trendspy-lite

TrendsPy Lite

A fork of TrendsPy which does not use pandas/numpy

Key Features

Explore

• Track popularity over time (interest_over_time)
• Analyze geographic distribution (interest_by_region)
• Compare interest across different timeframes and regions (multirange support)
• Get related queries and topics (related_queries, related_topics)

Trending Now

• Access current trending searches (trending_now, trending_now_by_rss)
• Get related news articles (trending_now_news_by_ids)
• Retrieve historical data for 500+ trending keywords with independent normalization (trending_now_showcase_timeline)

Search Utilities

• Find category IDs (categories)
• Search for location codes (geo)

Flexible Time Formats

• Custom intervals: 'now 123-H', 'today 45-d'
• Date-based offsets: '2024-02-01 10-d'
• Standard ranges: '2024-01-01 2024-12-31'

Installation

pip install trendspy

Basic Usage

from trendspy import Trends
tr = Trends()
df = tr.interest_over_time(['python', 'javascript'])
df.plot(title='Python vs JavaScript Interest Over Time', 
        figsize=(12, 6))

# Analyze geographic distribution
geo_df = tr.interest_by_region('python')

# Get related queries
related = tr.related_queries('python')

Advanced Features

Search Categories and Locations

# Find technology-related categories
categories = tr.categories(find='technology')
# Output: [{'name': 'Computers & Electronics', 'id': '13'}, ...]

# Search for locations
locations = tr.geo(find='york')
# Output: [{'name': 'New York', 'id': 'US-NY'}, ...]

# Use in queries
df = tr.interest_over_time(
    'python',
    geo='US-NY',      # Found location ID
    cat='13'          # Found category ID
)

Real-time Trending Searches and News

# Get current trending searches in the US
trends = tr.trending_now(geo='US')

# Get trending searches with news articles
trends_with_news = tr.trending_now_by_rss(geo='US')
print(trends_with_news[0])  # First trending topic
print(trends_with_news[0].news[0])  # Associated news article

# Get news articles for specific trending topics
news = tr.trending_now_news_by_ids(
    trends[0].news_tokens,  # News tokens from trending topic
    max_news=3  # Number of articles to retrieve
)
for article in news:
    print(f"Title: {article.title}")
    print(f"Source: {article.source}")
    print(f"URL: {article.url}\n")

Independent Historical Data for Multiple Keywords

from trendspy import BatchPeriod

# Unlike standard interest_over_time where data is normalized across all keywords,
# trending_now_showcase_timeline provides independent data for each keyword
# (up to 500+ keywords in a single request)

keywords = ['keyword1', 'keyword2', ..., 'keyword500']

# Get independent historical data
df_24h = tr.trending_now_showcase_timeline(
    keywords,
    timeframe=BatchPeriod.Past24H  # 16-minute intervals
)

# Each keyword's data is normalized only to itself
df_24h.plot(
    subplots=True,
    layout=(5, 2),
    figsize=(15, 20),
    title="Independent Trend Lines"
)

# Available time windows:
# - Past4H:  ~30 points (8-minute intervals)
# - Past24H: ~90 points (16-minute intervals)
# - Past48H: ~180 points (16-minute intervals)
# - Past7D:  ~42 points (4-hour intervals)

Geographic Analysis

# Country-level data
country_df = tr.interest_by_region('python')

# State-level data for the US
state_df = tr.interest_by_region(
    'python',
    geo='US',
    resolution='REGION'
)

# City-level data for California
city_df = tr.interest_by_region(
    'python',
    geo='US-CA',
    resolution='CITY'
)

Timeframe Formats

• Standard API timeframes: 'now 1-H', 'now 4-H', 'today 1-m', 'today 3-m', 'today 12-m'
• Custom intervals:
  • Short-term (< 8 days): 'now 123-H', 'now 72-H'
  • Long-term: 'today 45-d', 'today 90-d', 'today 18-m'
  • Date-based: '2024-02-01 10-d', '2024-03-15 3-m'
• Date ranges: '2024-01-01 2024-12-31'
• Hourly precision: '2024-03-25T12 2024-03-25T15' (for periods < 8 days)
• All available data: 'all'

Multirange Interest Over Time

Compare search interest across different time periods and regions:

# Compare different time periods
timeframes = [
    '2024-01-25 12-d',    # 12-day period
    '2024-06-20 23-d'     # 23-day period
]
geo = ['US', 'GB']        # Compare US and UK

df = tr.interest_over_time(
    'python',
    timeframe=timeframes,
    geo=geo
)

Note: When using multiple timeframes, they must maintain consistent resolution and the maximum timeframe cannot be more than twice the length of the minimum timeframe.

Proxy Support

TrendsPy supports the same proxy configuration as the requests library:

# Initialize with proxy
tr = Trends(proxy="http://user:pass@10.10.1.10:3128")
# or
tr = Trends(proxy={
    "http": "http://10.10.1.10:3128",
    "https": "http://10.10.1.10:1080"
})

# Configure proxy after initialization
tr.set_proxy("http://10.10.1.10:3128")

Documentation

For more examples and detailed API documentation, check out the Jupyter notebook in the repository: basic_usage.ipynb

License

MIT License - see the LICENSE file for details.

Disclaimer

This library is not affiliated with Google. Please ensure compliance with Google's terms of service when using this library.