One of our web developers reported a bug with my image api that shrunk large images to a thumbnail size. Basically looked like this img = ResizeImage(largeImage, 50); // shrink the image by 50%
The 'bug' was when he was passed in the thumbnail image and requesting a 300% increase, and the image was too pixelated.
I tried to explain that if you need the larger image, use the image from disk (since the images were already sized optimally for display) and the api was just for resizing downward.
Thinking I was done, the next day I was called into a large conference room with the company vice-president, two of the web-dev managers, and several of the web developers.
VP: "I received an alarming email saying you refused to fix that bug in your code. Is that correct?"
Me: "Bug? No, there is no bug. The image api is executing just as it is supposed to."
MGR1: "Uh...no it isn't. Images using *your* code is pixelated and unfit for our site and our customers."
MGR2: "Yes, I looked at your code and don't understand what the big deal is. Looks like a simple fix."
<web developers nodding their heads>
Me: "OK, I'll bite. What is the simple fix?"
<MGR2 looks over at one of the devs>
Dev1: "Well, for example, if we request an image resize of 300, and the image is only 50x50, only increase the size by 10. Maybe 15."
Me: "Wow..OK. So what if the image is, for example, 640x480?"
MGR1: "75. Maybe 80 if it's a picture of boots."
VP: "Oh yes, boots. We need good pictures of boots."
Me: "I'm not exactly sure how to break this to you, but my code doesn't do 'maybe'. I mean, you have the image from disk.
You obviously used the api to create the thumbnail, but are trying to use the thumbnail to go back to the regular size. Why not use the original image?"
<Web-Dev managers look awkwardly towards the web devs>
Dev3: "Yea, well uh...um...that would require us to create a variable or something to store the original image. The place in the code where we need the regular image, it's easier to call your method."
Me: "Um, not really. You still have to resolve the product name from the URL path. Deriving the original file name is what you are doing already. Just do the same thing in your part of the code."
Dev2: "But we'd have to change our code"
Mgr2: "I know..I know. How about if we, for example, send you 12345.jpg and request a resize greater than 100, you go to disk and look for that image?"
<VP, mgrs, and devs nod happily>
Me: "Um, no that won't work. All I see is the image stream. I have no idea what file is and the api shouldn't be guessing, going to disk or anything like that."
Dev1: "What if we pass you the file name?"
<VP, mgrs, and devs nod happily again>
Me: "No, that would break the API contract and ...uh..wait...I'm familiar with your code. How about I make the change? I'm pretty sure I'll only have to change one method"
VP: "What! No...it’s gotta be more than that. Our site is huge."
<Mgrs and devs grumble and shift around in their chairs>
Me: "I'm done talking about this. I can change your code for you or you can do it. There is no bug and I'm not changing the api because you can't use it correctly."

Later I discovered they stopped using the resize api and wrote dynamic html to 'resize' the images on the client (download the 5+ meg images, and use the length and width properties)

You know who sucks at developing APIs?

Facebook.

I mean, how are so high paid guys with so great ideas manage to come up with apis THAT shitty?

Let's have a look. They took MVC and invented flux. It was so complicated that there were so many overhyped articles that stated "Flux is just X", "Flux is just Y", and exactly when Redux comes to the stage, flux is forgotten. Nobody uses it anymore.

They took declarative cursors and created Relay, but again, Apollo GraphQL comes and relay just goes away. When i tried just to get started with relay, it seemed so complicated that i just closed the tab. I mean, i get the idea, it's simple yet brilliant, but the api...

Immutable.js. Shitload of fuck. Explain WHY should i mess with shit like getIn(path: Iterable<string | number>): any and class List<T> { push(value: T): this }? Clojurescript offers Om, the React wrapper that works about three times faster! How is it even possible? Clojure's immutable data structures! They're even opensourced as standalone library, Mori js, and api is great! Just use it! Why reinvent the wheel?

It seems like when i just need to develop a simple react app, i should configure webpack (huge fuckload of work by itself) to get hot reload, modern es and jsx to work, then add redux, redux-saga, redux-thunk, react-redux and immutable.js, and if i just want my simple component to communicate with state, i need to define a component, a container, fucking mapStateToProps and mapDispatchToProps, and that's all just for "hello world" to pop out. And make sure you didn't forget to type that this.handler = this.handler.bind(this) for every handler function. Or use ev closure fucked up hack that requires just a bit more webpack tweaks. We haven't even started to communicate to the server! Fuck!

I bet there is savage ass overengineer sitting there at facebook, and he of course knows everything about how good api should look, and he also has huge ass ego and he just allowed to ban everything that he doesn't like. And he just bans everything with good simple api because it "isn't flexible enough".

"React is heavier than preact because we offer isomorphic multiple rendering targets", oh, how hard want i to slap your face, you fuckface. You know what i offered your mom and she agreed?

They even created create-react-app, but state management is still up to you. And react-boierplate is just too complicated.

When i need web app, i type "lein new re-frame", then "lein dev", and boom, live reload server started. No config. Every action is just (dispatch) away, works from any component. State subscription? (subscribe). Isolated side-effects? (reg-fx). Organize files as you want. File size? Around 30k, maybe 60 if you use some clojure libs.

If you don't care about massive market support, just use hyperapp. It's way simpler.

Dear developers, PLEASE, don't forget about api. Take it serious, it's very important. You may even design api first, and only then implement the actual logic. That's even better.

And facebook, sincerelly,

Fuck you.

Ever wanted cheat codes to devRant? Well, that's weird. But here you go, I guess.

Since the avatars do not use any external assets (Such as images), all avatars are generated. To be friendly to people who want to make third-party devRant clients (such as devRantron), avatars are generated server-side, so that the assets don't need to be distributed, and third-party programmers don't need to work out rendering avatars.

But this allows you to cheat a little.

The devRant avatars API works like this: you request a really long URL from the API, specifying the IDs of each cosmetic item the user has active, and it returns a PNG file. But you don't need an auth token to generate an avatar (which makes sense), so the avatar API is essentially a sandbox you can play around with if you have the time and patience.

You can write a really good avatar previewer with this knowledge, and see your avatar with a white tiger, even if you don't have the ++s

M: Me
FAC : Fucking annoying colleague
1.
FAC: Hey how did you set up your microservices?
M: I used docke...
FAC: But docker is hard to setup, i want an easier option
2.
FAC: Which services do you have?
M: I have one service for the api, one with redi..
FAC: Redis is not a service
3.
FAC: Do you use AWS API gateway?
M: No, in set up my ow..
FAC: why would you set up your own? I just use the one from AWS.
4.
FAC: How many instances are you have running
M: I have 5 replic...
FAC: 5 replicas? That's why i hate microservices,they are costly
5.
FAC: How did you divide up your app?
M: Since I am starting, its better to run the monolithic and then break it up lat...
FAC: I knew it,you don't actually use microservices
6.
M:(thinking)* Fucker, if you know it well why are you fucking disturbing me?? *

Great news, our company's has a brand new security-first product, with an easy to use API and a beautiful web interface.

It is SQL-injection-enabled, XSS-compatible, logins are optional (if you do not provide a password, you are logged in as admin).

The json-api has custom-date formats, bools are any of "1", "0", 1, 0, false or null (but never true). Numbers are strings or numbers. Utf-8 is not supported. Most of our customers use special characters.

The web interface is using plain bootstrap, and because of XSS it is really easy to customize everything.

How the hell this product got launched is beyond me.

"How am I supposed to use this API? Do you expect the client to open that black thing [terminal and curl]? Why doesn't it work in the browser with clicking?"

🤷‍♀️

- Apparently, I do frontend now.

Client: "Hey we want you to integrate your product with our system."
Me: "Oh, OK. Where's your API?"
Client: "Here! We even have an outdated .Net SDK, we use XML."
Me: "Ok.. how do we authenticate? What's your OAuth 2.0 endpoint?"
Client: "O auth what?"
Me: " You know, the current standard for REST API authentication and authorisation"
Client: " What's REST?"
*Hungs up*

Although I do give some privacy related advice here and there on here, I'm planning on hiring a server dedicated to devRant regarding privacy/tiny simple tools.

I've got the folloing in mind:

- Host the privacy website
- Put a pi-hole server on it for everyone to use
- Own IP lookup API which would display it in a few data formats.

Any other ideas?

Marketing coworker: We MUST integrate with XYZ

Me: sure, they have a REST API and OAuth2. What part of our system do you want to integrate with XYZ?

coworker: Emmm... I don't know. Everything

Me: ...okay. What does XYZ even do?

coworker: Emmm... our customers use it, so we have to integrate with them.

Me: fine. I'm reading that XYZ has function ABC and PQR. Which one do our customers use?

coworker: ABC, definitely ABC

...

coworker: Our customers say nothing is working!

*looks into the problem*
Me: That's because they're using PQR and not ABC.

This kind of research is NOT my job, it's YOURS

Writing more infrastructure than product.

Look, my application requests and transforms data from a single external API endpoint, it's just one GET request...

But I made an intelligent response caching middleware to prevent downtime when the parent API goes down, I made mocks and tests for everything, the documentation is directly generated from the code and automatically hosted for every git branch using hooks, responses are translated into JSONschema notation which automatically generate integration tests on commit, and the transformations are set up as a modular collection of composable higher order lenses!

Boss: Please use less amphetamine.

My devGoals for 2018:
- Build a RESTful API with NodeJS just for learning.
- Finish my first product (electronics sideproject).
- Convert more people to use CraftCMS or at least not use Joomla or WP.
- Get a raise.
- Add Continuous Integration to more projects.
- Add more unit testing where appropriate.
- Create and release a mobile app.

To be continued...
*playing to be continued meme sound*

There is this service that I want to use (wont name it for privacy/legal reasons) and I have created a trial account which gives me a limited access to their apis. However the usage is where things are interesting.
The api access is restricted to some 1000 calls per trial account. But also they have a explorer option which lets to have the functionality as a web app like a dashboard and the explorer usage has unlimited access.

Now since I didnt want to exhaust my api limit, I let my service call the explorer apis instead. Is this ethically wrong or it is the fault of the service providers that they have such a big gaping hole in their licensing?

Manager: "We can't have new releases breaking older versions of the mobile app!!!!! We'll lose all our customers!!!!"

fullStackChris: "That's fine, we can do API versioning, but it will take some time to implement, I'll have to be quite careful and write some tests to implement it. Probably 2-3 weeks..."

Manager: "NO WAY, THAT TIME ESTIMATE IS WAY TOO LONG, WE DON'T HAVE TIME FOR THAT!!!"

fullStackChris: "So how do you wanna support multiple versions of the app without doing any sort of versioning?"

Manager: "...we'll think of something!"

Dev: "..."

And with 99% certainty, I expect to hear this in a week or two:

Manager: "fullStackChris, we'd like to introduce you to the highly technical concept, API versioning. It's a way to version the API so we can support multiple versions of the application our customers use! It's amazing! Please implement this immediately so we can support multiple versions of the application!"

Sigh... each day managers learn a bit more how physical reality works... you can't have your cake and eat it too.

Published a well documented and tested API with project examples for basic use cases

- "Yeah we didn't use yours because we didn't know how to use it"
- "Did you look at the documentation or code examples?"
- "What where?"
- "In the repo you just cloned"
- "Yeah no <random guy> found a hacky way of doing what we want, his thing just works"
- "I..."

I have in my head parts of a game, I'd love to build. A top-down space colonization game. You control various spacecrafts and build spacestations etc. to expand your control.
The catch is that you have limited direct controls in-game. Most of the "gameplay" consists of having your own piece of software running somewhere, communicating with the API and controlling your colony.
Similar to Screeps, but I don't like the limitations of writing JavaScript and uploading it to the service. I'd love for the user to have free control of what language to use. They just need to be able to create a TCP socket and communicate with the API

Not a rant about anything in particular. Just a summary of some feelings stored in the hateful part of my heart.

Developing for Android: Add this third-party library to your Gradle build. Use (this) built-in Android class to make the thing work.

*Clicks link

Deprecated since API version SUCKMYDICK-7. Use (this) instead

*Clicks link

Deprecated since API version LICKMYBALLS-32. Use...

Developing for Windows: Please use (this) API call. It was literally already available before Bill Gates was born. Carbon dating has placed this item to older than the universe itself and it is likely the entry point for the big bang. It is also still the best way to accomplish (task).

Developing for Linux: "Hmm, I wonder how to use this"

> > > Some shitty mailing list in small blue monospace font tells you to reference a man page that is three versions behind but the only version available.

What? Those three sentences didn't explain it enough? Well, maybe you aren't cut out for this type of thing.

JavaScript: you know how it is.

SQL: You expect a decent-quality answer from stack overflow but you always get an outdated and hacky response and it's using syntax from Microsoft SQL. You need MySQL.

C#: A surprising number of Microsoft forum results ranking high on Google. You click on one in hopes that it will be of any sort of quality. You quickly close the tab and wonder why you ever even had hope.

Literally any REST API: Is it "query" or "q"? "UserID" or "user_id"? Oh, fuck, where's the docs again?

You thought you escaped JavaScript, but it was a trick!: Some bullshit library you downloaded to make your other library work redefined one of the global variables in the project you inherited. Now you get 347 "<x> is not a function" errors in your console. Good luck, asshole.

FontAwesome/ Material fonts/ Any icon font pack: You search "Close" for a close button icon. No results. You search "Simplified railroad crossing sign without the railroad". You get a close icon.

I think that's all of my pent up rage. Each of them were too small for an individual rant so I had to do this essay.

Does anyone have a good and free translator api I could use? I’m doing small school project and I just wanted to create text translator app. I tried yandex already, and ibm sucks since it doesn’t allow ajax request

"I decided to write the core software for my Time Machine in Java. It's Date API is so error free and easy to use."

As the legend goes, the scientist is trapped in another dimension between Timezones.

#thanksjava

So I'm looking to buy a drone for my internship company to find people during floods. And damn these companies suck balls.
Closed source.
You want to use API for onboard image processing?
Buy a €3500 drone
Add €1100 processor stuff
Add €850 camera
ugh.

The proprietary API I have to use which uses XML but is totally dynamic and without a schema. They basically managed to successfully merge JSON and XML to get the disadvantages of both JSON and XML.

Web Developer with no common sense: “I’m going to query the currency calculator API for each of the 1000 records to convert.”.

Web Developer with common sense: “I’m going to query the currency rates API and use the calculation to convert each of the 1000 records.”

When can we expect to have sex robots that can also be used for domestic chores? I don't want to invest in a robot that only gets occasional use.

And a Roomba isn't very erotic.

If anyone is working on this, GraphQL API will be handy. I'm going to be doing a lot of things with it that you can't even start to imagine so needs to be fully configurable.

Can use Raspberry Pi...I don't need it to be too bright.

New project at work involving Google Nest Hubs, supervisor asks me to do the initial setup of one of them to start developing with it using its API.

I start looking throughout the documentation and realise that we need to setup a work google account in order to register the devices, pay a fee and only THEN be able to use Google's API for Smart Devices (damn, you, Google!). Supervisor is somewhat baffled by this, and in my head I'm also surprised by his reaction. I'd assume you'd research your devices before you buy them, right?

Later, he comes into the room I'm in (I'm still allowed to work on location), looks at the freshly setup Nest Hub, saying "wow, this sure is a much smaller screen than I was expecting". I mean, you did research these devices before you bought TWENTY of them. RIGHT?!

On my way to fight with this Google device-registration-API-thing now.

To be continued...

Finally I understand what it is like to use a language that does not have any extra resources like stackoverflow or tutorials.

Using ReasonML for my final year project and the only resource I have is the API doc ☹️

While working on my one of the first project in java i ended up using deprecated Calendar API for the date. Since deadline was near i thought it would be a good idea to use the JCalendar API for as date picker (which is a third party API).BAD IDEA. It was the night before the submission round about 11pm when i realized that there is no way to convert JCal object into Calendar and it turned out it is not working as expected you have to subtract a particular number from the year to get date right.

To convert JCal into Calendar i used the toString function to get the date in string sliced it using substring into year,month,day then had to assign date to Calendar object via constructor.
Had to write 70 lines of code just to convert JCal into Calendar...

And then there were other complications related to this problem. Had tu pull an all nighter just to solve date related problems
LESSONS LEARNED :
NEVER USE A DEPRECATED API
NEVER USE THIRD PARTY APIs WITHOUT RESEARCH

Client:
So we need to implement this entire eCommerce API so we can use this webshop in two days.

Me:

I used to work in a small agency that did websites and Phonegap apps, and the senior developer was awful.

He had over a decade of experience, but it was the same year of experience over and over again. His PHP was full of bad practices:

- He'd never used an MVC framework at all, and was resistant to the idea, claiming he was too busy. Instead he did everything as PHP pages
- He didn't know how to use includes, and would instead duplicate the database connection settings. In EVERY SINGLE FILE.
- He routinely stored passwords in plain text until I pretty much forced him to use the new PHP password hashing API
- He sent login details as query strings in a GET request
- He couldn't use version control, and he couldn't deploy applications using anything other than FTP

You have to integrate company A system with company B.
You look up B`s api documentation, write the necessary code, test it thoroughly. Done
__________________________________
A wild dev from B has appeared!!!
------------------------------------------------------

him > you cannot use this SDK to implement this solution. You must use a different one...

You > (coming back to him after scouring through the docs) . Ok which sdk is it then? I can't find it anywhere. Could you send it to me pls?
him > I don't kbow, my knowledge doesn't go that far. Send an email to the dev department.

After sending the email, and spending a couple of days trying to figure out which sdk was he referring to. His department answers that the sdk you should use, is the one you already used. The only one they have.

A couple of days lost, because certain smart-ass could not refrain himself from meddling in the project.

What would you do?

Note: Killing him is out of the question :)

I created a REST API for a customer that one of their customers should use. I sent some documentation and code samples to the "developer". He didn't understand why he should send css to the API. He obviously couldn't tell json from css...

Threaded API Calls, or rather:

How I killed five server instances in one shot..

Creating 50000 Threads sounds like a good idea in general, the bad part starts when you use start() in a for loop to run them all at once.

No further comments your honor...

Create an web application(a product of our company). Manager insisted on using X third party API instead of Y third party API.

Now the app is complete.
*Very slow, Shitty User Experience *

I feel shameful for creating such a disaster and also wasted 3 sprints on it. But couldn't do anything because I don't have an authority to take decision which API or technology stack I should use.

Business head and manager had a meeting. Now they want to use Y third party API.

So they called me for discussion, lets me know we will now use Y third party API and it should be completed within a week because we just need to change the API calls in code that's all, despite of knowing all the core logic is built around those API.

Don't know how to react to this :( Want to hit my head on a wall

Why do simple errors take the longest to fucking find!

Was using the geolocation api (js) to get the current longitude and latitude of my location. Stored them in an object to use in a fetch(). Every time I ran the fetch it was giving me the wrong location!

1hr later I realized I had used.

Fetch(https${longitude}blablabla${longitude})

After realizing this mistake and everything I that lead up to that moment I closed my MacBook and took my ass to sleep.

Moral of this story is...take fucking breaks.
Goodnight

Me: why are we paying for OCR when the API offers both json and pdf format for the data?
Manager: because we need to have the data in a PDF format for reporting to this 3rd party
Me: sure, but can we not just request both json and PDF from the vendor (it’s the same data). send the json for the automated workflow (save time, money and get better accuracy) and send the PDF to the 3rd party?
Manager: we made a commercial decision to use PDF, so we will use PDF as the format.
Me: but ...

The company I am currently working for is partnering with another startup. Nothing special about that. We should integrate their API into our system. I wasn't involved in the process when it came to checking there API and if it would work with our Systems. The Person who did that already left the company so I was left behind with some internal documentation. In that Documentation is already written that API is basically trash....

After I started integrating the API I found more and more flaws in the design. They are not sending any responses that would help, when a param is missing or the authentication isn't correct, only 500's . I got some documentation from the partner company so i thought it will be fine as long as the Documentation would be accurate. Turns out the documentation isn't even close to be up to date. Wrong content types wrong endpoints, wrong naming. Basically we could not work with that. We shortly contacted the partner Company. After a few WEEKS we got a response that they updated the Documentation what was right but still not everything was correct. At this point I lost my mind. I researched a little bit about them, the company is founded from 2 young people who basically came strait out of the University and doest have any experience or idea how to build an API. I investigated a little bit there websites.

They have an Admin panel on the base domain from their API but it is only accessible via HTTP. Like WTF , They use HTTP for an Admin Panel this must be a joke right?

They use Cloudflare without a HTTP to HTTPS redirection ???

I really had not that much time to research in there website but if I find these things in 5 minutes I don't want to know what I can find in like an hour.

At the end we will still use them as partners because surprise surprise our company already sold the product that uses their API.

I know that I will be the person who has to help fixing this shit when it breaks and it will break 1000% JUST FUCK THIS SHIT. FUCK THE PARTNER COMPANY. FUCK THERE API.

So I wrote an application that loads data from a 3rd party API. It allows the user to enter a record locator number and pull it up. By design, the value can be a partial match and it will pull up the record still.

The first API call I make only took 2-3 seconds, so I didn't see an issue as it's loading most of the data the app needs. I keep the filters/fields as they are and move on.

Fast forward 6 months. The user is complaining that the records are taking 30-45 seconds to load. Sure enough, load times are terrible. I've made lots of changes to what fields I'm loading through the API, and I'm calling several additional APIs, so I start pulling pieces of code out to see if anything improves. They all barely make any difference--still 30+ second load times. I end up removing everything except the first API call I developed that was taking 2-3 seconds before. Still taking 30+ seconds.

The 3rd party API allows you to filter using "starts with" or "contains". I used "contains" initially and had no issue, but I decided to try "starts with" since it should fit most use cases.

Load time is less than one second. I add back everything else. Load time is just over a second.

It seems that the 3rd party updated the API and multiplied load times by 10 when using that particular filter. I spent almost an hour on this since the platform doesn't support performance or debugging tools very well, and it all came down to a one line fix.

The GitHub graphql API is pretty neat, mostly because it's a great example of a product where graphql has advantages over REST. As a code reviewer for repos with hundreds of simultaneous PRs, I use it to filter through branches for stuff that needs my attention the most.

NewRelic's NRQL API is also quite nice, as it provides an unusual but very direct interface into the underlying application metrics.

I'm also a big fan of launchlibrary, purely because I love spaceflight, and their API is an extremely rich and actively maintained resource. This makes it a great data source for playing around with plotting & statistics libraries — when I'm learning new languages or tools, I prefer to make something "real" rather than following a tutorial, and I often use launchlibrary as a fun and useful data backend.

Have to use a 3rd party API which responds to all requests like

{
status: 200,
data:{
status: 'fail' / 'pass',
data: { data}
}

Should I be sad?

P.S. They ask for a 'userName'

>be my team
>developing a mobile app
>I'm responsible for developing a "RESTful" API to interface communications between the app and the database
>there's also an "admin" web application which the client themselves will use to manage some shit in the database
>I've developed the API, it works with the mobile app
>instead of just making it simply a front-end app that makes requests to the API like the mobile app does, the guy responsible for the admin app completely ignores my API and implements his own with a certain messy dollar symbol language and a certain bloated piece of server software, accessing the same database directly, and does some operations in his own special way that will break what I've implemented
>now data inserted via admin app is inaccessible to the server API, and I'm expected to "fix" my code so it's consistent with this guy's shit, but the only way to do it is introducing interdependency between the actual API and the admin app's back end

Fuck my life, now I'm the one responsible for the app being broken because no way the guy who's used to kludging unmaintainable shit together fast would ever fuck anything up

Tries to use SoundCloud API for a client

Docs say you need a client key

Wants to create one by signing up clients application

Signup-Form says:applicazion registration currently not available

Goes to soundcloud dev forums

Raging devs rage about that soundcloud has terminated their api registration for about 13 months now

Me thinking: That's probably the best way to make a conpany grow!

Managed to reduce a file from 1155 to 288 lines just by converting the 50 or so methods all calling the same API library with similar setups to use one line helper functions.

Amazing API's and SEX!

Alright people, now that I've got your attention, I'm getting to the point where I need to plan and roll out a solid API for a project. So after reading a lot of the horror stories written here, What are the finer points of what makes an amazing API experience to use and integrate with over a poor one?
And don't say documentation (If you do explain why) 😁

Working with Android DatePicker is such a pain in the ass.

You want to have your DatePicker appearing as a SpinnerView? Well, easy!

If you're under API 21, you can use the following method 'setSpinnerViewShown()'. If you're between API 21 and API 23 you need to add some style configuration. And if you're above of API 23 you can't use both of the methods above, you need to create a custom xml with the attribute "datePickerMode" (no, datePickerMode can't be set programmatically, it would be too easy to guess).

If you want to add a listener to it, you think it might be a method called 'setDateChangeListener' or something like this? Well no! You must use the 'init(year, month, day, Listener)' method, logic!

If you think you're finished with this bullshit, of course not. Their is a known bug on API 21 that you must take into account (but this bug isn't fixed, no, it's just documented somewhere on google forums).

I don't know the team that designed the DatePicker for Android, but it might a team of champanzee that randomly changed their minds to the phases of the moon!

Am I allowed to use an API from the government that they do not have publicly documented or explicitly said anyone can use BUT don't have any authentication on it?

I assigned a new task to an intern who has been with us for a month. He was supposed to prepare the testing environment and test the Geolocation API. When it works, then he can start integrating it with our platform and everything.

After a week, he emails me to say that he thinks the Geolocation API doesn't work. I was weirded out by that because a lot of people use it. We scheduled a meeting and asked him for a demo of his code to see what the error message is.

Him: *no Visual Studio, no code, nothing at all* So here it goes.

Me: ????

Him: *Goes to the API documentation, copies the base URL, pastes it to the browser and hits Enter* See? It says 404 not found.

Me: *literally facepalmed*

Now, he is working on sales management. We totally took him off every software developing projects.

Last week's Android development time breakdown:

21.9% Managing state
17.7% Referring to lifecycle diagrams
15.1% Waiting for Gradle
8.5% Reading the official docs on how to use component x
8.4% Reordering auto-generated ConstraintLayout XML
7.5% Swearing
4.2% Googling “Stack overflow component x is deprecated”
3.9% Googling “Stack overflow implement component x on API 24 or lower”
3.7% Googling “Stack overflow implement component x on API 21 or lower”
3.2% Googling “Stack overflow implement component x on API 19 or lower”
2.9% Googling “Stack overflow callback y called twice”, realising its a feature and not a bug, swearing a lot
2.0% Checking if Flutter is mature yet
1.0% Implementing business logic

Api-docs: Use the query parameter name_pattern to return results that contain name. Otherwise use name to return an exact match
Api: Returns *name* results when using name and everything when using name_pattern without a wildcard

So there’s this SOAP api I have to use (not by choice, and not the only one i have to use) that returns a bunch of XML nodes to confirm the data sent made it and checks out - pretty standard stuff yea.

Now every once in a while it doesn’t respond (as far as I could tell) so today I wrapped a debug around the soap call, error handler and responses and threw a bunch of messages it’s way to try and force it not to respond in order to be able to put some decent error handling in place.

Well it wouldn’t fail.
100 messages .... all responses good
100 more.... all responses good

And then 100 more.... all respond with “x”, plain text not XML as expected!

Wtf is this shit!!!!!

Boss"So, we need to get some data about the users using the APIs from this list of sites."
Me"Alright, sounds feasible enough"

Navigating to first site.

M"Hold on, where's the API?"
B"What do you mean? You're looking at it."
M"This is a website with a search bar, not an API"
B"Same thing. Get to scrapping that data."
M"I-It's written in a JS framework to be reactive in a half-assed way."
B"We need that data"
M"The data is not even consistent!"
B"That's why we need to join it with all these different sources."

The API was a lie. None of the sites had anything remotely similar to an API.

Having to use bloody selenium with chrome driver to scrap all the information because of course, it has to be done programatically every week from now on.

I just hope no captcha of any kind is installed before I finish this project.

Oh dear Apple,

so you're telling me that for the camera API, the requestAuthorization() method returns a bool and for the PhotoLibrary it returns an enum. Why are you going with an enum when there are 2 states and the camera api doesn't use an enum?

(The 2 methods serve the same purpose, the enum exists in the camera api as well and has the exact same states)

And why do I need access to users photos, if I only want to write photos?

Spending all weekend on research and testing for a new feature, finding the perfect framework for it, going trough the API just to come Monday to work and get told by Project Lead: we can't use it, it is not in our stack.

How can it be in our stack!? We've never done anything like that before!

React developers, What do you think about the new hooks api?

Will you refactor your codebases to use functional components or just stick with classes (even though the react community is moving towards removing classes completely)?

I think its awesome as it reduces the bundle sizes if you use function components though. I have been working on an awesome project for a while and I'm being tempted to refactor the whole codebase to use functions instead of classes. What do you think?

We use jira at my company. It's great for me, because no ticketing system's UI is worth a shit, but jira's API is excellent. But we're switching to a new system that is an absolute piece of garbage. Every page is 100% Javascript, so no source can ever be viewed, and the URL never changes to reflect what's onscreen. If you know a ticket number, no URL will ever get you straight to it. You have to navigate multiple slow-loading 25MB piles of Javascript to reach what you're seeking. And most damning of all: the new system has an API, but our highest management is withholding access to it, claiming it breeds laziness.

Is amazing the kind of shit you have to swallow when your management has regular meetings with really really super extremely good-looking sales people.

Facebook is a giant piece of shit. Not only is their platform a massive contributor to mental illness, even their API's are fucking garbage. I'm trying to use their ads API and what it does is it hijacks the entire fucking request so you can't even extract data from the request after calling it. Fuck Facebook and everything they've ever "contributed" to society.

Azure portal is all I hate about Windows in a single, convenient, accessible web page.

Worst part being there are some parts of my work requiring me to deal with it on a daily basis as there is no PowerShell equivalent command, nor any API I could use, to perform some tasks.

Job description: designing and building microservices and API contracts for enterprise use. Deep understanding of api/rest design, AWS, etc.

Interview: in this weird IDE while I stare over you, go through and parse this multi-dimensional primitive array using recursion.

...Wtf does this have to do with the role?

What's the most complex programming task you ever had to solve? Or: A "hack"* of a framework/API you came up with to make something work?

*undocumented or unintended use

I work in a dev company. One of our clients hired us to help them out as their devs are failing with their deadlines.

I had to expose app services via an api. I did it. Client company devs didn't like the way I did it as I rewrote their datamodels and declared them as api-use-only. I was demanded to return bkend services' data structures.

I didn't agree and waited until deadline to submit my code.

Now they are honestly thanking me for what I did as I've saved them from a forever-mutating-api-and-angry-integration-customers hell.

Not sure whether should I be happy or worried. I forced my solution onto them. It's not proffessional. But yhe customer is happier now than it would've been.

What do you do in such situations?

So this morning something wonderful happened. One of my teams deployed some cache for our API using Redis and they didn't config it!!!
So it's TROLL-time!!! And we have some really cool troll rules if you DOS the QA stack!! So if they resolve it in less than an hour they will only have to use a funny cap all day long (even for lunch) but if it takes more than 1 hour it will be a full cyclist outfit for the whole day (over their own cloth).

PS: if you ever use Redis don't mysql it (why config it? it'll work!!!), because Redis will murder your server!!!

!Rant

TLDR: What's your favorite REST API Documentation tool?

I'm about to start developing a really large REST API. I have never really had need to document my previous API's since they were small, self explanatory and had only me using them. Aside from this one being too large for me to keep track of there is also a remote team that will need to integrate with it.

Basically I need write exceptional documentation while using as little time as possible. I love postman and am planning to use it for documentation since I currently use it to test during development anyway but I have seen some really neat looking tools like swagger and apiary so I figured I would check for some other suggestions.

What is your current / preferred REST API documentation method?

step 1) open and browse producthunt for new dev tools to use and try out
.
step 2) opens dev tool/app's website *ooh nice landing page*
.
step 3) tries to find api and documentation, scrolls to bottom of the webpage
.
step 4) "we are still in private beta, sign up to be notified about the final release!"
.
step 5) lol *sighs* bookmarks tab before closing it
.
step 6) repeat step 1
.
.

Making games for my TI-Nspire CX CAS is so much fun!
It's so simple but you can do a lot with it. It's also a bit of a challenge because you don't have a huge API with lots of methods and events. You have to use what you have.
Oh btw you can program it with Lua!

Twitter bot is now up on GitHub, for anyone wondering how to use the Twitter4j API.

Dear API vendor,

Please get off your arse and learn about REST, OpenAPI, JSON Schema, XSD and basic documentation so that I don't have to guess how to use your shitty, inconsistent, RPC over HTTP service.

With Love,

Platypus

>uni project
>6 people in group
>3 devs (including me)
I am in charge of electronics and software to control it as well as the application that will use them.

2 other "devs" in charge of a simple website.
Literally, static pages, a login/registration, and a dump of data when users are logged in.

Took on writing the api for the data as well, since I didn't fully trust the other 2.

Finished api, soldered all electronics, 3d printed models.

Check on the website.

Ugly af, badly written html and css.
No function working yet.

Project is due next week Thursday.

Guess who's not having a weekend and gonna be pulling 2 all nighters

NodeJS 10.0.0 is out.

The biggest features I see is the release of N-API. I always wanted to use that but never did as it was in beta. It will be exiciting to use it with a new project!

Working on a new payment gateway for one of my customers, and it turns out that instead of just specifying the parameters for what to include in the API call they want you to use their drop-in module for it...which is still written in PHP 4 and hasn't been updated since 2011. Also turns out that they only accept data formatted in XML.

Not insurmountable, but more than I feel like dealing with right this moment...

Been writing some Python apps for the last weeks and I really enjoy the language. I got all the basic stuff down but not sure how to progress to learn the more advanced stuff. What even is advanced stuff in Python and where can I find information about it?

I'm thinking about creating an API, any advanced techniques I could benefit or use for that?

thank you @dfox! love the new avatar builder!

is it possible to link those avatars/expose an api to use on other sites?

Third party vendor whose shitty API we have to use as part of our application blocks our account IN PRODUCTION without even telling us every time they have a slowdown because of course it must be us. Yeah, I didn't want to go home anyway🖕

This is the story of the API documentation.
Which btw I couldn't find on the producent's website anywhere. I had the pdf shared with me by a coworker.

I knew the api was fucked up the moment I looked at endpoint documentation.

GET params? WHERE, ORDERBY etc. Literally make a SQL select in a GET request.

Returned stuff? The whole thing. Not some DTO, you literally get everything you can get.
Eg if you get IP in your response, you get it in several formats: dotted form, as hex, and as int. In 3 different json fields.

Oh, and regarding IP - one would imagine you can use masks or prefixes for subnets, right? Nope. The only param you can use there is the subnet size. So you have to calculate the power of 2 every time you want to make a request.

That's from the endpoint documentation. But what about some general info on the API, before all that?

As I was looking for something, I decided to read that intro and general info about the API.

Okay, so there was a change log between API versions. "removed [endpoint which sounds like correct REST design], please use [this generic thing with SQL-like GETs]"... Several of them.

And there was also this sentence which said that the API is not restful, "it's REST-like". <facepalm>

If it was a bad attempt at REST API, I would let it go. But this sentence clearly showed they knew they did everything wrong. And the changelog showed they didn't stop there, they were actively making it worse.

A "partner" company has created a "REST" API we use on an online shop we developed to send all shop-related requests to.
At least once a month, something fails on their end and the customer calls us every time, expecting that we did something wrong, but it has never been us.
These "partners" do exactly zero testing, are extremely slow in solving API bugs, have almost no logging and have no monitoring on the API at all.
Today at noon, suddenly no customers were able to order anything anymore for 4 hours.

How the fuck can you run a business so unbelievably brainless that this keeps repeating monthly?
Time they fire all their "devs" and everyone in charge of the company and operations. TERMINATE.

in 2017 i published my first website. it was basically a remake of google's translation telephone, because google shut it down. unfortunately, the translation api costs money, so rather than pay, i set up a gscript api endpoint that translates it for me.

apparently when you use gscript, translation is free. this was back when i was 14, which is crazy to think about.

Decrypt api responses in an iOS app which my “senior” dev thinks it is more secure to encrypt responses in stead of setting up a proper SSL cert (they use plain http to save money 🙄)

They disable the encryption since it does not function as we wanted and set up SSL instead🙄

When you click on the github wiki of the API you wanted to use and it shows you the "Create the first page" banner....

Guess I'll have to find another API with proper fucking documentation.

We're sorry, but you're not certified to use THAT API Call. Certification takes 6 - 10 weeks.

We have to use a 20 year old API that is half assed and doesn't even work right every time.
Every three months the same discussion comes up why something doesn't work that relies on that API. I have to explain the situation over and over again... And then my boss starts to give 'solutions' which we already use or are utterly stupid... >.<

In case someone is wondering: SOAP API on a Windows Server 2003 with timeouts every few minutes and XML output in a language that is not English (even the tags!).

why isnt there a standardized API for unsubscribing from newsletters that newsletter providers are bound to use? -.-

We are switching to an infinite scroll mode for our app. I told the backend dev in my project (we're just 2 people) I need an API to get more than one post at once.
He told me to use a loop to call the old URL 15 times.
...Not sure if stupid or just too lazy

I spent hours trying to enable CORS on AWS Lambda through API gateway (it was supposed to be simple and Amazon had a nice tutorial) but it turns out that there's a known bug that makes Lambda Proxy Integrations not adhere to any setting in the API Gateway, you have to respond with the headers through the Lambda yourself.

Amazon now mentions this in the tutorial, but if you click "Enable CORS" in API Gateway, it'll show you green check marks and tell you that everything went fine, but you'll find that the Lambda does not respond with the CORS headers. They shouldn't even have "Enable CORS" as an option when you use their Lambda Proxy Integration.

Arghhhh.

So I'm working alongside a very big development house that specialises in vehicle finance. They are huge - 60 plus developers, only work in pairs, have a minimum fee of £2500 per day and work with all the largest car manufacturers.

However, today they decided to completely change their api sub domain and all api routes that handle finance quotes. No warning at.

Of course sites I'm working on that consume this consume the said api are broken too.

To make things worse, my client pYs this company circa £19000 per year to use their api. I also recently discovered that the client is paying for their so called managed service. Insanity.

I mean seriously. This company has 4 layers of project manager and 3 forms of a senior developer. According to companies house their turnover is in excess of £4 million per year.

However, they get the basics wrong and do not warn in advance of major changes to their core api service.

Off I go to deal with 10 of project manAgers and support people in the hope if speaking to someone who can actually help.

If you're not going to update your API documentation please just delete it. I spent two hours trying to use the example only to discover SECRET REST PARAMETERS that solved my problem.

If it weren't for the hero bitching in the comments about the missing documentation I would never have gotten this to work.

FUCKING HELL.

It's already enough that this FUCKING API I have to work with is a mess of JSON and XML responses mixed together.
With mixed german and english keys and attributes all sprinkled over it.
And API access locked to Austria only for some reason.

And then they even manage to fuck up the little bit of JSON they use.

It's just a fucking array of strings (where one could easily use integers).

They can stick this fucking steaming pile of shit that they call API up their PHP infested assholes.

I hate web development sometimes.

I found an api that doesn't offer https when you use their free plan.

FFS, you have to send your api in plaintext... What dumbass of a developer didn't step up to marketing and told them to shut their fucking uneducated mouths

When your client sends a cropped screenshot of the API keys and you spend half an hour typing it out multiple times trying to match it and failing, then another 30min trying to explain the simple process of copying and pasting text.

You can send a cropped screenshot but not use Ctrl-C Ctrl-V or right-click copy right-click paste.

Fuck LinkedIn
Fuck their closed API
Microsoft always steals your PRIVATE info, but is scared of us using their PUBLIC info
Fucking morons
I can just use selenium and still have all the info I need
Then why the fuck would you close API to approved only
Can't you just track the traffic????

Ok so I have done some work with crypto currency mining pools and recently a client requested for me to make a splash page that showed data from multiple instances of these pools APIs. I went to find some documentation for this open source api and to my surprise there is none. I thought of querying the public API from the clients side and it worked, however it's so slow that the data shows up roughly 20 seconds after the page loads.

Easy fix right? Make a PHP server get the data every 5 seconds, cache it and serve the data with the page and use a websocket for live updates! Until I found out that there is no practical way in this garbage framework to get the damn API data without making an HTTP request or mutilating the original source code. I'm so done with this garbage framework. It literally loads pages based on a page and action parameter on the index.php. I quit.

I am writing an api to get data from another api, mix it with some data from a database and then send it back.
I am using nodejs (javascript) to write it but I would like to learn something new. Can you recommend me any languages that I could use for my backend? I was thinking maybe go but I am open for ideas.

When the API you are supposed to use for fetching data returns another item when the item you were looking for is not found! Dudes WTF? Ever heard of 404 or something

So we were making android application for our college festival.. we decided to use Firebase as our primary service for "everything". Decided to use Firestore as database as it wouldn't require much web API call, and mainly as it had a free plan.

We thought that we would never hit the limits of free plan... Needless to say, we started hitting the daily limits about a week before the fest. And it became more of an issue a few days before the fest when we started to hit the limits within 4 hours of the day!!!

But we were lucky enough that the app sustained on the day of festival, lucky enough!!

What if I told you that algebra, geometry, precalculus's (ect) laws are the equivalent of api/documentation. we learn them and use them when necessary to solve problems. I will use the quadratic formula in math to solve a problem just like I would use a particular function in programming.

It's a really interesting discussion, when your boss tells you that it's a perfectly fine idea to directly use a Firebase DB from an Angular web app by storing the Admin Auth Token in a variable in JS.
Thank the spaghetti monster, I was able to argue against it and use the already partially implemented RESTful API with the already used auth.
He basically wanted to save time and omit extra login routes.

It's OK to save time and not implement $randomFeatures.
BUT DON'T FUCKING TRY TO SAVE TIME ON SECURITY!

If it wasn't for me, this web app would turn into a bigger gaping (security) asshole than Sasha Grey's...

I'm new to javascript and I'm doing the freecodecamp. I was doing a challenge and I was looking everywhere how to fix my problem (I needed to use an API from another page) turns out all I needed to do was add the Jquery library, I've been stuck with this for HOURS FML

Android, the development side.

First it was cool to put stuff together and then i wanted to actually use the phone hardware and realized that the api is terrible and abstracted away in the worst way possible.

Like every java dev would make something like new Camera().photo("penis.jpg") and let the gc take care of the rest but nooooooooo you need persisted objects and datastreams and special permission checks.

Getting a location in android is so complicated:
First there's the permissions. Ok add it to the manifest. Oh wait, run-time permissions.
Gotta check if user has allowed the specific app to use location or ask for the permission.
Ok. That's done. Why am i not getting a location? Of course, user can turn it off from settings. Gotta check for that aswell. Or ask for it somehow.
Finally i should be able to get the location! Now, how to I use the Location service to get location in the most efficient way that suits for me? Or should I use the Google api.
Every answer in stackoverflow uses a different method. Oh well, gotta try out them all :).

My vocabulary is way to small to express my feelings when being forced to use .Net 4.0. Just spent like 2 hours searching why my Api requests failed.
Turns out it used TLS 1.0 which got rejected by the server. Then I spent another 2 hours finding out how to make it use TLS 1.2. Surprisingly it does work now (although it came out before TSL 1.2 specification). But yeah still a fucking pile of shit.

Public REST (-inspired) API. Should I skip numeric IDs because it's easy for consumers to snoop around?

Example:

POST api/foo
201 Created api/foo/69

Uh, I'll get 68 just because I can. Hopefully it returns Unauthorized, unless we some kind of bug.

Is it just security by obscurity if I use, like, guids or something instead of sequental IDs?

Why React is soo complicated? Seriously why do we always need to use some stupid shit like this

const [state, setState] = useState();

what the fuck is this? Who designed it that way, why it should always look like shit for no reason? Why its API feels like spaghetti? When i am comparing other front-end frameworks they feel like heaven in their structure.

AND WHAT IS WRONG WITH ME HAVING A QUOTE ON A PARAGRAPH ELEMENT, WHY DO I HAVE TO USE FUCKING HTMLENTITES FROM 2000?

FUCK&apos;ERE

A Bible app for Android called Tour the Bible. It uses the Google maps API to put a pin in each location as you read so you can follow along and virtually "tour" the Bible.

I had a basic version built but then I switched to Kotlin then I pretty much completely rewrote it to use an api for the Bible content instead of plain .txt files. I still try to work on it in my spare time but with a family, full time job, full time school and part time freelancing I don't get much spare time.

I would like to present new super API which I have "pleasure" to work with. Documentation (very poor written in *.docx without list of contents) says that communication is json <-> json which is not entirely true. I have to post request as x-www-form with one field which contains data encoded as json.

Response is json but they set Content-Type header as text/html and Postman didn't prettify body by default...

I'm attaching screenshot as a evidence.

I can't understand why people don't use frameworks and making other lives harder :-/

So company x decides that they're depreciating their REST API v1 in favor of v2 which came out like 2 months ago. But I figure: "I'm okay because I use their official Python module". Well v1 went offline two days ago and they still have not updated their own library to work with v2.

I asked on StackOverflow how to use Angular2 and NodeJS together, that's the foundations of MEAN stack. The answer was "use a generator". SERIOUSLY? It's the stack most famous ever and nobody knows how to serve a fucking static folder without using a generator? What if I've already my own generator? What if I want just to add API to my current project? What if I just want to know what fucking is happening under the hood? What the fuck? Seriously? A generator? Oh my God! A generator!

I would like to call out the moron who decided to control docker through HTTP when the maximum time the server can take to finish the task is longer than the HTTP timeout.
If you expect things might time out, you don't use a HTTP request. You use a resource and poll it, or Websockets, or possibly SSE.
Shoehorning your API into a frame that obviously doesn't fit doesn't help anyone. Just admit that you don't know what HTTP is and use a regular TCP socket with regular pings.

My lecturer gave us a piece of very buggy software with no API which we use to model computer networks. It literally says in the assignment specification "Your program must run for at least 60 seconds before crashing."

These people are setting us up for failure.

I'm making a devRant API wrapper for Dart. The point? None! Just to get better at the language and mostly make my own life easier. Open source or not, seems like I'll be the only one to use it. Or am I? Is anyone else (I know ewpratten is) using Dartlang?

We have to use a API from a customer that requires the XML we send to be sorted in a certain way.

NodeJS C++ add-on is one of the best and worst thing ever to exist in NodeJS.

Writing a native add-on is such a fucking pain. It's full of inconsistent API. They are trying to fix that by the introduction of N-API. But that shit is still in experimental mode.

I want to use nan but I know that that is also going to be deprecated once the N-API gets stable.

fml

Backend : *breaks feature in prod by changing api*

Me: Feature is broken in prod. Please fix. I was told that the API will not deprecate the old use case.

Backend: ... Fixing it in the backend will take time. Add support for it in the frontend.

Me: I'm not done with the new feature just yet. And it will take some time to have it reviewed and fully tested. Please fix the API.

Backend: .... Well, make a new PR and add support for it first. The new feature can come later.

Me. (-_-) Okay

Sometimes it feels like I'm a code janitor rather than a frontend intern

A few weeks back we ported our PHP Rest API into a couple of Go micro-services.

Incredibly _satisfying_ job.

Requests went from 20+ seconds to ~100-300ms.

There was still one bottle-neck, though, because we had to use most of the old cluster-fork of a database (because no way I'll be able to fix all that in a week).

And ooh, next we're thinking of switching to gRPC. Man, we have the best jobs.

Imagine you're building an app that users can make payments to each other(thats not the main feature anyway), what payment API would you use? that users won't have to leave your app to make the payments, they just supply a kind of payment id from the api and get paid ASAP

Seniority sucks sometimes.

Today I took the time to figure out the API for a shipping website.

I was super proud of this and figured that this could be useful for our application but the guy who has been there for 5+ years was like "no we have used an API for a previous client and caused the application to slow down by 1.2 seconds."

Of course in my mind I'm like, "but what was the API that you used? When I use it, it's running and returning in milliseconds. Surely you're not doing it right."

Of course because of his seniority I'm not going to say that.

It just sucks a lot because I spent a lot of time reverse engineering the API. It would have been nice to have been noticed for that work.

Having a vodka and coke and trying to learn how to use a JSON api on a android app (also learning that). I'm sure it will go well.

Google, please explain to me: Why the fuck would you create a hardcoded requirement in your libraries to use a plaintext json file with credentials to your API?

Credentials which give full access to all of the company email, addresses, cloud services, etc?

And why would you accompany this in your docs with example implementations which read as if they were an intern's first coding project — non psr compliant PHP, snippets of Go which won't compile due to type errors...

I'm starting to become convinced that the whole of the Google Cloud API was actually written by thirteen year old who found their parent's liquor cabinet.

Fuck this I'll build my own Google.

If you have an hard-to-use API, you need a fucking strong documentation. Otherwise, a fucking developer like me will get lost and will spend days and days trying to make it work. Man up that documentation, for God's sake.

Included a widget on a website (JS that adds its own code in a division, like Facebook like boxes). The script added its own CSS file that overwrites bootstrap classes, with !important and at the very end on the <head>. Even worse is that there's no alternative, its a widget from a government association and there's no API. And of course, the client insisted to have it so I had to rewrite some part to use custom classes instead of bootstrap's.

If your customer service is so bad that people talk about trying to use the occult to get in touch with you, your company is shit.

Dev related cause I only want to talk to the customer service because the api docs are that fucking bad.

C/C++ is so fucking unreadable. Who had the cancer idea to use abbreviations for EVERYTHING?? If a word is longer than 3 chars you can be sure those psychologically unstable devs will use a shortcut.
Pointer ==> ptr
test_and_set ==> fucking t_a_s
rollback ==> rb
I don't want to play a shitty word guessing game for every fucking function in your horribly documented api.

Before starting the project team lead told me this is the page of apidoc which will be updated after each new service call so that the communication is clear and i know the parameter of each api call well to use them.

1 month passed and no new service call appeared so i got curious and took a look at his code. He has created 8 f****g nrw calls but did not update the docs.

So, I've decided I'm going to completely redo my Discord Bot. I will probably still use Node.js to program it, but I haven't 100% decided yet. If I do use Node.js, I will probably end up using a different Discord API wrapper for it. I also don't know what I'm going to call the bot. NoVegBot 2.0 sounds too boring, but idk.

A Twitter API issue is allowing people to (once again) bootloop iPhones and iPads with the text parser. This was just submitted to Twitter's staff, I don't see anyone talking about it yet. Disable notifications or uninstall and use the mobile site.

Question: Does using cookies for user session handling hinder the scalability of your backend because all the API's have to live on the same domain. Basically if one API starts to get a lot of request and you want to add another server to off balance the load you would have to add an entire webserver rather than just a small micro webserver with the API running on it mainly because cookies are used to authenticate user request and cookies don't survive CORS request. Am I right or don't know what the hell i'm talking about lol need some opinions I suggested we make all API's micro services and use JWT for user sessions

For my peeps in the RoR arena, did y'all ever felt the need to change from ERB as yout server side rendering engine of choice?

I find it hard to use anything else, i would normally stick to it unless I was using Rails as an API and leave the frontend to React or Vue.

Asking about y'alls opinion because I knew about HAML from a while back. But never really used it and I find Rails with ERB to be really efficient.

Ruby pagebuilding with ERB is really flipping comfortable man.ERB has been my favorite for years.

Currently migrating a project to use Svelte and wanted to see what some of y'all think about Haml or erb. Just for the sake of curiosity. Don't know how many rails users we have in here.

Tired of chasing an elusive architecture and finding good community that helps promote it. Basically:

- Not CRUD
- Not MVC
- More like CQRS; commands and queries represent use cases
- Event Sourced; event log is source of truth, everything else is a cached projection
- Functional Domain Design; not DDD; focus on immutability and simplicity
- Functional in general; less OO
- More focus on domain concepts rather than tech concepts
- Domain can be used through CLI, API, or SDK
- UI is just another client to the API
- Authorization is ABAC, graph-based access control

I'm looking for a fucking unicorn.

So I was talking microservices architecture with some lead techs.
And I started asking how did they combine/connect their microservices.

And despite having a lot, they use HTTP as the main transporter.
So the put some API-Gateway, all inside traffic has to go through it, to connect to the final client.

And I said that I do meshing microservices, and we use Nats as man transporter, so our messages go through UDP and not TCP.

And they freaked out. Saying UDP is too low level and not useful...

My question: if you do microservices oriented architecture, and not SOA, do you use HTTP? Did you use it simply because "it works"?

Trying to use a coworker's new API endpoint and I keep getting an "OAuth2 Bearer Token missing" error, despite triple-checking that I set the Authorization header correctly... finally dig into the source code and I find out that all their endpoints require that the bearer token be put into the request body. The fuck?

I am a android developer. Mostly use java. I'm about to Develop an API for our next app. What should i use : php or python?

It's incredible how many sites don't check / handle localStorage permissions.. they usually just completely break.

That is they use a div to block the view despite the page usually still working ..

There literally is an API to check for permissions..

I'd like to make an open source data registry, with an easy to access API. For example, perhaps I'd like to list all cities and postal codes in France, or perhaps get the number of dogs in Canada. Any ideas how to structure the open source part? How can people contribute? Should it use external APs?

Been wanting to get into OOP and unit testing, haven't found a single semi big project that I can use to study these techniques...

Wrote myself a class when asked to do an API call last week, and I think it looks pretty decent...

Does anyone have a ressource to just see how to do it "properly" on medium sized projects? (100k loc)

Wrote a whole http request script to do direct calls to google woth the whole oauth, which where successfull, and after all this work the request responds with a fuck you, the account does not exists response. Apparently it does not give authorization for service accounts and i will be forced to use the google api request to make this fucking thing work. Fuck google.

There are a few constants in Software Development:

1) The requirements always change.
2) Don't trust input.

Silly me was so naive to ignore 1 and 2 and later I dealt with the consequences.

1) Oh, we have this new API and we're only going to build Google Maps interfaces with it. Nice, easy task. We won't have to address the other parts of the library, wooh! The next day: "Yeah guys, we kinda wanna use the other parts now". Me: sigh.

2) Simple task: I have my API accept CSV files so I can generate graphs out of them. What could go wrong? Provide wrong file? I caught that. Provide completely fucked up and garbled CSV? Whoops.

Many months into the project, we discovered that the client doesn't have a REST API for the data we needed from their side available at all - it's a legacy SOAP service.

I somehow got our Node.JS backend to do SOAP calls and use the XML results.

I'm not sure if I should feel proud or dirty at this point.

I was on Instagram and I saw a boy advertising to sell 5k followers and I was wondering, how do they do that? Can they programmically create and control those accounts? I know they can use the Instagram API to some degree but I feel like Instagram probably has security set up to detect that type of stuff (5k accounts following someone at the same time, etc). Does anybody know? I’m actually really curious

I'm so tired of fs issues with webpack/react. fucking useless piece of shit. I look online and it looks like it's a pain in the ass for anyone actually using a modern stack. Literally just trying to use mailjet's API to send emails from a React app and I've been solving dependency issues for fucking hours because of the MANY node modules it requires. requires fs, dns, tls, and dgram for a FUCKING post request because mailjet makes you use their node package.

The new twitter developer stuff is SHIT.

I need to apply now for my small app I use wanted to build this evening as an automation site project. Can't you cunts just give me access to this api so I can finish my projects.

How the fuck should I describe in 300 characters THAT I WILL BE TWEETING ONE TWEET PER WEEK NOTIFYING ABOUT A NEW EPISODE.

Now I need to fill 80 more characters with some dumb stuff just to be able to apply.....

I can understand why in technical interviews they use those algorithmic questions. It's an incredibly short period of time to assess someone's coding capabilities. BUT can't we find a better way to do it? I mean, I've never implemented dijkstra in a work environment, and I had never heard of someone that faces those kind of problems in a normal day of work.

I may be judging by my limited information, but wouldn't be more useful to actually ask to solve a more plausible problem?

For example, create a microservice that implements this API, send us the GitHub link and the API url.

¯\_(ツ)_/¯

Working on a project, got stuck, thought to call it a night but nooo. Fucking brain decided to come up with a solution 0.0000001 ms right after I shut my eyes. Now I am way too tired to open my laptop back again and I am forced to use Facebook Graph API on my cellphone. FML

For a Web project I usually start with copying a minimal boilerplate node app, find some api I wanna use, copy paste a single example of some data I wanna display and render something on a html page as fast as possible. Seeing stuff on a page motivates me to keep going and helps me figure out the most essential direction forward.

SMS vs telegram bot vs WhatsApp api ?
Do or die question

We're looking to change the domain provider we use to register new client's domains to one that has an API.
So far it looks like it'll either be namecheap or AWS's Route 53.
We're also looking for the same thing with mail inboxes.

Do you have any recommendations / experience with either of these?

I was hoping to find a solution that would provide both the domain registration and the mailboxes with no hosting and accessible via API but I've had no such luck.
(Except for maybe two, but neither looked up to date)

!rant

build me a little tool today, which make use of the spotify API.
Just want to thank the devs of this API. Well documented, good examples, easy to use.
There are some good devs Out there 😊

Who in their right mind would do this / think of this....

Salesforce has the option use their API. Either via SOAP or Rest. At my work we currently use SOAP and I wanted to rewrite that to Rest. Fine, you would say.

Their Rest API uses oAuth, nothing fancy you would think. But those motherfuckers, per default have the option enabled that the refresh tokens you get via the necessary API calls are being marked expired the moment the API gives them to you... Then why the hell give them in the first place.

It took me 2 hours of my life to figure out, why in godsname all my refresh tokens were marked as expired. Fuck you Salesforce, I want those 2 hours back! God fucking damn it... I really fed up with this type of bullshit!!

Hmm. I've been wondering how I'll deploy an api based on a microservice style the smartest way... The general plan was to use salt to setup the base server and install dependencies and add the configuration.. Doing updates would be a git pull and pm2 restart api. I would love to know how you deploy your software ?

Wonderful experience today

I'm scraping data from an old system, saving that data as json and my next step is transforming the data and pushing it to an api (thank god the new system has an api)

Now I stumbled upon an issue, I found it a bit hard to retrieve a file with the scraper library I'm using, it was also quite difficult to set specific headers to download the file I was looking for instead of navigating to the index of the website. Then I tried a built-in language function to retrieve the files that I needed during the scrape, no luck 'cause I had to login to the website first.

I didn't want to use a different library since I worked so hard and got so far.

My quick solution: Perform a get request to the website, borrow the session ID cookie and then use the built-in function's http headers functionality to retrieve the file.

Luckily this is a throwaway script so being dirty for this once is OK, it works now :)

Let me just say that I've been playing whack a mole with a new feature for while now. And it's becoming very tiring.

TLDR; CTO is changing the way we're going to implement this, every other day.

June 1st,
CEO: let's implement feature AAA,
CTO: we're going to have a call with Andy to tell us all about his product that will make this super easy, call will be June 4th.

Days before June 4th,
Me: Researchs product X, makes demo works flawlessly.

June 4th,
Call all good, few tips from Andy. We come to the pricing section of Product X

CTO: this will not work, pricing doesn't fit on our budget, fair enough.

June 7th -11th
Me: research altenative approach. Makes second demo.

CTO: Works good, seems to have too many moving parts, let's have call with Bob to check Product Y. It should make our lifes easier.

ME: Geee, ok let's check it out.

June 14th,

Call with Bob, all good, product has a fair price, stuff is experimental.

CTO: let's use Product Y, and just use what we get from their api now, and worry about changes later.

Me: Hmmm, that's a bit risky, but ok, you the boss, right?, starts again new demo. API doesn't work as documented.

Lots of trial and error to figure out how the api is working now, finally demo works well,

June 17th,

API changed, now it works as documented, (expected as it is experimental), previous demo doesn't work anymore.

June 18th,

Redoing research. inputs are completely different from Product Y now, need to redo all that is working and do and a lot more of research.

Go live is scheduled for end of next week, I hope that the API is stable now, and that I get to go live on schedule.

It is funny to see, that it would probably been the same if we just waited on the API to stabilize, and check the pricing section before choosing a product? Who knows.

Anyways, I actually feel happy that over the years I developed the patience to work with ever changing situations like this one.

fucking internet explorer asks whether to save the json response sent by my api a bug in it.
can't change the response to text/HTML can lead to xss.
why why do clients have to use ie

fucking web hosts blocking all SMTP ports outgoing, forcing me to use PHP mail from their shitty blacklisted IP's.

Since I can't use a web api to send the mail Iended up setting up my home server to forward port 53 back out to the mail server, alot of hassle to get mail working :(

i am indeed working for some fun litte projects with the Dev Rant Unofficial Wrapper API and it looks good to use it! Now. Lets make some Action!

Alright, if you don't make apply a little feature, which you already use in your webpage, in your API, I'll have to fucking parse that shit.

Screw you, I'm sacrificing my free time...

I want to start with Web development and for that I want to code a dashboard for finances with a connection to an Restful API.
I know HTML, CSS, TS and some JS. But I don't know which framework to use.

The framework should:
- have an easy way to separate HTML from JS or TS code.
- easy way to break down a single page into different html files.
- not have to use npm or Node.JS. Preffered is a CDN solution.
- HTML Templating

Maybe also tutorials on how to setup the coding enviroment.

The frontend developers in my company are the reason why I have anxiety. Here are few things that grinds my knees:
1) for a long time in projects, they deleted the auth token from their storage without integrating the logout api. They thought why use an API for that. :)
2) most of them had no clue that form fields could accept javascript as inputs and work as XSS vulnerabilities. This actually happened with a client, he got so fucking pissed.
3) One of them asked me to convert a PATCH request to DELETE cos fuck REST and HTTP methods.

For fuck’s sake. I need to get out of this place.

The c++ 'Eigen' library. It is by far the best library if you're in need of maths constructs, speed wise. And the only maths library that has an API that isn't a pain to use.

posting the question here, because it's quite esoteric.

for the guys familiar with the devrant api: how do i know what value to use for the app parameters? when i use '3' (as suggested) i get an error telling me that the app id is invalid

or do i have to request one now?

I keep forgetting what a massive pain in the ass it is to use dynamic libraries in anything other than C. I'm genuinely contemplating defining a serialization-based API and talking to plugins with a blob->blob function pair.

How the fuck is Firebase still a thing? I just spent hours debugging a random "not authorised" error, only to find out you need to enable a deprecated API even if you're only using the new (recommended) one. Do they tell you about it? Fuck no, they keep it disabled by default, they tell you to only use the new API, and they make it pretty much impossible to find the deprecated API you need to enable without a direct link.

And why the fuck does the official SDK send image URL as { "imageUrl": "http://..." }, when the endpoint expects it to be { "image": "http://..." }? Why the fuck does the documentation mention both options interchangeably, while only the latter one actually works?

So I was planning to use an REST API wrapper library and I included into my app spent over an hour working my logic... No errors... but then when I compile... I get a FUCKING DEPENDENCY COMPATIBILITY ERROR.... My NET Framework app isn't compatible with NET Standard libraries??? WTF.....

"Just start ahead"

I am supposed to transform calls from one api to another one. Yet there's no documentation, ambiguous code statements, no examples of what values are contained -- but sure, let me just start assuming how the whole thing is supposed to work. That won't lead us more into a murky waters at all.

Even more frustrating: We own the api. We should be able to tell by the access logs how we are queried. Yet for some reason, access logs cannot be accessed and I shall "just work from the swagger defintion".

Well, that swagger definition is broken, its example are shit (somebody liked to use undefined in optional fields, making me wonder even more what the heck is going on here), and I have no idea of what I am doing. Fun times.

I have implemented RESTful API using expressJS, and another React app which will use the API's to fetch data.
I'm getting a problem of Allow-Origin Header.
what's the proper way of calling a API ?

do I use a CORS middleware and allow all origin ('*') and use Api-key as way of check authorization to prevent mis-use. ?
any other tricks ?

Learning how to use AWS api getway to secure APIs. I'm enjoying it

I'm worried that this question will probably get downvoted on SO and I cannot post from my account as well, so here it is..

How to play youtube embedded video in mobile over a custom button / svg / image ?

One idea I'm currently working on is to trigger &autoplay=1 onclick but it is only appending that parameter to existing URL?

Please let me know if you have any thoughts, and I'm doing this for only mobile devices only

P.S I cannot use YouTube API

Motherfucking piece of shit tensorflow, I fucking hate this piece of trash API fucking impossible to fucking use

Soo... me and my best friend decided to go literally "wherever" by plane for as cheap as we can (<50€ per person for a return ticket) and I am looking for a service that offers a free API to search for plane tickets as it would make my and her life much easier if I could just write a program, let it run on my Raspberry Pi and make it send me an email whenever it finds some cheap tickets. I noticed that Google lets you search for cheap plane tickets, but is there an API (even unofficial) that I could use?

I'm trying to stand up a docker container to read a storage queue with dotnet and invoke ffmpeg to convert some videos. For a whole day I fought with this wrapper (FFMpegCore) which kept throwing file not found errors on the ffmpeg binary itself. Locally (windows) it worked fine.

I spent a ton of time trying to install the Debian package, trying to add it to the path manually, trying to just use the wrapper just to generate the arguments I wanted (I'm not an ffmpeg pro, so the fluent API the wrapper has is super useful) and running it manually, nothing worked. Finally, I realized it wasn't getting to the part where I ran it manually: just using the fluent API to get the arguments was invoking ffmpeg and throwing.

I took away the wrapper completely, start ffmpeg manually and it works...

Ay carumba. Things just can't be easy.

There is a project: multi step form. It has login form, text field, date, upload. It is written in jQuery, Zend and oracle. I need to rewrite. I can use my familar jQuery, PHP skills. Or I can go for react redux, build API with Zend. The issue is that I only have 2 months. Very basic knowledge of react, redux webpack. So I need to learn and build the project. Should I go for new tech. I am not sure I can reach deadline, as there are many things to learn. Advice? If i use the new tech, how do I learn quickly?

Having a look at electron again after giving up on it months ago.

I must say, the documentation has improved a lot since then, and it looks---dare I say---intuitive to use?

The electron api demos app is surely some help, but I'm not really all that sure how much it lives up to its name. It doesn't really demonstrate anything, and it doesn't cover the whole api, just small chunks of it.

Loving the event system though!

Does anyone use pushover.net or know a reliable / cheap push notification app with an api behind It?

Im trying to setup some monitoring alerts back to my phone, using slack / teams etc for personal projects seems pointless.

Emails are currently being sent but this relies on me actually looking at them and I tend to miss a few important ones. 😂

Not to worried about daily / monthly limits, it's going to be < 20 notifs a day kind of thing on a terribly bad day.

How to protect API endpoints from unauthorized usage by bots?

If the API end points are meant to be used by any incoming to CSR frontend user without prior registration?

So far, my the only idea is going from pure CSR React to something with partial SSR at least in Node.js, Django or any other backend framework. I would be able restricting some API endpoints usage to specific allowed server ip.

Next.js allows dynamically both things as well.

As alternative I have a guess to invent some scheme with temporally issued tokens... But all my scheme ideas I can break really easily so far.

Any options? If SSR is my only choice, what would you recommend as best option in already chosen Django and not decided fully front-end framework?

I have the most crazy idea to put some CSR frontend framework literally into my django backend and making initial SSR from it. The only thing its missing... my lack of skills how to use React, but perhaps I have enough time to get a hang of it.

SSRed frontend can be protected with captcha means at least.

Can someone give me good ideas of useful apps ? Preferably ones that i can use free api(s) ( i want to make it open source) . And please don't say todo app or weather app.

My most recent workaround occurred last week.
We have a demo very soon and I had to change our iOS app to use a new Web API endpoint for uploading content.
Long story short: The existing code is so awful and rigid and dependant on Core Data that I ended up having to completely bypass the service layer of the app and implement the new endpoint as a raw HTTP request. Its gonna take a long time to refactor the existing service layer. All because the new endpoint has a different content type.