence page to mark how many points I think we should aspire to achieve as part of our 'agile transformation'. This is in addition to the 3 sperate jira projects, approximately 1000 actual jiras, three jira dashboards and one kibana dashboard they've already created to track this. will we hire people to do the work needed to use the new deploy tools? lol of course not what were you thinking
|
# ? Sep 15, 2021 23:14 |
|
Powerful Two-Hander posted:our "it transformation project leadership team" asked me to manually fill out a confluence page to mark how many points I think we should aspire to achieve as part of our 'agile transformation'. This is in addition to the 3 sperate jira projects, approximately 1000 actual jiras, three jira dashboards and one kibana dashboard they've already created to track this. points as in story points? answer: idk, all of them?
|
# ? Sep 15, 2021 23:56 |
|
Powerful Two-Hander posted:our "it transformation project leadership team" asked me to manually fill out a confluence page to mark how many points I think we should aspire to achieve as part of our 'agile transformation'. This is in addition to the 3 sperate jira projects, approximately 1000 actual jiras, three jira dashboards and one kibana dashboard they've already created to track this. sounds like they want a random number. I recommend random.org
|
# ? Sep 16, 2021 02:17 |
|
12 rats tied together posted:i agree with him. if its any consolation, youd have an empty reflog because there arent any commits so it wouldnt have saved you anyway code:
|
# ? Sep 16, 2021 07:10 |
|
idk what’s so hard, git is just a monad on the blockchain
|
# ? Sep 16, 2021 07:30 |
|
That would actually be easier to understand.
|
# ? Sep 16, 2021 07:54 |
|
maybe i'm the terrible programmer but about 3/4 of my current team thinks nosql is more 'scalable' than our current postgres and anytime i ask when the last time the db went down because it couldn't scale they stop talking to me for a bit lol based on the number of PRs i've had to reject recently because they tried to add jsonb columns to our tables i suspect the real purpose of nosql is for devs too lazy to properly understand their data model who just want to dump whatever garbage data structure they have into the db without thinking about how to query it, what the performance impact is, or what kind of ugly hack code we'll have to write if the (de)serialization has to be done manually. i suspect i'm going to lose this fight anyway
|
# ? Sep 16, 2021 16:46 |
|
Joe Chip posted:based on the number of PRs i've had to reject recently because they tried to add jsonb columns to our tables i suspect the real purpose of nosql is for devs too lazy to properly understand their data model who just want to dump whatever garbage data structure they have into the db without thinking about how to query it, what the performance impact is, or what kind of ugly hack code we'll have to write if the (de)serialization has to be done manually. i suspect i'm going to lose this fight anyway its always this this is the larger extreme of "never learning your tools" i have read past the 2nd chapter in the git book now and still don't really care about clean commit histories but hey, i know how they work now
|
# ? Sep 16, 2021 16:47 |
|
Carthag Tuek posted:if its any consolation, youd have an empty reflog because there arent any commits so it wouldnt have saved you anyway thanks, i kind of had a hunch that this was the case but i was too lazy to check. fortunately for me, i had an old folder with all of the pyc files, which i turned back into normal python and restored like 90% of my work. just needed to spend like 20 minutes fixing formatting stuff normally i would never speak of this but we are in the terrible programming thread
|
# ? Sep 16, 2021 16:52 |
|
Joe Chip posted:maybe i'm the terrible programmer but about 3/4 of my current team thinks nosql is more 'scalable' than our current postgres and anytime i ask when the last time the db went down because it couldn't scale they stop talking to me for a bit lol No, they're talking out of their rear end and don't want to think about their schemas. That said with jsonb there is literally no reason to switch DBs if you do just need to serialize a json blob.
|
# ? Sep 16, 2021 18:11 |
|
JsonB is good and cool and makes me not have to think about how I store my data. Then again, I work in embedded and the most data I have to store in a db is at most a few hundred rows lol.
|
# ? Sep 16, 2021 18:18 |
|
I use jsonb somewhat liberally because it's super easy to take stuff out of it and put it into a real column / table if it becomes useful. alter table add skidmark text not nullable default poop->>'touch' my rule of thumb is: the moment it needs to appear in a WHERE or JOIN, it comes out of the semi-structured jsonb limbo and gets put into a proper relational structure
|
# ? Sep 16, 2021 18:38 |
|
my rule of thumb there is the same, if you only want the entire thing at once, you can put it in hstore/jsonb, and it's probably faster that way too if you ever need just a single thing from it, it's time for a real schema
|
# ? Sep 16, 2021 18:49 |
|
|
# ? Jun 10, 2024 02:15 |
|
does anyone ever pick json instead of jsonb to encourage extraction into proper columns? if you just want to store/retrieve the whole thing, jsonb sounds like not what you want
|
# ? Sep 16, 2021 19:09 |