Add support for text/markdown content type

internal/magic/text.go

@@ -121,6 +121,72 @@ var (
 	)
 	// Rtf matches a Rich Text Format file.
 	Rtf = prefix([]byte("{\\rtf"))
+
+	// Markdown matches a Markdown document
+	// Markdown matches a Markdown document. Detection looks for multiple instances
+	// of standard Markdown syntax patterns to avoid false positives with plain text.
+	Markdown = func(raw []byte, limit uint32) bool {
+		// Skip if empty
+		if len(raw) == 0 {
+			return false
+		}
+
+		// Check first 1024 bytes for Markdown patterns
+		checkLen := 1024
+		if len(raw) < checkLen {
+			checkLen = len(raw)
+		}
+		content := raw[:checkLen]
+
+		// Lines that definitively indicate Markdown syntax when found at start or after newline
+		mdPatterns := [][]byte{
+			[]byte("# "),   // ATX heading
+			[]byte("## "),  // Level 2 heading
+			[]byte("### "), // Level 3 heading
+			[]byte("- "),   // Unordered list
+			[]byte("* "),   // Alternate unordered list
+			[]byte("1. "),  // Ordered list
+			[]byte("> "),   // Blockquote
+			[]byte("```"),  // Code fence
+		}
+
+		// Special patterns that can appear anywhere
+		mdInlinePatterns := [][]byte{
+			[]byte("]("), // Link definition
+		}
+
+		// Count how many Markdown patterns we find
+		matches := 0
+
+		// Check patterns that must be at start or after newline
+		for _, pattern := range mdPatterns {
+			// Check at start of content
+			if bytes.HasPrefix(content, pattern) {
+				matches++
+			} else {
+				// Check after newlines
+				p := append([]byte{'\n'}, pattern...)
+				if bytes.Contains(content, p) {
+					matches++
+				}
+			}
+			if matches >= 2 {
+				return true
+			}
+		}
+
+		// Check inline patterns
+		for _, pattern := range mdInlinePatterns {
+			if bytes.Contains(content, pattern) {
+				matches++
+			}
+			if matches >= 2 {
+				return true
+			}
+		}
+
+		return false
+	}
 )
 
 // Text matches a plain text file.

mimetype_test.go

@@ -164,6 +164,9 @@ var testcases = []testcase{
 	{"audio mp4 MSNV", "\x00\x00\x00\x18ftypMSNV", "audio/mp4", false},
 	{"audio mp4 NDAS", "\x00\x00\x00\x18ftypNDAS", "audio/mp4", false},
 	{"lnk", "\x4C\x00\x00\x00\x01\x14\x02\x00", "application/x-ms-shortcut", true},
+	{"markdown full", "# Heading\n\nParagraph with [link](http://example.com)\n\n* List item\n", "text/markdown", true},
+	{"markdown code", "# Title\n\n```go\nfunc main() {\n    fmt.Println(\"Hello\")\n}\n```\n", "text/markdown", true},
+	{"markdown it is not", "This is just [some] plain text with * asterisks * and # hashes\nBut no actual markdown.", "text/plain; charset=utf-8", false},
 	{"mdb", offset(4, "Standard Jet DB"), "application/x-msaccess", true},
 	{"midi", "\x4D\x54\x68\x64", "audio/midi", true},
 	{"mkv", "\x1a\x45\xdf\xa3\x01\x00\x00\x00\x00\x00\x00\x23\x42\x86\x81\x01\x42\xf7\x81\x01\x42\xf2\x81\x04\x42\xf3\x81\x08\x42\x82\x88\x6d\x61\x74\x72\x6f\x73\x6b\x61", "video/x-matroska", true},

supported_mimes.md

@@ -1,4 +1,4 @@
-## 178 Supported MIME types
+## 179 Supported MIME types
 This file is automatically generated when running tests. Do not edit manually.
 
 Extension | MIME type | Aliases
@@ -181,3 +181,4 @@ Extension | MIME type | Aliases
 **.ics** | text/calendar | -
 **.warc** | application/warc | -
 **.vtt** | text/vtt | -
+**.md** | text/markdown | -

tree.go

@@ -80,20 +80,21 @@ var (
 		alias("application/x-ogg")
 	oggAudio = newMIME("audio/ogg", ".oga", magic.OggAudio)
 	oggVideo = newMIME("video/ogg", ".ogv", magic.OggVideo)
-	text     = newMIME("text/plain", ".txt", magic.Text, html, svg, xml, php, js, lua, perl, python, json, ndJSON, rtf, srt, tcl, csv, tsv, vCard, iCalendar, warc, vtt)
+	text     = newMIME("text/plain", ".txt", magic.Text, html, svg, xml, php, js, lua, perl, python, json, ndJSON, rtf, srt, tcl, csv, tsv, vCard, iCalendar, warc, vtt, markdown)
 	xml      = newMIME("text/xml", ".xml", magic.XML, rss, atom, x3d, kml, xliff, collada, gml, gpx, tcx, amf, threemf, xfdf, owl2).
 			alias("application/xml")
-	json    = newMIME("application/json", ".json", magic.JSON, geoJSON, har)
-	har     = newMIME("application/json", ".har", magic.HAR)
-	csv     = newMIME("text/csv", ".csv", magic.Csv)
-	tsv     = newMIME("text/tab-separated-values", ".tsv", magic.Tsv)
-	geoJSON = newMIME("application/geo+json", ".geojson", magic.GeoJSON)
-	ndJSON  = newMIME("application/x-ndjson", ".ndjson", magic.NdJSON)
-	html    = newMIME("text/html", ".html", magic.HTML)
-	php     = newMIME("text/x-php", ".php", magic.Php)
-	rtf     = newMIME("text/rtf", ".rtf", magic.Rtf).alias("application/rtf")
-	js      = newMIME("text/javascript", ".js", magic.Js).
-		alias("application/x-javascript", "application/javascript")
+	json     = newMIME("application/json", ".json", magic.JSON, geoJSON, har)
+	markdown = newMIME("text/markdown", ".md", magic.Markdown)
+	har      = newMIME("application/json", ".har", magic.HAR)
+	csv      = newMIME("text/csv", ".csv", magic.Csv)
+	tsv      = newMIME("text/tab-separated-values", ".tsv", magic.Tsv)
+	geoJSON  = newMIME("application/geo+json", ".geojson", magic.GeoJSON)
+	ndJSON   = newMIME("application/x-ndjson", ".ndjson", magic.NdJSON)
+	html     = newMIME("text/html", ".html", magic.HTML)
+	php      = newMIME("text/x-php", ".php", magic.Php)
+	rtf      = newMIME("text/rtf", ".rtf", magic.Rtf).alias("application/rtf")
+	js       = newMIME("text/javascript", ".js", magic.Js).
+			alias("application/x-javascript", "application/javascript")
 	srt = newMIME("application/x-subrip", ".srt", magic.Srt).
 		alias("application/x-srt", "text/x-srt")
 	vtt    = newMIME("text/vtt", ".vtt", magic.Vtt)