Move all awk functions to separate files

awk/blockquote.awk

@@ -0,0 +1,14 @@
+BEGIN { in_bq = 0 }
+/^>[[:space:]]?/ {
+    if (!in_bq) { print "<blockquote>"; in_bq = 1 }
+    sub(/^>[[:space:]]?/, "", $0)
+    print $0
+    next
+}
+{
+    if (in_bq) { print "</blockquote>"; in_bq = 0 }
+    print
+}
+END {
+    if (in_bq) print "</blockquote>"
+}

awk/blockquote_to_admonition.awk

@@ -0,0 +1,46 @@
+function cap(s) { return toupper(substr(s, 1, 1)) tolower(substr(s, 2)) }
+BEGIN { count = 0 }
+{ lines[++count] = $0 }
+END {
+    i = 1
+    while (i <= count) {
+        if (lines[i] == "<blockquote>") {
+            j = i + 1
+            while (j <= count && lines[j] != "</blockquote>") j++
+            if (j <= count) {
+                first = ""
+                first_idx = 0
+                for (k = i + 1; k < j; k++) {
+                    if (lines[k] != "") {
+                        first = lines[k]
+                        first_idx = k
+                        break
+                    }
+                }
+                if (first ~ /^\[![A-Za-z]+\]$/) {
+                    kind = first
+                    sub(/^\[!/, "", kind)
+                    sub(/\]$/, "", kind)
+                    lkind = tolower(kind)
+                    if (lkind == "note" || lkind == "tip" || lkind == "important" || lkind == "warning" || lkind == "caution") {
+                        print "<div class=\"admonition admonition-" lkind "\">"
+                        print "<p class=\"admonition-title\">" cap(lkind) "</p>"
+                        has_body = 0
+                        for (k = first_idx + 1; k < j; k++) {
+                            if (lines[k] != "") {
+                                print "<p>" lines[k] "</p>"
+                                has_body = 1
+                            }
+                        }
+                        if (!has_body) print "<p></p>"
+                        print "</div>"
+                        i = j + 1
+                        continue
+                    }
+                }
+            }
+        }
+        print lines[i]
+        i++
+    }
+}

awk/breaks.awk

@@ -0,0 +1,5 @@
+{
+    if ($0 == "" && prev == "") print "<br />"
+    else print $0
+    prev = $0
+}

awk/fenced_code.awk

@@ -0,0 +1,27 @@
+BEGIN { in_fence = 0; first_line = 0 }
+{
+    if (!in_fence && $0 ~ /^```/) {
+        in_fence = 1
+        first_line = 1
+        next
+    }
+    if (in_fence && $0 ~ /^```[[:space:]]*$/) {
+        print "</code></pre>"
+        in_fence = 0
+        next
+    }
+    if (in_fence) {
+        if (first_line) {
+            first_line = 0
+            if ($0 == "") next
+            print "<pre><code>" $0
+        } else {
+            print
+        }
+    } else {
+        print
+    }
+}
+END {
+    if (in_fence) print "</code></pre>"
+}

awk/generate_sidebar.awk

@@ -4,6 +4,34 @@ function title_from_name(name) {
     return name
 }
 
+function compare_paths(p1, p2,    parts1, parts2, n1, n2, i, name1, name2, lname1, lname2, w1, w2) {
+    n1 = split(p1, parts1, "/")
+    n2 = split(p2, parts2, "/")
+    for (i = 1; i <= n1 && i <= n2; i++) {
+        name1 = parts1[i]
+        name2 = parts2[i]
+        if (i == n1) gsub(/\.md$/, "", name1)
+        if (i == n2) gsub(/\.md$/, "", name2)
+        lname1 = tolower(name1)
+        lname2 = tolower(name2)
+
+        if (lname1 == "index" && i == n1 && lname2 != "index") return -1
+        if (lname2 == "index" && i == n2 && lname1 != "index") return 1
+
+        w1 = (lname1 in custom_order ? custom_order[lname1] : 999999)
+        w2 = (lname2 in custom_order ? custom_order[lname2] : 999999)
+
+        if (w1 < w2) return -1
+        if (w1 > w2) return 1
+
+        if (lname1 < lname2) return -1
+        if (lname1 > lname2) return 1
+    }
+    if (n1 < n2) return -1
+    if (n1 > n2) return 1
+    return 0
+}
+
 BEGIN {
     n_dlines = split(dinfo, dlines, "\n")
     for (i = 1; i <= n_dlines; i++) {
@@ -39,33 +67,6 @@ BEGIN {
         has_index[dir] = 1
     }
 }
-function compare_paths(p1, p2,    parts1, parts2, n1, n2, i, name1, name2, lname1, lname2, w1, w2) {
-    n1 = split(p1, parts1, "/")
-    n2 = split(p2, parts2, "/")
-    for (i = 1; i <= n1 && i <= n2; i++) {
-        name1 = parts1[i]
-        name2 = parts2[i]
-        if (i == n1) gsub(/\.md$/, "", name1)
-        if (i == n2) gsub(/\.md$/, "", name2)
-        lname1 = tolower(name1)
-        lname2 = tolower(name2)
-
-        if (lname1 == "index" && i == n1 && lname2 != "index") return -1
-        if (lname2 == "index" && i == n2 && lname1 != "index") return 1
-
-        w1 = (lname1 in custom_order ? custom_order[lname1] : 999999)
-        w2 = (lname2 in custom_order ? custom_order[lname2] : 999999)
-
-        if (w1 < w2) return -1
-        if (w1 > w2) return 1
-
-        if (lname1 < lname2) return -1
-        if (lname1 > lname2) return 1
-    }
-    if (n1 < n2) return -1
-    if (n1 > n2) return 1
-    return 0
-}
 
 END {
     for (i = 0; i < count - 1; i++) {
@@ -78,7 +79,6 @@ END {
         }
     }
 
-
     print "<ul>"
     if ("index.md" in all_paths) {
         print "<li><a href=\"/index.html\">Home</a></li>"

awk/headers.awk

@@ -0,0 +1,65 @@
+function strip_markdown(s) {
+    gsub(/<[^>]+>/, "", s)
+    gsub(/[*_`~]/, "", s)
+    gsub(/[\[\]]/, "", s)
+    gsub(/\([^\)]*\)/, "", s)
+    sub(/^[[:space:]]*/, "", s)
+    sub(/[[:space:]]*$/, "", s)
+    return s
+}
+BEGIN {
+    has_prev = 0
+    in_pre = 0
+}
+{
+    if ($0 ~ /^<pre><code>/) {
+        in_pre = 1
+        if (has_prev && prev != "") { print prev; has_prev = 0 }
+        print
+        next
+    }
+    if (in_pre) {
+        if ($0 ~ /<\/code><\/pre>/) in_pre = 0
+        print
+        next
+    }
+
+    if ($0 ~ /^=+$/ && has_prev && prev != "" && prev !~ /^<[a-z]/) {
+        print "<h1 id=\"" strip_markdown(prev) "\">" prev "</h1>"
+        has_prev = 0
+    } else if ($0 ~ /^-+$/ && has_prev && prev != "" && prev !~ /^<[a-z]/) {
+        print "<h2 id=\"" strip_markdown(prev) "\">" prev "</h2>"
+        has_prev = 0
+    } else {
+        if (has_prev) {
+            line = prev
+            if (line ~ /^# /) {
+                sub(/^# /, "", line); print "<h1 id=\"" strip_markdown(line) "\">" line "</h1>"
+            } else if (line ~ /^## /) {
+                sub(/^## /, "", line); print "<h2 id=\"" strip_markdown(line) "\">" line "</h2>"
+            } else if (line ~ /^### /) {
+                sub(/^### /, "", line); print "<h3 id=\"" strip_markdown(line) "\">" line "</h3>"
+            } else if (line ~ /^#### /) {
+                sub(/^#### /, "", line); print "<h4 id=\"" strip_markdown(line) "\">" line "</h4>"
+            } else if (line ~ /^##### /) {
+                sub(/^##### /, "", line); print "<h5 id=\"" strip_markdown(line) "\">" line "</h5>"
+            } else if (line ~ /^###### /) {
+                sub(/^###### /, "", line); print "<h6 id=\"" strip_markdown(line) "\">" line "</h6>"
+            } else {
+                print prev
+            }
+        }
+        prev = $0
+        has_prev = 1
+    }
+}
+END {
+    if (has_prev) {
+        line = prev
+        if (line ~ /^# /) {
+            sub(/^# /, "", line); print "<h1 id=\"" strip_markdown(line) "\">" line "</h1>"
+        } else {
+            print prev
+        }
+    }
+}

awk/indented_code.awk

@@ -0,0 +1,9 @@
+BEGIN { in_code = 0 }
+/^	|    / {
+    if (!in_code) { print "<pre><code>"; in_code = 1 }
+    sub(/^	|    /, "", $0)
+    gsub(/&/, "&amp;"); gsub(/</, "&lt;"); gsub(/>/, "&gt;")
+    print; next
+}
+{ if (in_code) { print "</code></pre>"; in_code = 0 } print }
+END { if (in_code) print "</code></pre>" }

awk/lists.awk

@@ -0,0 +1,66 @@
+BEGIN {
+    depth = 0
+    in_pre = 0
+}
+
+{
+    if ($0 ~ /^<pre>/) in_pre = 1
+    if (in_pre) {
+        while (depth > 0) { print "</" cur_type[depth] ">"; depth-- }
+        print
+        if ($0 ~ /<\/pre>/) in_pre = 0
+        next
+    }
+
+    line = $0
+    type = ""
+    # match list marker and its preceding spaces
+    if (line ~ /^[ \t]*[*+-] /) {
+        type = "ul"
+        match(line, /^[ \t]*[*+-] /)
+        marker_len = RLENGTH
+    } else if (line ~ /^[ \t]*[0-9]+\. /) {
+        type = "ol"
+        match(line, /^[ \t]*[0-9]+\. /)
+        marker_len = RLENGTH
+    }
+
+    if (type != "") {
+        content = substr(line, marker_len + 1)
+        # get indentation level
+        match(line, /^[ \t]*/)
+        indent = RLENGTH
+        
+        if (depth == 0 || indent > cur_indent[depth]) {
+            depth++
+            cur_indent[depth] = indent
+            cur_type[depth] = type
+            print "<" type ">"
+        } else {
+            while (depth > 1 && indent < cur_indent[depth]) {
+                print "</" cur_type[depth] ">"
+                depth--
+            }
+            if (type != cur_type[depth]) {
+                print "</" cur_type[depth] ">"
+                cur_type[depth] = type
+                print "<" type ">"
+            }
+        }
+
+        print "<li>" content "</li>"
+    } else {
+        while (depth > 0) {
+            print "</" cur_type[depth] ">"
+            depth--
+        }
+        print line
+    }
+}
+
+END {
+    while (depth > 0) {
+        print "</" cur_type[depth] ">"
+        depth--
+    }
+}

awk/markdown_embed.awk

@@ -119,11 +119,35 @@ function css_highlight_line(line,    m, prop, val) {
         return "<span class=\"tok-punc\">}</span>"
     }
 
-    if (match(line, /^([[:space:]]*)(--?[A-Za-z0-9_-]+)([[:space:]]*:[[:space:]]*)([^;]*)(;?[[:space:]]*)$/, m)) {
+    if (line ~ /^[[:space:]]*--?[A-Za-z0-9_-]+[[:space:]]*:[[:space:]]*[^;]*;?[[:space:]]*$/) {
-        prop = "<span class=\"tok-prop\">" m[2] "</span>"
+        match(line, /:[[:space:]]*/)
-        gsub(/var\(--[A-Za-z0-9_-]+\)/, "<span class=\"tok-var\">&</span>", m[4])
+        sep_pos = RSTART
-        val = "<span class=\"tok-val\">" m[4] "</span>"
+        sep_len = RLENGTH
-        return m[1] prop m[3] val m[5]
+        
+        pre_sep = substr(line, 1, sep_pos - 1)
+        sep = substr(line, sep_pos, sep_len)
+        post_sep = substr(line, sep_pos + sep_len)
+        
+        match(pre_sep, /--?[A-Za-z0-9_-]+/)
+        prop_pos = RSTART
+        prop_len = RLENGTH
+        
+        indent = substr(pre_sep, 1, prop_pos - 1)
+        prop_name = substr(pre_sep, prop_pos, prop_len)
+        
+        if (match(post_sep, /;[[:space:]]*$/)) {
+            val_part = substr(post_sep, 1, RSTART - 1)
+            suffix = substr(post_sep, RSTART)
+        } else {
+            val_part = post_sep
+            suffix = ""
+        }
+        
+        prop = "<span class=\"tok-prop\">" prop_name "</span>"
+        gsub(/var\(--[A-Za-z0-9_-]+\)/, "<span class=\"tok-var\">&</span>", val_part)
+        val = "<span class=\"tok-val\">" val_part "</span>"
+        
+        return indent prop sep val suffix
     }
 
     return line

awk/markdown_inline.awk

@@ -0,0 +1,173 @@
+BEGIN {
+    in_pre = 0
+}
+
+{
+    if ($0 ~ /<pre>/) {
+        in_pre = 1
+    }
+    
+    if (in_pre) {
+        print
+        if ($0 ~ /<\/pre>/) {
+            in_pre = 0
+        }
+        next
+    }
+
+    line = $0
+
+    # automatic links
+    while (match(line, /<https?:\/\/[^>]+>/)) {
+        start = RSTART; len = RLENGTH
+        url = substr(line, start + 1, len - 2)
+        repl = "<a href=\"" url "\">" url "</a>"
+        line = substr(line, 1, start - 1) repl substr(line, start + len)
+    }
+
+    # automatic email address links
+    while (match(line, /<[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\.[A-Za-z]{2,}>/)) {
+        start = RSTART; len = RLENGTH
+        email = substr(line, start + 1, len - 2)
+        repl = "<a href=\"mailto:" email "\">" email "</a>"
+        line = substr(line, 1, start - 1) repl substr(line, start + len)
+    }
+
+    # force-inline image syntax (double bang)
+    while (match(line, /!!\[[^\]]*\]\([^\)]+ "[^"]*"\)/)) {
+        start = RSTART; len = RLENGTH
+        token = substr(line, start, len)
+        match(token, /\[[^\]]*\]/); alt = substr(token, RSTART + 1, RLENGTH - 2)
+        match(token, /"[^"]*"/); title = substr(token, RSTART + 1, RLENGTH - 2)
+        match(token, /\([^\)]+/); inner = substr(token, RSTART + 1, RLENGTH - 1)
+        sub(/[[:space:]]*"[^"]*"/, "", inner); src = inner
+        repl = "<img data-force-inline=\"1\" alt=\"" alt "\" src=\"" src "\" title=\"" title "\" />"
+        line = substr(line, 1, start - 1) repl substr(line, start + len)
+    }
+    while (match(line, /!!\[[^\]]*\]\([^\)]+\)/)) {
+        start = RSTART; len = RLENGTH
+        token = substr(line, start, len)
+        match(token, /\[[^\]]*\]/); alt = substr(token, RSTART + 1, RLENGTH - 2)
+        match(token, /\([^\)]+/); src = substr(token, RSTART + 1, RLENGTH - 1)
+        repl = "<img data-force-inline=\"1\" alt=\"" alt "\" src=\"" src "\" />"
+        line = substr(line, 1, start - 1) repl substr(line, start + len)
+    }
+
+    # inline image
+    while (match(line, /!\[[^\]]*\]\([^\)]+ "[^"]*"\)/)) {
+        start = RSTART; len = RLENGTH
+        if (start > 1 && substr(line, start - 1, 1) == "\\") break
+        token = substr(line, start, len)
+        match(token, /\[[^\]]*\]/); alt = substr(token, RSTART + 1, RLENGTH - 2)
+        match(token, /"[^"]*"/); title = substr(token, RSTART + 1, RLENGTH - 2)
+        match(token, /\([^\)]+/); inner = substr(token, RSTART + 1, RLENGTH - 1)
+        sub(/[[:space:]]*"[^"]*"/, "", inner); src = inner
+        repl = "<img alt=\"" alt "\" src=\"" src "\" title=\"" title "\" />"
+        line = substr(line, 1, start - 1) repl substr(line, start + len)
+    }
+    while (match(line, /!\[[^\]]*\]\([^\)]+\)/)) {
+        start = RSTART; len = RLENGTH
+        if (start > 1 && substr(line, start - 1, 1) == "\\") break
+        token = substr(line, start, len)
+        match(token, /\[[^\]]*\]/); alt = substr(token, RSTART + 1, RLENGTH - 2)
+        match(token, /\([^\)]+/); src = substr(token, RSTART + 1, RLENGTH - 1)
+        repl = "<img alt=\"" alt "\" src=\"" src "\" />"
+        line = substr(line, 1, start - 1) repl substr(line, start + len)
+    }
+
+    # inline link
+    while (match(line, /\[[^\]]*\]\([^\)]+ "[^"]*"\)/)) {
+        start = RSTART; len = RLENGTH
+        if (start > 1 && (substr(line, start - 1, 1) == "\\" || substr(line, start - 1, 1) == "!")) break
+        token = substr(line, start, len)
+        match(token, /\[[^\]]*\]/); text = substr(token, RSTART + 1, RLENGTH - 2)
+        match(token, /"[^"]*"/); title = substr(token, RSTART + 1, RLENGTH - 2)
+        match(token, /\([^\)]+/); inner = substr(token, RSTART + 1, RLENGTH - 1)
+        sub(/[[:space:]]*"[^"]*"/, "", inner); href = inner
+        repl = "<a href=\"" href "\" title=\"" title "\">" text "</a>"
+        line = substr(line, 1, start - 1) repl substr(line, start + len)
+    }
+    while (match(line, /\[[^\]]*\]\([^\)]+\)/)) {
+        start = RSTART; len = RLENGTH
+        if (start > 1 && (substr(line, start - 1, 1) == "\\" || substr(line, start - 1, 1) == "!")) break
+        token = substr(line, start, len)
+        match(token, /\[[^\]]*\]/); text = substr(token, RSTART + 1, RLENGTH - 2)
+        match(token, /\([^\)]+/); href = substr(token, RSTART + 1, RLENGTH - 1)
+        repl = "<a href=\"" href "\">" text "</a>"
+        line = substr(line, 1, start - 1) repl substr(line, start + len)
+    }
+
+    # MFM font syntax
+    while (match(line, /\$\[font\.serif [^\]]+\]/)) {
+        start = RSTART; len = RLENGTH
+        content = substr(line, start + 13, len - 14)
+        line = substr(line, 1, start - 1) "<span style=\"font-family: serif;\">" content "</span>" substr(line, start + len)
+    }
+    while (match(line, /\$\[font\.monospace [^\]]+\]/)) {
+        start = RSTART; len = RLENGTH
+        content = substr(line, start + 17, len - 18)
+        line = substr(line, 1, start - 1) "<span style=\"font-family: monospace;\">" content "</span>" substr(line, start + len)
+    }
+    while (match(line, /\$\[font\.sans [^\]]+\]/)) {
+        start = RSTART; len = RLENGTH
+        content = substr(line, start + 12, len - 13)
+        line = substr(line, 1, start - 1) "<span style=\"font-family: sans-serif;\">" content "</span>" substr(line, start + len)
+    }
+
+    # Bold, Italic, Strikethrough (BRE-like logic in AWK)
+    # Strong Bold **
+    while (match(line, /\*\*[^*]+\*\*/)) {
+        start = RSTART; len = RLENGTH
+        content = substr(line, start + 2, len - 4)
+        repl = "<strong>" content "</strong>"
+        line = substr(line, 1, start - 1) repl substr(line, start + len)
+    }
+    # Strong Bold __
+    while (match(line, /__[^_]+__/)) {
+        start = RSTART; len = RLENGTH
+        content = substr(line, start + 2, len - 4)
+        repl = "<strong>" content "</strong>"
+        line = substr(line, 1, start - 1) repl substr(line, start + len)
+    }
+    # Italic *
+    while (match(line, /\*[^*]+\*/)) {
+        start = RSTART; len = RLENGTH
+        content = substr(line, start + 1, len - 2)
+        repl = "<em>" content "</em>"
+        line = substr(line, 1, start - 1) repl substr(line, start + len)
+    }
+    # Italic _
+    while (match(line, /_[^_]+_/)) {
+        start = RSTART; len = RLENGTH
+        if (start > 1 && substr(line, start - 1, 1) == "\\") break
+        content = substr(line, start + 1, len - 2)
+        repl = "<em>" content "</em>"
+        line = substr(line, 1, start - 1) repl substr(line, start + len)
+    }
+    # Strikethrough ~~
+    while (match(line, /~~[^~]+~~/)) {
+        start = RSTART; len = RLENGTH
+        content = substr(line, start + 2, len - 4)
+        repl = "<strike>" content "</strike>"
+        line = substr(line, 1, start - 1) repl substr(line, start + len)
+    }
+
+    # special characters
+    if (line !~ /&[A-Za-z0-9#]+;/) {
+        gsub(/&/, "&amp;", line)
+    }
+    
+    p = 1
+    while (match(substr(line, p), /</)) {
+        start = p + RSTART - 1
+        next_char = substr(line, start + 1, 1)
+        if (next_char !~ /^[\/A-Za-z]/) {
+            line = substr(line, 1, start - 1) "&lt;" substr(line, start + 1)
+            p = start + 4
+        } else {
+            p = start + 1
+        }
+    }
+
+    print line
+}

awk/mask_inline_code.awk

@@ -0,0 +1,85 @@
+function mask(s,    t) {
+    t = s
+    gsub(/\*/, "\034P0\034", t)
+    gsub(/_/, "\034P1\034", t)
+    gsub(/`/, "\034P2\034", t)
+    gsub(/\[/, "\034P3\034", t)
+    gsub(/\]/, "\034P4\034", t)
+    gsub(/\(/, "\034P5\034", t)
+    gsub(/\)/, "\034P6\034", t)
+    gsub(/!/, "\034P7\034", t)
+    gsub(/\$/, "\034P8\034", t)
+    gsub(/#/, "\034P9\034", t)
+    gsub(/\+/, "\034P10\034", t)
+    gsub(/-/, "\034P11\034", t)
+    gsub(/\\/, "\034P12\034", t)
+    gsub(/</, "\034P13\034", t)
+    gsub(/>/, "\034P14\034", t)
+    return t
+}
+{
+    # backslash escapes
+    gsub(/\\\*/, "\034P0\034")
+    gsub(/\\_/, "\034P1\034")
+    gsub(/\\`/, "\034P2\034")
+    gsub(/\\\[/, "\034P3\034")
+    gsub(/\\\]/, "\034P4\034")
+    gsub(/\\\(/, "\034P5\034")
+    gsub(/\\\)/, "\034P6\034")
+    gsub(/\\!/, "\034P7\034")
+    gsub(/\\\$/, "\034P8\034")
+    gsub(/\\#/, "\034P9\034")
+    gsub(/\\\+/, "\034P10\034")
+    gsub(/\\\-/, "\034P11\034")
+    gsub(/\\\\/, "\034P12\034")
+    gsub(/\\</, "\034P13\034")
+    gsub(/\\>/, "\034P14\034")
+
+    # inline code (1 or 2 backticks)
+    line = $0
+    if (line ~ /^```/) {
+        print line
+        next
+    }
+    out = ""
+    p = 1
+    while (match(substr(line, p), /`+/)) {
+        pstart = p + RSTART - 1
+        plen = RLENGTH
+        if (plen >= 3) {
+            out = out substr(line, p, pstart - p + plen)
+            p = pstart + plen
+            continue
+        }
+
+        # Found 1 or 2 backticks at pstart
+        # Search for closing marker
+        marker = substr(line, pstart, plen)
+        tail = substr(line, pstart + plen)
+        mpos = index(tail, marker)
+        if (mpos > 0) {
+            # Check if it is followed by more backticks
+            if (substr(tail, mpos + plen, 1) == "`") {
+                # Not a match, treat as literal
+                out = out substr(line, p, pstart - p + plen)
+                p = pstart + plen
+                continue
+            }
+            
+            # Found match!
+            content = substr(tail, 1, mpos - 1)
+            out = out substr(line, p, pstart - p)
+            if (plen == 2 && substr(content, 1, 1) == " " && substr(content, length(content), 1) == " ") {
+                content = substr(content, 2, length(content) - 2)
+            }
+            out = out "<code>" mask(content) "</code>"
+            p = pstart + plen + mpos + plen - 1
+        } else {
+            # No closing marker, treat as literal
+            out = out substr(line, p, pstart - p + plen)
+            p = pstart + plen
+        }
+    }
+    out = out substr(line, p)
+    print out
+}

awk/mask_plain.awk

@@ -0,0 +1,52 @@
+function find_unescaped_tag(s, tag,    p, off, pos) {
+    p = 1
+    while (1) {
+        off = index(substr(s, p), tag)
+        if (off == 0) return 0
+        pos = p + off - 1
+        if (pos == 1 || substr(s, pos - 1, 1) != "\\") return pos
+        p = pos + 1
+    }
+}
+
+function mask_plain(s,    t) {
+    t = s
+    gsub(/\*/, "\034P0\034", t)
+    gsub(/_/, "\034P1\034", t)
+    gsub(/`/, "\034P2\034", t)
+    gsub(/\[/, "\034P3\034", t)
+    gsub(/\]/, "\034P4\034", t)
+    gsub(/\(/, "\034P5\034", t)
+    gsub(/\)/, "\034P6\034", t)
+    gsub(/!/, "\034P7\034", t)
+    gsub(/\$/, "\034P8\034", t)
+    return t
+}
+BEGIN { in_plain = 0 }
+{
+    line = $0
+    out = ""
+    while (1) {
+        if (!in_plain) {
+            pos = find_unescaped_tag(line, "<plain>")
+            if (pos == 0) {
+                out = out line
+                break
+            }
+            out = out substr(line, 1, pos - 1) "<mfmplain>"
+            line = substr(line, pos + 7)
+            in_plain = 1
+        } else {
+            pos = find_unescaped_tag(line, "</plain>")
+            if (pos == 0) {
+                out = out mask_plain(line)
+                line = ""
+                break
+            }
+            out = out mask_plain(substr(line, 1, pos - 1)) "</mfmplain>"
+            line = substr(line, pos + 8)
+            in_plain = 0
+        }
+    }
+    print out
+}

awk/paragraphs.awk

@@ -0,0 +1,43 @@
+BEGIN {
+    in_p = 0
+    in_pre = 0
+}
+
+{
+    if ($0 ~ /^<pre>/) in_pre = 1
+    
+    if (in_pre) {
+        if (in_p) { print "</p>"; in_p = 0 }
+        print
+        if ($0 ~ /<\/pre>/) in_pre = 0
+        next
+    }
+
+    if ($0 ~ /^<\/?(div|table|p|[ou]l|h[1-6]|[bh]r|blockquote|li)/) {
+        if (in_p) {
+            print "</p>"
+            in_p = 0
+        }
+        print
+        next
+    }
+
+    if ($0 == "") {
+        if (in_p) {
+            print "</p>"
+            in_p = 0
+        }
+        print
+        next
+    }
+
+    if (!in_p) {
+        print "<p>"
+        in_p = 1
+    }
+    print
+}
+
+END {
+    if (in_p) print "</p>"
+}

awk/pipe_tables.awk

@@ -0,0 +1,143 @@
+function trim(s) {
+    sub(/^[[:space:]]+/, "", s)
+    sub(/[[:space:]]+$/, "", s)
+    return s
+}
+
+function is_table_row(line, t) {
+    t = line
+    return (t ~ /^[[:space:]]*\|/ && t ~ /\|[[:space:]]*$/)
+}
+
+function is_table_sep(line, t) {
+    if (!is_table_row(line)) return 0
+    t = line
+    gsub(/[|:\-[:space:]]/, "", t)
+    return (t == "" && line ~ /-/)
+}
+
+function split_row(line, out, n, i, raw) {
+    raw = line
+    sub(/^[[:space:]]*\|/, "", raw)
+    sub(/\|[[:space:]]*$/, "", raw)
+    n = split(raw, out, /\|/)
+    for (i = 1; i <= n; i++) out[i] = trim(out[i])
+    return n
+}
+
+function align_for(sep, t) {
+    t = trim(sep)
+    if (t ~ /^:-+:$/) return "center"
+    if (t ~ /^:-+$/) return "left"
+    if (t ~ /^-+:$/) return "right"
+    return ""
+}
+
+function render_cell(cell, inner) {
+    inner = trim(cell)
+    if (inner ~ /^```.*```$/) {
+        sub(/^```[[:space:]]*/, "", inner)
+        sub(/[[:space:]]*```$/, "", inner)
+        return "<pre><code>" inner "</code></pre>"
+    }
+    return inner
+}
+
+BEGIN { count = 0 }
+{ lines[++count] = $0 }
+
+END {
+    in_pre = 0
+    i = 1
+    while (i <= count) {
+        if (lines[i] ~ /^<pre><code>/) {
+            in_pre = 1
+            print lines[i]
+            i++
+            continue
+        }
+        if (in_pre) {
+            print lines[i]
+            if (lines[i] ~ /^<\/code><\/pre>/) in_pre = 0
+            i++
+            continue
+        }
+
+        if (i < count && is_table_row(lines[i]) && is_table_sep(lines[i + 1])) {
+            n_header = split_row(lines[i], header)
+            n_sep = split_row(lines[i + 1], sep)
+            n_cols = (n_header > n_sep ? n_header : n_sep)
+
+            print "<table>"
+            print "<thead>"
+            print "<tr>"
+            for (c = 1; c <= n_cols; c++) {
+                cell = (c <= n_header ? render_cell(header[c]) : "")
+                a = (c <= n_sep ? align_for(sep[c]) : "")
+                if (a != "") print "<th style=\"text-align: " a ";\">" cell "</th>"
+                else print "<th>" cell "</th>"
+            }
+            print "</tr>"
+            print "</thead>"
+
+            j = i + 2
+            print "<tbody>"
+            while (j <= count && is_table_row(lines[j])) {
+                n_body = split_row(lines[j], body)
+                print "<tr>"
+                for (c = 1; c <= n_cols; c++) {
+                    cell = (c <= n_body ? render_cell(body[c]) : "")
+                    a = (c <= n_sep ? align_for(sep[c]) : "")
+                    if (a != "") print "<td style=\"text-align: " a ";\">" cell "</td>"
+                    else print "<td>" cell "</td>"
+                }
+                print "</tr>"
+                j++
+            }
+            print "</tbody>"
+            print "</table>"
+
+            i = j
+            continue
+        }
+
+        if (is_table_sep(lines[i]) && i < count && is_table_row(lines[i + 1])) {
+            n_sep = split_row(lines[i], sep)
+            n_cols = n_sep
+
+            print "<table>"
+            print "<thead>"
+            print "<tr>"
+            for (c = 1; c <= n_cols; c++) {
+                a = align_for(sep[c])
+                if (a != "") print "<th style=\"text-align: " a ";\"></th>"
+                else print "<th></th>"
+            }
+            print "</tr>"
+            print "</thead>"
+
+            j = i + 1
+            print "<tbody>"
+            while (j <= count && is_table_row(lines[j])) {
+                n_body = split_row(lines[j], body)
+                print "<tr>"
+                for (c = 1; c <= n_cols; c++) {
+                    cell = (c <= n_body ? render_cell(body[c]) : "")
+                    a = (c <= n_sep ? align_for(sep[c]) : "")
+                    if (a != "") print "<td style=\"text-align: " a ";\">" cell "</td>"
+                    else print "<td>" cell "</td>"
+                }
+                print "</tr>"
+                j++
+            }
+            print "</tbody>"
+            print "</table>"
+
+            i = j
+            continue
+        }
+
+        print lines[i]
+        i++
+    }
+}

awk/update_site_conf.awk

@@ -0,0 +1,10 @@
+BEGIN { done = 0 }
+/^title[[:space:]]*=/ {
+    print "title = \"" new_title "\""
+    done = 1
+    next
+}
+{ print }
+END {
+    if (!done) print "title = \"" new_title "\""
+}

kewt.sh

@@ -82,18 +82,7 @@ create_new_site() {
     printf "# _kewt_ website\n" > "$new_dir/index.md"
 
     if [ -n "$new_title" ]; then
-        awk -v new_title="$new_title" '
+        awk -v new_title="$new_title" -f "$awk_dir/update_site_conf.awk" "$new_dir/site.conf" > "$new_dir/site.conf.tmp" && mv "$new_dir/site.conf.tmp" "$new_dir/site.conf"
-            BEGIN { done = 0 }
-            /^title[[:space:]]*=/ {
-                print "title = \"" new_title "\""
-                done = 1
-                next
-            }
-            { print }
-            END {
-                if (!done) print "title = \"" new_title "\""
-            }
-        ' "$new_dir/site.conf" > "$new_dir/site.conf.tmp" && mv "$new_dir/site.conf.tmp" "$new_dir/site.conf"
     fi
 
     echo "Created new site at '$new_dir'."
@@ -101,7 +90,7 @@ create_new_site() {
 }
 
 generate_nav() {
-    dinfo=$(find "$1" -not -path '*/.*' | sort | awk -v src="$1" -f "$awk_dir/collect_dir_info.awk")
+    dinfo=$(find "$1" ! -path '*/.*' | sort | awk -v src="$1" -f "$awk_dir/collect_dir_info.awk")
     find "$1" -name "*.md" | sort | awk -v src="$1" -v single_file_index="$single_file_index" -v flatten="$flatten" -v order="$order" -v dinfo="$dinfo" -f "$awk_dir/generate_sidebar.awk"
 }
 
@@ -314,9 +303,9 @@ find "$src" -type d | sort | while read -r dir; do
 
     if [ ! -f "$dir/index.md" ]; then
         if [ "$single_file_index" = "true" ]; then
-            md_count=$(find "$dir" -maxdepth 1 -name "*.md" | wc -l)
+            md_count=$(find "$dir" ! -name "$(basename "$dir")" -prune -name "*.md" | wc -l)
             if [ "$md_count" -eq 1 ]; then
-                md_file=$(find "$dir" -maxdepth 1 -name "*.md")
+                md_file=$(find "$dir" ! -name "$(basename "$dir")" -prune -name "*.md")
                 render_markdown "$md_file" > "$out_dir/index.html"
                 continue
             fi
@@ -327,7 +316,7 @@ find "$src" -type d | sort | while read -r dir; do
         [ -z "$display_dir" ] && display_dir="/"
         echo "# Index of $display_dir" > "$temp_index"
         echo "" >> "$temp_index"
-        find "$dir" -maxdepth 1 -not -path '*/.*' -not -path "$dir" | sort | while read -r entry; do
+        find "$dir" ! -name "$(basename "$dir")" -prune ! -path '*/.*' | sort | while read -r entry; do
             name="${entry##*/}"
             case "$name" in
                 template.html|site.conf|style.css|index.md) continue ;;
@@ -359,7 +348,7 @@ find "$src" -type f | sort | while IFS= read -r file; do
     esac
 
     if [ "$single_file_index" = "true" ] && [ "${file%.md}" != "$file" ] && [ ! -f "$(dirname "$file")/index.md" ]; then
-        md_count=$(find "$(dirname "$file")" -maxdepth 1 -name "*.md" | wc -l)
+        md_count=$(find "$(dirname "$file")" ! -name "$(basename "$(dirname "$file")")" -prune -name "*.md" | wc -l)
         [ "$md_count" -eq 1 ] && continue
     fi
 

markdown.sh

@@ -3,34 +3,6 @@
 script_dir=$(CDPATH= cd -- "$(dirname -- "$0")" && pwd)
 awk_dir="$script_dir/awk"
 
-sed_ere() {
-    if sed -E '' </dev/null >/dev/null 2>&1; then
-        sed -E "$@"
-    else
-        sed -r "$@"
-    fi
-}
-
-sed_ere_inplace() {
-    script="$1"
-    file="$2"
-    tmp="${file}.tmp.$$"
-    sed_ere "$script" "$file" > "$tmp" && mv "$tmp" "$file" || {
-        rm -f "$tmp"
-        return 1
-    }
-}
-
-sed_ere_inplace_n() {
-    script="$1"
-    file="$2"
-    tmp="${file}.tmp.$$"
-    sed_ere -n "$script" "$file" > "$tmp" && mv "$tmp" "$file" || {
-        rm -f "$tmp"
-        return 1
-    }
-}
-
 sed_inplace() {
     script="$1"
     file="$2"
@@ -44,711 +16,49 @@ sed_inplace() {
 temp_file="/tmp/markdown.$$"
 cat "$@" > "$temp_file"
 
-# backslash escapes for literal characters and inline code masking
+# Mask
-awk '
+awk -f "$awk_dir/mask_inline_code.awk" "$temp_file" > "$temp_file.tmp" && mv "$temp_file.tmp" "$temp_file"
-function mask(s,    t) {
+awk -f "$awk_dir/mask_plain.awk" "$temp_file" > "$temp_file.tmp" && mv "$temp_file.tmp" "$temp_file"
-    t = s
-    gsub(/\*/, "\034P0\034", t)
-    gsub(/_/, "\034P1\034", t)
-    gsub(/`/, "\034P2\034", t)
-    gsub(/\[/, "\034P3\034", t)
-    gsub(/\]/, "\034P4\034", t)
-    gsub(/\(/, "\034P5\034", t)
-    gsub(/\)/, "\034P6\034", t)
-    gsub(/!/, "\034P7\034", t)
-    gsub(/\$/, "\034P8\034", t)
-    gsub(/#/, "\034P9\034", t)
-    gsub(/\+/, "\034P10\034", t)
-    gsub(/-/, "\034P11\034", t)
-    gsub(/\\/, "\034P12\034", t)
-    gsub(/</, "\034P13\034", t)
-    gsub(/>/, "\034P14\034", t)
-    return t
-}
-{
-    # backslash escapes
-    gsub(/\\\*/, "\034P0\034")
-    gsub(/\\_/, "\034P1\034")
-    gsub(/\\`/, "\034P2\034")
-    gsub(/\\\[/, "\034P3\034")
-    gsub(/\\\]/, "\034P4\034")
-    gsub(/\\\(/, "\034P5\034")
-    gsub(/\\\)/, "\034P6\034")
-    gsub(/\\!/, "\034P7\034")
-    gsub(/\\\$/, "\034P8\034")
-    gsub(/\\#/, "\034P9\034")
-    gsub(/\\\+/, "\034P10\034")
-    gsub(/\\\-/, "\034P11\034")
-    gsub(/\\\\/, "\034P12\034")
-    gsub(/\\</, "\034P13\034")
-    gsub(/\\>/, "\034P14\034")
-
-    # inline code (1 or 2 backticks)
-    line = $0
-    if (line ~ /^```/) {
-        print line
-        next
-    }
-    out = ""
-    p = 1
-    while (match(substr(line, p), /`+/)) {
-        pstart = p + RSTART - 1
-        plen = RLENGTH
-        if (plen >= 3) {
-            out = out substr(line, p, pstart - p + plen)
-            p = pstart + plen
-            continue
-        }
-
-        # Found 1 or 2 backticks at pstart
-        # Search for closing marker
-        marker = substr(line, pstart, plen)
-        tail = substr(line, pstart + plen)
-        mpos = index(tail, marker)
-        if (mpos > 0) {
-            # Check if it is followed by more backticks
-            if (substr(tail, mpos + plen, 1) == "`") {
-                # Not a match, treat as literal
-                out = out substr(line, p, pstart - p + plen)
-                p = pstart + plen
-                continue
-            }
-            
-            # Found match!
-            content = substr(tail, 1, mpos - 1)
-            out = out substr(line, p, pstart - p)
-            if (plen == 2 && substr(content, 1, 1) == " " && substr(content, length(content), 1) == " ") {
-                content = substr(content, 2, length(content) - 2)
-            }
-            out = out "<code>" mask(content) "</code>"
-            p = pstart + plen + mpos + plen - 1
-        } else {
-            # No closing marker, treat as literal
-            out = out substr(line, p, pstart - p + plen)
-            p = pstart + plen
-        }
-    }
-    out = out substr(line, p)
-    print out
-}' "$temp_file" > "$temp_file.tmp" && mv "$temp_file.tmp" "$temp_file"
-
-awk '
-
-function find_unescaped_tag(s, tag,    p, off, pos) {
-    p = 1
-    while (1) {
-        off = index(substr(s, p), tag)
-        if (off == 0) return 0
-        pos = p + off - 1
-        if (pos == 1 || substr(s, pos - 1, 1) != "\\") return pos
-        p = pos + 1
-    }
-}
-
-function mask_plain(s,    t) {
-    t = s
-    gsub(/\*/, "\034P0\034", t)
-    gsub(/_/, "\034P1\034", t)
-    gsub(/`/, "\034P2\034", t)
-    gsub(/\[/, "\034P3\034", t)
-    gsub(/\]/, "\034P4\034", t)
-    gsub(/\(/, "\034P5\034", t)
-    gsub(/\)/, "\034P6\034", t)
-    gsub(/!/, "\034P7\034", t)
-    gsub(/\$/, "\034P8\034", t)
-    return t
-}
-BEGIN { in_plain = 0 }
-{
-    line = $0
-    out = ""
-    while (1) {
-        if (!in_plain) {
-            pos = find_unescaped_tag(line, "<plain>")
-            if (pos == 0) {
-                out = out line
-                break
-            }
-            out = out substr(line, 1, pos - 1) "<mfmplain>"
-            line = substr(line, pos + 7)
-            in_plain = 1
-        } else {
-            pos = find_unescaped_tag(line, "</plain>")
-            if (pos == 0) {
-                out = out mask_plain(line)
-                line = ""
-                break
-            }
-            out = out mask_plain(substr(line, 1, pos - 1)) "</mfmplain>"
-            line = substr(line, pos + 8)
-            in_plain = 0
-        }
-    }
-    print out
-}
-' "$temp_file" > "$temp_file.plain.$$" && mv "$temp_file.plain.$$" "$temp_file"
 
+# Reference links
+refs=$(cat "$@" | awk '/^\[[^\]]+\]: +/')
 IFS='
 '
-refs=$(sed_ere -n "/^\[.+\]: +/p" "$@")
+for ref in $refs; do
-for ref in $refs
+    ref_id=$(echo "$ref" | sed 's/^\[\(.*\)\]: .*/\1/')
-do
+    ref_url=$(echo "$ref" | sed 's/^\[.*\]: \([^ ]*\).*/\1/')
-    ref_id=$(printf %s "$ref" | sed_ere -n "s/^\[(.+)\]: .*/\1/p" | tr -d '\n')
+    ref_title=$(echo "$ref" | sed -n 's/^\[.*\]: [^ ]* "\(.*\)"/\1/p' | sed 's@|@!@g')
-    ref_url=$(printf %s "$ref" | sed_ere -n "s/^\[.+\]: (.+)/\1/p" | cut -d' ' -f1 | tr -d '\n')
+    sed_inplace "s|!\[\([^]]*\)\]\[$ref_id\]|<img src=\"$ref_url\" title=\"$ref_title\" alt=\"\1\" />|g" "$temp_file"
-    ref_title=$(printf %s "$ref" | sed_ere -n "s/^\[.+\]: (.+) \"(.+)\"/\2/p" | sed 's@|@!@g' | tr -d '\n')
+    sed_inplace "s|\[\([^]]*\)\]\[$ref_id\]|<a href=\"$ref_url\" title=\"$ref_title\">\1</a>|g" "$temp_file"
+    sed_inplace "s|!\[$ref_id\]\[\]|<img src=\"$ref_url\" title=\"$ref_title\" alt=\"$ref_id\" />|g" "$temp_file"
+    sed_inplace "s|\[$ref_id\]\[\]|<a href=\"$ref_url\" title=\"$ref_title\">$ref_id</a>|g" "$temp_file"
+done
+sed_inplace "/^\[[^\]]*\]: +/d" "$temp_file"
 
-    # reference-style image using the label
+# Blocks
-    sed_ere_inplace "s|!\[([^]]+)\]\[($ref_id)\]|<img src=\"$ref_url\" title=\"$ref_title\" alt=\"\1\" />|g" "$temp_file"
+sed_inplace "s/^>!\[/> [!/g" "$temp_file"
-    # reference-style link using the label
+sed_inplace "s/^>\[!/> [!/g" "$temp_file"
-    sed_ere_inplace "s|\[([^]]+)\]\[($ref_id)\]|<a href=\"$ref_url\" title=\"$ref_title\">\1</a>|g" "$temp_file"
 
-    # implicit reference-style
+while grep '^>' "$temp_file" >/dev/null; do
-    sed_ere_inplace "s|!\[($ref_id)\]\[\]|<img src=\"$ref_url\" title=\"$ref_title\" alt=\"\1\" />|g" "$temp_file"
+    awk -f "$awk_dir/blockquote.awk" "$temp_file" > "$temp_file.tmp" && mv "$temp_file.tmp" "$temp_file"
-    # implicit reference-style
-    sed_ere_inplace "s|\[($ref_id)\]\[\]|<a href=\"$ref_url\" title=\"$ref_title\">\1</a>|g" "$temp_file"
 done
 
-# delete the reference lines
+awk -f "$awk_dir/blockquote_to_admonition.awk" "$temp_file" > "$temp_file.tmp" && mv "$temp_file.tmp" "$temp_file"
-sed_ere_inplace "/^\[.+\]: +/d" "$temp_file"
+awk -f "$awk_dir/fenced_code.awk" "$temp_file" > "$temp_file.tmp" && mv "$temp_file.tmp" "$temp_file"
-
+awk -f "$awk_dir/indented_code.awk" "$temp_file" > "$temp_file.tmp" && mv "$temp_file.tmp" "$temp_file"
-# normalize GitHub admonition shorthand in blockquotes
+awk -f "$awk_dir/pipe_tables.awk" "$temp_file" > "$temp_file.tmp" && mv "$temp_file.tmp" "$temp_file"
-sed_ere_inplace '
+awk -f "$awk_dir/headers.awk" "$temp_file" > "$temp_file.tmp" && mv "$temp_file.tmp" "$temp_file"
-/^>!\[/s/^>!\[/> [!/
+awk -f "$awk_dir/lists.awk" "$temp_file" > "$temp_file.tmp" && mv "$temp_file.tmp" "$temp_file"
-/^>\[!/s/^>\[!/> [!/
+
-s/^>([^[:space:]>])/> \1/
+sed_inplace "s/^\*\*\*+$/<hr \/>/g" "$temp_file"
-' "$temp_file"
+sed_inplace "s/^---+$/<hr \/>/g" "$temp_file"
-
+sed_inplace "s/^___+$/<hr \/>/g" "$temp_file"
-# blockquotes
+
-# use grep to find all the nested blockquotes
+# Spacing
-while grep '^> ' "$temp_file" >/dev/null
+awk -f "$awk_dir/breaks.awk" "$temp_file" > "$temp_file.tmp" && mv "$temp_file.tmp" "$temp_file"
-do
+awk -f "$awk_dir/paragraphs.awk" "$temp_file" > "$temp_file.tmp" && mv "$temp_file.tmp" "$temp_file"
-    sed_ere_inplace_n '
+
-/^$/b blockquote
+# Inline styles
-
+awk -f "$awk_dir/markdown_inline.awk" "$temp_file" > "$temp_file.tmp" && mv "$temp_file.tmp" "$temp_file"
-H
-$ b blockquote
-b
-
-:blockquote
-x
-s/(\n+)(> .*)/\1<blockquote>\n\2\n<\/blockquote>/ # wrap the tags in a blockquote
-p
-' "$temp_file"
-
-    sed_inplace '1 d' "$temp_file" # cleanup superfluous first line
-
-    # cleanup blank lines and remove subsequent blockquote characters
-    sed_ere_inplace '
-/^> /s/^> (.*)/\1/
-' "$temp_file"
-done
-
-# convert [!TYPE] blockquotes into admonition blocks
-awk '
-function cap(s) { return toupper(substr(s, 1, 1)) tolower(substr(s, 2)) }
-BEGIN { count = 0 }
-{ lines[++count] = $0 }
-END {
-    i = 1
-    while (i <= count) {
-        if (lines[i] == "<blockquote>") {
-            j = i + 1
-            while (j <= count && lines[j] != "</blockquote>") j++
-            if (j <= count) {
-                first = ""
-                first_idx = 0
-                for (k = i + 1; k < j; k++) {
-                    if (lines[k] != "") {
-                        first = lines[k]
-                        first_idx = k
-                        break
-                    }
-                }
-                if (first ~ /^\[![A-Za-z]+\]$/) {
-                    kind = first
-                    sub(/^\[!/, "", kind)
-                    sub(/\]$/, "", kind)
-                    lkind = tolower(kind)
-                    if (lkind == "note" || lkind == "tip" || lkind == "important" || lkind == "warning" || lkind == "caution") {
-                        print "<div class=\"admonition admonition-" lkind "\">"
-                        print "<p class=\"admonition-title\">" cap(lkind) "</p>"
-                        has_body = 0
-                        for (k = first_idx + 1; k < j; k++) {
-                            if (lines[k] != "") {
-                                print "<p>" lines[k] "</p>"
-                                has_body = 1
-                            }
-                        }
-                        if (!has_body) print "<p></p>"
-                        print "</div>"
-                        i = j + 1
-                        continue
-                    }
-                }
-            }
-        }
-        print lines[i]
-        i++
-    }
-}
-' "$temp_file" > "$temp_file.admon.$$" && mv "$temp_file.admon.$$" "$temp_file"
-
-# Setext-style headers
-sed_ere_inplace_n '
-# Setext-style headers need to be wrapped around newlines
-/^$/ b print
-
-# else, append to holding area
-H
-$ b print
-b
-
-:print
-x
-/=+$/{
-s/\n(.*)\n=+$/\n<h1>\1<\/h1>/
-p
-b
-}
-/\-+$/{
-s/\n(.*)\n\-+$/\n<h2>\1<\/h2>/
-p
-b
-}
-p
-' "$temp_file"
-
-sed_inplace '1 d' "$temp_file" # cleanup superfluous first line
-
-# atx-style headers and other block styles
-sed_ere_inplace '
-/^#+ /s/ #+$// # kill all ending header characters
-/^# /s/# ([A-Za-z0-9 ]*)(.*)/<h1 id="\1">\1\2<\/h1>/g # H1
-/^#{2} /s/#{2} ([A-Za-z0-9 ]*)(.*)/<h2 id="\1">\1\2<\/h2>/g # H2
-/^#{3} /s/#{3} ([A-Za-z0-9 ]*)(.*)/<h3 id="\1">\1\2<\/h3>/g # H3
-/^#{4} /s/#{4} ([A-Za-z0-9 ]*)(.*)/<h4 id="\1">\1\2<\/h4>/g # H4
-/^#{5} /s/#{5} ([A-Za-z0-9 ]*)(.*)/<h5 id="\1">\1\2<\/h5>/g # H5
-/^#{6} /s/#{6} ([A-Za-z0-9 ]*)(.*)/<h6 id="\1">\1\2<\/h6>/g # H6
-
-/^\*\*\*+$/s/\*\*\*+/<hr \/>/ # hr with *
-/^---+$/s/---+/<hr \/>/ # hr with -
-/^___+$/s/___+/<hr \/>/ # hr with _
-
-' "$temp_file"
-
-# unordered lists
-# use grep to find all the nested lists
-while grep '^[\*\+\-] ' "$temp_file" >/dev/null
-do
-sed_ere_inplace_n '
-# wrap the list
-/^$/b list
-
-# wrap the li tags then add to the hold buffer
-# use uli instead of li to avoid collisions when processing nested lists
-/^[\*\+\-] /s/[\*\+\-] (.*)/<\/uli>\n<uli>\n\1/
-
-H
-$ b list # if at end of file, check for the end of a list
-b # else, branch to the end of the script
-
-# this is where a list is checked for the pattern
-:list
-# exchange the hold space into the pattern space
-x
-# look for the list items, if there wrap the ul tags
-/<uli>/{
-s/(.*)/\n<ul>\1\n<\/uli>\n<\/ul>/ # close the ul tags
-s/\n<\/uli>// # kill the first superfluous closing tag
-p
-b
-}
-p
-' "$temp_file"
-
-sed_inplace '1 d' "$temp_file" # cleanup superfluous first line
-
-# convert to the proper li to avoid collisions with nested lists
-sed_inplace 's/uli>/li>/g' "$temp_file"
-
-# prepare any nested lists
-sed_ere_inplace '/^[\*\+\-] /s/(.*)/\n\1\n/' "$temp_file"
-done
-
-# ordered lists
-# use grep to find all the nested lists
-while grep -E '^[1-9]+\. ' "$temp_file" >/dev/null
-do
-sed_ere_inplace_n '
-# wrap the list
-/^$/b list
-
-# wrap the li tags then add to the hold buffer
-# use oli instead of li to avoid collisions when processing nested lists
-/^[1-9]+\. /s/[1-9]+\. (.*)/<\/oli>\n<oli>\n\1/
-
-H
-$ b list # if at end of file, check for the end of a list
-b # else, branch to the end of the script
-
-:list
-# exchange the hold space into the pattern space
-x
-# look for the list items, if there wrap the ol tags
-/<oli>/{
-s/(.*)/\n<ol>\1\n<\/oli>\n<\/ol>/ # close the ol tags
-s/\n<\/oli>// # kill the first superfluous closing tag
-p
-b
-}
-p
-' "$temp_file"
-
-sed_inplace '1 d' "$temp_file" # cleanup superfluous first line
-
-# convert list items into proper list items to avoid collisions with nested lists
-sed_inplace 's/oli>/li>/g' "$temp_file"
-
-# prepare any nested lists
-sed_ere_inplace '/^[1-9]+\. /s/(.*)/\n\1\n/' "$temp_file"
-done
-
-# make escaped periods literal
-sed_ere_inplace '/^[1-9]+\\. /s/([1-9]+)\\. /\1\. /' "$temp_file"
-
-# fenced code blocks (triple backticks)
-awk '
-BEGIN { in_fence = 0; first_line = 0 }
-{
-    if (!in_fence && $0 ~ /^```/) {
-        printf "<pre><code>"
-        in_fence = 1
-        first_line = 1
-        next
-    }
-    if (in_fence && $0 ~ /^```[[:space:]]*$/) {
-        print "</code></pre>"
-        in_fence = 0
-        next
-    }
-    if (in_fence) {
-        if (first_line) {
-            first_line = 0
-            if ($0 == "") next
-        }
-        print
-    } else {
-        print
-    }
-}
-END {
-    if (in_fence) print "</code></pre>"
-}
-' "$temp_file" > "$temp_file.fence.$$" && mv "$temp_file.fence.$$" "$temp_file"
-
-
-# code blocks
-sed_ere_inplace_n '
-# if at end of file, append the current line to the hold buffer and print it
-${
-H
-b code
-}
-
-# wrap the code block on any non code block lines
-/^\t| {4}/!b code
-
-# else, append to the holding buffer and do nothing
-H
-b # else, branch to the end of the script
-
-:code
-# exchange the hold space with the pattern space
-x
-# look for the code items, if there wrap the pre-code tags
-/\t| {4}/{
-s/(\t| {4})(.*)/<pre><code>\n\1\2\n<\/code><\/pre>/ # wrap the ending tags
-p
-b
-}
-p
-' "$temp_file"
-
-sed_inplace '1 d' "$temp_file" # cleanup superfluous first line
-
-# convert html characters inside pre-code tags into printable representations
-sed_ere_inplace '
-# get inside pre-code tags
-/^<pre><code>/{
-:inside
-n
-# if you found the end tags, branch out
-/^<\/code><\/pre>/!{
-s/&/\&amp;/g # ampersand
-s/</\&lt;/g # less than
-s/>/\&gt;/g # greater than
-b inside
-}
-}
-' "$temp_file"
-
-# remove the first tab (or 4 spaces) from the code lines
-sed_ere_inplace 's/^\t| {4}(.*)/\1/' "$temp_file"
-
-# markdown pipe tables
-awk '
-function trim(s) {
-    sub(/^[[:space:]]+/, "", s)
-    sub(/[[:space:]]+$/, "", s)
-    return s
-}
-
-function is_table_row(line, t) {
-    t = line
-    return (t ~ /^[[:space:]]*\|/ && t ~ /\|[[:space:]]*$/)
-}
-
-function is_table_sep(line, t) {
-    if (!is_table_row(line)) return 0
-    t = line
-    gsub(/[|:\-[:space:]]/, "", t)
-    return (t == "" && line ~ /-/)
-}
-
-function split_row(line, out, n, i, raw) {
-    raw = line
-    sub(/^[[:space:]]*\|/, "", raw)
-    sub(/\|[[:space:]]*$/, "", raw)
-    n = split(raw, out, /\|/)
-    for (i = 1; i <= n; i++) out[i] = trim(out[i])
-    return n
-}
-
-function align_for(sep, t) {
-    t = trim(sep)
-    if (t ~ /^:-+:$/) return "center"
-    if (t ~ /^:-+$/) return "left"
-    if (t ~ /^-+:$/) return "right"
-    return ""
-}
-
-function render_cell(cell, inner) {
-    inner = trim(cell)
-    if (inner ~ /^```.*```$/) {
-        sub(/^```[[:space:]]*/, "", inner)
-        sub(/[[:space:]]*```$/, "", inner)
-        return "<pre><code>" inner "</code></pre>"
-    }
-    return inner
-}
-
-BEGIN { count = 0 }
-{ lines[++count] = $0 }
-
-END {
-    in_pre = 0
-    i = 1
-    while (i <= count) {
-        if (lines[i] ~ /^<pre><code>/) {
-            in_pre = 1
-            print lines[i]
-            i++
-            continue
-        }
-        if (in_pre) {
-            print lines[i]
-            if (lines[i] ~ /^<\/code><\/pre>/) in_pre = 0
-            i++
-            continue
-        }
-
-        if (i < count && is_table_row(lines[i]) && is_table_sep(lines[i + 1])) {
-            n_header = split_row(lines[i], header)
-            n_sep = split_row(lines[i + 1], sep)
-            n_cols = (n_header > n_sep ? n_header : n_sep)
-
-            print "<table>"
-            print "<thead>"
-            print "<tr>"
-            for (c = 1; c <= n_cols; c++) {
-                cell = (c <= n_header ? render_cell(header[c]) : "")
-                a = (c <= n_sep ? align_for(sep[c]) : "")
-                if (a != "") print "<th style=\"text-align: " a ";\">" cell "</th>"
-                else print "<th>" cell "</th>"
-            }
-            print "</tr>"
-            print "</thead>"
-
-            j = i + 2
-            print "<tbody>"
-            while (j <= count && is_table_row(lines[j])) {
-                n_body = split_row(lines[j], body)
-                print "<tr>"
-                for (c = 1; c <= n_cols; c++) {
-                    cell = (c <= n_body ? render_cell(body[c]) : "")
-                    a = (c <= n_sep ? align_for(sep[c]) : "")
-                    if (a != "") print "<td style=\"text-align: " a ";\">" cell "</td>"
-                    else print "<td>" cell "</td>"
-                }
-                print "</tr>"
-                j++
-            }
-            print "</tbody>"
-            print "</table>"
-
-            i = j
-            continue
-        }
-
-        if (is_table_sep(lines[i]) && i < count && is_table_row(lines[i + 1])) {
-            n_sep = split_row(lines[i], sep)
-            n_cols = n_sep
-
-            print "<table>"
-            print "<thead>"
-            print "<tr>"
-            for (c = 1; c <= n_cols; c++) {
-                a = align_for(sep[c])
-                if (a != "") print "<th style=\"text-align: " a ";\"></th>"
-                else print "<th></th>"
-            }
-            print "</tr>"
-            print "</thead>"
-
-            j = i + 1
-            print "<tbody>"
-            while (j <= count && is_table_row(lines[j])) {
-                n_body = split_row(lines[j], body)
-                print "<tr>"
-                for (c = 1; c <= n_cols; c++) {
-                    cell = (c <= n_body ? render_cell(body[c]) : "")
-                    a = align_for(sep[c])
-                    if (a != "") print "<td style=\"text-align: " a ";\">" cell "</td>"
-                    else print "<td>" cell "</td>"
-                }
-                print "</tr>"
-                j++
-            }
-            print "</tbody>"
-            print "</table>"
-
-            i = j
-            continue
-        }
-
-        print lines[i]
-        i++
-    }
-}
-' "$temp_file" > "$temp_file.table.$$" && mv "$temp_file.table.$$" "$temp_file"
-
-# br tags
-sed_ere_inplace '
-# if an empty line, append it to the next line, then check on whether there is two in a row
-/^$/ {
-N
-N
-/^\n{2}/s/(.*)/\n<br \/>\1/
-}
-' "$temp_file"
-
-# emphasis and strong emphasis and strikethrough
-sed_ere_inplace_n '
-# batch up the entire stream of text until a line break in the action
-/^$/b emphasis
-
-H
-$ b emphasis
-b
-
-:emphasis
-x
-s/\*\*([^\n]+)\*\*/<strong>\1<\/strong>/g
-s/__([^_\n]+)__/<strong>\1<\/strong>/g
-s/\*([^\*\n]+)\*/<em>\1<\/em>/g
-s/([^\\])_([^_\n]+)_/\1<em>\2<\/em>/g
-s/\~\~([^\n]+)\~\~/<strike>\1<\/strike>/g
-p
-' "$temp_file"
-
-sed_inplace '1 d' "$temp_file" # cleanup superfluous first line
-
-# paragraphs
-sed_ere_inplace_n '
-# if an empty line, check the paragraph
-/^$/ b para
-# else append it to the hold buffer
-H
-# at end of file, check paragraph
-$ b para
-# now branch to end of script
-b
-# this is where a paragraph is checked for the pattern
-:para
-# return the entire paragraph into the pattern space
-x
-# look for non block-level elements, if there - print the p tags
-/\n<(div|table|pre|p|[ou]l|h[1-6]|[bh]r|blockquote|li)/!{
-s/(\n+)(.*)/\1<p>\n\2\n<\/p>/
-p
-b
-}
-p
-' "$temp_file"
-
-sed_inplace '1 d' "$temp_file" # cleanup superfluous first line
-
-# cleanup area where P tags have broken nesting
-sed_ere_inplace_n '
-# if the line looks like like an end tag
-/^<\/(div|table|pre|p|[ou]l|h[1-6]|[bh]r|blockquote)>/{
-h
-# if EOF, print the line
-$ {
-x
-b done
-}
-# fetch the next line and check on whether or not it is a P tag
-n
-/^<\/p>/{
-G
-b done
-}
-# else, append the line to the previous line and print them both
-H
-x
-}
-:done
-p
-' "$temp_file"
-
-# inline styles and special characters
-sed_ere_inplace '
-/^<pre><code>/,/^<\/code><\/pre>/b
-
-s/<(http[s]?:\/\/.*)>/<a href=\"\1\">\1<\/a>/g # automatic links
-s/<(.*@.*\..*)>/<a href=\"mailto:\1\">\1<\/a>/g # automatic email address links
-
-# force-inline image syntax (double bang)
-s/!!\[([^]]*)\]\(([^)]*) \"([^\"]*)\"\)/<img data-force-inline=\"1\" alt=\"\1\" src=\"\2\" title=\"\3\" \/>/g
-s/!!\[([^]]*)\]\(([^)]*)\)/<img data-force-inline=\"1\" alt=\"\1\" src=\"\2\" \/>/g
-
-s/(^|[^\\])!\[([^]]*)\]\(([^)]*) \"([^\"]*)\"\)/\1<img alt=\"\2\" src=\"\3\" title=\"\4\" \/>/g # inline image with title
-s/(^|[^\\])!\[([^]]*)\]\(([^)]*)\)/\1<img alt=\"\2\" src=\"\3\" \/>/g # inline image without title
-
-s/(^|[^\\!])\[([^]]*)\]\(([^)]*) \"([^\"]*)\"\)/\1<a href=\"\3\" title=\"\4\">\2<\/a>/g # inline link with title
-s/(^|[^\\!])\[([^]]*)\]\(([^)]*)\)/\1<a href=\"\3\">\2<\/a>/g # inline link
-
-# MFM font syntax
-s/\$\[font\.serif ([^]]+)\]/<span style=\"font-family: serif;\">\1<\/span>/g
-s/\$\[font\.monospace ([^]]+)\]/<span style=\"font-family: monospace;\">\1<\/span>/g
-s/\$\[font\.sans ([^]]+)\]/<span style=\"font-family: sans-serif;\">\1<\/span>/g
-
-# special characters
-/&.+;/!s/&/\&amp;/g # ampersand
-/<[\/a-zA-Z]/!s/</\&lt;/g# less than bracket
-' "$temp_file"
-
-# display and cleanup
 awk -v input_file="$1" -v site_root="$MARKDOWN_SITE_ROOT" -v fallback_file="$MARKDOWN_FALLBACK_FILE" -f "$awk_dir/markdown_embed.awk" "$temp_file"
 rm "$temp_file"