1
0
Fork 0
mirror of https://git.sr.ht/~seirdy/seirdy.one synced 2024-12-24 17:52:11 +00:00

Quit using HTML-Tidy

See https://github.com/htacg/tidy-html5/issues/1094. The most recent
commit without that regression can't handle `dl` elements with `div`
children.
This commit is contained in:
Rohan Kumar 2023-11-15 09:21:26 -08:00
parent dcb19779ce
commit 4910a7c2c1
No known key found for this signature in database
GPG key ID: 1E892DB2A5F84479
2 changed files with 7 additions and 6 deletions

View file

@ -29,7 +29,7 @@ run_tidy () {
# delete the stylesheet from the html file; we'll re-insert it later.
# Also remove two indentation levels
sed 7d "$html_file" | xmllint --format --encode UTF-8 --noent - | tail -n +2 | run_tidy >"$tmp_file"
sed 7d "$html_file" | xmllint --format --encode UTF-8 --noent - | tail -n +2 >"$tmp_file"
{
head -n7 "$tmp_file"
cat "${OUTPUT_DIR:?}/tmp.css"
@ -37,12 +37,11 @@ sed 7d "$html_file" | xmllint --format --encode UTF-8 --noent - | tail -n +2 | r
#shellcheck source=/home/rkumar/Executables/ghq/git.sr.ht/~seirdy/seirdy.one/scripts/xhtmlize.sh
sed \
-e '1,7d' \
-e "s|name=\"generator\" />|name=\"generator\" />\n${TIDY:?}|" \
-e 's|\.svg" width="16" /><span|.svg" width="16" /> <span|' \
-e 's|"u-photo photo" ?/><span|"u-photo photo" ?/>&#160;<span|' \
-e 's|</span>(&nbsp;)?.span itemprop="familyName|</span>&#160;<span itemprop="familyName"|' \
-E \
-e 's|([a-z])<data|\1 <data|' \
-e 's#</span>(<a[^>]*rel="(nofollow ugc|ugc nofollow)"([^>]*)?>liked</a>)#</span> \1#' \
-e 's#<pre( tabindex="0")?>\n(\t|\s)*<(code|samp)( |>)#<pre tabindex="0"><\3\4#' \
"$tmp_file"
"$tmp_file" \
| awk '/^<\/code>/{printf "%s",$0;next}7'
} >"$html_file"

View file

@ -11,7 +11,9 @@ script_dir="$(dirname "$0")"
tidy_version="$(tidy -version)"
export TIDY="<meta content=\"$tidy_version\" name=\"generator\" />"
sed -e '7q;d' "$output_dir/index.html" | tr -d '\t' >"$output_dir/tmp.css"
{
printf '\t' && sed -e '7q;d' "$output_dir/index.html"
} >"$output_dir/tmp.css"
cleanup() {
rm -f "$output_dir/tmp.css"
}