serJD/speckleAggregateBranches

1import copy2 3import gradio as gr4from huggingface_hub import webhook_endpoint, WebhookPayload5from fastapi import Request6 7 8#import other libaries9from specklepy.api.client import SpeckleClient10from specklepy.api.credentials import get_default_account, get_local_accounts11from specklepy.transports.server import ServerTransport12from specklepy.api import operations13from specklepy.objects.geometry import Polyline, Point14 15import pandas as pd16import numpy as np17import json18import os19from utils import * 20 21speckle_token = os.environ.get("SPECKLE_TOKEN")22 23current_directory = os.path.dirname(os.path.abspath(__file__))24# Path to the config.json file25config_file_path = os.path.join(current_directory, "config.json")26with open(config_file_path, 'r') as f:27        config = json.load(f)28 29 30 31CLIENT = SpeckleClient(host="https://speckle.xyz/")32CLIENT.authenticate_with_token(token=speckle_token)33 34 35print(config.keys())36branchA = config["branchA"]37branchB = config["branchB"]38STREAM_ID = config["streamID"]39 40 41@webhook_endpoint42async def mergeStreams(request: Request):43 44    # Initialize flag45    should_continue = False46 47    # Read the request body as JSON48    payload = await request.json()49 50    print("============= payload =============")51    print(payload)52    print("============= config =============")53    print(config)54    payload = payload["payload"]55 56 57    # webhook calls can come from different sources 58    if payload.get('source') == 'notionTrigger':59        action = payload.get('action')60        streamName = payload.get('streamName')61        branchName = payload.get('branchName')62        update_source = "notionTrigger"63 64        should_continue = True65 66 67    else:68        update_source = "speckleWebhook"69        event_name = payload["event"]["event_name"]70        streamid = payload.get("stream", {}).get("id")71        # Extract branchName for commit_update events from the "old" commit data72        if event_name == "commit_update":73            branchName = payload.get("event", {}).get("data", {}).get("old", {}).get("branchName")74        else:75            branchName = payload.get("event", {}).get("data", {}).get("commit", {}).get("branchName")76    77        # List of valid event types78        valid_event_types = ["commit_create", "commit_delete", "commit_update"]79        80        if event_name in valid_event_types:81            if streamid == STREAM_ID:82                if branchName == branchA:83                    should_continue = True84                else:85                    print(f"Branch name {branchName} not found in config.")86            else:87                print(f"Stream name {streamid} not found in config.")88        else:89            print(f"Event type {event_name} is not one of the specified types.")90 91    # If the flag is True, continue running the main part of the code92    if should_continue:93        # get stream94        stream = getSpeckleStream(STREAM_ID,95                                branchA,96                                CLIENT,97                                commit_id = "")98 99        # navigate to list with speckle objects of interest100        try:101            stream_data = stream["@Data"]["@{0}"]102        except:103            print("something went wrong, try again with non-capital d")104        try:105            stream_data = stream["@data"]["@{0}"]106        except:107            print("check on speckle.com how to access the data")108 109        # transform stream_data to dataframe (create a backup copy of this dataframe)110        df = get_dataframe(stream_data, return_original_df=False)111        df_A = df.copy()112 113        # get stream114        stream = getSpeckleStream(STREAM_ID,115                                    branchB,116                                    CLIENT,117                                    commit_id = "")118 119        # navigate to list with speckle objects of interest120        try:121            stream_data = stream["@Data"]["@{0}"]122        except:123            print("something went wrong, try again with non-capital d")124        try:125            stream_data = stream["@data"]["@{0}"]126        except:127            print("check on speckle.com how to access the data")128 129        # transform stream_data to dataframe (create a backup copy of this dataframe)130        df = get_dataframe(stream_data, return_original_df=False)131        df_B = df.copy()132        133        excludeCol = config["EXCLUDE_COLS"]134        uuidCol = config["UUID_COL"]135        refCol = config["REFERENCE_COL"]136        aggregated_df_b, log_dict = aggregate_data_optimized(df_A.copy(), df_B.copy(), uuidCol, refCol,excludeCol)137 138 139        # additional cleanups, remove geometry and fill  na140        try: 141          aggregated_df_b_noGeo = aggregated_df_b.drop(columns=['@geometry'])142        except:143          pass144        try: 145          aggregated_df_b_noGeo = aggregated_df_b.drop(columns=['@Geometry'])146        except:147          pass148        aggregated_df_b_noGeo = aggregated_df_b.fillna("NA")149        150        151        print (aggregated_df_b_noGeo)152        commit_id = updateStreamAnalysisFast(153                      client = CLIENT,154                      stream_id =STREAM_ID,155                      branch_name = branchB,156                      new_data=aggregated_df_b_noGeo ,157                      geometryGroupPath=["@Data", "@{0}"],158                      match_by_id="id",159                      #openai_key =None,160                      return_original = False,161                      comm_message="auto commit from HF; Triggered by:" + update_source)162        163        return "https://speckle.xyz/streams/" + STREAM_ID + "/commits/" + commit_id