From b58b1fa962bdfeaeb6ef76d8af53baa9b14a01ae Mon Sep 17 00:00:00 2001
From: liuhao1024 <sunsky.lau@gmail.com>
Date: Mon, 13 Jul 2026 15:04:33 +0800
Subject: [PATCH] fix(gateway): allow [[as_document]] glued directly to MEDIA
 path (#63632)

MEDIA_TAG_CLEANUP_RE failed to match when a directive like [[as_document]]
was concatenated directly to the file extension without whitespace
(e.g., MEDIA:/home/user/report.xlsx[[as_document]]). This caused the
file to be silently not delivered while the gateway reported success.

Root cause: the lookahead character class [\s`",;:)\}\]|$) did not
include [, so [[as_document]] immediately after the extension broke
the lookahead assertion.

Fix: add \[ to the lookahead class so directives can follow the path
without whitespace. This is safe because [ is already stripped elsewhere
in the same file via .replace("[[as_document]]", "").

Added regression tests covering:
- Directive glued to extension (issue case)
- Directive with whitespace (baseline)
- Tag at end of string ($ anchor)
---
 gateway/platforms/base.py               |  2 +-
 tests/gateway/test_media_tag_cleanup.py | 61 +++++++++++++++++++++++++
 2 files changed, 62 insertions(+), 1 deletion(-)
 create mode 100644 tests/gateway/test_media_tag_cleanup.py

diff --git a/gateway/platforms/base.py b/gateway/platforms/base.py
index 4f2c0e98fa4f..af74265d9a17 100644
--- a/gateway/platforms/base.py
+++ b/gateway/platforms/base.py
@@ -1514,7 +1514,7 @@ MEDIA_TAG_CLEANUP_RE = re.compile(
     r'''[`"']?MEDIA:\s*'''
     r'''(?P<path>`[^`\n]+`|"[^"\n]+"|'[^'\n]+'|'''
     r'''(?:~/|/|[A-Za-z]:[/\\])\S+(?:[^\S\n]+\S+)*?\.(?:''' + _MEDIA_EXT_ALTERNATION + r'''))'''
-    r'''(?=[\s`"',;:)\]}]|$)[`"']?''',
+    r'''(?=[\s`"',;:)\]}\[]|$)[`"']?''',
     re.IGNORECASE,
 )
 
diff --git a/tests/gateway/test_media_tag_cleanup.py b/tests/gateway/test_media_tag_cleanup.py
new file mode 100644
index 000000000000..2bf77845c232
--- /dev/null
+++ b/tests/gateway/test_media_tag_cleanup.py
@@ -0,0 +1,61 @@
+"""Tests for MEDIA_TAG_CLEANUP_RE regex matching behavior (#63632)."""
+
+
+class TestMediaTagCleanup:
+    """Tests for MEDIA_TAG_CLEANUP_RE regex matching behavior."""
+
+    def test_media_tag_with_directive_glued_to_extension(self):
+        """Regression: MEDIA:<path>[[as_document]] must match when directive is glued
+        directly to the extension without whitespace (#63632).
+
+        The fix adds `\\[` to the lookahead character class in MEDIA_TAG_CLEANUP_RE.
+        """
+        from gateway.platforms.base import MEDIA_TAG_CLEANUP_RE
+
+        # Issue case: [[as_document]] glued directly to .xlsx
+        text = "Готово. MEDIA:/home/hermes/report.xlsx[[as_document]]"
+        assert MEDIA_TAG_CLEANUP_RE.search(text) is not None
+        stripped = MEDIA_TAG_CLEANUP_RE.sub("", text)
+        assert "MEDIA:" not in stripped
+        assert "/home/hermes/report.xlsx" not in stripped
+
+        # Same with whitespace (should still work)
+        text_with_space = "Готово. MEDIA:/home/hermes/report.xlsx [[as_document]]"
+        assert MEDIA_TAG_CLEANUP_RE.search(text_with_space) is not None
+        stripped = MEDIA_TAG_CLEANUP_RE.sub("", text_with_space)
+        assert "MEDIA:" not in stripped
+        assert "/home/hermes/report.xlsx" not in stripped
+
+        # Other directives ([[as_image]]) should also work
+        text_image = "Done. MEDIA:/tmp/chart.png[[as_image]]"
+        assert MEDIA_TAG_CLEANUP_RE.search(text_image) is not None
+        stripped = MEDIA_TAG_CLEANUP_RE.sub("", text_image)
+        assert "MEDIA:" not in stripped
+        assert "/tmp/chart.png" not in stripped
+
+    def test_media_tag_with_whitespace_still_works(self):
+        """Baseline: MEDIA tags with whitespace before/after still match."""
+        from gateway.platforms.base import MEDIA_TAG_CLEANUP_RE
+
+        # Space before closing quote
+        text = "Here is your report: MEDIA:/tmp/report.md "
+        stripped = MEDIA_TAG_CLEANUP_RE.sub("", text).strip()
+        assert "MEDIA:" not in stripped
+        assert "/tmp/report.md" not in stripped
+
+        # Multiple spaces (regex removes tag but preserves surrounding whitespace)
+        text = "Report at MEDIA:/tmp/data.pdf  done"
+        stripped = MEDIA_TAG_CLEANUP_RE.sub("", text)
+        assert "MEDIA:" not in stripped
+        assert "/tmp/data.pdf" not in stripped
+        assert "Report at" in stripped and "done" in stripped
+
+    def test_media_tag_at_end_of_string(self):
+        """MEDIA tags at the end of a string should match ($ anchor)."""
+        from gateway.platforms.base import MEDIA_TAG_CLEANUP_RE
+
+        text = "Here is the file: MEDIA:/tmp/file.docx"
+        stripped = MEDIA_TAG_CLEANUP_RE.sub("", text).strip()
+        assert "MEDIA:" not in stripped
+        assert "/tmp/file.docx" not in stripped
+        assert "Here is the file:" in stripped