-- Leo's gemini proxy

-- Connecting to bbs.geminispace.org:1965...

-- Connected

-- Sending request

-- Meta line: 20 text/gemini; charset=utf-8

Exporting Bubble data in a "raw" format?


In addition to the human-readable Gempub archives, I've been considering various options for exporting your data out of Bubble in case someone wants to get their data out for other purposes.


For example, one might want to use their posts as a basis for a personal capsule, or run some scripts to post-process them somehow.


What would be the best "raw" machine-readable export format? Any good alternatives not listed?


Poll Results


1. SQLite3 database file

████▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁ 15%

2. SQL database dump (text file with SQL)

█▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁ 5%

3. JSON file with structured data

██████████████████▁▁▁▁▁▁ 75%

4. YAML file

█▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁ 5%

5. TOML file

▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁ 0%


20 votes were cast.


🗳️

Posted in: s/Bubble

🕹️ skyjake [mod, sysop]

2023-05-25 · 1 year ago


10 Comments ↓


🕹️ skyjake [OP/mod...] · 2023-05-25 at 08:30:

JSON is probably the easiest to parse and process, but the database ones would enable advanced queries out of the box.


(Considered including XML as a humor option, but I really don't want to deal with XML. 🤢)


🤖 alexlehm · 2023-05-25 at 08:34:

I am kind of for XML, even though that is not popular anymore, reading a JSON file is not my favorite thing since it uses more quoting than an XML file with UTF8


🤖 alexlehm · 2023-05-25 at 08:35:

not sure how xml works in python for an export


🍵 michaelnordmeyer · 2023-05-25 at 11:54:

Doesn't Bubble require a MariaDB? I would expect an export to be importable without much effort again. Otherwise I would prefer JSON.


🕹️ skyjake [OP/mod...] · 2023-05-25 at 12:22:

Yeah I'm using MariaDB. Importing is not straightforward regardless of the format, because there's a need to remap ID numbers so they don't conflict with existing ones, and one has to check whether the same data is already present in the database to avoid duplication.


👤 AnoikisNomads · 2023-05-25 at 18:52:

JSON is fine. It is easy enough to reimport well-structured JSON into a relational database if needed -- or use import raw JSON into a RDBMS and use JSON_* SQL functions directly.


— MariaDB JSON functions

— MS-SQL


seriously though, I believe that is a usecase you do not optimize for


😎 flipperzero · 2023-05-28 at 00:10:

i'd like to contribute and agree w/ some ppl in the comments in reference to XML


🐝 Addison · 2023-05-28 at 00:54:

JSON (and to a lesser extent CSV) is pretty much ubiquitous for this kind of thing in my opinion, and is use-case-agnostic (which SQL is not)


I'm in favor of JSON for the sake of its simplicity and universality.


🦀 jeang3nie · 2023-05-28 at 13:30:

@Addison I agree, and it's the same reasoning I had for voting `json`. Either json or csv are both easy to parse in any language so making sense of the raw data outside the context of a database is possible.


🕹️ skyjake [OP/mod...] · 2023-05-29 at 20:37:

I think the results are starting to look pretty clear. I will do a JSON export and consider an SQLite file as a secondary option.

-- Response ended

-- Page fetched on Sun May 19 20:21:49 2024