fix: Improve URL / tag / mention extraction when composing (#564)

Previous code used custom regular expressions to extract URLs, hashtags, and mentions from text while the user was writing a post. These were inconsistent with the ones that Mastodon uses so the derived character count could be wrong. As well as being visually incorrect this could prevent the user from posting a status that was within the length limit, or allow them to attempt to post a status that was over the length limit (which would then fail). Fix this by dropping the homegrown regular expressions and using the same text parsing library that Mastodon users; twitter-text. This has been converted to Kotlin and the functionality related to Twitter specific features has been removed. The hashtag handling has been adjusted, as Mastodon is more permissive about the positions where hashtags can appear than Twitter is, in particular, a hashtag does not need to be preceded with whitespace if the tag appears after some scripts, such as Hirigana.
pachli · Mar 23, 2024 · 2a4126a · 2a4126a
1 parent 62fd47e
commit 2a4126a
Show file tree

Hide file tree

Showing 8 changed files with 2,266 additions and 166 deletions.
diff --git a/app/src/main/java/app/pachli/util/SpanUtils.kt b/app/src/main/java/app/pachli/util/SpanUtils.kt
@@ -1,68 +1,33 @@
+/*
+ * Copyright 2024 Pachli Association
+ *
+ * This file is a part of Pachli.
+ *
+ * This program is free software; you can redistribute it and/or modify it under the terms of the
+ * GNU General Public License as published by the Free Software Foundation; either version 3 of the
+ * License, or (at your option) any later version.
+ *
+ * Pachli is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even
+ * the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General
+ * Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along with Pachli; if not,
+ * see <http://www.gnu.org/licenses>.
+ */
+
 package app.pachli.util
 
 import android.text.Spannable
 import android.text.Spanned
-import android.text.style.CharacterStyle
 import android.text.style.ForegroundColorSpan
 import android.text.style.URLSpan
 import app.pachli.core.ui.MentionSpan
 import app.pachli.core.ui.NoUnderlineURLSpan
-import java.util.regex.Pattern
-import kotlin.math.max
-
-/**
- * @see <a href="https://github.com/tootsuite/mastodon/blob/master/app/models/tag.rb">
- *     Tag#HASHTAG_RE</a>.
- */
-private const val HASHTAG_SEPARATORS = "_\\u00B7\\u200c"
-private const val UNICODE_WORD = "\\p{L}\\p{Mn}\\p{Nd}\\p{Nl}\\p{Pc}" // Ugh, java ( https://stackoverflow.com/questions/4304928/unicode-equivalents-for-w-and-b-in-java-regular-expressions )
-private const val TAG_REGEX = "(?:^|[^/)\\w])#(([${UNICODE_WORD}_][$UNICODE_WORD$HASHTAG_SEPARATORS]*[\\p{Alpha}$HASHTAG_SEPARATORS][$UNICODE_WORD$HASHTAG_SEPARATORS]*[${UNICODE_WORD}_])|([${UNICODE_WORD}_]*[\\p{Alpha}][${UNICODE_WORD}_]*))"
-
-/**
- * @see <a href="https://github.com/tootsuite/mastodon/blob/master/app/models/account.rb">
- *     Account#MENTION_RE</a>
- */
-private const val USERNAME_REGEX = "[\\w]+([\\w\\.-]+[\\w]+)?"
-private const val MENTION_REGEX = "(?<=^|[^\\/$UNICODE_WORD])@(($USERNAME_REGEX)(?:@[$UNICODE_WORD\\.\\-]+[$UNICODE_WORD]+)?)"
-
-private const val HTTP_URL_REGEX = "(?:(^|\\b)http://[^\\s]+)"
-private const val HTTPS_URL_REGEX = "(?:(^|\\b)https://[^\\s]+)"
-
-/**
- * Dump of android.util.Patterns.WEB_URL
- */
-private val STRICT_WEB_URL_PATTERN = Pattern.compile("(((?:(?i:http|https|rtsp)://(?:(?:[a-zA-Z0-9\\\$\\-\\_\\.\\+\\!\\*\\'\\(\\)\\,\\;\\?\\&\\=]|(?:\\%[a-fA-F0-9]{2})){1,64}(?:\\:(?:[a-zA-Z0-9\\\$\\-\\_\\.\\+\\!\\*\\'\\(\\)\\,\\;\\?\\&\\=]|(?:\\%[a-fA-F0-9]{2})){1,25})?\\@)?)?(?:(([a-zA-Z0-9[ -\uD7FF豈-\uFDCFﷰ-\uFFEF\uD800\uDC00-\uD83F\uDFFD\uD840\uDC00-\uD87F\uDFFD\uD880\uDC00-\uD8BF\uDFFD\uD8C0\uDC00-\uD8FF\uDFFD\uD900\uDC00-\uD93F\uDFFD\uD940\uDC00-\uD97F\uDFFD\uD980\uDC00-\uD9BF\uDFFD\uD9C0\uDC00-\uD9FF\uDFFD\uDA00\uDC00-\uDA3F\uDFFD\uDA40\uDC00-\uDA7F\uDFFD\uDA80\uDC00-\uDABF\uDFFD\uDAC0\uDC00-\uDAFF\uDFFD\uDB00\uDC00-\uDB3F\uDFFD\uDB44\uDC00-\uDB7F\uDFFD&&[^ [ - ]\u2028\u2029 　]]](?:[a-zA-Z0-9[ -\uD7FF豈-\uFDCFﷰ-\uFFEF\uD800\uDC00-\uD83F\uDFFD\uD840\uDC00-\uD87F\uDFFD\uD880\uDC00-\uD8BF\uDFFD\uD8C0\uDC00-\uD8FF\uDFFD\uD900\uDC00-\uD93F\uDFFD\uD940\uDC00-\uD97F\uDFFD\uD980\uDC00-\uD9BF\uDFFD\uD9C0\uDC00-\uD9FF\uDFFD\uDA00\uDC00-\uDA3F\uDFFD\uDA40\uDC00-\uDA7F\uDFFD\uDA80\uDC00-\uDABF\uDFFD\uDAC0\uDC00-\uDAFF\uDFFD\uDB00\uDC00-\uDB3F\uDFFD\uDB44\uDC00-\uDB7F\uDFFD&&[^ [ - ]\u2028\u2029 　]]_\\-]{0,61}[a-zA-Z0-9[ -\uD7FF豈-\uFDCFﷰ-\uFFEF\uD800\uDC00-\uD83F\uDFFD\uD840\uDC00-\uD87F\uDFFD\uD880\uDC00-\uD8BF\uDFFD\uD8C0\uDC00-\uD8FF\uDFFD\uD900\uDC00-\uD93F\uDFFD\uD940\uDC00-\uD97F\uDFFD\uD980\uDC00-\uD9BF\uDFFD\uD9C0\uDC00-\uD9FF\uDFFD\uDA00\uDC00-\uDA3F\uDFFD\uDA40\uDC00-\uDA7F\uDFFD\uDA80\uDC00-\uDABF\uDFFD\uDAC0\uDC00-\uDAFF\uDFFD\uDB00\uDC00-\uDB3F\uDFFD\uDB44\uDC00-\uDB7F\uDFFD&&[^ [ - ]\u2028\u2029 　]]]){0,1}\\.)+(xn\\-\\-[\\w\\-]{0,58}\\w|[a-zA-Z[ -\uD7FF豈-\uFDCFﷰ-\uFFEF\uD800\uDC00-\uD83F\uDFFD\uD840\uDC00-\uD87F\uDFFD\uD880\uDC00-\uD8BF\uDFFD\uD8C0\uDC00-\uD8FF\uDFFD\uD900\uDC00-\uD93F\uDFFD\uD940\uDC00-\uD97F\uDFFD\uD980\uDC00-\uD9BF\uDFFD\uD9C0\uDC00-\uD9FF\uDFFD\uDA00\uDC00-\uDA3F\uDFFD\uDA40\uDC00-\uDA7F\uDFFD\uDA80\uDC00-\uDABF\uDFFD\uDAC0\uDC00-\uDAFF\uDFFD\uDB00\uDC00-\uDB3F\uDFFD\uDB44\uDC00-\uDB7F\uDFFD&&[^ [ - ]\u2028\u2029 　]]]{2,63})|((25[0-5]|2[0-4][0-9]|[0-1][0-9]{2}|[1-9][0-9]|[1-9])\\.(25[0-5]|2[0-4][0-9]|[0-1][0-9]{2}|[1-9][0-9]|[1-9]|0)\\.(25[0-5]|2[0-4][0-9]|[0-1][0-9]{2}|[1-9][0-9]|[1-9]|0)\\.(25[0-5]|2[0-4][0-9]|[0-1][0-9]{2}|[1-9][0-9]|[0-9]))))(?:\\:\\d{1,5})?)([/\\?](?:(?:[a-zA-Z0-9[ -\uD7FF豈-\uFDCFﷰ-\uFFEF\uD800\uDC00-\uD83F\uDFFD\uD840\uDC00-\uD87F\uDFFD\uD880\uDC00-\uD8BF\uDFFD\uD8C0\uDC00-\uD8FF\uDFFD\uD900\uDC00-\uD93F\uDFFD\uD940\uDC00-\uD97F\uDFFD\uD980\uDC00-\uD9BF\uDFFD\uD9C0\uDC00-\uD9FF\uDFFD\uDA00\uDC00-\uDA3F\uDFFD\uDA40\uDC00-\uDA7F\uDFFD\uDA80\uDC00-\uDABF\uDFFD\uDAC0\uDC00-\uDAFF\uDFFD\uDB00\uDC00-\uDB3F\uDFFD\uDB44\uDC00-\uDB7F\uDFFD&&[^ [ - ]\u2028\u2029 　]];/\\?:@&=#~\\-\\.\\+!\\*'\\(\\),_\\\$])|(?:%[a-fA-F0-9]{2}))*)?(?:\\b|\$|^))")
+import com.twitter.twittertext.Extractor
 
 private val spanClasses = listOf(ForegroundColorSpan::class.java, URLSpan::class.java)
-private val finders = mapOf(
-    FoundMatchType.HTTP_URL to PatternFinder(':', HTTP_URL_REGEX, 5, Character::isWhitespace),
-    FoundMatchType.HTTPS_URL to PatternFinder(':', HTTPS_URL_REGEX, 6, Character::isWhitespace),
-    FoundMatchType.TAG to PatternFinder('#', TAG_REGEX, 1, ::isValidForTagPrefix),
-    // TODO: We also need a proper validator for mentions
-    FoundMatchType.MENTION to PatternFinder('@', MENTION_REGEX, 1, Character::isWhitespace),
-)
 
-private enum class FoundMatchType {
-    HTTP_URL,
-    HTTPS_URL,
-    TAG,
-    MENTION,
-}
-
-private class FindCharsResult {
-    lateinit var matchType: FoundMatchType
-    var start: Int = -1
-    var end: Int = -1
-}
-
-private class PatternFinder(
-    val searchCharacter: Char,
-    regex: String,
-    val searchPrefixWidth: Int,
-    val prefixValidator: (Int) -> Boolean,
-) {
-    val pattern: Pattern = Pattern.compile(regex, Pattern.CASE_INSENSITIVE)
-}
+private val extractor = Extractor().apply { isExtractURLWithoutProtocol = false }
 
 /**
  * Takes text containing mentions and hashtags and urls and makes them the given colour.
@@ -75,18 +40,16 @@ fun highlightSpans(text: Spannable, colour: Int) {
 
     // Colour the mentions and hashtags.
     val string = text.toString()
-    val length = text.length
-    var start = 0
-    var end = 0
-    while (end in 0 until length && start >= 0) {
-        // Search for url first because it can contain the other characters
-        val found = findPattern(string, end)
-        start = found.start
-        end = found.end
-        if (start in 0 until end) {
-            text.setSpan(getSpan(found.matchType, string, colour, start, end), start, end, Spanned.SPAN_INCLUSIVE_EXCLUSIVE)
-            start += finders[found.matchType]!!.searchPrefixWidth
+
+    val entities = extractor.extractEntitiesWithIndices(string)
+
+    for (entity in entities) {
+        val span = when (entity.type) {
+            Extractor.Entity.Type.URL -> NoUnderlineURLSpan(string.substring(entity.start, entity.end))
+            Extractor.Entity.Type.HASHTAG -> ForegroundColorSpan(colour)
+            Extractor.Entity.Type.MENTION -> MentionSpan(string.substring(entity.start, entity.end))
         }
+        text.setSpan(span, entity.start, entity.end, Spanned.SPAN_INCLUSIVE_EXCLUSIVE)
     }
 }
 
@@ -95,87 +58,3 @@ private fun <T> clearSpans(text: Spannable, spanClass: Class<T>) {
         text.removeSpan(span)
     }
 }
-
-private fun findPattern(string: String, fromIndex: Int): FindCharsResult {
-    val result = FindCharsResult()
-    for (i in fromIndex..string.lastIndex) {
-        val c = string[i]
-        for (matchType in FoundMatchType.entries) {
-            val finder = finders[matchType]
-            if (finder!!.searchCharacter == c &&
-                (
-                    (i - fromIndex) < finder.searchPrefixWidth ||
-                        finder.prefixValidator(string.codePointAt(i - finder.searchPrefixWidth))
-                    )
-            ) {
-                result.matchType = matchType
-                result.start = max(0, i - finder.searchPrefixWidth)
-                findEndOfPattern(string, result, finder.pattern)
-                if (result.start + finder.searchPrefixWidth <= i + 1 && // The found result is actually triggered by the correct search character
-                    result.end >= result.start
-                ) { // ...and we actually found a valid result
-                    return result
-                }
-            }
-        }
-    }
-    return result
-}
-
-private fun findEndOfPattern(string: String, result: FindCharsResult, pattern: Pattern) {
-    val matcher = pattern.matcher(string)
-    if (matcher.find(result.start)) {
-        // Once we have API level 26+, we can use named captures...
-        val end = matcher.end()
-        result.start = matcher.start()
-        when (result.matchType) {
-            FoundMatchType.TAG -> {
-                if (isValidForTagPrefix(string.codePointAt(result.start))) {
-                    if (string[result.start] != '#' ||
-                        (string[result.start] == '#' && string[result.start + 1] == '#')
-                    ) {
-                        ++result.start
-                    }
-                }
-            }
-            else -> {
-                if (Character.isWhitespace(string.codePointAt(result.start))) {
-                    ++result.start
-                }
-            }
-        }
-        when (result.matchType) {
-            FoundMatchType.HTTP_URL, FoundMatchType.HTTPS_URL -> {
-                // Preliminary url patterns are fast/permissive, now we'll do full validation
-                if (STRICT_WEB_URL_PATTERN.matcher(string.substring(result.start, end)).matches()) {
-                    result.end = end
-                }
-            }
-            else -> result.end = end
-        }
-    }
-}
-
-private fun getSpan(matchType: FoundMatchType, string: String, colour: Int, start: Int, end: Int): CharacterStyle {
-    return when (matchType) {
-        FoundMatchType.HTTP_URL -> NoUnderlineURLSpan(string.substring(start, end))
-        FoundMatchType.HTTPS_URL -> NoUnderlineURLSpan(string.substring(start, end))
-        FoundMatchType.MENTION -> MentionSpan(string.substring(start, end))
-        else -> ForegroundColorSpan(colour)
-    }
-}
-
-private fun isWordCharacters(codePoint: Int): Boolean {
-    return (codePoint in 0x30..0x39) || // [0-9]
-        (codePoint in 0x41..0x5a) || // [A-Z]
-        (codePoint == 0x5f) || // _
-        (codePoint in 0x61..0x7a) // [a-z]
-}
-
-private fun isValidForTagPrefix(codePoint: Int): Boolean {
-    return !(
-        isWordCharacters(codePoint) || // \w
-            (codePoint == 0x2f) || // /
-            (codePoint == 0x29)
-        ) // )
-}
diff --git a/app/src/test/java/app/pachli/SpanUtilsTest.kt b/app/src/test/java/app/pachli/SpanUtilsTest.kt
@@ -4,9 +4,11 @@ import app.pachli.core.testing.fakes.FakeSpannable
 import app.pachli.util.highlightSpans
 import org.junit.Assert
 import org.junit.Test
+import org.junit.experimental.runners.Enclosed
 import org.junit.runner.RunWith
 import org.junit.runners.Parameterized
 
+@RunWith(Enclosed::class)
 class SpanUtilsTest {
     @Test
     fun matchesMixedSpans() {
@@ -19,8 +21,8 @@ class SpanUtilsTest {
 
     @Test
     fun doesntMergeAdjacentURLs() {
-        val firstURL = "http://first.thing"
-        val secondURL = "https://second.thing"
+        val firstURL = "http://first.bar"
+        val secondURL = "https://second.bar"
         val inputSpannable = FakeSpannable("$firstURL $secondURL")
         highlightSpans(inputSpannable, 0xffffff)
         val spans = inputSpannable.spans
@@ -71,14 +73,6 @@ class SpanUtilsTest {
             Assert.assertTrue(spans.isEmpty())
         }
 
-        @Test
-        fun doesNotMatchSpanEmbeddedInAnotherSpan() {
-            val inputSpannable = FakeSpannable("@aa${thingToHighlight}aa")
-            highlightSpans(inputSpannable, 0xffffff)
-            val spans = inputSpannable.spans
-            Assert.assertEquals(1, spans.size)
-        }
-
         @Test
         fun spansDoNotOverlap() {
             val begin = "@begin"

diff --git a/app/src/test/java/app/pachli/components/compose/ComposeActivityTest.kt b/app/src/test/java/app/pachli/components/compose/ComposeActivityTest.kt
@@ -288,7 +288,7 @@ class ComposeActivityTest {
 
     @Test
     fun whenTextContainsUrl_onlyEllipsizedURLIsCounted() {
-        val url = "https://www.google.dk/search?biw=1920&bih=990&tbm=isch&sa=1&ei=bmDrWuOoKMv6kwWOkIaoDQ&q=indiana+jones+i+hate+snakes+animated&oq=indiana+jones+i+hate+snakes+animated&gs_l=psy-ab.3...54174.55443.0.55553.9.7.0.0.0.0.255.333.1j0j1.2.0....0...1c.1.64.psy-ab..7.0.0....0.40G-kcDkC6A#imgdii=PSp15hQjN1JqvM:&imgrc=H0hyE2JW5wrpBM:"
+        val url = "https://www.google.dk/search?biw=1920&bih=990&tbm=isch&sa=1&ei=bmDrWuOoKMv6kwWOkIaoDQ&q=indiana+jones+i+hate+snakes+animated&oq=indiana+jones+i+hate+snakes+animated&gs_l=psy-ab.3...54174.55443.0.55553.9.7.0.0.0.0.255.333.1j0j1.2.0....0...1c.1.64.psy-ab..7.0.0....0.40G-kcDkC6A#imgdii=PSp15hQjN1JqvM:&imgrc=H0hyE2JW5wrpBM%3A"
         val additionalContent = "Check out this @image #search result: "
         rule.launch()
         rule.getScenario().onActivity {
@@ -303,7 +303,7 @@ class ComposeActivityTest {
     @Test
     fun whenTextContainsShortUrls_allUrlsGetEllipsized() {
         val shortUrl = "https://pachli.app"
-        val url = "https://www.google.dk/search?biw=1920&bih=990&tbm=isch&sa=1&ei=bmDrWuOoKMv6kwWOkIaoDQ&q=indiana+jones+i+hate+snakes+animated&oq=indiana+jones+i+hate+snakes+animated&gs_l=psy-ab.3...54174.55443.0.55553.9.7.0.0.0.0.255.333.1j0j1.2.0....0...1c.1.64.psy-ab..7.0.0....0.40G-kcDkC6A#imgdii=PSp15hQjN1JqvM:&imgrc=H0hyE2JW5wrpBM:"
+        val url = "https://www.google.dk/search?biw=1920&bih=990&tbm=isch&sa=1&ei=bmDrWuOoKMv6kwWOkIaoDQ&q=indiana+jones+i+hate+snakes+animated&oq=indiana+jones+i+hate+snakes+animated&gs_l=psy-ab.3...54174.55443.0.55553.9.7.0.0.0.0.255.333.1j0j1.2.0....0...1c.1.64.psy-ab..7.0.0....0.40G-kcDkC6A#imgdii=PSp15hQjN1JqvM:&imgrc=H0hyE2JW5wrpBM%3A"
         val additionalContent = " Check out this @image #search result: "
         rule.launch()
         rule.getScenario().onActivity {
@@ -317,7 +317,7 @@ class ComposeActivityTest {
 
     @Test
     fun whenTextContainsMultipleURLs_allURLsGetEllipsized() {
-        val url = "https://www.google.dk/search?biw=1920&bih=990&tbm=isch&sa=1&ei=bmDrWuOoKMv6kwWOkIaoDQ&q=indiana+jones+i+hate+snakes+animated&oq=indiana+jones+i+hate+snakes+animated&gs_l=psy-ab.3...54174.55443.0.55553.9.7.0.0.0.0.255.333.1j0j1.2.0....0...1c.1.64.psy-ab..7.0.0....0.40G-kcDkC6A#imgdii=PSp15hQjN1JqvM:&imgrc=H0hyE2JW5wrpBM:"
+        val url = "https://www.google.dk/search?biw=1920&bih=990&tbm=isch&sa=1&ei=bmDrWuOoKMv6kwWOkIaoDQ&q=indiana+jones+i+hate+snakes+animated&oq=indiana+jones+i+hate+snakes+animated&gs_l=psy-ab.3...54174.55443.0.55553.9.7.0.0.0.0.255.333.1j0j1.2.0....0...1c.1.64.psy-ab..7.0.0....0.40G-kcDkC6A#imgdii=PSp15hQjN1JqvM:&imgrc=H0hyE2JW5wrpBM%3A"
         val additionalContent = " Check out this @image #search result: "
         rule.launch()
         rule.getScenario().onActivity {
@@ -331,7 +331,7 @@ class ComposeActivityTest {
 
     @Test
     fun whenTextContainsUrl_onlyEllipsizedURLIsCounted_withCustomConfiguration() {
-        val url = "https://www.google.dk/search?biw=1920&bih=990&tbm=isch&sa=1&ei=bmDrWuOoKMv6kwWOkIaoDQ&q=indiana+jones+i+hate+snakes+animated&oq=indiana+jones+i+hate+snakes+animated&gs_l=psy-ab.3...54174.55443.0.55553.9.7.0.0.0.0.255.333.1j0j1.2.0....0...1c.1.64.psy-ab..7.0.0....0.40G-kcDkC6A#imgdii=PSp15hQjN1JqvM:&imgrc=H0hyE2JW5wrpBM:"
+        val url = "https://www.google.dk/search?biw=1920&bih=990&tbm=isch&sa=1&ei=bmDrWuOoKMv6kwWOkIaoDQ&q=indiana+jones+i+hate+snakes+animated&oq=indiana+jones+i+hate+snakes+animated&gs_l=psy-ab.3...54174.55443.0.55553.9.7.0.0.0.0.255.333.1j0j1.2.0....0...1c.1.64.psy-ab..7.0.0....0.40G-kcDkC6A#imgdii=PSp15hQjN1JqvM:&imgrc=H0hyE2JW5wrpBM%3A"
         val additionalContent = "Check out this @image #search result: "
         val customUrlLength = 16
         getInstanceCallback = { getInstanceWithCustomConfiguration(configuration = getCustomInstanceConfiguration(charactersReservedPerUrl = customUrlLength)) }
@@ -348,7 +348,7 @@ class ComposeActivityTest {
     @Test
     fun whenTextContainsShortUrls_allUrlsGetEllipsized_withCustomConfiguration() {
         val shortUrl = "https://pachli.app"
-        val url = "https://www.google.dk/search?biw=1920&bih=990&tbm=isch&sa=1&ei=bmDrWuOoKMv6kwWOkIaoDQ&q=indiana+jones+i+hate+snakes+animated&oq=indiana+jones+i+hate+snakes+animated&gs_l=psy-ab.3...54174.55443.0.55553.9.7.0.0.0.0.255.333.1j0j1.2.0....0...1c.1.64.psy-ab..7.0.0....0.40G-kcDkC6A#imgdii=PSp15hQjN1JqvM:&imgrc=H0hyE2JW5wrpBM:"
+        val url = "https://www.google.dk/search?biw=1920&bih=990&tbm=isch&sa=1&ei=bmDrWuOoKMv6kwWOkIaoDQ&q=indiana+jones+i+hate+snakes+animated&oq=indiana+jones+i+hate+snakes+animated&gs_l=psy-ab.3...54174.55443.0.55553.9.7.0.0.0.0.255.333.1j0j1.2.0....0...1c.1.64.psy-ab..7.0.0....0.40G-kcDkC6A#imgdii=PSp15hQjN1JqvM:&imgrc=H0hyE2JW5wrpBM%3A"
         val additionalContent = " Check out this @image #search result: "
         val customUrlLength = 18 // The intention is that this is longer than shortUrl.length
         getInstanceCallback = { getInstanceWithCustomConfiguration(configuration = getCustomInstanceConfiguration(charactersReservedPerUrl = customUrlLength)) }
@@ -364,7 +364,7 @@ class ComposeActivityTest {
 
     @Test
     fun whenTextContainsMultipleURLs_allURLsGetEllipsized_withCustomConfiguration() {
-        val url = "https://www.google.dk/search?biw=1920&bih=990&tbm=isch&sa=1&ei=bmDrWuOoKMv6kwWOkIaoDQ&q=indiana+jones+i+hate+snakes+animated&oq=indiana+jones+i+hate+snakes+animated&gs_l=psy-ab.3...54174.55443.0.55553.9.7.0.0.0.0.255.333.1j0j1.2.0....0...1c.1.64.psy-ab..7.0.0....0.40G-kcDkC6A#imgdii=PSp15hQjN1JqvM:&imgrc=H0hyE2JW5wrpBM:"
+        val url = "https://www.google.dk/search?biw=1920&bih=990&tbm=isch&sa=1&ei=bmDrWuOoKMv6kwWOkIaoDQ&q=indiana+jones+i+hate+snakes+animated&oq=indiana+jones+i+hate+snakes+animated&gs_l=psy-ab.3...54174.55443.0.55553.9.7.0.0.0.0.255.333.1j0j1.2.0....0...1c.1.64.psy-ab..7.0.0....0.40G-kcDkC6A#imgdii=PSp15hQjN1JqvM:&imgrc=H0hyE2JW5wrpBM%3A"
         val additionalContent = " Check out this @image #search result: "
         val customUrlLength = 16
         getInstanceCallback = { getInstanceWithCustomConfiguration(configuration = getCustomInstanceConfiguration(charactersReservedPerUrl = customUrlLength)) }

diff --git a/app/src/test/java/app/pachli/components/compose/StatusLengthTest.kt b/app/src/test/java/app/pachli/components/compose/StatusLengthTest.kt
@@ -40,9 +40,12 @@ class StatusLengthTest(
                 // "@user@server" should be treated as "@user"
                 arrayOf("123 @[email protected]", 12),
                 // URLs under 23 chars are treated as 23 chars
-                arrayOf("123 http://example.url", 27),
+                arrayOf("123 http://example.org", 27),
                 // URLs over 23 chars are treated as 23 chars
                 arrayOf("123 http://urlthatislongerthan23characters.example.org", 27),
+                // URLs end when they should (the ")." should be part of the status
+                // length, not considered to be part of the URL)
+                arrayOf("test (https://example.com). test", 36),
                 // Short hashtags are treated as is
                 arrayOf("123 #basictag", 13),
                 // Long hashtags are *also* treated as is (not treated as 23, like URLs)