batch insert totals

This commit is contained in:
Ravi Prasad
2025-06-19 22:13:11 -05:00
parent d8e3ff34b6
commit 1a2a4c1b73
3 changed files with 199 additions and 203 deletions
+24 -24
View File
@@ -9,33 +9,33 @@ import (
"github.com/nprasad2077/NBA_Go/utils"
)
// importPlayerAdvanced fetches and stores advanced stats for seasons 20172025
func importPlayerAdvanced(db *gorm.DB) {
for season := 2015; season <= 2019; season++ {
if err := services.FetchAndStorePlayerAdvancedScrapedStats(db, season, false); err != nil {
log.Printf("advanced import failed for %d: %v", season, err)
}
log.Printf("Advanced import for season: %d", season)
time.Sleep(1100 * time.Millisecond)
utils.SleepWithJitter(1000 * time.Millisecond)
}
}
// // importPlayerAdvanced fetches and stores advanced stats for seasons 20172025
// func importPlayerAdvanced(db *gorm.DB) {
// for season := 2013; season <= 2014; season++ {
// if err := services.FetchAndStorePlayerAdvancedScrapedStats(db, season, false); err != nil {
// log.Printf("advanced import failed for %d: %v", season, err)
// }
// log.Printf("Advanced import for season: %d", season)
// time.Sleep(1100 * time.Millisecond)
// utils.SleepWithJitter(1000 * time.Millisecond)
// }
// }
// importPlayerAdvancedPlayoffs fetches and stores advanced stats for playoffs seasons 20232025
func importPlayerAdvancedPlayoffs(db *gorm.DB) {
for season := 2015; season <= 2019; season++ {
if err := services.FetchAndStorePlayerAdvancedScrapedStats(db, season, true); err != nil {
log.Printf("advanced import failed for %d: %v", season, err)
}
log.Printf("Advanced Playoffs import for season: %d", season)
time.Sleep(1100 * time.Millisecond)
utils.SleepWithJitter(1250 * time.Millisecond)
}
}
// // importPlayerAdvancedPlayoffs fetches and stores advanced stats for playoffs seasons 20232025
// func importPlayerAdvancedPlayoffs(db *gorm.DB) {
// for season := 2013; season <= 2014; season++ {
// if err := services.FetchAndStorePlayerAdvancedScrapedStats(db, season, true); err != nil {
// log.Printf("advanced import failed for %d: %v", season, err)
// }
// log.Printf("Advanced Playoffs import for season: %d", season)
// time.Sleep(1100 * time.Millisecond)
// utils.SleepWithJitter(1250 * time.Millisecond)
// }
// }
// importPlayerTotalsScrape fetches & stores scraped regular-season total stats
func importPlayerTotalsScrape(db *gorm.DB) {
for season := 2015; season <= 2019; season++ {
for season := 2013; season <= 2014; season++ {
if err := services.FetchAndStorePlayerTotalScrapedStats(db, season, false); err != nil {
log.Printf("scraped totals import failed for %d: %v", season, err)
}
@@ -47,7 +47,7 @@ func importPlayerTotalsScrape(db *gorm.DB) {
// importPlayerPlayoffsScrape fetches & stores scraped playoff total stats
func importPlayerTotalsPlayoffsScrape(db *gorm.DB) {
for season := 2015; season <= 2019; season++ {
for season := 2013; season <= 2014; season++ {
if err := services.FetchAndStorePlayerTotalScrapedStats(db, season, true); err != nil {
log.Printf("scraped playoffs import failed for %d: %v", season, err)
}
+4 -4
View File
@@ -45,11 +45,11 @@ func main() {
// Run all migrations + import steps exactly once
db := config.InitDB(true)
importPlayerAdvanced(db)
log.Println("🎉 Player Advanced Import completed successfully")
// importPlayerAdvanced(db)
// log.Println("🎉 Player Advanced Import completed successfully")
importPlayerAdvancedPlayoffs(db)
log.Println("🎉 Player Advanced Playoffs Import completed successfully")
// importPlayerAdvancedPlayoffs(db)
// log.Println("🎉 Player Advanced Playoffs Import completed successfully")
importPlayerTotalsScrape(db)
log.Println("🎉 Player Totals (scraped) Import completed successfully")
+171 -175
View File
@@ -3,209 +3,205 @@
package services
import (
"bytes"
"fmt"
"io"
"log"
"net/http"
"strings"
"bytes"
"fmt"
"io"
"log"
"net/http"
"strings"
"github.com/PuerkitoBio/goquery"
"github.com/nprasad2077/NBA_Go/models"
"gorm.io/gorm"
"gorm.io/gorm/clause"
"github.com/PuerkitoBio/goquery"
"github.com/nprasad2077/NBA_Go/models"
"gorm.io/gorm"
"gorm.io/gorm/clause"
)
const (
regularURLFmt = "https://www.basketball-reference.com/leagues/NBA_%d_totals.html"
playoffURLFmt = "https://www.basketball-reference.com/playoffs/NBA_%d_totals.html"
regularURLFmt = "https://www.basketball-reference.com/leagues/NBA_%d_totals.html"
playoffURLFmt = "https://www.basketball-reference.com/playoffs/NBA_%d_totals.html"
)
// urlForSeason chooses regular vs. playoff URL.
func urlForSeason(season int, isPlayoff bool) string {
if isPlayoff {
return fmt.Sprintf(playoffURLFmt, season)
}
return fmt.Sprintf(regularURLFmt, season)
if isPlayoff {
return fmt.Sprintf(playoffURLFmt, season)
}
return fmt.Sprintf(regularURLFmt, season)
}
// FetchAndStorePlayerTotalScrapedStats scrapes BR totals (regular or playoffs)
// and upserts into PlayerTotalStat.
// and batch upserts them into PlayerTotalStat for significantly better performance.
func FetchAndStorePlayerTotalScrapedStats(db *gorm.DB, season int, isPlayoff bool) error {
url := urlForSeason(season, isPlayoff)
req, err := http.NewRequest("GET", url, nil)
if err != nil {
return err
}
req.Header.Set("User-Agent", "Mozilla/5.0 (compatible)")
resp, err := http.DefaultClient.Do(req)
if err != nil {
return err
}
defer resp.Body.Close()
url := urlForSeason(season, isPlayoff)
req, err := http.NewRequest("GET", url, nil)
if err != nil {
return err
}
req.Header.Set("User-Agent", "Mozilla/5.0 (compatible)")
resp, err := http.DefaultClient.Do(req)
if err != nil {
return err
}
defer resp.Body.Close()
body, err := io.ReadAll(resp.Body)
if err != nil {
return err
}
body, err := io.ReadAll(resp.Body)
if err != nil {
return err
}
doc, err := goquery.NewDocumentFromReader(bytes.NewReader(body))
if err != nil {
return err
}
doc, err := goquery.NewDocumentFromReader(bytes.NewReader(body))
if err != nil {
return err
}
table := doc.Find("table#totals_stats")
if table.Length() == 0 {
return fmt.Errorf("could not find table#totals_stats")
}
table := doc.Find("table#totals_stats")
if table.Length() == 0 {
return fmt.Errorf("could not find table#totals_stats for season %d", season)
}
// table := doc.Find("table#totals_stats")
// if table.Length() == 0 {
// commentSel := doc.
// Find("div#all_totals_stats").
// Contents().
// FilterFunction(func(i int, s *goquery.Selection) bool {
// return goquery.NodeName(s) == "#comment"
// })
// 1) Collect the data-stat keys in header order.
var headers []string
table.Find("thead tr th").Each(func(i int, th *goquery.Selection) {
if stat, ok := th.Attr("data-stat"); ok && stat != "" {
headers = append(headers, stat)
}
})
headers = append(headers, "player-additional") // Add the appended-player column.
// if commentSel.Length() == 0 {
// return fmt.Errorf("could not find table#totals_stats (even inside comment)")
// }
// --- BATCHING LOGIC START ---
// Create a slice to hold all the player stats parsed from the page.
var statsToUpsert []models.PlayerTotalStat
// commentedHTML := commentSel.Nodes[0].FirstChild.Data
// innerDoc, err := goquery.NewDocumentFromReader(strings.NewReader(commentedHTML))
// if err != nil {
// return fmt.Errorf("failed to parse commented totals_stats HTML: %w", err)
// }
// table = innerDoc.Find("table#totals_stats")
// if table.Length() == 0 {
// return fmt.Errorf("could not find table#totals_stats after uncommenting")
// }
// }
// 2) Iterate rows and collect all player stats into the slice.
table.Find("tbody tr").Each(func(_ int, tr *goquery.Selection) {
if cl, _ := tr.Attr("class"); strings.Contains(cl, "thead") {
return // Skip repeated header rows inside the table body.
}
// 1) collect the data-stat keys in header order
var headers []string
table.Find("thead tr th").Each(func(i int, th *goquery.Selection) {
if stat, ok := th.Attr("data-stat"); ok && stat != "" {
headers = append(headers, stat)
}
})
// add the appended-player column
headers = append(headers, "player-additional")
cells := tr.Find("th, td")
data := make(map[string]string, len(headers))
var playerID string
// 2) iterate rows
table.Find("tbody tr").Each(func(_ int, tr *goquery.Selection) {
if cl, _ := tr.Attr("class"); strings.Contains(cl, "thead") {
return // skip header rows
}
cells.Each(func(i int, cell *goquery.Selection) {
text := strings.TrimSpace(cell.Text())
key := headers[i]
data[key] = text
if id, ok := cell.Attr("data-append-csv"); ok {
playerID = id
}
})
cells := tr.Find("th, td")
data := make(map[string]string, len(headers))
var playerID string
// Skip rows that aren't actual player data rows.
if playerID == "" {
return
}
data["player-additional"] = playerID
cells.Each(func(i int, cell *goquery.Selection) {
text := strings.TrimSpace(cell.Text())
key := headers[i]
data[key] = text
if id, ok := cell.Attr("data-append-csv"); ok {
playerID = id
}
})
if playerID == "" {
return
}
data["player-additional"] = playerID
// 3a) Pick the right “ExternalID” key.
extID := mustAtoi(data["rk"])
if extID == 0 {
extID = mustAtoi(data["ranker"])
}
// 3a) pick the right “ExternalID” key (playoffs use “rk”; season uses “ranker”)
extID := mustAtoi(data["rk"])
if extID == 0 {
extID = mustAtoi(data["ranker"])
}
// 3b) Pick the right “PlayerName” key.
playerName := data["player"]
if playerName == "" {
playerName = data["name_display"]
}
// 3b) pick the right “PlayerName” key (playoffs use “player”; season uses “name_display”)
playerName := data["player"]
if playerName == "" {
playerName = data["name_display"]
}
// 3c) Pick the right “Team” key.
teamID := data["team_id"]
if teamID == "" {
teamID = data["team_name_abbr"]
}
// 3c) pick the right “Team” key (playoffs use “team_id”; season uses “team_name_abbr”)
teamID := data["team_id"]
if teamID == "" {
teamID = data["team_name_abbr"]
}
// Pick “games”.
g := mustAtoi(data["games"])
if g == 0 {
g = mustAtoi(data["g"])
}
// pick “games” → fallback to “g” if empty
g := mustAtoi(data["games"])
if g == 0 {
g = mustAtoi(data["g"])
}
// Pick “games_started”.
gs := mustAtoi(data["games_started"])
if gs == 0 {
gs = mustAtoi(data["gs"])
}
// pick “games_started” → fallback to “gs” if empty
gs := mustAtoi(data["games_started"])
if gs == 0 {
gs = mustAtoi(data["gs"])
}
stat := models.PlayerTotalStat{
ExternalID: extID,
PlayerID: playerID,
PlayerName: playerName,
Position: data["pos"],
Age: mustAtoi(data["age"]),
Games: g,
GamesStarted: gs,
MinutesPG: mustParseFloat(data["mp"]),
FieldGoals: mustAtoi(data["fg"]),
FieldAttempts: mustAtoi(data["fga"]),
FieldPercent: mustParseFloat(data["fg_pct"]),
ThreeFG: mustAtoi(data["fg3"]),
ThreeAttempts: mustAtoi(data["fg3a"]),
ThreePercent: mustParseFloat(data["fg3_pct"]),
TwoFG: mustAtoi(data["fg2"]),
TwoAttempts: mustAtoi(data["fg2a"]),
TwoPercent: mustParseFloat(data["fg2_pct"]),
EffectFGPercent: mustParseFloat(data["efg_pct"]),
FT: mustAtoi(data["ft"]),
FTAttempts: mustAtoi(data["fta"]),
FTPercent: mustParseFloat(data["ft_pct"]),
OffensiveRB: mustAtoi(data["orb"]),
DefensiveRB: mustAtoi(data["drb"]),
TotalRB: mustAtoi(data["trb"]),
Assists: mustAtoi(data["ast"]),
Steals: mustAtoi(data["stl"]),
Blocks: mustAtoi(data["blk"]),
Turnovers: mustAtoi(data["tov"]),
PersonalFouls: mustAtoi(data["pf"]),
Points: mustAtoi(data["pts"]),
Team: teamID,
Season: season,
IsPlayoff: isPlayoff,
}
stat := models.PlayerTotalStat{
ExternalID: extID,
PlayerID: playerID,
PlayerName: playerName,
Position: data["pos"],
Age: mustAtoi(data["age"]),
Games: g,
GamesStarted: gs,
MinutesPG: mustParseFloat(data["mp"]),
FieldGoals: mustAtoi(data["fg"]),
FieldAttempts: mustAtoi(data["fga"]),
FieldPercent: mustParseFloat(data["fg_pct"]),
ThreeFG: mustAtoi(data["fg3"]),
ThreeAttempts: mustAtoi(data["fg3a"]),
ThreePercent: mustParseFloat(data["fg3_pct"]),
TwoFG: mustAtoi(data["fg2"]),
TwoAttempts: mustAtoi(data["fg2a"]),
TwoPercent: mustParseFloat(data["fg2_pct"]),
EffectFGPercent: mustParseFloat(data["efg_pct"]),
FT: mustAtoi(data["ft"]),
FTAttempts: mustAtoi(data["fta"]),
FTPercent: mustParseFloat(data["ft_pct"]),
OffensiveRB: mustAtoi(data["orb"]),
DefensiveRB: mustAtoi(data["drb"]),
TotalRB: mustAtoi(data["trb"]),
Assists: mustAtoi(data["ast"]),
Steals: mustAtoi(data["stl"]),
Blocks: mustAtoi(data["blk"]),
Turnovers: mustAtoi(data["tov"]),
PersonalFouls: mustAtoi(data["pf"]),
Points: mustAtoi(data["pts"]),
Team: teamID,
Season: season,
IsPlayoff: isPlayoff,
}
// Add the parsed stat object to our slice instead of writing to the DB immediately.
statsToUpsert = append(statsToUpsert, stat)
})
// 4) upsert on (player_id, season, team, is_playoff)
if err := db.Clauses(clause.OnConflict{
Columns: []clause.Column{
{Name: "player_id"},
{Name: "season"},
{Name: "team"},
{Name: "is_playoff"},
},
DoUpdates: clause.AssignmentColumns([]string{
"external_id", "player_name", "position", "age",
"games", "games_started", "minutes_pg",
"field_goals", "field_attempts", "field_percent",
"three_fg", "three_attempts", "three_percent",
"two_fg", "two_attempts", "two_percent",
"effect_fg_percent",
"ft", "ft_attempts", "ft_percent",
"offensive_rb", "defensive_rb", "total_rb",
"assists", "steals", "blocks", "turnovers",
"personal_fouls", "points",
}),
}).Create(&stat).Error; err != nil {
log.Printf("Failed upsert for %s: %v", stat.PlayerID, err)
}
})
// 3) Perform the batch upsert operation after collecting all rows.
if len(statsToUpsert) > 0 {
log.Printf("Attempting to batch upsert %d player total stats for season %d...", len(statsToUpsert), season)
return nil
// GORM's OnConflict clause works with slices, performing the batch operation efficiently.
if err := db.Clauses(clause.OnConflict{
Columns: []clause.Column{
{Name: "player_id"},
{Name: "season"},
{Name: "team"},
{Name: "is_playoff"},
},
DoUpdates: clause.AssignmentColumns([]string{
"external_id", "player_name", "position", "age",
"games", "games_started", "minutes_pg",
"field_goals", "field_attempts", "field_percent",
"three_fg", "three_attempts", "three_percent",
"two_fg", "two_attempts", "two_percent",
"effect_fg_percent",
"ft", "ft_attempts", "ft_percent",
"offensive_rb", "defensive_rb", "total_rb",
"assists", "steals", "blocks", "turnovers",
"personal_fouls", "points",
}),
}).Create(&statsToUpsert).Error; err != nil {
// If the batch operation fails, log the error and return it.
log.Printf("Failed to batch upsert player total stats: %v", err)
return err
}
log.Printf("✅ Successfully batch upserted %d records for season %d.", len(statsToUpsert), season)
} else {
log.Printf("No player data found to import for season %d.", season)
}
// --- BATCHING LOGIC END ---
return nil
}