For non-trivial reviews, when there are files with several changes, I tend to do the following in Git:
- Create local branch for the pr to review
- Squash if necessary to get everything in the one commit
- Soft reset, so all the changes are modifications in the working tree
- Go thru the modificiations “in situ” so to speak, so I get the entire context, with changes marked in the IDE, instead of just a few lines on either side.
Just curious if this is “a bit weird”, or something others do as well?
(ed: as others mentioned, a squash-merge and reset, or reset back without squashing, is the same, so step 2 isn’t necessary:))
You must log in or register to comment.
- Make branch for the ticket
- Make periodic commits to branch
- Open PR from branch to main
- (optional) if the changes are big, I have a meeting with devs to discuss it. If I can’t easily explain to the junior devs what I did and why, it means I did something wrong.
As a senior dev, I’ve found “can the junior devs grok wtf I did/made” to be an excellent “did I overengineer?” Litmus test.
A good implementation should be not too hard to explain to the juniors, and they should be able to “get it” in a single short 20-30 minute meeting at most.
If they are curious/interested and ask questions, that’s a good sign I made something useful and worthwhile.
If I get a lot of “I’m not sure I get it” and blank stares, I probably have overcomplicated the solution.
If that “ooooh, okay!” Comes quickly, then we are good!
If you need a whole meeting to explain what’s going on in your PR then it’s already too complicated, IMO. That explanation should be done as a combination of PR description, commit messages, documentation, comments, and the code itself.
Everyone will forget a meeting. The stuff I described will pay dividends for future maintainers.
The meetings should happen at design and planning phase.
Do you not do demo meetings after introducing entirely new features?
Sometimes a PR can be quite large as it involves an entirely brand new feature that simply didn’t exist before.
And if it’s an internal tool/service for fellow devs to use to make their lives easier, yes, it likely deserves a meeting so the devs can have a chance to QnA about it. Usually 5-10 minutes going over the who/what/why/where/how, then 20 mins or whatever of any needed QnA if devs are curious for more info about specifics, like performance or extensibility or etc.
If you create a new tool like that abd then just hand it off with all the devs have to go on being “here’s the manual, figure it out” you know what happens?
Almost no one reads it, and pretty much no one uses it, because parsing a giant manual of info is difficult co.pared to seeing a live demo
Do you not do demo meetings after introducing entirely new features?
Sure but it’s rare that a single PR introduces anything but a small slice of a feature, and the demo is usually something that can just be recorded in under a minute and sent over chat. The PR reviewer is likely to try building and using the new functionality too.
And if it’s an internal tool/service for fellow devs to use to make their lives easier, yes, it likely deserves a meeting so the devs can have a chance to QnA about it.
Just document the tool or service. That must exist anyway if it’s going to be used by many people. And we’re not really talking about a PR review at this point, we’re talking about an internal release ceremony, which may involve parties that are not going to look at the PR.
If you create a new tool like that abd then just hand it off with all the devs have to go on being “here’s the manual, figure it out” you know what happens?
Almost no one reads it, and pretty much no one uses it
You’re still straying from the topic of conversation: PR review process. But regardless… If no one uses it then it’s probably not useful. Why did you build something that didn’t compel people to use it from a one-paragraph internal PSA?
it’s rare that a single PR introduces anything but a small slice of a feature
Yes, I never said this was a common occurrence. This is indeed rare but it dies happen.
And we’re not really talking about a PR review at this point
No, this is 100% still part of the PR review, I was pretty explicit about that. This process should happen for a large feature/service/etc as this simply cannot be covered by a “LGTM!” Comment on a PR.
These meetings primarily cover when you’ve established something that needs to be followed or used moving forward. IE: “This new feature makes our lives a lot easier and now (annoying thing) is much easier to implement. This is how I used it in my PR and I wanna make sure all the rest of the team is on the same page about it, and everyone else starts using it in the future.”
This 100% comes up here and there and it absolutely necessitates an actual meeting, because any dev worth their salt knows what happens if you dont get everyone on the same page on it…
The inevitable “why didn’t you use (thing I made explicitly for this purpose)?” Convo comes up a month or two later, because all you did was document the new thing and open a PR, get a couple "LGTM!"s, and you posted a blurb about it in the chat and pinged some folks.
And if you’ve gone through this process before you will know that simply just never works, full stop. Doesn’t matter the team, doesn’t matter how well documented, doesn’t matter how good your write up is… People. Don’t. Read.
There’s a reason “RTFM” is such a meme in linux communities, people don’t fuckin read mate, abd that’s why critical big PRs 100% need a 20-30 min QnA meeting so people can actually talk about it.
I’d estimate tops 10-20% of devs that should read your docs, will actually do it.
Let’s step back. We are trying to talk about code review, right? That’s when you are getting a final check that your code looks roughly correct and does what you already planned for it to do with your colleagues, give or take a few details you couldn’t plan for. So at this point, usually at least one other person knows what you’re trying to do, is bought in, and should be able to review it.
I’ll concede that if you are desperate to merge a PR and people are on vacation or something, then you might want to grab someone else to review. I’d send them the PR and say, “let me know if you have questions.” Then if they do, you can have a quick chat or meet one on one.
That’s code review.
You are talking about some other thing. I’ve been in internal release meetings with lots of people that needed to understand the basics of a feature in order to support it or use it. That happens way after the PR is merged. And I’ve been in meetings to figure out how we should solve a problem, which happens before the PR is reviewed. I have gotten PSA emails or slack messages about cool new tools that make my life easier or changed our coding standards in the organization.
I have never been asked to meet with multiple people about what a PR is about before reviewing it.
IME those meetings only happen if the PR is objectionable upon review and it’s hard to negotiate what must be changed.
That’s code review.
Only on very small projects.
As the team grows, having just 1 person review your code is simply not sufficient.
Even 2-3 may not be enough to sanity check if a larger new feature on a massive project is good. If it impacts the team and means a fundamental shift in methodology, then you need more voices weighing in.
Now, can you merge the PR in first, then have the meeting after? Sure, I guess, but why would you?
If the team reacts negatively to what you built, finds flaws in it, or simply just doesn’t get it because you overengineered, now you have to revert the PR and go back to the drawing board.
It makes tremendously more sense to bring folks impacted in on a swarmed live PR review as you go over what you did, where you did it, and why… before you merge it in.
At this point you can QnA, get suggestions, feedback, etc.
Now typically most of my response to live chat feedback is “aight, can you add that as a comment on the PR itself so I can see it after?” And then they go and do that asap, usually typing it up as I answer other folks questions. Because at this stage the PR is literally the perfect place for feedback to go.
I’ve been down the path of “1 person LGTMs the PR, huge new feature is added, I document it on the wiki rigorously, I post the new feature to chat… 1/10 people bother to use it and 9/10 give blank glassy stares when I bring it up”
It. Doesn’t. Work.
Period, lol. And don’t get me wrong, I wish it did, but people just don’t RTFM mate, that’s a fact of life a d the sooner you accept that, the easier the job gets.
K what do I know, it’s just what I’ve done on code bases with ~5 million lines of code, at the point where no one understands even half of it.
If you were reviewing a “non-trivial” PR from me, I’d recommend not squashing because I would’ve broken it up into readable atomic commits.
This should be much more wide-spread. The hardest part of programming is reading someone else’s code.
More people should learn to do
git rebase -i, it’s a simple way to re-organise your commits to make sure that they tell a story to someone going through the PR commit by commit. It only takes a minute and can save your colleagues so much time and increase the quality of the review process.Unfortunately it’s uncommon now that GitHub’s PR workflow dominates, so people think in terms of (often squashed) PRs and talk about “stacking PRs”. At least GitHub supports viewing PRs commit by commit.
If PRs are just how it’s going to be, I wish GitHub could auto cut stacked PRs from a linear branch of commits.
It’s surprising how Github is sorely lacking in git features and even encourages bad practices.
-
There is only a linear history view of commits without any child/parent relationships visualized. I have coworkers that tell me it’s annoying how I use merge commits because they just see a big list of small commits in the history, and they’d rather see one commit per PR. Well… I sympathize, but also, fuck that! I’m not squashing my carefully crafted commits because Github has a shitty UI. Use
git log --first-parentor something likelazygit. It’s sad that many open source projects have adopted a “squash and merge” mantra for PRs. All of the merge methods have a valid use case, but people are so beholden to the damned Github history view. -
PRs have no concept of what a patch is. If you rebase, the entire history disappears. Reviewers cannot see changes made to individual patches unless they are applied first in new commits and then manually squashed into old patches after the review is done. Phabricator does this better IMO; each patch has its own history. I’ve even heard devs at other companies tell me they never make PRs bigger than 100 LOC. That just seems like a huge waste of time to me, unless you have some special tools to support that bizarre workflow.
-
Merge queues only support a single merge method, one of merge commit, rebase, or squash. So we just don’t use this desirable feature because of this limitation.
Ooh yeah PR as patches, persistent despite rebases, would be nice.
Many git operations fundamentally have three SHAs as parameters (tree operations after all), and GitHub’s model simplifies it down to two.
-
Or use a tool like StackedGit which makes the atomic commit workflow incredibly simple. Build atomic commits as you go instead of after you’ve written all of the code.
I made a branch, make commits, and then make a PR. I don’t care about the number of commits because sometimes a reviewer might be able to make more sense of a PR if they view each commit instead of all the changes at once.
For us we just make sure that the branch builds and passes tests before merging it in, and just do a general look over to make sure everything looks correct, follows best practices, etc. if the UI was changed I usually add screenshots of before/after or a screen recording of me using the feature. Sometimes these can really help a reviewer understand what all the changes mean.
In my experience, I prefer to review or contribute commits which are logical changes that are compartmentalized enough that if needed, they could be reverted without impacting something completely differently. This doesn’t mean 1 commit is always the right number of commits in a PR.
For example, if you have a feature addition which requires you to update the version of a dependency in the project, and this dependency update breaks existing code, I would have two commits, being:
- Update dependency and fix issues because of the upgrade
- Add new feature using new dependency
When stepping through the commits in the PR or looking at a
git blame, it’s clear which changes were needed because of the new dependency, and which were feature additions.Obviously this isn’t a one size fits all, but if someone submitted a PR with 12 commits of trial and error, and the overall changes are like +2 lines -3 lines, I’d ask them to clean that up before it gets merged.
You can squash merge so it goes into the main branch as one commit, that’s usually how I do it.
Right, for junior devs or trivial changes, that’s fine. My take is if I’m going to make someone take the time to review my work, I take the time to make sure that it’s cleaned up and would be something I would merge if I were reviewing it. Most of this comes from working on some larger Open Source projects which still require patches be submitted via email which I know is a real “back in my day” moment, but it did instil good practices which I try to carry on.
It’s possible to do a diff across all the commits in the PR (rather, a diff between the last commit before, and the final commit of), so no, I do not squash them all for review. Otherwise, yes, I will definitely clone and review the code locally for large PRs.
Squashing seems like you’d potentially lose out on info and have a harder time isolating the changes you’re looking through. I guess it depends on how much has been changed and whether some of the commits along the branch were more important than others.
I also don’t think the reset is necessary, you should be able to diff the branch head against whatever you want.
Yeah I had the same thoughts, had no idea people even bothered to do that
Sometimes the info lost is just a typo or a revert. I’d say heavily depends on the workflow of the people involved. Some like long history, some like rebasing, others, something in between. How you review those approaches changes a lot
Sure, that’s fine. I use interactive rebase for “cleaning” a lot. I’m just saying it doesn’t make a difference for diffing (as you can diff any commit against any other) and doing it as a matter of routine sounds like it could skip potentially useful history.
I mostly rebase but if a branch has things happen in a sequence that matters, I would merge it instead, for example.
I just comment “Looks good!” and approve.
As a manager I wait until another engineer approves then comment “lgtm” and approve so the PR hits its two review minimum.
Same here, unless it’s a small PR then I like to nitpick 👍
Different things will work for different organizations. More important than this is whether everyone is on the same page with their workflow in checking PRs. One commit prs might be easy for some workflows and bad for others
This does strike me as odd, your commits should be cleaned up if they are a mess of “reverted X”, “fix typo”, “saved days work”, etc. on the other hand, you don’t usually have to explain your modifications if you didn’t squash your commits.
I hadn’t yet tried squashing + reset, that seems like a smart idea, but yeah, I don’t particularly like the usual PR review UIs.
I do the merge via CLI anyways, so might as well check out the code for the review and then view it in my IDE + be able to run it and such.
If you’re using the CLI and cleaning up a branch for a PR, the interactive rebase is a godsend. Just run
git rebase -i origin/main(or whatever your target branch is) and you can reorder/squash/reword commits.
Not too weird. Personally I use a tool called reviewable.io which essentially does the same thing but bundled into a nice UI. It depends on the branch though, if there are a lot of changes and it’s a senior developer then the changes are usually broken up into meaningful commits but usually the commits are just a ball of mud.
There are IDE extensions that show the diff of the entire PR locally without having to squash anything. So yes, it’s weird to reinvent a square wheel.
I usually find web gui good enough to get context of code. Doing all that and then needing to go back and forth from IDE to PR to leave comments sounds like it’d take a bunch more time.
Why squash all the commits if you’re just going to reset? Unless I’m crazy you can reset across multiple commits.
Yeah
git reset --softthen the sha of the last commit you want included in reset.
I’ll usually do steps 1 and 2 for a reasonably complex review. I’ll reference the diff from the website (for work, this is Azure DevOps, for personal, GitHub or similar) while I inspect and run the modified code locally.
Squash merge into the main branch. It’s the only way to fly. (just my 2c!)
