480 lines
15 KiB
Kotlin
480 lines
15 KiB
Kotlin
package jp.juggler.subwaytooter.util
|
||
|
||
import android.content.Context
|
||
import android.content.SharedPreferences
|
||
import android.text.SpannableStringBuilder
|
||
import android.text.Spanned
|
||
import android.util.SparseBooleanArray
|
||
import androidx.annotation.DrawableRes
|
||
import jp.juggler.emoji.EmojiMap
|
||
import jp.juggler.subwaytooter.App1
|
||
import jp.juggler.subwaytooter.Pref
|
||
import jp.juggler.subwaytooter.R
|
||
import jp.juggler.subwaytooter.api.entity.CustomEmoji
|
||
import jp.juggler.subwaytooter.pref
|
||
import jp.juggler.subwaytooter.span.EmojiImageSpan
|
||
import jp.juggler.subwaytooter.span.HighlightSpan
|
||
import jp.juggler.subwaytooter.span.NetworkEmojiSpan
|
||
import jp.juggler.subwaytooter.span.createSpan
|
||
import jp.juggler.subwaytooter.table.HighlightWord
|
||
import jp.juggler.util.LogCategory
|
||
import jp.juggler.util.asciiPattern
|
||
import jp.juggler.util.codePointBefore
|
||
import java.util.*
|
||
import java.util.regex.Pattern
|
||
import kotlin.math.min
|
||
|
||
object EmojiDecoder {
|
||
|
||
private val log = LogCategory("EmojiDecoder")
|
||
|
||
private const val cpColon = ':'.toInt()
|
||
|
||
private const val cpZwsp = '\u200B'.toInt()
|
||
|
||
var handleUnicodeEmoji = true
|
||
|
||
fun customEmojiSeparator(pref: SharedPreferences) = if (Pref.bpCustomEmojiSeparatorZwsp(pref)) {
|
||
'\u200B'
|
||
} else {
|
||
' '
|
||
}
|
||
|
||
// タンス側が落ち着いたら [^[:almun:]_] から [:space:]に切り替える
|
||
// private fun isHeadOrAfterWhitespace( s:CharSequence,index:Int):Boolean {
|
||
// val cp = s.codePointBefore(index)
|
||
// return cp == -1 || CharacterGroup.isWhitespace(cp)
|
||
// }
|
||
|
||
fun canStartShortCode(s: CharSequence, index: Int): Boolean {
|
||
return when (val cp = s.codePointBefore(index)) {
|
||
-1 -> true
|
||
cpColon -> false
|
||
cpZwsp -> true
|
||
// rubyの (Letter | Mark | Decimal_Number) はNG
|
||
// ftp://unicode.org/Public/5.1.0/ucd/UCD.html#General_Category_Values
|
||
else -> when (Character.getType(cp).toByte()) {
|
||
// Letter
|
||
// LCはエイリアスなので文字から得られることはないはず
|
||
Character.UPPERCASE_LETTER,
|
||
Character.LOWERCASE_LETTER,
|
||
Character.TITLECASE_LETTER,
|
||
Character.MODIFIER_LETTER,
|
||
Character.OTHER_LETTER -> false
|
||
// Mark
|
||
Character.NON_SPACING_MARK,
|
||
Character.COMBINING_SPACING_MARK,
|
||
Character.ENCLOSING_MARK -> false
|
||
// Decimal_Number
|
||
Character.DECIMAL_DIGIT_NUMBER -> false
|
||
|
||
else -> true
|
||
}
|
||
}
|
||
// https://mastodon.juggler.jp/@tateisu/99727683089280157
|
||
// https://github.com/tootsuite/mastodon/pull/5570 がマージされたらこっちに切り替える
|
||
// return cp == -1 || CharacterGroup.isWhitespace(cp)
|
||
}
|
||
|
||
fun canStartHashtag(s: CharSequence, index: Int): Boolean {
|
||
val cp = s.codePointBefore(index)
|
||
// HASHTAG_RE = /(?:^|[^\/\)\w])#(#{HASHTAG_NAME_RE})/i
|
||
return if (cp >= 0x80) {
|
||
true
|
||
} else when (cp.toChar()) {
|
||
'/' -> false
|
||
')' -> false
|
||
'_' -> false
|
||
in 'a'..'z' -> false
|
||
in 'A'..'Z' -> false
|
||
in '0'..'9' -> false
|
||
else -> true
|
||
}
|
||
}
|
||
|
||
private class EmojiStringBuilder(val options: DecodeOptions) {
|
||
|
||
val sb = SpannableStringBuilder()
|
||
var normal_char_start = -1
|
||
|
||
private fun openNormalText() {
|
||
if (normal_char_start == -1) {
|
||
normal_char_start = sb.length
|
||
}
|
||
}
|
||
|
||
fun closeNormalText() {
|
||
if (normal_char_start != -1) {
|
||
val end = sb.length
|
||
applyHighlight(normal_char_start, end)
|
||
normal_char_start = -1
|
||
}
|
||
}
|
||
|
||
private fun applyHighlight(start: Int, end: Int) {
|
||
val list = options.highlightTrie?.matchList(sb, start, end) ?: return
|
||
for (range in list) {
|
||
val word = HighlightWord.load(range.word) ?: continue
|
||
sb.setSpan(
|
||
HighlightSpan(word.color_fg, word.color_bg),
|
||
range.start,
|
||
range.end,
|
||
Spanned.SPAN_EXCLUSIVE_EXCLUSIVE
|
||
)
|
||
|
||
if (word.sound_type != HighlightWord.SOUND_TYPE_NONE) {
|
||
if (options.highlightSound == null) options.highlightSound = word
|
||
}
|
||
|
||
if (word.speech != 0) {
|
||
if (options.highlightSpeech == null) options.highlightSpeech = word
|
||
}
|
||
|
||
if (options.highlightAny == null) options.highlightAny = word
|
||
}
|
||
}
|
||
|
||
fun addNetworkEmojiSpan(text: String, url: String) {
|
||
closeNormalText()
|
||
val start = sb.length
|
||
sb.append(text)
|
||
val end = sb.length
|
||
sb.setSpan(
|
||
NetworkEmojiSpan(url, scale = options.enlargeCustomEmoji),
|
||
start,
|
||
end,
|
||
Spanned.SPAN_EXCLUSIVE_EXCLUSIVE
|
||
)
|
||
}
|
||
|
||
fun addImageSpan(text: String, @DrawableRes res_id: Int) {
|
||
val context = options.context
|
||
if (context == null) {
|
||
openNormalText()
|
||
sb.append(text)
|
||
} else {
|
||
closeNormalText()
|
||
val start = sb.length
|
||
sb.append(text)
|
||
val end = sb.length
|
||
sb.setSpan(
|
||
EmojiImageSpan(context, res_id, scale = options.enlargeEmoji),
|
||
start,
|
||
end,
|
||
Spanned.SPAN_EXCLUSIVE_EXCLUSIVE
|
||
)
|
||
}
|
||
}
|
||
|
||
fun addImageSpan(text: String, er: EmojiMap.EmojiResource) {
|
||
val context = options.context
|
||
if (context == null) {
|
||
openNormalText()
|
||
sb.append(text)
|
||
} else {
|
||
closeNormalText()
|
||
val start = sb.length
|
||
sb.append(text)
|
||
val end = sb.length
|
||
sb.setSpan(
|
||
er.createSpan(context, scale = options.enlargeEmoji),
|
||
start,
|
||
end,
|
||
Spanned.SPAN_EXCLUSIVE_EXCLUSIVE
|
||
)
|
||
}
|
||
}
|
||
|
||
fun addUnicodeString(s: String) {
|
||
|
||
if(!handleUnicodeEmoji){
|
||
openNormalText()
|
||
sb.append(s)
|
||
return
|
||
}
|
||
|
||
var i = 0
|
||
val end = s.length
|
||
|
||
// 絵文字ではない部分をコピーする
|
||
fun normalCopy(initialJ: Int): Boolean {
|
||
var j = initialJ
|
||
while (j < end && !EmojiMap.isStartChar(s[j])) {
|
||
j += min(end - j, Character.charCount(s.codePointAt(j)))
|
||
}
|
||
if (j <= i) return false
|
||
// https://github.com/tateisu/SubwayTooter/issues/69
|
||
val text = s.substring(i, j).replace('\u00AD', '-')
|
||
openNormalText()
|
||
sb.append(text)
|
||
i = j
|
||
return true
|
||
}
|
||
|
||
while (i < end) {
|
||
// 絵文字ではない部分をコピーする
|
||
if (normalCopy(i) && i >= end) break
|
||
|
||
// 絵文字コードを探索
|
||
val result = EmojiMap.utf16Trie.get(s,i,end)
|
||
if (result == null) {
|
||
// 見つからなかったら、通常テキストを1文字以上コピーする
|
||
normalCopy(i + min(end-i, Character.charCount(s.codePointAt(i))))
|
||
continue
|
||
}
|
||
|
||
val nextChar = if (result.endPos >= end ) null else s[result.endPos].toInt()
|
||
|
||
// 絵文字バリエーション・シーケンス（EVS）のU+FE0E（VS-15）が直後にある場合
|
||
// その文字を絵文字化しない
|
||
if (nextChar == 0xFE0E) {
|
||
normalCopy(result.endPos+1)
|
||
continue
|
||
}
|
||
|
||
val emoji = if (nextChar == 0xFE0F && s[result.endPos-1].toInt() != 0xFE0F) {
|
||
// 絵文字の最後が 0xFE0F でない
|
||
// 直後にU+0xFE0F (絵文字バリエーション・シーケンスEVSのVS-16）がある
|
||
// 直後のそれまで含めて絵文字として表示する
|
||
s.substring(i, result.endPos + 1)
|
||
} else {
|
||
s.substring(i, result.endPos)
|
||
}
|
||
addImageSpan(emoji, result.data)
|
||
i += emoji.length
|
||
}
|
||
}
|
||
}
|
||
|
||
private const val codepointColon = ':'.toInt()
|
||
// private const val codepointAtmark = '@'.toInt()
|
||
|
||
private val shortCodeCharacterSet =
|
||
SparseBooleanArray().apply {
|
||
for (c in 'A'..'Z') put(c.toInt(), true)
|
||
for (c in 'a'..'z') put(c.toInt(), true)
|
||
for (c in '0'..'9') put(c.toInt(), true)
|
||
for (c in "+-_@:") put(c.toInt(), true)
|
||
for (c in ".") put(c.toInt(), true)
|
||
}
|
||
|
||
private interface ShortCodeSplitterCallback {
|
||
fun onString(part: String) // shortcode以外の文字列
|
||
fun onShortCode(
|
||
prevCodePoint: Int,
|
||
part: String,
|
||
name: String
|
||
) // part : ":shortcode:", name : "shortcode"
|
||
}
|
||
|
||
private val reUrl = """https?://[\w/:%#@${'$'}&?!()\[\]~.=+\-]+"""
|
||
.asciiPattern()
|
||
|
||
private fun splitShortCode(
|
||
s: String,
|
||
callback: ShortCodeSplitterCallback
|
||
) {
|
||
val urlList = ArrayList<IntRange>().apply {
|
||
val m = reUrl.matcher(s)
|
||
while (m.find()) {
|
||
add(m.start()..m.end())
|
||
}
|
||
}
|
||
|
||
val end = s.length
|
||
var i = 0
|
||
while (i < end) {
|
||
|
||
// ":"以外を読み飛ばす
|
||
// URL中のコロンも読み飛ばす
|
||
var start = i
|
||
loop@ while (i < end) {
|
||
val c = s.codePointAt(i)
|
||
if (c == codepointColon && null == urlList.find { i in it }) break@loop
|
||
i += Character.charCount(c)
|
||
}
|
||
if (i > start) callback.onString(s.substring(start, i))
|
||
|
||
if (i >= end) break
|
||
|
||
start = i++ // start=コロンの位置 i=その次の位置
|
||
|
||
// 閉じるコロンを探す
|
||
var posEndColon = -1
|
||
while (i < end) {
|
||
val cp = s.codePointAt(i)
|
||
if (cp == codepointColon) {
|
||
posEndColon = i
|
||
break
|
||
} else if (!shortCodeCharacterSet.get(cp, false))
|
||
break
|
||
|
||
i += Character.charCount(cp)
|
||
}
|
||
|
||
// 閉じるコロンが見つからないか、shortcodeが短すぎるなら
|
||
// startの位置のコロンだけを処理して残りは次のループで処理する
|
||
if (posEndColon == -1 || posEndColon - start < 2) {
|
||
callback.onString(":")
|
||
i = start + 1
|
||
continue
|
||
}
|
||
|
||
val prevCodePoint = when {
|
||
start <= 0 -> 0x20
|
||
else -> s.codePointBefore(start)
|
||
}
|
||
|
||
callback.onShortCode(
|
||
prevCodePoint,
|
||
s.substring(start, posEndColon + 1), // ":shortcode:"
|
||
s.substring(start + 1, posEndColon) // "shortcode"
|
||
)
|
||
|
||
i = posEndColon + 1 // コロンの次の位置
|
||
}
|
||
}
|
||
|
||
private val reNicoru = """\Anicoru\d*\z""".asciiPattern(Pattern.CASE_INSENSITIVE)
|
||
private val reHohoemi = """\Ahohoemi\d*\z""".asciiPattern(Pattern.CASE_INSENSITIVE)
|
||
|
||
fun decodeEmoji(options: DecodeOptions, s: String): SpannableStringBuilder {
|
||
|
||
val builder = EmojiStringBuilder(options)
|
||
|
||
val emojiMapCustom = options.emojiMapCustom
|
||
val emojiMapProfile = options.emojiMapProfile
|
||
|
||
val useEmojioneShortcode = when (val context = options.context) {
|
||
null -> false
|
||
else -> Pref.bpEmojioneShortcode(context.pref())
|
||
}
|
||
|
||
splitShortCode(s, callback = object : ShortCodeSplitterCallback {
|
||
override fun onString(part: String) {
|
||
builder.addUnicodeString(part)
|
||
}
|
||
|
||
override fun onShortCode(prevCodePoint: Int, part: String, name: String) {
|
||
// フレニコのプロフ絵文字
|
||
if (emojiMapProfile != null && name.length >= 2 && name[0] == '@') {
|
||
val emojiProfile = emojiMapProfile[name] ?: emojiMapProfile[name.substring(1)]
|
||
if (emojiProfile != null) {
|
||
val url = emojiProfile.url
|
||
if (url.isNotEmpty()) {
|
||
builder.addNetworkEmojiSpan(part, url)
|
||
return
|
||
}
|
||
}
|
||
}
|
||
|
||
// カスタム絵文字
|
||
val emojiCustom = emojiMapCustom?.get(name)
|
||
if (emojiCustom != null) {
|
||
val url = when {
|
||
Pref.bpDisableEmojiAnimation(App1.pref) && emojiCustom.static_url?.isNotEmpty() == true -> emojiCustom.static_url
|
||
else -> emojiCustom.url
|
||
}
|
||
builder.addNetworkEmojiSpan(part, url)
|
||
return
|
||
}
|
||
|
||
// 通常の絵文字
|
||
if (useEmojioneShortcode) {
|
||
val info =
|
||
EmojiMap.shortNameToEmojiInfo[name.toLowerCase(Locale.JAPAN).replace('-', '_')]
|
||
if (info != null) {
|
||
builder.addImageSpan(part, info.er)
|
||
return
|
||
}
|
||
}
|
||
|
||
when {
|
||
reHohoemi.matcher(name).find() -> builder.addImageSpan(
|
||
part,
|
||
R.drawable.emoji_hohoemi
|
||
)
|
||
reNicoru.matcher(name).find() -> builder.addImageSpan(
|
||
part,
|
||
R.drawable.emoji_nicoru
|
||
)
|
||
else -> builder.addUnicodeString(part)
|
||
}
|
||
|
||
}
|
||
})
|
||
|
||
builder.closeNormalText()
|
||
|
||
return builder.sb
|
||
}
|
||
|
||
// 投稿などの際、表示は不要だがショートコード=>Unicodeの解決を行いたい場合がある
|
||
// カスタム絵文字の変換も行わない
|
||
fun decodeShortCode(
|
||
s: String,
|
||
emojiMapCustom: HashMap<String, CustomEmoji>? = null
|
||
): String {
|
||
|
||
val sb = StringBuilder()
|
||
|
||
splitShortCode(s, callback = object : ShortCodeSplitterCallback {
|
||
override fun onString(part: String) {
|
||
sb.append(part)
|
||
}
|
||
|
||
override fun onShortCode(prevCodePoint: Int, part: String, name: String) {
|
||
|
||
// カスタム絵文字にマッチするなら変換しない
|
||
val emojiCustom = emojiMapCustom?.get(name)
|
||
if (emojiCustom != null) {
|
||
sb.append(part)
|
||
return
|
||
}
|
||
|
||
// カスタム絵文字ではなく通常の絵文字のショートコードなら絵文字に変換する
|
||
val info =
|
||
EmojiMap.shortNameToEmojiInfo[name.toLowerCase(Locale.JAPAN).replace('-', '_')]
|
||
sb.append(info?.unified ?: part)
|
||
}
|
||
})
|
||
|
||
return sb.toString()
|
||
}
|
||
|
||
// 入力補完用。絵文字ショートコード一覧を部分一致で絞り込む
|
||
internal fun searchShortCode(
|
||
context: Context,
|
||
prefix: String,
|
||
limit: Int
|
||
): ArrayList<CharSequence> {
|
||
val dst = ArrayList<CharSequence>()
|
||
for (shortCode in EmojiMap.shortNameList) {
|
||
if (dst.size >= limit) break
|
||
if (!shortCode.contains(prefix)) continue
|
||
|
||
val info = EmojiMap.shortNameToEmojiInfo[shortCode] ?: continue
|
||
|
||
val sb = SpannableStringBuilder()
|
||
val start = 0
|
||
sb.append(' ')
|
||
val end = sb.length
|
||
|
||
sb.setSpan(
|
||
info.er.createSpan(context),
|
||
start,
|
||
end,
|
||
Spanned.SPAN_EXCLUSIVE_EXCLUSIVE
|
||
)
|
||
|
||
sb.append(' ')
|
||
.append(':')
|
||
.append(shortCode)
|
||
.append(':')
|
||
|
||
dst.add(sb)
|
||
}
|
||
return dst
|
||
}
|
||
}
|