opencode/picoclaw
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

perf: pre-compile regex patterns in markdownToTelegramHTML

Browse source 
Move 10 regexp.MustCompile() calls from inside functions to package-level
var declarations. Eliminates ~20 KB of heap allocation per outbound message.

- markdownToTelegramHTML: 8 patterns pre-compiled
- extractCodeBlocks: 1 pattern pre-compiled
- extractInlineCodes: 1 pattern pre-compiled
- removeBotMention @username regex stays inline (runtime-dependent)

regexp.Regexp is goroutine-safe, so pre-compiled patterns are safe to share.

Ref: https://github.com/chillum-codeX/picoclaw-under-the-hood
This commit is contained in:
chillum-codeX 2026-02-26 07:54:45 +05:30
parent 3c58f6af4c
commit 16d9338606
1 changed files with 27 additions and 14 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

41
pkg/channels/telegram/telegram.go
View file

@ -511,6 +511,22 @@ func parseChatID(chatIDStr string) (int64, error) {
return id, err return id, err
} }
// Pre-compiled regex patterns for markdownToTelegramHTML.
// Compiled once at package init, reused across all calls (regexp is goroutine-safe).
// This eliminates ~20 KB of heap allocation per outbound message.
var (
reHeader = regexp.MustCompile(`^#{1,6}\s+(.+)$`)
reBlockquote = regexp.MustCompile(`^>\s*(.*)$`)
reLink = regexp.MustCompile(`\[([^\]]+)\]\(([^)]+)\)`)
reBold = regexp.MustCompile(`\*\*(.+?)\*\*`)
reBoldAlt = regexp.MustCompile(`__(.+?)__`)
reItalic = regexp.MustCompile(`_([^_]+)_`)
reStrike = regexp.MustCompile(`~~(.+?)~~`)
reListItem = regexp.MustCompile(`^[-*]\s+`)
reCodeBlock = regexp.MustCompile("```[\\w]*\\n?([\\s\\S]*?)```")
reInlineCode = regexp.MustCompile("`([^`]+)`")
)
func markdownToTelegramHTML(text string) string { func markdownToTelegramHTML(text string) string {
if text == "" { if text == "" {
return "" return ""
@ -522,19 +538,18 @@ func markdownToTelegramHTML(text string) string {
inlineCodes := extractInlineCodes(text) inlineCodes := extractInlineCodes(text)
text = inlineCodes.text text = inlineCodes.text
text = regexp.MustCompile(`^#{1,6}\s+(.+)$`).ReplaceAllString(text, "$1") text = reHeader.ReplaceAllString(text, "$1")
text = regexp.MustCompile(`^>\s*(.*)$`).ReplaceAllString(text, "$1") text = reBlockquote.ReplaceAllString(text, "$1")
text = escapeHTML(text) text = escapeHTML(text)
text = regexp.MustCompile(`\[([^\]]+)\]\(([^)]+)\)`).ReplaceAllString(text, `<a href="$2">$1</a>`) text = reLink.ReplaceAllString(text, `<a href="$2">$1</a>`)
text = regexp.MustCompile(`\*\*(.+?)\*\*`).ReplaceAllString(text, "<b>$1</b>") text = reBold.ReplaceAllString(text, "<b>$1</b>")
text = regexp.MustCompile(`__(.+?)__`).ReplaceAllString(text, "<b>$1</b>") text = reBoldAlt.ReplaceAllString(text, "<b>$1</b>")
reItalic := regexp.MustCompile(`_([^_]+)_`)
text = reItalic.ReplaceAllStringFunc(text, func(s string) string { text = reItalic.ReplaceAllStringFunc(text, func(s string) string {
match := reItalic.FindStringSubmatch(s) match := reItalic.FindStringSubmatch(s)
if len(match) < 2 { if len(match) < 2 {
@ -543,9 +558,9 @@ func markdownToTelegramHTML(text string) string {
return "<i>" + match[1] + "</i>" return "<i>" + match[1] + "</i>"
}) })
text = regexp.MustCompile(`~~(.+?)~~`).ReplaceAllString(text, "<s>$1</s>") text = reStrike.ReplaceAllString(text, "<s>$1</s>")
text = regexp.MustCompile(`^[-*]\s+`).ReplaceAllString(text, "• ") text = reListItem.ReplaceAllString(text, "• ")
for i, code := range inlineCodes.codes { for i, code := range inlineCodes.codes {
escaped := escapeHTML(code) escaped := escapeHTML(code)
@ -570,8 +585,7 @@ type codeBlockMatch struct {
} }
func extractCodeBlocks(text string) codeBlockMatch { func extractCodeBlocks(text string) codeBlockMatch {
re := regexp.MustCompile("```[\\w]*\\n?([\\s\\S]*?)```") matches := reCodeBlock.FindAllStringSubmatch(text, -1)
matches := re.FindAllStringSubmatch(text, -1)
codes := make([]string, 0, len(matches)) codes := make([]string, 0, len(matches))
for _, match := range matches { for _, match := range matches {
@ -579,7 +593,7 @@ func extractCodeBlocks(text string) codeBlockMatch {
} }
i := 0 i := 0
text = re.ReplaceAllStringFunc(text, func(m string) string { text = reCodeBlock.ReplaceAllStringFunc(text, func(m string) string {
placeholder := fmt.Sprintf("\x00CB%d\x00", i) placeholder := fmt.Sprintf("\x00CB%d\x00", i)
i++ i++
return placeholder return placeholder
@ -594,8 +608,7 @@ type inlineCodeMatch struct {
} }
func extractInlineCodes(text string) inlineCodeMatch { func extractInlineCodes(text string) inlineCodeMatch {
re := regexp.MustCompile("`([^`]+)`") matches := reInlineCode.FindAllStringSubmatch(text, -1)
matches := re.FindAllStringSubmatch(text, -1)
codes := make([]string, 0, len(matches)) codes := make([]string, 0, len(matches))
for _, match := range matches { for _, match := range matches {
@ -603,7 +616,7 @@ func extractInlineCodes(text string) inlineCodeMatch {
} }
i := 0 i := 0
text = re.ReplaceAllStringFunc(text, func(m string) string { text = reInlineCode.ReplaceAllStringFunc(text, func(m string) string {
placeholder := fmt.Sprintf("\x00IC%d\x00", i) placeholder := fmt.Sprintf("\x00IC%d\x00", i)
i++ i++
return placeholder return placeholder