Just the way we likes it.

Comments

• Saganaki@lemmy.zip ⁨2⁩ ⁨months⁩ ago

  “Wait, why is this not a CSV file?”

  source

  • Jason2357@lemmy.ca ⁨2⁩ ⁨months⁩ ago

    God I hate csv with the fire of a thousand suns.

    Contractors never seem to know how to write them correctly. Last year, one even provided “csv”s that were just Oracle error messages. lol. Another told me their system could not quote string columns nor escape commas or use anything but commas as their separator, so there were unpredictable numbers of commas in the rows when the actual data contained commas. Total nightmare. And so much of my data has special character issues because somewhere in the pipeline a text encoding was wrong and there is exactly one mangled character in 5 million lines for me to find.

    Give me the data as closely to the source data as you can. If it is a database, then a database dump or access to a clone of your database is the best option by far. I don’t care how obscure your shit is, Ill do the conversion myself.

    For intermediate data, something like parquet or language specific formats like Rdata or pickle files. Maaaaybe very carefully created csv files for archival purposes, but even then, I think parquet is safe for the long haul nowadays.

    source

    • kernelle@lemmy.dbzer0.com ⁨2⁩ ⁨months⁩ ago

      I can’t tell you how many scripts I’ve written to format poorly made CSV files

      source

      • -> View More Comments
    • vithigar@lemmy.ca ⁨2⁩ ⁨months⁩ ago

      After many years of being a developer I’ve come to the conclusion that the single strongest indicator of a person’s competence is how they handle CSV when asked to produce or consume it.

      source

      • -> View More Comments
    • GTG3000@programming.dev ⁨2⁩ ⁨months⁩ ago

      Reminds me of writing my own csv parser that implemented escapes properly. The one everyone else went with of course was written in regex, so it was faster… But broke if there were escaped newlines.

      source
    • espurr@sopuli.xyz ⁨2⁩ ⁨months⁩ ago

      What delimiter should I be using instead of commas?

      source

      • -> View More Comments
    • Jason2357@lemmy.ca ⁨2⁩ ⁨months⁩ ago

      P.s. in the above quagmire, the only solution is choose to keep only the most important un-clean column per csv, and make it the last column in the file so you have predictable columns. If you need more, then write separate csvs. Computers are stupid.

      source

      • -> View More Comments
• harmbugler@piefed.social ⁨2⁩ ⁨months⁩ ago

  But, please, not wriggling.

  source
• Berengaria_of_Navarre@lemmy.world ⁨2⁩ ⁨months⁩ ago

  Oh baby I like it raw🎵

  Oh baby I like it raaAAWWWW 🎵

  source

  • fossilesque@mander.xyz ⁨2⁩ ⁨months⁩ ago

    I’m glad I wasn’t then only one lol.

    source
• Chakravanti@monero.town ⁨2⁩ ⁨months⁩ ago

  I’m not sure you understand what Golem is…

  source
• diabetic_porcupine@lemmy.world ⁨2⁩ ⁨months⁩ ago

  I like to raw dog my data exports

  source