feat(scripts): rewrite parser as modular Python CLI
Replace monolithic scraping scripts with sportstime_parser package: - Multi-source scrapers with automatic fallback for 7 sports - Canonical ID generation for games, teams, and stadiums - Fuzzy matching with configurable thresholds for name resolution - CloudKit Web Services uploader with JWT auth, diff-based updates - Resumable uploads with checkpoint state persistence - Validation reports with manual review items and suggested matches - Comprehensive test suite (249 tests) CLI: sportstime-parser scrape|validate|upload|status|retry|clear Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
Trey t
245
Scripts/sportstime_parser/tests/fixtures/nhl/espn_scoreboard.json
vendored
Normal file
245
Scripts/sportstime_parser/tests/fixtures/nhl/espn_scoreboard.json
vendored
Normal file
@@ -0,0 +1,245 @@
|
||||
{
|
||||
"leagues": [
|
||||
{
|
||||
"id": "90",
|
||||
"uid": "s:70~l:90",
|
||||
"name": "National Hockey League",
|
||||
"abbreviation": "NHL"
|
||||
}
|
||||
],
|
||||
"season": {
|
||||
"type": 2,
|
||||
"year": 2026
|
||||
},
|
||||
"day": {
|
||||
"date": "2025-10-08T00:00:00Z"
|
||||
},
|
||||
"events": [
|
||||
{
|
||||
"id": "401671901",
|
||||
"uid": "s:70~l:90~e:401671901",
|
||||
"date": "2025-10-08T23:00:00Z",
|
||||
"name": "Pittsburgh Penguins at Boston Bruins",
|
||||
"shortName": "PIT @ BOS",
|
||||
"competitions": [
|
||||
{
|
||||
"id": "401671901",
|
||||
"uid": "s:70~l:90~e:401671901~c:401671901",
|
||||
"date": "2025-10-08T23:00:00Z",
|
||||
"attendance": 17850,
|
||||
"type": {
|
||||
"id": "1",
|
||||
"abbreviation": "STD"
|
||||
},
|
||||
"venue": {
|
||||
"id": "1823",
|
||||
"fullName": "TD Garden",
|
||||
"address": {
|
||||
"city": "Boston",
|
||||
"state": "MA"
|
||||
},
|
||||
"capacity": 17850,
|
||||
"indoor": true
|
||||
},
|
||||
"competitors": [
|
||||
{
|
||||
"id": "1",
|
||||
"uid": "s:70~l:90~t:1",
|
||||
"type": "team",
|
||||
"order": 0,
|
||||
"homeAway": "home",
|
||||
"team": {
|
||||
"id": "1",
|
||||
"uid": "s:70~l:90~t:1",
|
||||
"location": "Boston",
|
||||
"name": "Bruins",
|
||||
"abbreviation": "BOS",
|
||||
"displayName": "Boston Bruins"
|
||||
},
|
||||
"score": "4",
|
||||
"winner": true
|
||||
},
|
||||
{
|
||||
"id": "5",
|
||||
"uid": "s:70~l:90~t:5",
|
||||
"type": "team",
|
||||
"order": 1,
|
||||
"homeAway": "away",
|
||||
"team": {
|
||||
"id": "5",
|
||||
"uid": "s:70~l:90~t:5",
|
||||
"location": "Pittsburgh",
|
||||
"name": "Penguins",
|
||||
"abbreviation": "PIT",
|
||||
"displayName": "Pittsburgh Penguins"
|
||||
},
|
||||
"score": "2",
|
||||
"winner": false
|
||||
}
|
||||
],
|
||||
"status": {
|
||||
"clock": 0,
|
||||
"displayClock": "0:00",
|
||||
"period": 3,
|
||||
"type": {
|
||||
"id": "3",
|
||||
"name": "STATUS_FINAL",
|
||||
"state": "post",
|
||||
"completed": true
|
||||
}
|
||||
}
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"id": "401671902",
|
||||
"uid": "s:70~l:90~e:401671902",
|
||||
"date": "2025-10-09T00:00:00Z",
|
||||
"name": "Toronto Maple Leafs at Montreal Canadiens",
|
||||
"shortName": "TOR @ MTL",
|
||||
"competitions": [
|
||||
{
|
||||
"id": "401671902",
|
||||
"uid": "s:70~l:90~e:401671902~c:401671902",
|
||||
"date": "2025-10-09T00:00:00Z",
|
||||
"type": {
|
||||
"id": "1",
|
||||
"abbreviation": "STD"
|
||||
},
|
||||
"venue": {
|
||||
"id": "1918",
|
||||
"fullName": "Bell Centre",
|
||||
"address": {
|
||||
"city": "Montreal",
|
||||
"state": "QC"
|
||||
},
|
||||
"capacity": 21302,
|
||||
"indoor": true
|
||||
},
|
||||
"competitors": [
|
||||
{
|
||||
"id": "8",
|
||||
"uid": "s:70~l:90~t:8",
|
||||
"type": "team",
|
||||
"order": 0,
|
||||
"homeAway": "home",
|
||||
"team": {
|
||||
"id": "8",
|
||||
"uid": "s:70~l:90~t:8",
|
||||
"location": "Montreal",
|
||||
"name": "Canadiens",
|
||||
"abbreviation": "MTL",
|
||||
"displayName": "Montreal Canadiens"
|
||||
},
|
||||
"score": "3",
|
||||
"winner": false
|
||||
},
|
||||
{
|
||||
"id": "10",
|
||||
"uid": "s:70~l:90~t:10",
|
||||
"type": "team",
|
||||
"order": 1,
|
||||
"homeAway": "away",
|
||||
"team": {
|
||||
"id": "10",
|
||||
"uid": "s:70~l:90~t:10",
|
||||
"location": "Toronto",
|
||||
"name": "Maple Leafs",
|
||||
"abbreviation": "TOR",
|
||||
"displayName": "Toronto Maple Leafs"
|
||||
},
|
||||
"score": "5",
|
||||
"winner": true
|
||||
}
|
||||
],
|
||||
"status": {
|
||||
"clock": 0,
|
||||
"displayClock": "0:00",
|
||||
"period": 3,
|
||||
"type": {
|
||||
"id": "3",
|
||||
"name": "STATUS_FINAL",
|
||||
"state": "post",
|
||||
"completed": true
|
||||
}
|
||||
}
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"id": "401671903",
|
||||
"uid": "s:70~l:90~e:401671903",
|
||||
"date": "2025-10-09T02:00:00Z",
|
||||
"name": "Vegas Golden Knights at Los Angeles Kings",
|
||||
"shortName": "VGK @ LAK",
|
||||
"competitions": [
|
||||
{
|
||||
"id": "401671903",
|
||||
"uid": "s:70~l:90~e:401671903~c:401671903",
|
||||
"date": "2025-10-09T02:00:00Z",
|
||||
"type": {
|
||||
"id": "1",
|
||||
"abbreviation": "STD"
|
||||
},
|
||||
"venue": {
|
||||
"id": "1816",
|
||||
"fullName": "Crypto.com Arena",
|
||||
"address": {
|
||||
"city": "Los Angeles",
|
||||
"state": "CA"
|
||||
},
|
||||
"capacity": 18230,
|
||||
"indoor": true
|
||||
},
|
||||
"competitors": [
|
||||
{
|
||||
"id": "26",
|
||||
"uid": "s:70~l:90~t:26",
|
||||
"type": "team",
|
||||
"order": 0,
|
||||
"homeAway": "home",
|
||||
"team": {
|
||||
"id": "26",
|
||||
"uid": "s:70~l:90~t:26",
|
||||
"location": "Los Angeles",
|
||||
"name": "Kings",
|
||||
"abbreviation": "LAK",
|
||||
"displayName": "Los Angeles Kings"
|
||||
},
|
||||
"score": null,
|
||||
"winner": null
|
||||
},
|
||||
{
|
||||
"id": "54",
|
||||
"uid": "s:70~l:90~t:54",
|
||||
"type": "team",
|
||||
"order": 1,
|
||||
"homeAway": "away",
|
||||
"team": {
|
||||
"id": "54",
|
||||
"uid": "s:70~l:90~t:54",
|
||||
"location": "Vegas",
|
||||
"name": "Golden Knights",
|
||||
"abbreviation": "VGK",
|
||||
"displayName": "Vegas Golden Knights"
|
||||
},
|
||||
"score": null,
|
||||
"winner": null
|
||||
}
|
||||
],
|
||||
"status": {
|
||||
"clock": 0,
|
||||
"displayClock": "0:00",
|
||||
"period": 0,
|
||||
"type": {
|
||||
"id": "1",
|
||||
"name": "STATUS_SCHEDULED",
|
||||
"state": "pre",
|
||||
"completed": false
|
||||
}
|
||||
}
|
||||
}
|
||||
]
|
||||
}
|
||||
]
|
||||
}
|
||||
Reference in New Issue
Block a user