About

Raw text data from logs, CSVs, or note dumps needs structure before publication. This converter parses plain text line-by-line, splitting each entry by a configurable delimiter (pipe |, tab, comma, or custom string) into discrete fields: title, description, link, and date. It then generates clean, semantic HTML using <article> elements with proper heading hierarchy and microdata-ready markup. The output is not a simulation - it produces valid, copy-paste-ready HTML that renders correctly in any browser or email client.

Incorrect feed markup causes broken rendering in RSS readers, malformed Open Graph previews, and accessibility failures for screen reader users. This tool escapes all HTML entities in your input to prevent XSS injection, validates URL patterns, and auto-detects date strings. The delimiter-based parsing handles edge cases like empty fields, trailing whitespace, and lines with fewer columns than expected. Note: this tool assumes one feed item per line. Multi-line entries require manual concatenation before input.

Formulas

The conversion pipeline applies a deterministic sequence of transformations to each line L_i of the input text.

L_i → trim(L_i) → split(L_i, d) → [f₁, f₂, f₃, f₄] → escapeHTML(f_k) → render(<article>)

Where d is the user-selected delimiter string, f₁ = title, f₂ = description, f₃ = URL, f₄ = date. Empty lines where trim(L_i) = "" are discarded. URL validation uses the pattern ^https?:\/\/.+ applied via RegExp.test(). The total item count N = n∑i=1 1 for each L_i where trim(L_i) ≠ "".

Reference Data

Delimiter	Symbol	Common Source	Escape Needed	Notes
Pipe	\|	Database exports, wiki markup	No	Recommended default - rare in natural text
Tab	\t	Spreadsheets (TSV), clipboard paste	No	Invisible character - use preview to verify
Comma	,	CSV files, form data	Yes, if in content	Conflicts with natural punctuation
Semicolon	;	European CSV, SQL exports	Rare	Good alternative to comma
Double colon	::	Config files, custom logs	No	Multi-char delimiter - very safe
Arrow	=>	Key-value logs, debug output	No	Two-char, unlikely in content
Tilde	~	Unix paths, custom formats	No	Rare in body text
Hash	#	Comments, tags	No	May conflict with Markdown headings
Newline only	\n	Simple lists, one field per line	N/A	Each line becomes a title-only item
Custom	User-defined	Any proprietary format	Depends	Enter any string as delimiter
Field Mapping Order
Field 1	Title (required) - rendered as feed item heading
Field 2	Description (optional) - rendered as paragraph body
Field 3	URL (optional) - auto-linked if valid https?:// pattern detected
Field 4	Date (optional) - parsed and displayed as <time> element
HTML Entity Escaping
&	&	Ampersand - most common injection vector
<	<	Less-than - prevents tag injection
>	>	Greater-than - closes injected tags
"	"	Double quote - attribute injection
'	'	Single quote - JS string escape

Frequently Asked Questions

The parser maps fields positionally: field 1 always becomes the title. If fewer fields exist than the 4-slot schema (title, description, URL, date), the missing fields are left empty. A line with only one field renders as a title-only feed item with no description, link, or date. No error is thrown - the tool handles sparse data gracefully.

Field 3 is tested against the regular expression ^https?:\/\/.+ which requires the string to begin with http:// or https:// followed by at least one character. Strings that fail this test are rendered as plain text rather than anchor tags. FTP, mailto, and protocol-relative URLs (//example.com) are intentionally excluded to prevent malformed links in the output feed.

The output generates semantic HTML article elements, not XML-compliant RSS/Atom markup. To use in RSS, you would wrap each item's HTML inside a CDATA block within an element. The tool escapes all entities correctly, so the HTML is safe for CDATA embedding. However, the , , and XML wrapper must be added separately.

All HTML special characters in the input are escaped: < becomes <, > becomes >, & becomes &, and quotes are escaped. This is a security measure to prevent XSS injection. If your source text contains intentional HTML (e.g., bold), it will render as visible text, not as formatted markup. This is by design - the tool treats all input as untrusted plain text.

The parser operates in O(n) time complexity where n is the number of lines. In testing, 10,000 lines parse and render in under 200ms on modern hardware. Beyond 50,000 lines, DOM rendering becomes the bottleneck rather than parsing. For very large datasets (100k+ lines), consider processing in chunks or using the download feature rather than live preview, as the browser's layout engine may throttle.

The tool does not force a specific date format. It wraps the raw date string inside an HTML element. If the string matches ISO 8601 patterns (YYYY-MM-DD or YYYY-MM-DDTHH:MM:SS), it is also set as the datetime attribute for machine readability. Other formats (e.g., "March 5, 2024" or '05/03/24') are displayed as-is but the datetime attribute is omitted since ambiguous formats cannot be reliably parsed without locale context.