#!/usr/bin/env python3 import requests from csv import DictReader import sqlite3 from configparser import ConfigParser import re import json import os config_files = [f for f in os.listdir() if f.startswith('.taguette_gdocs')] for file_path in config_files: config = ConfigParser() config.read(file_path) # this is project ID from the configuration project_id = int(config['General']['taguette_project_id']) taguette_database_file = config['General']['taguette_database_file'] ## connect to sqlite3 con = sqlite3.connect(taguette_database_file) cur = con.cursor() # load the googgle sheet ID from the configuration gsheet_id = config['General']['gsheet_id'] gsheet_gid = config['General']['gsheet_gid'] # get the spreadsheet data axial_url = f"https://docs.google.com/spreadsheets/d/{gsheet_id}/export?format=csv&id={gsheet_id}&gid={gsheet_gid}" rv = requests.get(axial_url) csv_text = rv.content.decode('utf-8') # import taguette.database as tagdb # db = tagdb.connect("sqlite:////home/mako/taguette-snapshot-20210422-1.sqlite3") # ditionary to ensure that we don't have duplicate tags and to handle merges seen_paths = {} for row in DictReader(csv_text.splitlines(), delimiter=","): #print(row) tag_id = row['id'] new_name = row['tag'] axial_code = row['axial codes'] category = row['category'] description = row['description'] sql_stmt_get = "SELECT id, path, description from TAGS where id = ? AND project_id = ?" cur.execute(sql_stmt_get, (tag_id, project_id)) tag_info = cur.fetchall() if len(tag_info) > 1: print(f"ERROR: '{tag_id}' is not unique, SKIPPING") elif len(tag_info) == 0: print(f"ERROR: 'tag with ID {tag_id}' does not exist, SKIPPING") else: oldname = tag_info[0][1] old_description = tag_info[0][2] # delete any code and associated snipits if new_name == "DELETE": # delete all highlights sql_stmt_update = "DELETE FROM highlight_tags WHERE tag_id = ?" cur.execute(sql_stmt_update, (tag_id,)) # delete the tag itself sql_stmt_update = "DELETE FROM tags WHERE project_id = ? AND id = ?" cur.execute(sql_stmt_update, (project_id, tag_id)) print(f"DELETE TAG FROM {project_id}: {tag_id}") continue if axial_code: new_name = axial_code.lower() + "_" + new_name.lower() else: new_name = new_name.lower() new_description = description if description and category: new_description += " " if category: new_description += json.dumps({'category' : category}) # merge tags if we have two with the same name if new_name in seen_paths: # identify what we're merging into merge_keep_id = seen_paths[new_name] merge_delete_id = tag_id # reassociate all the highlight associated with tag 2 so they are associated with tag 1 intsead sql_stmt_update = "UPDATE highlight_tags SET tag_id = ? WHERE tag_id = ?" cur.execute(sql_stmt_update, (merge_keep_id, merge_delete_id)) # reassociate all the highlight associated with tag 2 so they are associated with tag 1 intsead sql_stmt_update = "DELETE FROM tags WHERE project_id = ? AND id = ?" cur.execute(sql_stmt_update, (project_id, merge_delete_id)) print(f"MERGE DUPLICATE TAGS for {project_id}: {new_name}") continue else: seen_paths[new_name] = tag_id if not oldname == new_name: sql_stmt_update = "UPDATE tags SET path = ? WHERE project_id = ? AND id = ?" print(f"UPDATE TAG for {project_id}: {oldname} → {new_name}") cur.execute(sql_stmt_update, (new_name, project_id, tag_id)) if new_description.strip() != old_description.strip(): sql_stmt_update = "UPDATE tags SET description = ? WHERE project_id = ? AND id = ?" print(f"UPDATE DESC for {project_id}: {old_description} → {new_description}") cur.execute(sql_stmt_update, (new_description, project_id, tag_id)) con.commit() con.close()