southfox/miniflux
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages 1 Wiki Activity Actions

Divide the sanitization time by 3

Browse source 
Instead of having to allocate a ~100 keys map containing possibly dynamic
values (at least to the go compiler), allocate it once in a global variable.
This significantly speeds things up, by reducing the garbage
collector/allocator involvements.

Local synthetic benchmarks have shown a improvements from 38% of wall time to only
12%.
This commit is contained in:
jvoisin 2024-02-29 03:18:21 +01:00 committed by Frédéric Guillot
parent 1f5c8ce353
commit f12d5131b0
1 changed files with 58 additions and 62 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

120
internal/reader/sanitizer/sanitizer.go
View file

@ -20,6 +20,62 @@ import (
var (
youtubeEmbedRegex = regexp.MustCompile(`//www\.youtube\.com/embed/(.*)`)
tagAllowList = map[string][]string{
"a": {"href", "title", "id"},
"abbr": {"title"},
"acronym": {"title"},
"audio": {"src"},
"blockquote": {},
"br": {},
"caption": {},
"cite": {},
"code": {},
"dd": {"id"},
"del": {},
"dfn": {},
"dl": {"id"},
"dt": {"id"},
"em": {},
"figcaption": {},
"figure": {},
"h1": {"id"},
"h2": {"id"},
"h3": {"id"},
"h4": {"id"},
"h5": {"id"},
"h6": {"id"},
"iframe": {"width", "height", "frameborder", "src", "allowfullscreen"},
"img": {"alt", "title", "src", "srcset", "sizes", "width", "height"},
"ins": {},
"kbd": {},
"li": {"id"},
"ol": {"id"},
"p": {},
"picture": {},
"pre": {},
"q": {"cite"},
"rp": {},
"rt": {},
"rtc": {},
"ruby": {},
"s": {},
"samp": {},
"source": {"src", "type", "srcset", "sizes", "media"},
"strong": {},
"sub": {},
"sup": {"id"},
"table": {},
"td": {"rowspan", "colspan"},
"tfooter": {},
"th": {"rowspan", "colspan"},
"thead": {},
"time": {"datetime"},
"tr": {},
"ul": {"id"},
"var": {},
"video": {"poster", "height", "width", "src"},
"wbr": {},
}
)
// Sanitize returns safe HTML.
@ -184,14 +240,14 @@ func getExtraAttributes(tagName string) ([]string, []string) {
}
func isValidTag(tagName string) bool {
if _, ok := getTagAllowList()[tagName]; ok {
if _, ok := tagAllowList[tagName]; ok {
return true
}
return false
}
func isValidAttribute(tagName, attributeName string) bool {
if attributes, ok := getTagAllowList()[tagName]; ok {
if attributes, ok := tagAllowList[tagName]; ok {
return inList(attributeName, attributes)
}
return false
@ -350,66 +406,6 @@ func isValidIframeSource(baseURL, src string) bool {
return strings.HasPrefix(src, prefix)
})
}
func getTagAllowList() map[string][]string {
return map[string][]string{
"a": {"href", "title", "id"},
"abbr": {"title"},
"acronym": {"title"},
"audio": {"src"},
"blockquote": {},
"br": {},
"caption": {},
"cite": {},
"code": {},
"dd": {"id"},
"del": {},
"dfn": {},
"dl": {"id"},
"dt": {"id"},
"em": {},
"figcaption": {},
"figure": {},
"h1": {"id"},
"h2": {"id"},
"h3": {"id"},
"h4": {"id"},
"h5": {"id"},
"h6": {"id"},
"iframe": {"width", "height", "frameborder", "src", "allowfullscreen"},
"img": {"alt", "title", "src", "srcset", "sizes", "width", "height"},
"ins": {},
"kbd": {},
"li": {"id"},
"ol": {"id"},
"p": {},
"picture": {},
"pre": {},
"q": {"cite"},
"rp": {},
"rt": {},
"rtc": {},
"ruby": {},
"s": {},
"samp": {},
"source": {"src", "type", "srcset", "sizes", "media"},
"strong": {},
"sub": {},
"sup": {"id"},
"table": {},
"td": {"rowspan", "colspan"},
"tfooter": {},
"th": {"rowspan", "colspan"},
"thead": {},
"time": {"datetime"},
"tr": {},
"ul": {"id"},
"var": {},
"video": {"poster", "height", "width", "src"},
"wbr": {},
}
}
func inList(needle string, haystack []string) bool {
return slices.Contains(haystack, needle)
}