Open Side Menu Go to the Top
Register
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** ** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD **

12-20-2018 , 12:12 PM
Quote:
Originally Posted by well named
JIRA tickets are kind of handy for this
Not really. They persist, but once the ticket is "done" you tend to start with a new ticket for the next enhancement so you end up with a handful of tickets to look through to see the conversation on the feature.

If you want to try a different software solution you might take a look at Jama, they are an automated requirements tracking system that maintains comment streams about features. No idea what the cost is or if the team would want to learn a new application more than just writing readme's...
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-20-2018 , 12:22 PM
Yea it’d be nice to have a software solution but if *I* dont even want to do it there’s 0 chance of getting my team to. You wouldnt even believe the resistance to Slack there was.
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-20-2018 , 01:00 PM
If your company is already using Jira then use Confluence.
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-20-2018 , 01:17 PM
Quote:
Originally Posted by KatoKrazy
If your company is already using Jira then use Confluence.
Wikis are where information goes to die
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-20-2018 , 01:20 PM
As you noticed with Slack, remote employees are a forcing function for some communication changes. Maybe that can help (although don't throw the poor remote person under the bus as making people change).
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-20-2018 , 02:27 PM
Quote:
Originally Posted by RustyBrooks
Wikis are where information goes to die
That's different from my experience at the last two companies I have worked at.
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-20-2018 , 03:41 PM
It didn’t so much die where I worked, it just hid amongst half a dozen versions of the products to sow confusion or provide a siren song to people trying to figure out how stuff works.
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-20-2018 , 09:34 PM
i like social intranet software, I've been browsing a few different products. Would cost us like $100/month. Selling it would be hard though.
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-20-2018 , 11:13 PM
If I'm logged out of github or bitbucket, and I logged out of github desktop as well, and I push to origin master from command line on a windows machine how does it authenticate me? (I do a push like this and it works and I'm not asked for username/password)
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-20-2018 , 11:16 PM
The only way it doesnt ask for password is if you’ve set up your credentials on those sites. you need your machines public ssh key associated with your github account. Some IDE probably did it for you if you didnt already.
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-20-2018 , 11:51 PM
Thanks. I found the public/private key on my laptop, but not on my desktop. It's probably in a different place. When I took them out temporarily I had to use password. The commits I pushed had me down with the wrong name (came from my .gitconfig file) and someone was worried that anyone could push without authentication so I was just making sure that wasn't the case.
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-21-2018 , 12:15 AM
How did you end up with a .gitconfig file with someone else’s name in it?
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-21-2018 , 12:25 AM
Quote:
Originally Posted by kerowo
How did you end up with a .gitconfig file with someone else’s name in it?
It was a nickname I used when I set up an account on github.
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-21-2018 , 03:12 AM
[blog]I got **** working on react/react-native/python/flask/mariadb local/linode development stack - a herculean amount of sysad for me[/blog]
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-21-2018 , 03:19 AM
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-21-2018 , 08:51 AM
heres a stackoverflow for Suzz.

so I need validate phone numbers. figure I will use a regex but Ive rarely worked with them. obv SO will have this.

sure enough, first hit on google: https://stackoverflow.com/questions/...ber-validation

the top answer, with almost 500 votes does not provide any sort of regex. instead, it tells the user not to use a regex.

2nd answer is ok

now the 3rd answer, with 279 votes. this one is so ludicrous, that I gotta quote it.

Quote:
If the user wants to give you his phone number, then trust him to get it right. If he does not want to give it to you then forcing him to enter a valid number will either send him to a competitor's site or make him enter a random string that fits your regex. I might even be tempted to look up the number of a premium rate sex line and enter that instead.
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-21-2018 , 09:15 AM
Do you have to account for international numbers?
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-21-2018 , 09:18 AM
its really simple actually, I just have used regex like twice and it was over a year ago.

numbers can be 1112223333 or 111-222-3333. thats it. we arent accepting any other formats. so googling or regex phone numbers was returning way too broad and complicated answers. but I just thought that the main SO was like beyond absurd, like every other SO phone# regex links to that one too.
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-21-2018 , 09:37 AM
Sure so I would say in that case you arent validating phone numbers. You are validating 10 numbers or 12 numbers where the 4th and 8th are "-".

I actually see where the responses are coming from in that case. It can create an incredibly frustrating experience as a user if something like that doesn't work for you.
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-21-2018 , 09:44 AM
I read that SO link and actually thought it was pretty good.

The first post tried to address the root problem with what I thought was actually a pretty good approach. I guess its sort of implicit that they're saying once you have the digits you can validate that its the right number of digits but trying to validate every possible approach of phone number formatting seems painful.

The second post gives a solid answer to the specific question being asked.

The third post is unnecessarily dickish, but also gives a valid perspective on the root problem the user is asking.

The fourth post references a library, a previously unmentioned alternative to solving the root problem the original poster is asking about.

Maybe I'm just old and remember days pre-SO where you would absolutely kill for a selection of answers like this.

Edit: SO is just one case of the general problem that engineers have where people often ask about a specific solution to a problem and there's some amount of tension between helping with or building the chosen solution or trying to dig deeper to the root problem and suggesting a better solution.
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-21-2018 , 10:51 AM
If it's just US numbers why not use an input mask on the text field so the user can only enter 10 digits? The ones I've seen that do this use grayed out text for the parens and dashes. That way you know you're only getting numbers.

As far as validating that it is the correct phone number for the user, I don't find the quoted text that outlandish. Are you ever going to call the user? If not, why bother saving the phone number? If it's part of the service the user will be motivated to get it right. If it's a big part of the service ask them to enter it twice like emails.
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-21-2018 , 11:39 AM
Quote:
Originally Posted by Larry Legend
Sure so I would say in that case you arent validating phone numbers. You are validating 10 numbers or 12 numbers where the 4th and 8th are "-".

I actually see where the responses are coming from in that case. It can create an incredibly frustrating experience as a user if something like that doesn't work for you.
right I realize this. I wasnt criticizing SO for not answering my questions.

Quote:
Originally Posted by jjshabado
I read that SO link and actually thought it was pretty good.

The first post tried to address the root problem with what I thought was actually a pretty good approach. I guess its sort of implicit that they're saying once you have the digits you can validate that its the right number of digits but trying to validate every possible approach of phone number formatting seems painful.

The second post gives a solid answer to the specific question being asked.

The third post is unnecessarily dickish, but also gives a valid perspective on the root problem the user is asking.

The fourth post references a library, a previously unmentioned alternative to solving the root problem the original poster is asking about.

Maybe I'm just old and remember days pre-SO where you would absolutely kill for a selection of answers like this.

Edit: SO is just one case of the general problem that engineers have where people often ask about a specific solution to a problem and there's some amount of tension between helping with or building the chosen solution or trying to dig deeper to the root problem and suggesting a better solution.
I dunno. I disagree with this. the guy asked for a way to regex a bunch of phone numbers and 2 out of the top 3 answers do not give any way to do that.
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-21-2018 , 11:42 AM
Quote:
Originally Posted by kerowo
If it's just US numbers why not use an input mask on the text field so the user can only enter 10 digits? The ones I've seen that do this use grayed out text for the parens and dashes. That way you know you're only getting numbers.

As far as validating that it is the correct phone number for the user, I don't find the quoted text that outlandish. Are you ever going to call the user? If not, why bother saving the phone number? If it's part of the service the user will be motivated to get it right. If it's a big part of the service ask them to enter it twice like emails.
if it was up to me I would do it a lot different.

but business does not want input masks or front end validation. they want the server to handle it and they want to allow the user to put in any crap they desire and hit the send button and then the server gives back the message.

I am just going to use C# attribute with a regex and an error message.
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-21-2018 , 03:03 PM
This seems kinda cool as a starting guide if anyone wants to learn more about using Go as a web server: Example web app with Go and Vue.js

Demo page
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote
12-21-2018 , 05:38 PM
Ok you *never mutate anything* nazis - here's one for you. Basically I get a user object back from Cognito that looks like this:

Code:
{
  "Username": "a0ca870d-a77b-4222-978b-6f7a0ced58cb",
  "Attributes": [
    {
      "Name": "custom:BlackBaudId",
      "Value": "SomeLookupId"
    },
    {
      "Name": "sub",
      "Value": "a0ca870d-a77b-4222-978b-6f7a0ced58cb"
    },
    {
      "Name": "email_verified",
      "Value": "true"
    },
    {
      "Name": "email",
      "Value": "joe@gmail.com"
    }
  ],
  "UserCreateDate": "2018-12-18T01:22:30.423Z",
  "UserLastModifiedDate": "2018-12-18T01:22:40.370Z",
  "Enabled": true,
  "UserStatus": "CONFIRMED"
}
But that's kind of painful to always be looping or plucking to get access to Attributes like email. So I want to massage it into something more useful at the point of extraction - like this:
Code:
{
  "Username": "a0ca870d-a77b-4222-978b-6f7a0ced58cb",
  "Attributes": {
     "custom:BlackBaudId": "SomeLookupId",
     "sub": "a0ca870d-a77b-4222-978b-6f7a0ced58cb",
     "email_verified": "true",
     "email": "joe@gmail.com"
   },
  "UserCreateDate": "2018-12-18T01:22:30.423Z",
  "UserLastModifiedDate": "2018-12-18T01:22:40.370Z",
  "Enabled": true,
  "UserStatus": "CONFIRMED"
}
If I'm allowed to mutate my argument, then I can make a method like this:
Code:
const processUser = user => {
  const attrObj = {};
  user.Attributes.forEach(attr => { attrObj[attr.Name] = attr.Value; });
  user.Attributes = attrObj;
};
If I am not allowed to mutate my argument, then I have to use either something like lodash deep clone, or the JSON.parse(JSON.stringify(user)) trick - which isn't the best for performance or code simplicity. Also the JSON trick doesn't work when the object could have functions or circular references - which we don't have to worry about here, but could in other cases.

Thoughts? Am I approaching this wrong?
** UnhandledExceptionEventHandler :: OFFICIAL LC / CHATTER THREAD ** Quote

      
m