Hi Markzus,
Issue #5253 <https://code.djangoproject.com/ticket/5253> that you pointed me to was indeed very helpful. What I suggested is along the lines of what Adam suggested and tried to accomplish 12 years ago, which clearly shows that there has been space for its implementation for a very long time. At the same time, all of the points brought up by Russ also made sense but I do think that we can resolve those issues to some extent. Allow me to address the little details one by one. Russ pointed out the need for a standardized format in csv files, and in turn Adam suggested that the first line begin the table name. I have a slightly different suggestion. It is a well known convention to put the column names as comma separated values in the first line of the csv file. I suggest we leave it at that. The column names of the first line will correspond to the fieldnames of the model created by the user, which instead of being referenced in the file itself can be referenced as a command line argument. This of course operates under the assumption that only one table of data will be in a single file. Which isn't a very bold assumption to make. I'll elaborate on why in just a bit. I'm going to try and do a better job of explain the context of the addition I speak of. Say you get your data off of a random site on the internet in the form of an Excel sheet. Now if you want to use it in django as a model, then right now this is what you'll have to do: Convert it to CSV from Excel. Write a script somethink like the one https://groups.google.com/g/django-developers/c/o1dFA31YwOk/m/XchFFWjnBQAJ talks about (This views file is the code I'm referring to) <https://github.com/pmutua/drf_csv_xlsx_file_upload/blob/master/patients/views.py>In which we read line by line and create the objects with a python script and make model objects with it. Creating a sub app to only be able to upload your dataset into your model seems like a stretch. This process could be made seem less if we just made this on the backend of django, which is going to take effort but I do think that it will be useful to the user. So the whole could simply boil down to `*python manage.py loaddata dataset.csv <app>.<ModelName>*` If you consider this then the assumption that one table will correspond to a single csv file really is pretty reasonable I think. So I'm not suggesting that we re-open the issue, I do however think that for django which is so user friendly and time efficient, uploading a csv file into a table really shouldn't be so long a process. Let me know if I made any sense.. On Friday, 27 November 2020 at 10:55:24 UTC+5:30 Mariusz Felisiak wrote: > Hi, > > Please take a look at existing ticket #5253 > <https://code.djangoproject.com/ticket/5253> which was rejected. We > should reach a strong consensus on the mailing list to reopen a closed > ticket (see triaging guidelines with regards to wontfix tickets > <https://docs.djangoproject.com/en/stable/internals/contributing/triaging-tickets/#closing-tickets>), > > but I think that Russ' comment > <https://code.djangoproject.com/ticket/5253#comment:25> is still valid. > > Best, > Mariusz > -- You received this message because you are subscribed to the Google Groups "Django developers (Contributions to Django itself)" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-developers+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/django-developers/fce8a7f9-ef07-4ca1-b32c-d5dbc3345209n%40googlegroups.com.
