Syntax Analyzer

Below are the API specifications for the Syntax Analyzer project, offering features such as syntax analysis and random sentence generation.

Analyze Sentence Syntax

This endpoint uses specified AI models to analyze the syntax of tokenized sentences, breaking them down into grammatical elements for detailed structural insights.

Endpoint

POST /analyzer

Request Body

Field	Type	Options	Required	Description	Constraints
model	string	gpt-3.5, gpt-4	Yes	AI model for analysis	One of: gpt-3.5, gpt-4
sentence	Array<string>	N/A	Yes	Tokens for analysis	Range: 2~20 tokens

Example Request

{
  "model": "gpt-4",
  "sentence": ["My", "name", "is", "John", "."]
}

Response

Analysis Data Model

type AnalysisSource = 'user' | 'sample';
type ISODateString = string;
type ConstituentType = 'clause' | 'phrase' | 'token'

type TConstituent = {
  id: number;                   // Random 9-digit number
  elementId: number;            // Constituent ID
  label: string;                // Grammatical label, e.g., "subject"
  abbreviation: string;         // Abbreviation, e.g., "s"
  type: ConstituentType;        // Type of constituent
};

type TSegment = {
  id: number;                   // Random 9-digit number
  begin: number;                // Start index
  end: number;                  // End index
  constituents: TConstituent[]; // Segment constituents
  children: TSegment[];         // Sub-segments
};

type TAnalysis = {
  id: string;                   // Random 21-byte string
  source: AnalysisSource;       // Data source
  createdAt: ISODateString;     // ISO 8601 timestamp
  sentence: string[];           // Tokenized sentence
  rootSegment: TSegment;        // Root segment
  isAnalyzedByGPT: boolean;     // AI-analyzed status
};

200 OK

{
  "id": "YW6AX-AOZb2-xYXt05xkn",
  "source": "user",
  "createdAt": "2023-05-10T23:08:08.000Z",
  "sentence": ["My", "name", "is", "John", "."],
  "rootSegment": {
    "id": 840296172,
    "begin": 0,
    "end": 5,
    "constituents": [],
    "children": [
      {
        "id": 987654321,
        "begin": 0,
        "end": 2,
        "constituents": [
          {
            "id": 456789123,
            "elementId": 1,
            "label": "subject",
            "abbreviation": "s",
            "type": "token"
          }
        ],
        "children": []
      },
      {
        "id": 789123456,
        "begin": 2,
        "end": 3,
        "constituents": [
          {
            "id": 321654987,
            "elementId": 2,
            "label": "verb",
            "abbreviation": "v",
            "type": "token"
          }
        ],
        "children": []
      },
      {
        "id": 654321789,
        "begin": 3,
        "end": 4,
        "constituents": [
          {
            "id": 654987321,
            "elementId": 5,
            "label": "object",
            "abbreviation": "o",
            "type": "token"
          }
        ],
        "children": []
      },
      {
        "id": 321789654,
        "begin": 4,
        "end": 5,
        "constituents": [],
        "children": []
      }
    ]
  },
  "isAnalyzedByGPT": true
}

400 Bad Request

{
  "errors": [
    {
      "type": "field",
      "value": "value that caused the error",
      "msg": "Error description",
      "path": "field_name",
      "location": "body"
    }
  ]
}

Generate random sentences

This endpoint generates random sentences, optionally constrained by sentence count, character limit, and topics.

Endpoint

GET /analyzer/random-sentences

Request Parameters

Field	Type	Required	Default	Description	Constraints
sent_count	number	No	5	Quantity of sentences	Range: 1~5
max_chars	number	No	80	Character limit per sentence	Range: 10~80
topics	Array<string>	No	[]	Topics to include in sentences	Max 3 topics

Response

200 OK

[
  "Apples are a popular fruit that come in many different varieties.",
  "Apple trees are typically grown in orchards for commercial production.",
  "Apple cider is a refreshing beverage made from pressed apples.",
]

400 Bad Request

{
  "errors": [
    {
      "type": "field",
      "value": "value that caused the error",
      "msg": "Error description",
      "path": "field_name",
      "location": "query"
    }
  ]
}

Check remaining counts

This endpoint provides remaining user quotas for syntax analysis and random sentences, resetting daily to 10 and 20, respectively.

Endpoint

GET /analyzer/remaining-counts

Response

200 OK

{
  "analysis": 10,
  "random_sentence": 20
}

• analysis : Remaining count for syntax analysis
• random_sentence : Remaining count for random sentence generation

400 Bad Request

{
  "status": "error",
  "message": "Error description"
}