Add rainbow highlighting support for tab-separated (TSV) files

[Serhioromano]

Problem

Files opened with tab characters as delimiters are not highlighted with rainbow colors. Tab-separated files (TSV) should support the same rainbow highlighting as comma-separated CSV files.

Use Case

I have a file open in VS Code where the columns are separated by tab characters (not commas). The Rainbow CSV extension does not apply rainbow syntax highlighting to this file.

Expected Behavior

When a tab-separated file is opened, each column should be highlighted with a different color, just like comma-separated CSV files.

Actual Behavior

Tab-separated files receive no rainbow highlighting — the extension appears to only work with comma-delimited files.

Steps to Reproduce

1. Open a .tsv (or other tab-delimited) file in VS Code
2. Observe that no rainbow column highlighting is applied
3. The file displays with default syntax highlighting only

Environment

• Extension: mechatroner.rainbow-csv
• Editor: Visual Studio Code

[Serhioromano]

Code Analysis

After reviewing the repository source code, here is what I found:

TSV Support Already Exists

The extension already supports TSV files. Here is the evidence from the codebase:

1. Language registration (package.json): TSV is registered as a language with id "tsv", file extensions .tsv and .tab.

2. Syntax grammar (syntaxes/tsv.tmLanguage.json): A dedicated TextMate grammar exists that maps tab-separated columns to the 10 rainbow color scopes (rainbow1 through rainbow10).

3. Dialect map (extension.js:127): 'tsv': ['\t', SIMPLE_POLICY] — tab is a first-class delimiter.

4. Autodetection (extension.js:1969): "TAB" is in the default autodetect_separators list, meaning opening a plain-text file with tab-separated content should auto-switch it to tsv language.

5. Activation (package.json): "onLanguage:tsv" is in activation events.

Why It Might Not Work

The likely scenarios where a tab-delimited file won't get rainbow highlighting:

1. File has no .tsv / .tab extension and autodetection fails

If a file doesn't have a .tsv or .tab extension, VS Code opens it as plaintext. The extension then tries to auto-detect the separator. Autodetection can fail if:

• The file has fewer than 10 non-comment lines (default of rainbow_csv.autodetection_min_line_count)
• The separator autodetection is disabled in settings (rainbow_csv.enable_separator_autodetection)
• The file is opened with a non-plaintext language that isn't already a rainbow dialect (the check at extension.js:1980 skips autodetection for non-plaintext, non-csv files)

2. Semantic highlighting disabled by theme

The configurationDefaults section in package.json sets "editor.semanticHighlighting.enabled": true for [tsv], but some themes override this. The grammar-based highlighting (TextMate) should still work, but if it doesn't, the theme might be the culprit.

Proposed Solutions

Quick Workaround (no code changes)

1. Open the file → click the language mode indicator in the bottom-right → select "TSV"
2. Or manually select a separator: Ctrl+Shift+P → "Rainbow CSV: Set rainbow separator" → choose Tab

Code Fix

There are two improvements that could be made:

A. Add ".txt" to TSV language extensions in package.json (optional, may be too aggressive):

{
    "id": "tsv",
    "extensions": [".tsv", ".tab", ".txt"]
}

B. Improve autodetection for files opened as other non-plaintext languages — currently try_autodetect_and_set_rainbow_filetype() at extension.js:1980 rejects files that aren't plaintext or .csv:

let is_default_csv = (file_path.endsWith('.csv') || file_path.endsWith('.CSV')) && original_language_id == 'csv';
if (original_language_id != PLAINTEXT && !is_default_csv) {
    return [active_doc, false];  // Skips autodetection entirely
}

This means a .log or .dat file that VS Code opens as "log" or another language will never be auto-detected as TSV even if it contains tab-separated data. A fix would be to expand this condition to allow autodetection for more file types, or add a user-facing command like "Detect Separator" that can be run on any file regardless of language mode.

C. Add a status bar quick-pick for separator selection on any file — currently the "Choose Separator" button only appears for dynamic csv dialect. Extending it to plaintext files (or all files) would let users quickly activate rainbow highlighting for any delimiter without navigating through language menus.