Pandas speichert eine Spalte als Float, obwohl ich mit .astype in str konvertiert habe ⇐ Python

[Anonymous]

Ich versuche, eine Zeichenfolge zu speichern, die als Formel in Google Sheets verwendet wird.

Code: Select all

import os
import re
from pathlib import Path
import pathlib
import glob

import gspread

from pydrive2.auth import GoogleAuth
from pydrive2.drive import GoogleDrive

import pandas as pd
import numpy as np

## authorization
gauth = GoogleAuth()
gauth.LocalWebserverAuth()

drive = GoogleDrive(gauth)

# list containing all folders in drive
folder_list = drive.ListFile({'q': "mimeType = 'application/vnd.google-apps.folder'"}).GetList()

# find PICTURES id
PICTURES = [file for file in folder_list if file['title'] == 'PICTURES']
PICTURES = PICTURES[0]
pics_id = '1dM4X2VCSBEB-yjqnAPBkmLoXdrMwBF9j'

rows = drive.ListFile({'q': f"'{pics_id}' in parents"}).GetList()
row_data = {}

for row in rows:
id = row['id']
row_title = row['title']
row_title_formatted = row_title.split()
if len(row_title_formatted) == 2:
row_title_formatted = row_title_formatted[-1]
if len(row_title_formatted) == 3:
row_title_formatted = f'{row_title_formatted[1]}R'
# print(row_title)
# print(f'formatted:{row_title_formatted}')
# print(id)

pictures = drive.ListFile({'q': f"'{id}' in parents"}).GetList()

pics_dict = {}

for p in pictures:
pic_title = p['title'].split('.')
if len(pic_title) == 4:
r, row_idx, pic_idx, j = pic_title

if len(pic_idx) == 1:
pic_idx = f'0{pic_idx}'

formatted_title = f'{row_idx}_{pic_idx}'
rear = False

if len(pic_title) == 5:
r, row_idx, rear, pic_idx, j = pic_title

if len(pic_idx) == 1:
pic_idx = f'0{pic_idx}'

formatted_title = f'{row_idx}R_{pic_idx}'

format_url = f"https://drive.google.com/uc?export=view&id={p['id']}"
p_dict = {
'title'  : formatted_title,
'id'     : p['id'],
'row_idx': row_idx,
'rear'   : rear,
'pic_idx': pic_idx,
'url_str': f'=IMAGE("{format_url}", 4, 300, 400)'
}

print(p_dict['url_str'])

pics_dict.update({f'{formatted_title}': p_dict})

row_data.update({f'{row_title_formatted}': pics_dict})

dfs = []
for row, pics in row_data.items():
for k, v in pics.items():
columns = v.keys()
df = pd.DataFrame.from_dict(pics, orient='index', dtype=str, columns=columns)
dfs.append(df)

combined_dfs = pd.concat(dfs, ignore_index=True)

# combined_dfs['url_str'] = combined_dfs['url_str'].astype(str)

combined_dfs.to_excel('pic_data.xlsx')
print(row_data)
print(dfs)
print()

Wenn ich „combined_dfs“ in meiner IDE überprüfe, scheinen die Ergebnisse in Ordnung zu sein


Wenn ich jedoch die Tabelle erneut mit
lade

Code: Select all

df = pd.read_excel('pic_data.xlsx', dtype=str, keep_default_na=False)

Dies ist das Ergebnis, das ich erhalte:


Es scheint, dass wir die Spalte „url_str“ als Float-Standbild speichern, obwohl ich sie als Zeichenfolge festgelegt habe. Irgendeine Idee, wo ich das vermassele?