tags to markdown code blocks
-/^/ {
- gsub(/^/, "```")
- gsub(/<\/pre>$/, "```")
- print
- next
-}
-
-# Fix tags that don't close properly
-/]*>/ {
- # If it has content after the tag and ends with ```, it's malformed
- if (/]*>[^<]*```$/) {
- # Replace content``` with just content (already has ```)
- gsub(/]*>/, "", $0)
- print
- next
- }
- # If it has content after the tag, it's likely malformed
- if (/]*>[^<]*$/) {
- gsub(/]*>/, "```", $0)
- }
-}
-
-# Remove
tags that appear at the end of lines
-{
- gsub(/<\/pre>$/, "```", $0)
-}
-
-# Remove anchor parts from headings (e.g., ## Title {:#anchor})
-/^#+ .* \{:#[^}]*\}$/ {
- heading = $0
- gsub(/\s*\{:#[^}]*\}$/, "", heading)
- gsub(/[ \t]+$/, "", heading)
- print heading
- next
-}
-
-# Remove anchor syntax like {:flag--deleted_packages} or {: .external}
-{
- gsub(/\{:[^}]*\}/, "", $0)
-}
-
-# Fix common Jekyll/Kramdown patterns that break MDX
-{
- # Remove {# ... #} Jekyll comments - be more aggressive
- while (match($0, /\{#[^}]*#\}/)) {
- gsub(/\{#[^}]*#\}/, "", $0)
- }
-}
-
-# Fix problematic {{ "" }} and {{ '' }} patterns
-{
- # These double curly braces with quotes break acorn parser
- gsub(/\{\{ *"" *\}\}/, "", $0)
- gsub(/\{\{ *"<\/var>" *\}\}/, "", $0)
- gsub(/\{\{ *'' *\}\}/, "", $0)
- gsub(/\{\{ *'<\/var>' *\}\}/, "", $0)
-}
-
-# Fix < and > that should be escaped differently in MDX
-{
- # In attribute values, convert < and > to actual < >
- # This is safer in JSX/MDX
- while (match($0, /="[^"]*&[lg]t;[^"]*"/)) {
- gsub(/</, "<", $0)
- gsub(/>/, ">", $0)
- }
-}
-
-# Fix empty thead tags - should be
-{
- # Multiple variations of broken thead
- gsub(/<\/th>/, "", $0)
- gsub(/<\/thead>/, "", $0)
-
- # Also fix lines that are ONLY after a thead
- if ($0 ~ /^<\/th>$/ || $0 ~ /^[ \t]*<\/th>[ \t]*$/) {
- next # Skip this line entirely
- }
-}
-
-# Fix malformed ![]()
tags with align attribute without quotes
-{
- # align=right should be align="right"
- gsub(/align=right/, "align=\"right\"", $0)
- gsub(/align=left/, "align=\"left\"", $0)
- gsub(/align=center/, "align=\"center\"", $0)
-}
-
-# Fix malformed
tags
- gsub(/
/, "
", $0)
-
- # Fix ![]()
tags - ensure they're self-closing
- while (match($0, /]*[^\/]>/)) {
- pre = substr($0, 1, RSTART - 1)
- tag = substr($0, RSTART, RLENGTH)
- post = substr($0, RSTART + RLENGTH)
- # Remove the trailing > and add />
- tag = substr(tag, 1, length(tag) - 1) " />"
- $0 = pre tag post
- }
-
- # Fix
tags
- gsub(/
/, "
", $0)
-}
-
-# Fix unclosed tags
-{
- # If we have
but no closing tag on the same line, and line ends with text
- if (/]*>/ && !/<\/p>/ && $0 !~ /<\/(div|table|ul|ol|blockquote)>$/) {
- # Check if it's just a with content and no closing
- if ($0 ~ /]*>[^<]+$/) {
- $0 = $0 "
"
- }
- }
-}
-
-# Fix unclosed tags in patterns
-{
- # Fix escaped underscores in code tags like \_
- # These appear in patterns like noimplicit\_deps
- if (/.*\\_.*<\/code>/) {
- gsub(/\\_/, "_", $0)
- }
-
- # If we have opening but line doesn't end with
- if (/]*>/ && !/<\/code>/) {
- # Look for the pattern and close it properly
- if ($0 ~ /]*>[^<]*$/) {
- $0 = $0 " "
- }
- }
-}
-
-# Fix unclosed and tags at end of lines (common in tables)
-/]*>[^<]*$/ {
- # Only add closing tag if there isn't already one
- if ($0 !~ /<\/td>$/) {
- $0 = $0 ""
- }
-}
-
-# Be more careful with - don't match
-/^[^<]*]*>[^<]*$/ {
- if ($0 !~ /<\/th>$/) {
- $0 = $0 ""
- }
-}
-
-# Fix malformed tags
-{
- # Ensure href has quotes
- while (match($0, /]*)href=([^"'][^ >]+)/)) {
- pre = substr($0, 1, RSTART - 1)
- match_str = substr($0, RSTART, RLENGTH)
- post = substr($0, RSTART + RLENGTH)
- gsub(/href=([^"'][^ >]+)/, "href=\"\\1\"", match_str)
- $0 = pre match_str post
- }
-
- # Fix broken )
-{
- # Escape angle brackets in text that looks like C++ includes
- # Pattern: #include or e.g. #include
- if ($0 ~ /#include and escape it too
- gsub(/\.h>/, ".h\\>", $0)
- }
-}
-
-# Skip blank lines before first H1
-/^[ \t]*$/ && before_first_h1 == 1 { next }
-
-# Convert first H1 to front-matter
-/^# / && first_h1_found == 0 {
- title = substr($0, 3) # Remove "# " prefix
- gsub(/^[ \t]+|[ \t]+$/, "", title) # Trim whitespace
- # Escape single quotes in title by doubling them for YAML
- gsub(/'/, "''", title)
- print "---"
- print "title: '" title "'"
- print "---"
- print ""
- first_h1_found = 1
- before_first_h1 = 0
- next
-}
-
-# Print all other lines
-{
- print
-}
\ No newline at end of file