1 files changed, 36 insertions, 5 deletions

diff --git a/app/src/main/java/org/terst/feedreader/data/FeedParser.kt b/app/src/main/java/org/terst/feedreader/data/FeedParser.kt
index cd337d6..735a707 100644
--- a/app/src/main/java/org/terst/feedreader/data/FeedParser.kt
+++ b/app/src/main/java/org/terst/feedreader/data/FeedParser.kt
@@ -16,7 +16,8 @@ data class ParsedItem(
     val link: String,
     val author: String?,
     val publishedAt: Long?,
-    val contentHtml: String?
+    val contentHtml: String?,
+    val imageUrl: String? = null
 )
 
 data class ParsedFeed(
@@ -65,6 +66,7 @@ object FeedParser {
         var itemDate: Long? = null
         var itemContent: String? = null
         var itemDescription: String? = null
+        var itemImageUrl: String? = null
 
         while (eventType != XmlPullParser.END_DOCUMENT && depth > 0) {
             when (eventType) {
@@ -74,7 +76,7 @@ object FeedParser {
                         "item" -> {
                             inItem = true
                             itemGuid = null; itemTitle = ""; itemLink = ""; itemAuthor = null
-                            itemDate = null; itemContent = null; itemDescription = null
+                            itemDate = null; itemContent = null; itemDescription = null; itemImageUrl = null
                         }
                         "title" -> {
                             val text = parser.nextText()
@@ -90,19 +92,31 @@ object FeedParser {
                         "pubdate" -> if (inItem) itemDate = parseRfc822Date(parser.nextText())
                         "content:encoded" -> if (inItem) itemContent = parser.nextText()
                         "description" -> if (inItem) itemDescription = parser.nextText()
+                        "media:thumbnail", "media:content" -> if (inItem && itemImageUrl == null) {
+                            val url = parser.getAttributeValue(null, "url")
+                            val type = parser.getAttributeValue(null, "type")
+                            if (url != null && (type == null || type.startsWith("image/"))) itemImageUrl = url
+                        }
+                        "enclosure" -> if (inItem && itemImageUrl == null) {
+                            val url = parser.getAttributeValue(null, "url")
+                            val type = parser.getAttributeValue(null, "type")
+                            if (url != null && type?.startsWith("image/") == true) itemImageUrl = url
+                        }
                     }
                 }
                 XmlPullParser.END_TAG -> {
                     depth--
                     if (parser.name.lowercase() == "item") {
                         inItem = false
+                        val content = itemContent ?: itemDescription
                         items += ParsedItem(
                             guid = itemGuid ?: itemLink.ifBlank { null },
                             title = itemTitle,
                             link = itemLink,
                             author = itemAuthor,
                             publishedAt = itemDate,
-                            contentHtml = itemContent ?: itemDescription
+                            contentHtml = content,
+                            imageUrl = itemImageUrl ?: extractFirstImageUrl(content)
                         )
                     }
                 }
@@ -127,6 +141,7 @@ object FeedParser {
         var entryDate: Long? = null
         var entryContent: String? = null
         var entrySummary: String? = null
+        var entryImageUrl: String? = null
 
         while (eventType != XmlPullParser.END_DOCUMENT && depth > 0) {
             when (eventType) {
@@ -136,7 +151,7 @@ object FeedParser {
                         "entry" -> {
                             inEntry = true
                             entryId = null; entryTitle = ""; entryLink = ""; entryAuthor = null
-                            entryDate = null; entryContent = null; entrySummary = null
+                            entryDate = null; entryContent = null; entrySummary = null; entryImageUrl = null
                         }
                         "title" -> {
                             val text = parser.nextText()
@@ -145,9 +160,13 @@ object FeedParser {
                         "link" -> {
                             val rel = parser.getAttributeValue(null, "rel")
                             val href = parser.getAttributeValue(null, "href")
+                            val type = parser.getAttributeValue(null, "type")
                             if (href != null && (rel == null || rel == "alternate")) {
                                 if (inEntry) entryLink = href else if (siteUrl == null) siteUrl = href
                             }
+                            if (inEntry && rel == "enclosure" && type?.startsWith("image/") == true && href != null) {
+                                entryImageUrl = href
+                            }
                         }
                         "id" -> if (inEntry) entryId = parser.nextText()
                         "name" -> if (inEntry && entryAuthor == null) entryAuthor = parser.nextText()
@@ -155,19 +174,24 @@ object FeedParser {
                         "published" -> if (inEntry) entryDate = parseIso8601Date(parser.nextText()) ?: entryDate
                         "content" -> if (inEntry) entryContent = parser.nextText()
                         "summary" -> if (inEntry) entrySummary = parser.nextText()
+                        "media:thumbnail" -> if (inEntry && entryImageUrl == null) {
+                            entryImageUrl = parser.getAttributeValue(null, "url")
+                        }
                     }
                 }
                 XmlPullParser.END_TAG -> {
                     depth--
                     if (parser.name.lowercase() == "entry") {
                         inEntry = false
+                        val content = entryContent ?: entrySummary
                         items += ParsedItem(
                             guid = entryId ?: entryLink.ifBlank { null },
                             title = entryTitle,
                             link = entryLink,
                             author = entryAuthor,
                             publishedAt = entryDate,
-                            contentHtml = entryContent ?: entrySummary
+                            contentHtml = content,
+                            imageUrl = entryImageUrl ?: extractFirstImageUrl(content)
                         )
                     }
                 }
@@ -177,6 +201,13 @@ object FeedParser {
         return ParsedFeed(feedTitle, siteUrl, items)
     }
 
+    private val imgTagRegex = Regex("""<img[^>]+src=["']([^"']+)["']""", RegexOption.IGNORE_CASE)
+
+    // Fallback for feeds with no explicit media:thumbnail/enclosure image --
+    // most blog feeds just embed the post's lead image directly in the content.
+    private fun extractFirstImageUrl(html: String?): String? =
+        html?.let { imgTagRegex.find(it)?.groupValues?.get(1) }
+
     private val rfc822Formatters = listOf(
         DateTimeFormatter.RFC_1123_DATE_TIME,
         DateTimeFormatterBuilder()