initial
This commit is contained in:
saidsurucu
@@ -0,0 +1,194 @@
|
||||
# danistay_mcp_module/client.py
|
||||
|
||||
import httpx
|
||||
from bs4 import BeautifulSoup
|
||||
from typing import Dict, Any, List, Optional
|
||||
import logging
|
||||
import html
|
||||
import re
|
||||
import tempfile
|
||||
import os
|
||||
from markitdown import MarkItDown
|
||||
|
||||
from .models import (
|
||||
DanistayKeywordSearchRequest,
|
||||
DanistayDetailedSearchRequest,
|
||||
DanistayApiResponse,
|
||||
DanistayDocumentMarkdown,
|
||||
DanistayKeywordSearchRequestData,
|
||||
DanistayDetailedSearchRequestData
|
||||
)
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
if not logger.hasHandlers():
|
||||
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(name)s - %(levelname)s - %(message)s')
|
||||
|
||||
class DanistayApiClient:
|
||||
BASE_URL = "https://karararama.danistay.gov.tr"
|
||||
KEYWORD_SEARCH_ENDPOINT = "/aramalist"
|
||||
DETAILED_SEARCH_ENDPOINT = "/aramadetaylist"
|
||||
DOCUMENT_ENDPOINT = "/getDokuman"
|
||||
|
||||
def __init__(self, request_timeout: float = 30.0):
|
||||
self.http_client = httpx.AsyncClient(
|
||||
base_url=self.BASE_URL,
|
||||
headers={
|
||||
"Content-Type": "application/json; charset=UTF-8", # Arama endpoint'leri için
|
||||
"Accept": "application/json, text/plain, */*", # Arama endpoint'leri için
|
||||
# /getDokuman HTML döndürdüğü için Accept header'ı GET isteğinde farklı olabilir
|
||||
# ama httpx genellikle bunu yönetir. Gerekirse özel header eklenebilir.
|
||||
"X-Requested-With": "XMLHttpRequest",
|
||||
},
|
||||
timeout=request_timeout,
|
||||
verify=False
|
||||
)
|
||||
|
||||
def _prepare_keywords_for_api(self, keywords: List[str]) -> List[str]:
|
||||
return [f'"{k.strip("\"")}"' for k in keywords if k and k.strip()]
|
||||
|
||||
async def search_keyword_decisions(
|
||||
self,
|
||||
params: DanistayKeywordSearchRequest
|
||||
) -> DanistayApiResponse:
|
||||
data_for_payload = DanistayKeywordSearchRequestData(
|
||||
andKelimeler=self._prepare_keywords_for_api(params.andKelimeler),
|
||||
orKelimeler=self._prepare_keywords_for_api(params.orKelimeler),
|
||||
notAndKelimeler=self._prepare_keywords_for_api(params.notAndKelimeler),
|
||||
notOrKelimeler=self._prepare_keywords_for_api(params.notOrKelimeler),
|
||||
pageSize=params.pageSize,
|
||||
pageNumber=params.pageNumber
|
||||
)
|
||||
final_payload = {"data": data_for_payload.model_dump(exclude_none=True)}
|
||||
logger.info(f"DanistayApiClient: Performing KEYWORD search via {self.KEYWORD_SEARCH_ENDPOINT} with payload: {final_payload}")
|
||||
return await self._execute_api_search(self.KEYWORD_SEARCH_ENDPOINT, final_payload)
|
||||
|
||||
async def search_detailed_decisions(
|
||||
self,
|
||||
params: DanistayDetailedSearchRequest
|
||||
) -> DanistayApiResponse:
|
||||
data_for_payload = DanistayDetailedSearchRequestData(
|
||||
daire=params.daire or "",
|
||||
esasYil=params.esasYil or "",
|
||||
esasIlkSiraNo=params.esasIlkSiraNo or "",
|
||||
esasSonSiraNo=params.esasSonSiraNo or "",
|
||||
kararYil=params.kararYil or "",
|
||||
kararIlkSiraNo=params.kararIlkSiraNo or "",
|
||||
kararSonSiraNo=params.kararSonSiraNo or "",
|
||||
baslangicTarihi=params.baslangicTarihi or "",
|
||||
bitisTarihi=params.bitisTarihi or "",
|
||||
mevzuatNumarasi=params.mevzuatNumarasi or "",
|
||||
mevzuatAdi=params.mevzuatAdi or "",
|
||||
madde=params.madde or "",
|
||||
siralama=params.siralama,
|
||||
siralamaDirection=params.siralamaDirection,
|
||||
pageSize=params.pageSize,
|
||||
pageNumber=params.pageNumber
|
||||
)
|
||||
final_payload = {"data": data_for_payload.model_dump(exclude_defaults=False, exclude_none=False)}
|
||||
logger.info(f"DanistayApiClient: Performing DETAILED search via {self.DETAILED_SEARCH_ENDPOINT} with payload: {final_payload}")
|
||||
return await self._execute_api_search(self.DETAILED_SEARCH_ENDPOINT, final_payload)
|
||||
|
||||
async def _execute_api_search(self, endpoint: str, payload: Dict) -> DanistayApiResponse:
|
||||
try:
|
||||
response = await self.http_client.post(endpoint, json=payload)
|
||||
response.raise_for_status()
|
||||
response_json_data = response.json()
|
||||
logger.debug(f"DanistayApiClient: Raw API response from {endpoint}: {response_json_data}")
|
||||
api_response_parsed = DanistayApiResponse(**response_json_data)
|
||||
if api_response_parsed.data and api_response_parsed.data.data:
|
||||
for decision_item in api_response_parsed.data.data:
|
||||
if decision_item.id:
|
||||
decision_item.document_url = f"{self.BASE_URL}{self.DOCUMENT_ENDPOINT}?id={decision_item.id}"
|
||||
return api_response_parsed
|
||||
except httpx.RequestError as e:
|
||||
logger.error(f"DanistayApiClient: HTTP request error during search to {endpoint}: {e}")
|
||||
raise
|
||||
except Exception as e:
|
||||
logger.error(f"DanistayApiClient: Error processing or validating search response from {endpoint}: {e}")
|
||||
raise
|
||||
|
||||
def _convert_html_to_markdown_danistay(self, direct_html_content: str) -> Optional[str]:
|
||||
"""
|
||||
Converts direct HTML content (assumed from Danıştay /getDokuman) to Markdown.
|
||||
"""
|
||||
if not direct_html_content:
|
||||
return None
|
||||
|
||||
# Basic HTML unescaping and fixing common escaped characters
|
||||
# This step might be less critical if MarkItDown handles them, but good for pre-cleaning.
|
||||
processed_html = html.unescape(direct_html_content)
|
||||
processed_html = processed_html.replace('\\"', '"') # If any such JS-escaped strings exist
|
||||
# Danistay HTML doesn't seem to have \\r\\n etc. from the example, but keeping for robustness
|
||||
processed_html = processed_html.replace('\\r\\n', '\n').replace('\\n', '\n').replace('\\t', '\t')
|
||||
|
||||
# For simplicity and to leverage MarkItDown's capability to handle full docs,
|
||||
# we pass the pre-processed full HTML.
|
||||
html_input_for_markdown = processed_html
|
||||
|
||||
markdown_text = None
|
||||
temp_file_path = None
|
||||
try:
|
||||
md_converter = MarkItDown(enable_plugins=False) # Basic conversion
|
||||
|
||||
with tempfile.NamedTemporaryFile(mode="w", delete=False, suffix=".html", encoding="utf-8") as tmp_file:
|
||||
tmp_file.write(html_input_for_markdown) # Write the full HTML string
|
||||
temp_file_path = tmp_file.name
|
||||
|
||||
conversion_result = md_converter.convert(temp_file_path)
|
||||
markdown_text = conversion_result.text_content
|
||||
logger.info("DanistayApiClient: HTML to Markdown conversion successful.")
|
||||
except Exception as e:
|
||||
logger.error(f"DanistayApiClient: Error during MarkItDown HTML to Markdown conversion: {e}")
|
||||
finally:
|
||||
if temp_file_path and os.path.exists(temp_file_path):
|
||||
os.remove(temp_file_path)
|
||||
|
||||
return markdown_text
|
||||
|
||||
async def get_decision_document_as_markdown(self, document_id: str) -> DanistayDocumentMarkdown:
|
||||
"""
|
||||
Retrieves a specific Danıştay decision by ID and returns its content as Markdown.
|
||||
The /getDokuman endpoint for Danıştay returns direct HTML.
|
||||
"""
|
||||
document_api_url = f"{self.DOCUMENT_ENDPOINT}?id={document_id}"
|
||||
source_url = f"{self.BASE_URL}{document_api_url}"
|
||||
logger.info(f"DanistayApiClient: Fetching Danistay document for Markdown (ID: {document_id}) from {source_url}")
|
||||
|
||||
try:
|
||||
# For direct HTML response, we might want different headers if the API is sensitive,
|
||||
# but httpx usually handles basic GET requests well.
|
||||
response = await self.http_client.get(document_api_url)
|
||||
response.raise_for_status()
|
||||
|
||||
# Danıştay /getDokuman directly returns HTML text
|
||||
html_content_from_api = response.text
|
||||
|
||||
if not isinstance(html_content_from_api, str) or not html_content_from_api.strip():
|
||||
logger.warning(f"DanistayApiClient: Received empty or non-string HTML content for ID {document_id}.")
|
||||
# Return with None markdown_content if HTML is effectively empty
|
||||
return DanistayDocumentMarkdown(
|
||||
document_id=document_id,
|
||||
markdown_content=None,
|
||||
source_url=source_url
|
||||
)
|
||||
|
||||
markdown_content = self._convert_html_to_markdown_danistay(html_content_from_api)
|
||||
|
||||
return DanistayDocumentMarkdown(
|
||||
document_id=document_id,
|
||||
markdown_content=markdown_content,
|
||||
source_url=source_url
|
||||
)
|
||||
except httpx.RequestError as e:
|
||||
logger.error(f"DanistayApiClient: HTTP error fetching Danistay document (ID: {document_id}): {e}")
|
||||
raise
|
||||
# Removed ValueError for JSON as Danistay /getDokuman returns direct HTML
|
||||
except Exception as e: # Catches other errors like MarkItDown issues if they propagate
|
||||
logger.error(f"DanistayApiClient: General error processing Danistay document (ID: {document_id}): {e}")
|
||||
raise
|
||||
|
||||
async def close_client_session(self):
|
||||
"""Closes the HTTPX client session."""
|
||||
if self.http_client and not self.http_client.is_closed:
|
||||
await self.http_client.aclose()
|
||||
logger.info("DanistayApiClient: HTTP client session closed.")
|
||||
@@ -0,0 +1,116 @@
|
||||
# danistay_mcp_module/models.py
|
||||
|
||||
from pydantic import BaseModel, Field, HttpUrl
|
||||
from typing import List, Optional, Dict, Any
|
||||
|
||||
class DanistayBaseSearchRequest(BaseModel):
|
||||
"""Base model for common search parameters for Danistay."""
|
||||
pageSize: int = Field(default=10, ge=1, le=100)
|
||||
pageNumber: int = Field(default=1, ge=1)
|
||||
# siralama and siralamaDirection are part of detailed search, not necessarily keyword search
|
||||
# as per user's provided payloads.
|
||||
|
||||
class DanistayKeywordSearchRequestData(BaseModel):
|
||||
"""Internal data model for the keyword search payload's 'data' field."""
|
||||
andKelimeler: List[str] = Field(default_factory=list)
|
||||
orKelimeler: List[str] = Field(default_factory=list)
|
||||
notAndKelimeler: List[str] = Field(default_factory=list)
|
||||
notOrKelimeler: List[str] = Field(default_factory=list)
|
||||
pageSize: int
|
||||
pageNumber: int
|
||||
|
||||
class DanistayKeywordSearchRequest(BaseModel): # This is the model the MCP tool will accept
|
||||
"""Model for keyword-based search request for Danistay."""
|
||||
andKelimeler: List[str] = Field(default_factory=list, description="Keywords for AND logic, e.g., ['word1', 'word2']")
|
||||
orKelimeler: List[str] = Field(default_factory=list, description="Keywords for OR logic.")
|
||||
notAndKelimeler: List[str] = Field(default_factory=list, description="Keywords for NOT AND logic.")
|
||||
notOrKelimeler: List[str] = Field(default_factory=list, description="Keywords for NOT OR logic.")
|
||||
pageSize: int = Field(default=10, ge=1, le=100)
|
||||
pageNumber: int = Field(default=1, ge=1)
|
||||
|
||||
class DanistayDetailedSearchRequestData(BaseModel): # Internal data model for detailed search payload
|
||||
"""Internal data model for the detailed search payload's 'data' field."""
|
||||
daire: Optional[str] = "" # API expects empty string for None
|
||||
esasYil: Optional[str] = ""
|
||||
esasIlkSiraNo: Optional[str] = ""
|
||||
esasSonSiraNo: Optional[str] = ""
|
||||
kararYil: Optional[str] = ""
|
||||
kararIlkSiraNo: Optional[str] = ""
|
||||
kararSonSiraNo: Optional[str] = ""
|
||||
baslangicTarihi: Optional[str] = ""
|
||||
bitisTarihi: Optional[str] = ""
|
||||
mevzuatNumarasi: Optional[str] = ""
|
||||
mevzuatAdi: Optional[str] = ""
|
||||
madde: Optional[str] = ""
|
||||
siralama: str # Seems mandatory in detailed search payload
|
||||
siralamaDirection: str # Seems mandatory
|
||||
pageSize: int
|
||||
pageNumber: int
|
||||
# Note: 'arananKelime' is not in the detailed search payload example provided by user.
|
||||
# If it can be included, it should be added here.
|
||||
|
||||
class DanistayDetailedSearchRequest(DanistayBaseSearchRequest): # MCP tool will accept this
|
||||
"""Model for detailed search request for Danistay."""
|
||||
daire: Optional[str] = Field(None, description="Chamber/Department name (e.g., '1. Daire').")
|
||||
esasYil: Optional[str] = Field(None, description="Case year for 'Esas No'.")
|
||||
esasIlkSiraNo: Optional[str] = Field(None, description="Starting sequence for 'Esas No'.")
|
||||
esasSonSiraNo: Optional[str] = Field(None, description="Ending sequence for 'Esas No'.")
|
||||
kararYil: Optional[str] = Field(None, description="Decision year for 'Karar No'.")
|
||||
kararIlkSiraNo: Optional[str] = Field(None, description="Starting sequence for 'Karar No'.")
|
||||
kararSonSiraNo: Optional[str] = Field(None, description="Ending sequence for 'Karar No'.")
|
||||
baslangicTarihi: Optional[str] = Field(None, description="Start date for decision (DD.MM.YYYY).")
|
||||
bitisTarihi: Optional[str] = Field(None, description="End date for decision (DD.MM.YYYY).")
|
||||
mevzuatNumarasi: Optional[str] = Field(None, description="Legislation number.")
|
||||
mevzuatAdi: Optional[str] = Field(None, description="Legislation name.")
|
||||
madde: Optional[str] = Field(None, description="Article number.")
|
||||
siralama: str = Field("1", description="Sorting criteria (e.g., 1: Esas No, 3: Karar Tarihi).")
|
||||
siralamaDirection: str = Field("desc", description="Sorting direction ('asc' or 'desc').")
|
||||
# Add a general keyword field if detailed search also supports it
|
||||
# arananKelime: Optional[str] = Field(None, description="General keyword for detailed search.")
|
||||
|
||||
|
||||
class DanistayApiDecisionEntry(BaseModel):
|
||||
"""Model for an individual decision entry from the Danistay API search response.
|
||||
Based on user-provided response samples for both keyword and detailed search.
|
||||
"""
|
||||
id: str
|
||||
# The API response for keyword search uses "daireKurul", detailed search example uses "daire".
|
||||
# We use an alias to handle both and map to a consistent field name "chamber".
|
||||
chamber: Optional[str] = Field(None, alias="daire", alt_alias="daireKurul", description="The chamber or board.")
|
||||
esasNo: Optional[str] = Field(None)
|
||||
kararNo: Optional[str] = Field(None)
|
||||
kararTarihi: Optional[str] = Field(None)
|
||||
arananKelime: Optional[str] = Field(None, description="Matched keyword if provided in response.")
|
||||
# index: Optional[int] = None # Present in response, can be added if needed by MCP tool
|
||||
# siraNo: Optional[int] = None # Present in detailed response, can be added
|
||||
|
||||
document_url: Optional[HttpUrl] = Field(None, description="URL to the full document, constructed by the client.")
|
||||
|
||||
class Config:
|
||||
populate_by_name = True # Important for alias to work
|
||||
extra = 'ignore' # Ignore any extra fields from API not defined in model
|
||||
|
||||
class DanistayApiResponseInnerData(BaseModel):
|
||||
"""Model for the inner 'data' object in the Danistay API search response."""
|
||||
data: List[DanistayApiDecisionEntry]
|
||||
recordsTotal: int
|
||||
recordsFiltered: int
|
||||
draw: Optional[int] = Field(None, description="Draw counter from API, usually for DataTables.")
|
||||
|
||||
class DanistayApiResponse(BaseModel):
|
||||
"""Model for the complete search response from the Danistay API."""
|
||||
data: DanistayApiResponseInnerData
|
||||
metadata: Optional[Dict[str, Any]] = Field(None, description="Optional metadata from API.")
|
||||
|
||||
class DanistayDocumentMarkdown(BaseModel):
|
||||
"""Model for a Danistay decision document, containing only Markdown content."""
|
||||
document_id: str
|
||||
markdown_content: Optional[str] = Field(None, description="The decision content converted to Markdown.")
|
||||
source_url: HttpUrl
|
||||
|
||||
class CompactDanistaySearchResult(BaseModel):
|
||||
"""A compact search result model for the MCP tool to return."""
|
||||
decisions: List[DanistayApiDecisionEntry]
|
||||
total_records: int
|
||||
requested_page: int
|
||||
page_size: int
|
||||
Reference in New Issue
Block a user