Implement CSV Storage,

[panarch]

Add CSVStorage Functionality

Introducing CSVStorage: a utility to process *.csv files, enabling SQL-like query operations such as SELECT, INSERT, and UPDATE.

Key Features:

1. SQL Queries on CSV: Directly parse and operate on *.csv files using familiar SQL query operations.

2. Optional Schema Support: An associated schema can be provided for each CSV file. For instance, for a data file named Book.csv, its corresponding schema file should be named Book.sql.

   • If an associated schema file is found, it will be read and applied.
   • In the absence of a schema file, the first row of the data file will be treated as column headers and all column types will default to TEXT.

3. Type Info File for Schemaless Data: An auxiliary types file (*.types.csv) can be used to support data type recognition for schemaless data.

   • For a CSV data file named Book.csv, its corresponding types file will be Book.types.csv.
   • The types file will have a 1:1 mapping with the CSV data file entries, specifying the data type for each entry in alignment with the GlueSQL conventions.

[coveralls]

Pull Request Test Coverage Report for Build 5864011608

• 797 of 800 (99.63%) changed or added relevant lines in 11 files are covered.
• 1 unchanged line in 1 file lost coverage.
• Overall coverage decreased (-0.02%) to 99.18%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
storages/csv-storage/src/store_mut.rs	155	158	98.1%

Files with Coverage Reduction	New Missed Lines	%
core/src/executor/validate.rs	1	99.24%

Totals
Change from base Build 5857086559:	-0.02%
Covered Lines:	46791
Relevant Lines:	47178

---

💛 - Coveralls

[devgony]

Looks like it is okay to use just immutable and variable

let fetch_data_header_columns = data_rdr
    .headers()
    .map_storage_err()?
    .into_iter()
    .map(|header| header.to_string())
    .collect::<Vec<_>>();

[devgony]

Oh... If it was intended to run only in 2 of 3 conditions, mutable closure would be right.

[devgony]

How about using mut and extend?

let mut appended_rows = prev_rows
    .map(|result| Ok(result?.1))
    .collect::<Result<Vec<_>>>()?;

appended_rows.extend(rows);

self.write(table_name, columns, appended_rows)

[panarch]

how about

let rows = prev_rows
    .map(|item| item.map(|(_, row)| row))
    .chain(rows.into_iter().map(Ok))
    .collect::<Result<Vec<_>>>()?;

this? then we can merge two rows with a single collect.