After three years since my post saying I'd do PHP and MySQL, I'd now also use C# with SQL Server. I would however use MVC on the client-side such as Angular or jQuery and only use C# for the API. That's if I'm only planning on using the server for the website. Since my VPS runs Ubuntu and I use it for other apps, I'd probably just scrap out C# with either Java or Python (I'd likely go with Python) and scrap SQL Server with MySQL. One thing that is guaranteed is I am not touching PHP ever again.
They do. That's what they came up with in 2007 and kept it since. What they should've done in 2007 and what they should be doing now that's another matter.
I do all programming in Perl/bash with the least amount of JS possible on website pages (I do use d3) The data from the games goes into MongoDB. The data for the website is summarized from Mongo into MySQL.
I think we're both referring to two separate pages. There's that bland HTML page which shows all the game data, which I'm assuming that's what you're referring to. If so, then you're definitely right. I'm talking about the Gamecenter data, which refreshes automatically.
Oh, yes, indeed. I get most of my data from the bland HTML. It does have that self-refreshing JS though.
Is all of that data on the Gamecenter page? If so, there may be JSON files there you could scrape the data from, because those JSONs must be getting updated after corrections.
You need to collect from both. Each has data the other does not. My "holy grail" is a single feed that has everything. Presumably it would be a structured document like JSON. (Also, whoever designed the JSON schema needs to be slapped.)
As I am rewriting my scraper I took a thorough look, and I couldn't find anything but the stuff you listed. One thing I didn't do was to trace the loading of the webpage during the match itself, I will have to do it for the next game on June 5th. It looks like the Roster json is indeed missing (and not really necessary because the HTML roster report is fairly easy to parse). I couldn't find any use for gcbx and gcsb and the Scoreboard and the live content (but my needs might've been different). So for me it all practically boiling down to: * PlayByPlay.json * live feed * PL.html, GS.html, ES.html (TV/TH.html for you, I guess - they do get updated live) Comparing these I find that the live feed doesn't have On-Ice information, PBP.json has it but only has Shot, Hit, Penalty and Goal events, and PL.html has everything but the parser is tougher. The live feed may yield GS/ES unnecessary though.
Looks like the PlayByPlay feed and the Roster feed are not at these addresses anymore. Has anyone figured out where they are now?
Thanks Again, My results for daily scores has been updated, the old source is not available. My attempt to mimic TSN/SportsNet score board... NHL Scores Regards PuckNuts
I just started last night redoing my now-shutdown website which calculated NHL standings using different point systems, such as the three-point system. To start, I made a small .NET Core library whose only method returns the records for each team. It's only parameter is the season, in a format like "20172018". Feel free to use and make any suggestions. Github: GitHub - parkerrappaport/NhlStandings NuGet: NhlStandings 1.0.0.0 Let me know if there are any issues.
I'm looking for a data source for this year's schedule, my old data source either hasn't been posted yet or is not being added for the upcoming season. It was games per calendar day: http://live.nhle.com/GameData/GCScoreboard/<Y-m-d>.jsonp The page with full schedule for the season as a whole also isn't available: http://live.nhle.com/GameData/SeasonSchedule-20182019.json All I need are game ID, teams, day and time. I have a home made goal light, but I need the schedule for games in order for it to work.
https://statsapi.web.nhl.com/api/v1/schedule?startDate=2018-09-21&endDate=2018-12-21 Change the start and end dates to the range you want.
Current URL I am using for NHL game scores. http://live.nhle.com/GameData/RegularSeasonScoreboardv3.jsonp PuckNuts