Do all the things like ++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatarSign Up
From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple APILearn More
Search - "wk279"
Simple 1 day task. This idiot takes two weeks and after 7 days of hounding finally opens a pull request.
I go in to review the code. Should be a simple 10-15 line patch.
13,000 lines of code changed.
"I fixed a bunch of formatting mistakes and replaced all instances of single quotes to double. Consistency is important you know."20
My code review nightmare part 3
Performed a review on/against a workplace 'nemesis'. I didn't follow the department standards document (cause I could care less about spacing, sorted usings, etc) and identified over 80 bugs, logic errors, n+1 patterns, memory leaks (yes, even in .net devs can cause em'), and general bad behavior (ex.'eating' exceptions that should be handled or at least logged)
Because 'Jeff' was considered a golden child (that's another long TL;DR), his boss and others took a major offense and demanded I justify my review, item by item.
About 2 hours into the meeting, our department mgr realized embarrassing Jeff any further wasn't doing anyone any good and decided to take matters into his own hands. Thinking 'well, its about time he did his job', I go back to my desk. About an hour later..
Mgr: "I need you in the conference room, RIGHT NOW!"
Mgr: "I spoke to Jeff and I think I know what the problem is. Did you ever train him on any of the problems you identified in the review?"
Me: "Um, no. Why would I?"
Mgr: "Ha!..I was right. So lets agree the problems are partially your fault, OK?"
Me: "Finding the bugs in his code is somehow my fault?"
Mgr: "Yes! For example, the n+1 problem in using the WCF service, you never trained him on how to use the service. You wrote the service, correct?"
Me: "Yes, but it's not my job to teach him how to write C#. I documented the process and have examples in the document to avoid n+1. All he had to do was copy/paste."
Mgr: "But you never sat with Jeff and talked to him like a human being? You sit over there in your silo and are oblivious to the problems you cause. This ends today!"
Me: "What the...I have no idea what you are talking about. What in the world did Jeff tell you?"
Mgr: "He told me enough and I'm putting an end to it. I want a compressive training class developed on how to use your service. I'll give you a month to get your act together and properly train these developers."
3 days later, I submit the power-point presentation and accompanying docs. It was only one WCF with a handful of methods. Mgr approved the training, etc..etc. execute the 'training', and Jeff submits a code review a couple of weeks later. From over 80 issues to around 50. The poop hits the fan again.
Mgr: "What's your problem? When are you going to take your responsibility seriously?"
Me: "Its pretty clear I don't have the problem. All the review items were also verified by other devs. Its not me trying to be an asshole."
Mgr: "Enough with the excuses. If you think you can do a better job *you* make the code changes and submit them for Jeff for review. No More Excuses!"
Couple of days later, I make the changes, submit them for review, and Jeff really couldn't say too much other than "I don't see this as an improvement"
TL;DR, I had been tracking the errors generated by the site due to the bugs prior to my changes. After deployment, # of errors went from thousands per hour to maybe hundreds per day (that's another story) and the site saw significant performance increases, fewer customer complaints, etc..etc.
At a company event, the department VP hands out special recognition awards:
VP: "This award is especially well earned. Not only does this individual exemplify the company's focus on teamwork, he also went above and beyond the call of duty to serve our customers. Jeff, come on up and get this well deserved award."19
My code review nightmare is reviewing C++ code that was written by Matlab people. Letting them commit code without reviewing it will surely bring Armageddon, but having to watch these horrors...
Functions signature like 'void func(int arr)'. Nononononono.21
My code review nightmare part 2
Team responsible for code 'quality' dictated in their 18+ page coding standard document that all the references in the 'using' block be sorted alphabetically. Easy enough in Visual Studio with the right-click -> 'Remove and Sort Usings', so I thought.
Called into a conference room with other devs and the area manager (because 'Toby' needed an audience) focusing on my lack of code quality and not adhering to the coding standard.
The numerous files in question were unit tests files
<the rest of the usings>
T: "As you can see, none of these files' usings are in alphabetical order"
Me: "Um, I think they are. M comes before S"
T: "The standards clearly dictate system level references are to be sorted first."
Mgr: "Yes, why didn't you sort before checking this code in? T couldn't have made the standards any easier to follow. All you had to do is right-click and sort."
Me: "I did. M comes before S."
T: "No You Didn't! That is not a system reference!"
Me: "I disagree. MSTest references are considered a system level reference, but whatever, I'll move that one line if it upsets you that much."
Mgr: "OK smartass, that's enough disrespect. Just follow the fucking standard."
T: "And learn to sort. It's easy. You should have learned that in college"
<Mgr and T have a laugh>
Me: "Are all your unit tests up to standard? I mean, are the usings sorted correctly?"
T:"Um..well..of course they are!"
Me: "Lets take a look."
I had no idea, a sorted usings seems like a detail no one cares about that much and something people do when bored. I navigate to project I knew T was working on and found nearly all the file's usings weren't sorted. I pick on one..
<the rest of the usings>
Me: "These aren't sorted..."
T: "Uh..um...hey...this file is sorted. N comes before M!"
Me: "Say that again. A little louder please."
Mgr: "NUnit is a system level nuget package. It's fine. We're not wasting time fixing some bug in how Visual Studio sorts"
Me: "Bug? What?..wait...and having me update 10 or so files isn't a waste of time?"
Mgr: "No! Coding standards are never a waste of time! We're done here. This meeting is to review your code and not T's. Fix your bugs and re-submit the code for review..today!"17
The nightmare continues.
Currently dealing with a code review from a “principal” dev (one step above senior), who is unironically called a “legendary dev” by some coworkers. It’s painfully obvious he didn’t read the code, and just started complaining and nitpicking.
It’s full of requests to do things that make absolutely no sense, and would make the code an unmaintainable mess.
• Ex: moving the logic and data collection from the module’s many callers into the module instead of just passing in the data.
• Ex: hiding api endpoint declarations by placing them in the module itself, and using magic instance variables to pass data to it. Basically: using global functions and variables instead of explicit declarations and calls.
• Ex: moving the logic to determine which api endpoint to use, for all callers, into the view.
More comments about methods being “too complex” (barely holds water) right next to comments saying “why are these separate? merge them together!”
Incredulously asking how many times I’m checking permissions and how ridiculous it all is. (The answer? Twice.)
Conflating my “permissions” param and method names with a supposedly forthcoming permissions system overhaul, and saying I shouldn’t use permissions because my code will all have to get rewritten. Even if that were true, and it’s likely not, the ticket still needs to use the current permissions. I can’t just ignore them because they might be rewritten someday.
Requests to revert some code cleanup because the reviewer thought the previous heavily-nested and uncommented versions (with code duplication) were easier to read. Unsurprisingly, he wrote them.
On the same ticket, my boss wants me to remove all styling and clientside validation, debouncing, and error messages from a form. Says “success” and “connection failed” messages are good enough. The form in question sends SMS and email using arbitrary user input for addresses. He also says it shouldn’t be denounced on the server, and doesn’t want me to bother checking permissions. Hello, spam!
Related: the legendary dev reviewer says he can’t think of a reason why we would want to disable the feature for consumers, so I should remove the consumer feature flag.
You can’t make this stuff up.7
Having to review an offshore C++ codebase made in Romania that the company I worked for they bought to control a wifi module on a complex RF mobile tech device that I can't legally give more details on.
If I could legally post this masterpiece, or should I say masterpiece-of-shit, all of you C++ dev would instantly get AIDS and all the existing types of cancer upon browsing it for 2 minutes.
It's laughably bad and unmaintainable. One of my colleague called it "the perfect example of human obfuscation" and it fits perfectly.
Think of a 100k LoC main function with nested loops and ifs with random sleep values, 1000 values of hardcoded 32 bits arrays declared globally in the first 10k lines for unknown reasons. Comments in Romanian mixed with english. Somehow, this shit works by some miracle.
The worst intern you can think of, while being piss drunk, could do better and it's no joke.5
My code review nightmare?
All of the reviews that consisted of a group of devs+managers in a conference room and a big screen micro-analyzing every line of code.
"Why did you call the variable that? Wouldn't be be more efficient to use XYZ components? You should switch everything to use ServiceBus."
and/or using the 18+ page coding standard document as a weapon.
PHB:"On page 5, paragraph 9, sub-section A-123, the standards dictate to select all the necessary data from the database. Your query is only selecting 5 fields from the 15 field field table. You might need to access more data in the future and this approach reduces the amount of code change."
Me: "Um, if the data requirements change, wouldn't we have change code anyway?"
PHB: "Application requirements are determined by our users, not you. That's why we have standards."
Me: "Um, that's not what I ..."
PHB: "Next file, oh boy, this one is a mess. On page 9, paragraph 2, sub-section Z-987, the standards dictate to only select the absolute minimum amount of the data from the database. Your query is selecting 3 fields, but the application is only using 2."
Me: "Yes, the application not using the field right now, but the user stated they might need the data for additional review."
PHB: "Did they fill out the proper change request form?"
Me: "No, they ...wait...Aren't the standards on page 9 contradictory to the standards on page 5?"
PHB: "NO! You'll never break your cowboy-coding mindset if you continue to violate standards. You see, standards are our promise to customers to ensure quality. You don't want to break our promises...do you?"7
Let junior dev design module.
Make code review.
What junior dev says: "It is a totally flexible concept!"
What junior dev means: "It is extremely shitty to use for the one use case it was meant to do, but it will be equally shitty to use for all the use cases we will never have."
Back to square one.9
Wasting about an hour trying to leave detailed comments and suggestions for a fellow student on an university project's code.
Said person just respond on the team group with "If anyone does not like something, he/she can go ahead to change it, so we would move on faster"
THEN WHY THE FUCK ARE WE STILL DOING CODE REVIEWS???6
Oh gee whiz fellas. I lived through my nightmare. Recently too.
(Multiple rants over last few months are merged in this one. Couldn't rant earlier because my login didn't work.)
I joined a new shithole recently.
It was a huge change because my whole tech stack changed, and on top of that the application domain was new too.
Boss: ho hey newbie, here take this task which is a core service redesign and implementation and finish it in two weeks because it has to be in production for a client.
Normally I'd be able to provide a reasonable analysis and estimate. But being new and unaware of how things work here, I just said 'cool, I'll try my best.' (I was aware that it was a big undertaking but didn't realize the scope and the alarming lack of support I'd get and the bullshit egos I'd have to deal with)
Like a mad man I worked 17+ hours a day with barely a day off every week and changed and produced a lot of code, most of it of decent quality.
Deadline came and went by. Got extended because it was impossible (and fake).
All the time my manager is continuously building pressure on me. When I asked questions I never got any direct/clear answers. On asking for help, I'd get an elaborate word vomit of what was already known/visible. Yet I finally managed to have an implementation ready.
Reviewer: You haven't added parameter comments on your functions and there aren't enough comments in code. We follow standards. Clean code and whatnot. Care for the craft verbal diarrhea.
Boss: Ho hey anux, do you think we'll be able to push the code to production?
Me: Nope. We care for the craft and have standards. We need to add redundant comments to self documented code first, because that is of utmost importance as Nuthead reviewer explained.
(what I wish I had said)
What I actually said: No, code is not reviewed yet.
And despite examples of functions which were not documented (which were written by the reviewer nut), I added 6-7 lines of comments for my single line functions describing how e.g. Sum takes two input integers and returns their sum and asked for a review again.
Reviewer: See this comment is better written as this same-meaning-but-slightly-longer way. Can we please add full stops everywhere even though they were not there to begin with? Can we please not follow this pattern and instead promote our anti-pattern? Thanks.
Me: Changed the comments. Added full stops. Here's a link for why this anti-pattern is bad.
Reviewer: you have written such beautiful code with such little gems. Brilliant. It's great to see how my mentoring has honed your skills.
I swear I would have broken a CRT on his stupid face if we weren't working remotely (and if I had a CRT).
It infuriates me how the solution to every problem with this guy is 'add a comment'.
What enrages me more is that I actually thought I could learn from this guy (in the beginning). My self doubt just made me burnout for little in return.
Thankfully this living nightmare will soon be over.9
When the pull request
1. has 100s of files with commit messages that make a very little sense
2. has the source files sprinkled with duplicate looking code having enough differences to screw my visual diffing ability
3. has too many changes combined when they can be independently reviewed and merged
4. includes me as the sole approver when I don't have enough context on the changes
5. includes references to tickets without any description, justification for the change, testing methodologies, test results, performance metrics comparison, etc. Literally none. It is as if the developer wants me to work with them from the Beginning1
NIGHT MARE? MY CODING NIGHT MARE IS USING IPAY88 for the project. Ipay88 is the well-known payment integration in my country but their service is fucked up, I wrote an article of comparing Ipay88 and stripe. IPAY88 is fucked up. So for those who would like to programme a backend system or application to Southeast Asian customers , please don't use ipay88! use alternatives.
Read it here https://johnmelodyme.wordpress.com/...2
Having only 1 colleague who can approve them and they are a lazy ass, always taking 2 days for revievs as hard as… removing an empty line and fixing a typo.
How tf did they give this person a degree.