I will create a local python script to convert CSV to json


About this gig
Repeat the same CSV-to-JSON conversion locally with a small Python command-line script configured for your confirmed header structure. You receive reusable source code, not a data-cleaning service.
Basic covers one agreed CSV schema: up to 10 columns, 1,000 data rows and 1 MiB (1,048,576 bytes) per input. UTF-8, optional BOM, and one agreed comma, semicolon or tab separator. Output is one flat JSON array. All values stay strings; IDs, spaces, blank fields, row order and duplicates are retained.
You receive the script, README, synthetic input/output sample and short tests. Existing output files are refused and inputs are not edited.
Send an anonymised sample, exact headers and expected JSON before ordering. Confirm your Python version and OS: the work sample was tested only on Python 3.13.0 on macOS. Other environments require review before acceptance.
No cleaning, type/date inference, nested JSON, APIs, hosting, EXE packaging or Python installation. I use AI-assisted tools, check your agreed examples and arrange separate QA and personal review before delivery. The portfolio is my own synthetic demo, not paid client work.
Get to know Keyan Dogan
Excel Data Cleaning, Templates, Calculators and Python CSV to JSON
- FromGermany
- Member sinceNov 2026
- Avg. response time1 hour
Languages
German, English, Turkish
My Portfolio
FAQ
How is this different from your data-cleaning service?
This delivery is a reusable local Python script that converts the same confirmed CSV structure to JSON on future runs. It keeps the field values and duplicate records. My separate cleaning service changes supplied product data using agreed cleanup rules.
What JSON structure will I receive?
One flat array of objects. Your exact confirmed headers become keys, and every field value stays a string. Blank fields remain empty strings. No nesting, inferred numbers, dates, grouping or key renaming is included.
Will it work on my computer?
Please provide your OS and Python version before ordering. The synthetic work sample was executed only on Python 3.13.0 on macOS. I do not promise other environments without reviewing and agreeing the required checks. Python installation, remote access and executable packaging are excluded.
What happens if the input is invalid or an output exists?
The script rejects incompatible headers, inconsistent field counts, invalid UTF-8 and parser-detected quoting errors with a readable message. It does not guess or clean values. A failed input check creates no output. Existing output paths are refused, and the original CSV is not edited.
Do you use AI, and is the sample a client project?
I use AI-assisted tools with checks against the confirmed brief, separate QA and personal review before delivery. The portfolio is my own synthetic demonstration, not a paid client project. Please disclose any AI restriction before acceptance.
What does one revision cover?
One consolidated correction against the same agreed headers, separator and JSON rules. A new schema, feature, format, environment requirement or integration needs a new agreement and capacity check. Unlimited support is not included.
Can you clean data or connect the script to an app?
This package only converts one confirmed local CSV schema. Data cleaning, type inference, web scraping, APIs, hosting, scheduling and database imports are excluded. Target-application acceptance must be checked against your expected JSON; an import result is not promised.

