-
Notifications
You must be signed in to change notification settings - Fork 18.4k
Warning for everyone who use the CSV files. #1553
Comments
this dataset is becoming unusable every day they change something |
@devBaldio looks like the changes were reversed. |
Just out of curiousity, I'm wondering how a date format change would be so inconvenient for a user. Are y'all bringing the data in thru Excel? Are you processing it with some library that can't recognize either/or date format? Not trying to be a smart Alec, just investigating if I can be some assistance to you. Thank you. |
@davefer Hey, we automatically scrape the data. In the code, it makes an object from the date text, so any change in the format makes it not working. Our website is downloading the data directly from this github on client side, so we dont have to do any server stuff. So when CSSE changes the format, it throws an error in the browser, because it can't parse the date. This makes our website not working, at all. |
@kukosek, if you improve your code to parse dates, then you'll be fine. Took me about 7 minutes to adapt to the new format, and I welcome the extra data it provides. Would have taken me longer to complain about it. |
@peterdrier Well I actually did it, and they rolled it back just as someone said here after a few minutes.. You know, its pretty stressful to continuosly check if your site is working because JHO is changing their formats like every day which is so unnecessary. |
@kukosek, Takes me <5 minutes each day to pull in the new file. They're not changing their file without reason, I am thankful for their efforts and all the work they've invested. I find it better to be supportive and build my own thing if I want a change, than to criticize their work.. Chances are they're like 2 people working 80 hr weeks to get this done on top of whatever else in their life is going on. Give them a break, they're not intentionally trying to stress you out. And the don't work for you. And maybe, just maybe, you should find somewhere else to spend your time if parsing dates causes you stress... just sayin. |
@peterdrier, what shoud I say... this is becoming stackoverflow. I appreciate their work, but this isn't a reason to shut my mouth about issues that I and many other people are currently dealing with. I know it must be very hard work for them and it seems they can't handle perfectly (probably a small team as you say). I just don't see a good reason why should they change the filenames (in time series), have different time format in one file than the other (time series recovered vs confirmed/deaths). Change my mind, maybe there is a reason that i don't know about. |
@kukosek, So they're not likely seasoned developers making these files. Looking at the types of changes, I can literally see them figuring things out slowly and without enough coffee.. And yea, people are complaining instead of being constructive. You want it better, perhaps start off with what could be better and why that would help. And if a change they made is obviously for the better, 3/24/20 -> 3/24/2020, don't complain about it, just let it be. Gosh, just wait till they start using 2020-03-24... |
Sorry but |
They've just changed the year format (ex: not 3/24/20 anymore but 3/24/2020).
The text was updated successfully, but these errors were encountered: