Awesome Story Generation

https://github.com/yingpengma/awesome-story-generation
Admin
A curated collection of research papers and resources on story generation and storytelling using...
Tokens:3,762
Snippets:30
Trust Score:8.3
Update:8 months ago
Show doc for...
Tokens:
Raw
### Run Full Citation Workflow - Command Line

Source: https://context7.com/yingpengma/awesome-story-generation/llms.txt

Executes the complete citation workflow from the command line, encompassing paper synchronization, citation fetching, and README badge updates in a single operation. This script automates the entire process.

```bash
# Run the full citation workflow
python citation_workflow.py

# Expected output:
# ⏰ 启动时间: 2025-07-12 16:30:00
# --------------------------------------------------------------------------------
#                              论文引用数管理工具
# --------------------------------------------------------------------------------
# [步骤 1/3] 同步论文列表
# ✅ 成功从 README.md 中提取了 85 篇论文
# ...
# [步骤 2/3] 获取最新引用数
# ...
# [步骤 3/3] 更新README.md中的引用标签
# ...
# ✅ 步骤3 - README更新: 完成
# ⏰ 完成时间: 2025-07-12 16:35:00
```

--------------------------------

### Run Citation Workflow

Source: https://context7.com/yingpengma/awesome-story-generation/llms.txt

Initializes and runs the complete citation management workflow. This includes synchronizing paper lists between README.md and a JSON file, fetching citation counts from the Semantic Scholar API, and updating citation badges in the README. It handles configuration for file paths, API delays, and skipping recent updates.

```python
from citation_workflow import CitationWorkflow

# Initialize the workflow
workflow = CitationWorkflow()

# Configure settings (defaults shown)
workflow.readme_file = 'README.md'      # Source markdown file
workflow.citations_file = 'citations.json'  # Citation data storage
workflow.api_delay = 1                   # Seconds between API requests
workflow.skip_recent_hours = 24          # Skip papers updated within this window

# Run the complete workflow
workflow.run_workflow()

# Output:
# [步骤 1/3] 同步论文列表
# ✅ 成功从 README.md 中提取了 85 篇论文
# ✅ 成功加载 citations.json，现有 85 篇论文
# 📋 开始同步论文列表...
# ✅ 论文列表同步完成！
# [步骤 2/3] 获取最新引用数
# ...
# [步骤 3/3] 更新README.md中的引用标签
# ✅ 内容完整性验证通过

```

--------------------------------

### Synchronize Paper List

Source: https://context7.com/yingpengma/awesome-story-generation/llms.txt

Synchronizes the list of papers between the README.md file and the citations.json database. It identifies new papers in the README to add, preserves existing citation data, and removes entries from the JSON that are no longer present in the README.

```python
from citation_workflow import CitationWorkflow

workflow = CitationWorkflow()
citations_data = workflow.sync_papers()

# Output:
# [步骤 1/3] 同步论文列表
# ✅ 成功从 README.md 中提取了 85 篇论文
# ✅ 成功加载 citations.json，现有 82 篇论文
# 📋 开始同步论文列表...
# ➕ 添加新论文：New Paper Title 1
# ➕ 添加新论文：New Paper Title 2
# ✅ 论文列表同步完成！
#    📊 当前总计：85 篇论文
#    ➕ 新增：3 篇
#    📋 数据清理：0 篇

print(f"Total papers: {len(citations_data['papers'] অ্যাপ্লিকেশন)}")

```

--------------------------------

### Load and Save Citation Data

Source: https://context7.com/yingpengma/awesome-story-generation/llms.txt

Manages the loading and saving of citation data to and from a JSON file. The data is structured as a dictionary mapping paper titles to their citation counts and last updated timestamps. It includes error handling for missing files, creating an empty structure if necessary.

```python
from citation_workflow import CitationWorkflow

workflow = CitationWorkflow()

# Load existing citations
data = workflow.load_citations()
# Output: ✅ 成功加载 citations.json，现有 85 篇论文

# Access paper data
paper = data['papers']['AutoAgents: A Framework for Automatic Agent Generation']
print(f"Citations: {paper['citations']}")
print(f"Last updated: {paper['last_updated']}")
# Citations: 129
# Last updated: 2025-07-12-04:09:30

# Modify and save
data['papers']['New Paper Title'] = {
    'title': 'New Paper Title',
    'citations': 0,
    'last_updated': '1970-01-01-00:00:00'
}
workflow.save_citations(data)

```

--------------------------------

### Batch Fetch Citations for Papers - Python

Source: https://context7.com/yingpengma/awesome-story-generation/llms.txt

Fetches citation counts for multiple papers, prioritizing those least recently updated. It skips papers updated within the last 24 hours to minimize API calls and adhere to rate limits. This function is part of the CitationWorkflow class.

```python
from citation_workflow import CitationWorkflow

workflow = CitationWorkflow()
citations_data = workflow.sync_papers()

# Get citations for all papers
updated_data, updated_papers_list = workflow.get_citations(citations_data)

# Output:
# [步骤 2/3] 获取最新引用数
# 📊 论文更新优先级排序（共85篇，按最后更新时间升序）:
#    1. 引用: 0, 最后更新: 1970-01-01-00:00:00
#    2. 引用: 7, 最后更新: 2025-07-12-04:08:46
#    ... 还有 80 篇论文
# 📊 需要更新: 5 篇，跳过: 80 篇（24小时内已更新）
# ...
# ✅ 成功更新: 5 (5.9%)
# ⏭️ 跳过（24小时内已更新）: 80 (94.1%)

print(f"Updated {len(updated_papers_list)} papers")
```

--------------------------------

### Extract Paper Titles from Markdown

Source: https://context7.com/yingpengma/awesome-story-generation/llms.txt

Extracts paper titles from a README.md file using regular expressions. It specifically looks for titles formatted in bold markdown (**Title**) followed by links. The function returns a cleaned list of titles, filtering out any that are too short.

```python
from citation_workflow import CitationWorkflow

workflow = CitationWorkflow()
titles = workflow.extract_titles_from_md()

# Output:
# ✅ 成功从 README.md 中提取了 85 篇论文

print(f"Found {len(titles)} papers")
print(f"First paper: {titles[0]}")
# Found 85 papers
# First paper: What Makes a Good Story and How Can We Measure It? A Comprehensive Survey of Story Evaluation

```

--------------------------------

### Extract Authors from README for a Paper - Python

Source: https://context7.com/yingpengma/awesome-story-generation/llms.txt

Extracts author information for a specific paper from the README.md file. This is useful for manual verification when Semantic Scholar returns a non-matching title. This function is part of the CitationWorkflow class.

```python
from citation_workflow import CitationWorkflow

workflow = CitationWorkflow()

# Extract authors for a paper
title = "AutoAgents: A Framework for Automatic Agent Generation"
authors = workflow.extract_authors_from_readme(title)

print(f"Authors: {authors}")
# Authors: Guangyao Chen, Siwei Dong, Yu Shu, Ge Zhang, Jaward Sesay, Börje F. Karlsson, Jie Fu, Yemin Shi
```

--------------------------------

### Update Citation Badges in README - Python

Source: https://context7.com/yingpengma/awesome-story-generation/llms.txt

Updates citation badges in the README.md file for papers that have been successfully updated. It performs strict content validation to ensure only badge values are modified, preserving all other content. This function is part of the CitationWorkflow class.

```python
from citation_workflow import CitationWorkflow

workflow = CitationWorkflow()
citations_data = workflow.sync_papers()
updated_data, updated_papers_list = workflow.get_citations(citations_data)

# Update README badges
success = workflow.update_readme_citations(updated_data, updated_papers_list)

# Output:
# [步骤 3/3] 更新README.md中的引用标签
# 📄 开始更新 5 篇论文的citation badges...
#    ✅ 成功更新: 'RecurrentGPT: Interactive Generation of (Arbitrarily...' → 64 引用
#    ✅ 成功更新: 'AutoAgents: A Framework for Automatic Agent Genera...' → 129 引用
# 🔍 验证内容完整性...
# ✅ 内容完整性验证通过
# 💾 README.md 更新完成！
#    ✅ 成功更新: 5 个citation badges

if success:
    print("README updated successfully")
```

--------------------------------

### Update Single Paper Citation Count

Source: https://context7.com/yingpengma/awesome-story-generation/llms.txt

Fetches the citation count for a specific paper from the Semantic Scholar API. It uses exact title matching and includes logic for handling API rate limits through retries. Updates are immediately saved to the JSON citation file.

```python
from citation_workflow import CitationWorkflow

workflow = CitationWorkflow()
citations_data = workflow.load_citations()

# Update a specific paper
title = "RecurrentGPT: Interactive Generation of (Arbitrarily) Long Text"
success, result = workflow.update_paper_citation(title, citations_data, retry_limit=3)

# Output:
# 📖 论文: 'RecurrentGPT: Interactive Generation of (Arbitrarily) Long Text'
#    ✅ 更新成功: 60 → 64 (📈 +4)
#    🔗 论文链接: https://www.semanticscholar.org/paper/...

if success:
    print(f"New citation count: {result}")
else:
    print(f"Update failed: {result}")

```

--------------------------------

### Citations Data JSON Structure

Source: https://context7.com/yingpengma/awesome-story-generation/llms.txt

Defines the JSON structure for storing paper metadata, including citation counts and last updated timestamps. Each paper entry contains its title, current citation count, and the last update time in ISO format.

```json
{
  "papers": {
    "AutoAgents: A Framework for Automatic Agent Generation": {
      "title": "AutoAgents: A Framework for Automatic Agent Generation",
      "citations": 129,
      "last_updated": "2025-07-12-04:09:30"
    },
    "RecurrentGPT: Interactive Generation of (Arbitrarily) Long Text": {
      "title": "RecurrentGPT: Interactive Generation of (Arbitrarily) Long Text",
      "citations": 64,
      "last_updated": "2025-07-12-04:10:16"
    }
  }
}
```

--------------------------------

### Check if Paper was Recently Updated - Python

Source: https://context7.com/yingpengma/awesome-story-generation/llms.txt

Checks if a paper's last update timestamp falls within a configurable skip window, defaulting to 24 hours. Returns True if the paper should be skipped to prevent redundant API calls. This function is part of the CitationWorkflow class.

```python
from citation_workflow import CitationWorkflow

workflow = CitationWorkflow()
workflow.skip_recent_hours = 24  # Skip papers updated in last 24 hours

# Check various timestamps
recent = workflow.is_recently_updated("2025-07-12-16:00:00")
old = workflow.is_recently_updated("2025-07-10-00:00:00")
invalid = workflow.is_recently_updated("1970-01-01-00:00:00")

print(f"Recent update (skip): {recent}")   # True - skip this paper
print(f"Old update (process): {old}")       # False - needs update
print(f"Invalid/new (process): {invalid}")  # False - needs update
```
Tokens:
Raw