app.py

import copy

import gradio as gr
from huggingface_hub import webhook_endpoint, WebhookPayload
from fastapi import Request


#import other libaries
from specklepy.api.client import SpeckleClient
from specklepy.api.credentials import get_default_account, get_local_accounts
from specklepy.transports.server import ServerTransport
from specklepy.api import operations
from specklepy.objects.geometry import Polyline, Point

import pandas as pd
import numpy as np
import json
import os
from utils import * 

speckle_token = os.environ.get("SPECKLE_TOKEN")

current_directory = os.path.dirname(os.path.abspath(__file__))
# Path to the config.json file
config_file_path = os.path.join(current_directory, "config.json")
with open(config_file_path, 'r') as f:
        config = json.load(f)



CLIENT = SpeckleClient(host="https://speckle.xyz/")
CLIENT.authenticate_with_token(token=speckle_token)


print(config.keys())
branchA = config["branchA"]
branchB = config["branchB"]
STREAM_ID = config["streamID"]


@webhook_endpoint
async def mergeStreams(request: Request):

    # Initialize flag
    should_continue = False

    # Read the request body as JSON
    payload = await request.json()

    print("============= payload =============")
    print(payload)
    print("============= config =============")
    print(config)
    payload = payload["payload"]


    # webhook calls can come from different sources 
    if payload.get('source') == 'notionTrigger':
        action = payload.get('action')
        streamName = payload.get('streamName')
        branchName = payload.get('branchName')
        update_source = "notionTrigger"

        should_continue = True


    else:
        update_source = "speckleWebhook"
        event_name = payload["event"]["event_name"]
        streamid = payload.get("stream", {}).get("id")
        # Extract branchName for commit_update events from the "old" commit data
        if event_name == "commit_update":
            branchName = payload.get("event", {}).get("data", {}).get("old", {}).get("branchName")
        else:
            branchName = payload.get("event", {}).get("data", {}).get("commit", {}).get("branchName")
    
        # List of valid event types
        valid_event_types = ["commit_create", "commit_delete", "commit_update"]
        
        if event_name in valid_event_types:
            if streamid == STREAM_ID:
                if branchName == branchA:
                    should_continue = True
                else:
                    print(f"Branch name {branchName} not found in config.")
            else:
                print(f"Stream name {streamid} not found in config.")
        else:
            print(f"Event type {event_name} is not one of the specified types.")

    # If the flag is True, continue running the main part of the code
    if should_continue:
        # get stream
        stream = getSpeckleStream(STREAM_ID,
                                branchA,
                                CLIENT,
                                commit_id = "")

        # navigate to list with speckle objects of interest
        try:
            stream_data = stream["@Data"]["@{0}"]
        except:
            print("something went wrong, try again with non-capital d")
        try:
            stream_data = stream["@data"]["@{0}"]
        except:
            print("check on speckle.com how to access the data")

        # transform stream_data to dataframe (create a backup copy of this dataframe)
        df = get_dataframe(stream_data, return_original_df=False)
        df_A = df.copy()

        # get stream
        stream = getSpeckleStream(STREAM_ID,
                                    branchB,
                                    CLIENT,
                                    commit_id = "")

        # navigate to list with speckle objects of interest
        try:
            stream_data = stream["@Data"]["@{0}"]
        except:
            print("something went wrong, try again with non-capital d")
        try:
            stream_data = stream["@data"]["@{0}"]
        except:
            print("check on speckle.com how to access the data")

        # transform stream_data to dataframe (create a backup copy of this dataframe)
        df = get_dataframe(stream_data, return_original_df=False)
        df_B = df.copy()
        
        excludeCol = config["EXCLUDE_COLS"]
        uuidCol = config["UUID_COL"]
        refCol = config["REFERENCE_COL"]
        aggregated_df_b, log_dict = aggregate_data_optimized(df_A.copy(), df_B.copy(), uuidCol, refCol,excludeCol)


        # additional cleanups, remove geometry and fill  na
        try: 
          aggregated_df_b_noGeo = aggregated_df_b.drop(columns=['@geometry'])
        except:
          pass
        try: 
          aggregated_df_b_noGeo = aggregated_df_b.drop(columns=['@Geometry'])
        except:
          pass
        aggregated_df_b_noGeo = aggregated_df_b.fillna("NA")
        
        
        print (aggregated_df_b_noGeo)
        commit_id = updateStreamAnalysisFast(
                      client = CLIENT,
                      stream_id =STREAM_ID,
                      branch_name = branchB,
                      new_data=aggregated_df_b_noGeo ,
                      geometryGroupPath=["@Data", "@{0}"],
                      match_by_id="id",
                      #openai_key =None,
                      return_original = False,
                      comm_message="auto commit from HF; Triggered by:" + update_source)
        
        return "https://speckle.xyz/streams/" + STREAM_ID + "/commits/" + commit_id