---

🧾 README

Overview

This Python project allows you to scrape both the match history and the upcoming matches of a team or a competition from oddsPortal.com.

It provides detailed information for a given sport, team, or competition during a specific season, according to user-defined parameters.

---

📊 Collected Data

For a competition

In addition to the general information provided by the user (season, bookmaker, market, region, competition, sport), the script collects for each match:

• Match score
• Home team
• Away team
• Opening and closing odds (value, date, and time) for outcome 1, X, and 2 in the 1X2 market

Depending on the typegame and spread options, the scraper adapts its behavior:

Key	Description	Possible Values	Default
typegame	Defines whether to scrape upcoming or historical matches	"upcoming" or "historical"	"historical"
spread	Defines the scraping scope for competitions	"none" (only the selected competition) / "team" (also fetches team histories) / "completly" (fetches all teams + all competitions they played in)	"none"

For example:

• typegame="historical", spread="team" → fetches the competition’s match history + all participating teams’ histories.
• typegame="upcoming" → fetches only upcoming matches of the competition.

When spread is "team" or "completly", already-scraped data in scraped_data/ are skipped to avoid redundancy and save time.

For a team

The script collects the same information as for competitions, plus:

• The region where the competition takes place
• The competition name

⚠️ The odds are retrieved according to the specified bookmaker.

---

⚙️ Requirements

• Python 3.x
• Playwright

---

💻 Installation

git clone https://github.com/djibril-marega/oddsportal_scraper.git
cd oddsportal_scraper
pip install -r requirements.txt

---

🚀 Usage

Method 1 — Recommended (modular and reproducible)

Create a .json configuration file containing teams or competitions.

Example for a team:

[
  {
    "sport": "Football",
    "region": "France",
    "team": "PSG",
    "teamid": "CjhkPw0k",
    "season": "2024/2025",
    "bookmaker": "Betclic"
  }
]

Example for a competition (historical data):

[
  {
    "sport": "Football",
    "region": "England",
    "competition": "Premier League",
    "season": "2024/2025",
    "bookmaker": "Betclic"
  }
]

Example for upcoming matches of a competition:

[
  {
    "sport": "Football",
    "region": "France",
    "competition": "Ligue 1",
    "season": "2025/2026",
    "bookmaker": "Betclic",
    "typegame": "upcoming"
  }
]

To execute the script:

python .\run_parallel_tests.py

• Logs are stored in the logs/ directory — one file per team/competition, plus a global summary log.

• Scraped data are saved in scraped_data/ — one file per team/competition.

• To enable verbose output:

  python .\run_parallel_tests.py -v

---

Method 2 — Quick test (less reproducible)

Run directly with pytest:

pytest test_oddsportal.py --sport=Football --region=France --competition="Ligue 1" --season=2025/2026 --bookmaker=Betclic --typegame=upcoming -v --tb=short

Options:

• --typegame: "upcoming" or "historical"
• --spread: "none", "team", or "completly"
• -v: verbose mode
• --tb=short: concise traceback

---

📁 Output Structure

project/
│
├── scraped_data/        # One file per team or competition
├── logs/                # Detailed and summary logs
├── run_parallel_tests.py
└── test_oddsportal.py

---

🧠 Notes

• The scraper relies on Playwright, so the first run may download browsers automatically.
• Using Method 1 is recommended for scalability and reproducibility.
• Already collected teams or competitions (for the same season) are automatically skipped to avoid redundant scraping.
• typegame and spread allow flexible control of scraping scope — from a single competition’s upcoming games to a full seasonal network of related competitions and teams.

---