Fix with unwrap_html + update tests

closeio · Jun 28, 2024 · 273addf · 273addf
1 parent 89176b6
commit 273addf
Show file tree

Hide file tree

Showing 9 changed files with 194 additions and 48 deletions.
diff --git a/pyproject.toml b/pyproject.toml
@@ -55,14 +55,16 @@ max-branches = 16
 
 [tool.ruff.lint.per-file-ignores]
 "tests/test_quotequail.py" = ["E501", "PT009"]
+"tests/test_unwrap_html.py" = ["E501"]
 
 [tool.mypy]
-python_version = "3.10"
+enable_error_code = "possibly-undefined"
+files = "quotequail"
+follow_imports = "normal"
 ignore_missing_imports = true
 no_implicit_optional = true
+pretty = true
+python_version = "3.10"
+show_error_context = true
 strict_equality = true
-follow_imports = "normal"
 warn_unreachable = true
-show_error_context = true
-pretty = true
-files = "quotequail"
diff --git a/quotequail/__init__.py b/quotequail/__init__.py
@@ -141,46 +141,48 @@ def unwrap_html(html: str) -> dict[str, str] | None:
 
     unwrap_result = _internal.unwrap(lines, 1, _patterns.MIN_HEADER_LINES, 1)
 
-    if unwrap_result:
-        typ, top_range, hdrs, main_range, bottom_range, needs_unindent = (
-            unwrap_result
+    if not unwrap_result:
+        return None
+
+    typ, top_range, hdrs, main_range, bottom_range, needs_unindent = (
+        unwrap_result
+    )
+
+    result = {
+        "type": typ,
+    }
+
+    top_range_slice = _html.trim_slice(lines, top_range)
+    main_range_slice = _html.trim_slice(lines, main_range)
+    bottom_range_slice = _html.trim_slice(lines, bottom_range)
+
+    if top_range_slice:
+        top_tree = _html.slice_tree(
+            tree, start_refs, end_refs, top_range_slice, html_copy=html
+        )
+        html_top = _html.render_html_tree(top_tree)
+        if html_top:
+            result["html_top"] = html_top
+
+    if bottom_range_slice:
+        bottom_tree = _html.slice_tree(
+            tree, start_refs, end_refs, bottom_range_slice, html_copy=html
         )
+        html_bottom = _html.render_html_tree(bottom_tree)
+        if html_bottom:
+            result["html_bottom"] = html_bottom
 
-        result = {
-            "type": typ,
-        }
-
-        top_range_slice = _html.trim_slice(lines, top_range)
-        main_range_slice = _html.trim_slice(lines, main_range)
-        bottom_range_slice = _html.trim_slice(lines, bottom_range)
-
-        if top_range_slice:
-            top_tree = _html.slice_tree(
-                tree, start_refs, end_refs, top_range_slice, html_copy=html
-            )
-            html_top = _html.render_html_tree(top_tree)
-            if html_top:
-                result["html_top"] = html_top
-
-        if bottom_range_slice:
-            bottom_tree = _html.slice_tree(
-                tree, start_refs, end_refs, bottom_range_slice, html_copy=html
-            )
-            html_bottom = _html.render_html_tree(bottom_tree)
-            if html_bottom:
-                result["html_bottom"] = html_bottom
-
-        if main_range_slice:
-            main_tree = _html.slice_tree(
-                tree, start_refs, end_refs, main_range_slice
-            )
-            if needs_unindent:
-                _html.unindent_tree(main_tree)
-            html = _html.render_html_tree(main_tree)
-            if html:
-                result["html"] = html
-
-        if hdrs:
-            result.update(hdrs)
+    if main_range_slice:
+        main_tree = _html.slice_tree(
+            tree, start_refs, end_refs, main_range_slice
+        )
+        if needs_unindent:
+            _html.unindent_tree(main_tree)
+        html = _html.render_html_tree(main_tree)
+        if html:
+            result["html"] = html
+
+    if hdrs:
+        result.update(hdrs)
 
     return result
diff --git a/quotequail/_html.py b/quotequail/_html.py
@@ -187,10 +187,8 @@ def slice_tree(
     else:
         new_tree = tree
 
-    if start_ref:
-        include_start = start_ref[1] is Position.Begin
-    if end_ref:
-        include_end = end_ref[1] is Position.End
+    include_start = start_ref[1] is Position.Begin if start_ref else False
+    include_end = end_ref[1] is Position.End if end_ref else False
 
     # If start_ref is the same as end_ref, and we don't include the element,
     # we are removing the entire tree. We need to handle this separately,

diff --git a/tests/conftest.py b/tests/conftest.py
@@ -0,0 +1,14 @@
+import os
+
+import pytest
+
+
+@pytest.fixture()
+def read_file():
+    def _read_file(name: str) -> str:
+        with open(
+            os.path.join(os.path.dirname(__file__), "files", name), "rb"
+        ) as f:
+            return f.read().decode("utf8")
+
+    return _read_file
diff --git a/tests/files/apple_forward.html b/tests/files/apple_forward.html
@@ -0,0 +1 @@
+<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">test<div class=""><br class=""></div><div class="">blah<br class=""><div><br class=""><div><br class=""><blockquote type="cite" class=""><div class="">Begin forwarded message:</div><br class="Apple-interchange-newline"><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px;" class=""><span style="font-family: -webkit-system-font, Helvetica Neue, Helvetica, sans-serif; color:rgba(0, 0, 0, 1.0);" class=""><b class="">From: </b></span><span style="font-family: -webkit-system-font, Helvetica Neue, Helvetica, sans-serif;" class="">Foo Bar &lt;<a href="mailto:foo@bar.example" class="">foo@bar.example</a>&gt;<br class=""></span></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px;" class=""><span style="font-family: -webkit-system-font, Helvetica Neue, Helvetica, sans-serif; color:rgba(0, 0, 0, 1.0);" class=""><b class="">Subject: </b></span><span style="font-family: -webkit-system-font, Helvetica Neue, Helvetica, sans-serif;" class=""><b class="">The Subject</b><br class=""></span></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px;" class=""><span style="font-family: -webkit-system-font, Helvetica Neue, Helvetica, sans-serif; color:rgba(0, 0, 0, 1.0);" class=""><b class="">Date: </b></span><span style="font-family: -webkit-system-font, Helvetica Neue, Helvetica, sans-serif;" class="">March 24, 2016 at 20:16:25 GMT+1<br class=""></span></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px;" class=""><span style="font-family: -webkit-system-font, Helvetica Neue, Helvetica, sans-serif; color:rgba(0, 0, 0, 1.0);" class=""><b class="">To: </b></span><span style="font-family: -webkit-system-font, Helvetica Neue, Helvetica, sans-serif;" class="">John Doe &lt;<a href="mailto:john@doe.example" class="">john@doe.example</a>&gt;<br class=""></span></div><br class=""><div class=""><div dir="ltr" class="">Text of the original email</div>
diff --git a/tests/files/apple_reply.html b/tests/files/apple_reply.html
@@ -0,0 +1 @@
+<html><head><meta http-equiv="Content-Type" content="text/html charset=us-ascii"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">Foo<div class=""><br class=""></div><div class="">Bar</div><div class=""><br class=""></div><div class=""><div><blockquote type="cite" class=""><div class="">On 2016-03-25, at 23:01, John Doe &lt;<a href="mailto:john@doe.example" class="">john@doe.example</a>&gt; wrote:</div><br class="Apple-interchange-newline"><div class=""><meta http-equiv="Content-Type" content="text/html charset=us-ascii" class=""><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">Some <b class="">important</b> email<br class=""></div></div></blockquote></div><br class=""></div></body></html>
diff --git a/tests/files/gmail_forward.html b/tests/files/gmail_forward.html
@@ -0,0 +1 @@
+<html><head></head><body><div dir="ltr">test<div><br></div><div>blah</div><div><br><div class="gmail_quote">---------- Forwarded message ----------<br>From: <b class="gmail_sendername">Foo Bar</b> <span dir="ltr">&lt;<a href="mailto:foo@bar.example">foo@bar.example</a>&gt;</span><br>Date: Thu, Mar 24, 2016 at 5:17 PM<br>Subject: The Subject<br>To: John Doe &lt;<a href="mailto:john@doe.example">john@doe.example</a>&gt;<br><br><br><div dir="ltr">Some text<div><br></div><div><br></div></div></div><br></div></div></body></html>
diff --git a/tests/files/gmail_reply.html b/tests/files/gmail_reply.html
@@ -0,0 +1,9 @@
+<html><head></head><body><div dir="ltr">foo<div><br></div><div>bar</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Mar 16, 2016 at 12:49 AM, Foo Bar <span dir="ltr">&lt;<a href="mailto:foo@bar.example" target="_blank">foo@bar.example</a>&gt;</span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi,<br>
+<br>This is the reply<br>
+<br>
+Thanks a lot!<br>
+<span class="HOEnZb"><font color="#888888">Foo<br>
+<br>
+</font></span></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature"><div dir="ltr"><div><div dir="ltr"><b>John Doe</b></div><div dir="ltr"><b>Senior Director</b><div>Some Company</div></div></div></div></div>
+</div>
+</body></html>
diff --git a/tests/test_unwrap_html.py b/tests/test_unwrap_html.py
@@ -0,0 +1,118 @@
+import pytest
+
+from quotequail import unwrap_html
+
+
+@pytest.mark.parametrize(
+    ("html", "expected"),
+    [
+        ("<p>html text</p>", None),
+        (
+            "Begin forwarded message:<br>\n<br>\nFrom: someone@example.com<br>\nTo: anyone@example.com<br>\nSubject: You won<br>\n",
+            {
+                "type": "forward",
+                "from": "someone@example.com",
+                "to": "anyone@example.com",
+                "subject": "You won",
+            },
+        ),
+    ],
+)
+def test_unwrap_html_simple(html, expected):
+    assert unwrap_html(html) == expected
+
+
+@pytest.mark.parametrize(
+    ("file", "expected"),
+    [
+        (
+            "apple_forward.html",
+            {
+                "type": "forward",
+                "subject": "The Subject",
+                "date": "March 24, 2016 at 20:16:25 GMT+1",
+                "from": "Foo Bar <foo@bar.example>",
+                "to": "John Doe <john@doe.example>",
+                "html_top": '<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">test<div class=""><br class=""></div><div class="">blah</div></body></html>',
+                "html": '<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class=""><div class=""><div><div><div><div class=""><div dir="ltr" class="">Text of the original email</div></div></div></div></div></div></body></html>',
+            },
+        ),
+        (
+            "gmail_forward.html",
+            {
+                "type": "forward",
+                "subject": "The Subject",
+                "date": "Thu, Mar 24, 2016 at 5:17 PM",
+                "from": "Foo Bar <foo@bar.example>",
+                "to": "John Doe <john@doe.example>",
+                "html_top": '<html><head></head><body><div dir="ltr">test<div><br></div><div>blah</div></div></body></html>',
+                "html": '<html><head></head><body><div dir="ltr"><div><div class="gmail_quote"><div dir="ltr">Some text</div></div></div></div></body></html>',
+            },
+        ),
+        (
+            "apple_reply.html",
+            {
+                "type": "reply",
+                "from": "John Doe <john@doe.example>",
+                "date": "2016-03-25, at 23:01",
+                "html": '<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class=""><div class=""><div><div><div class=""><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">Some <b class="">important</b> email</div></div></div></div></div></body></html>',
+                "html_top": '<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">Foo<div class=""><br class=""></div><div class="">Bar</div></body></html>',
+            },
+        ),
+        (
+            "gmail_reply.html",
+            {
+                "type": "reply",
+                "from": "Foo Bar <foo@bar.example>",
+                "date": "Wed, Mar 16, 2016 at 12:49 AM",
+                "html_top": '<html><head></head><body><div dir="ltr">foo<div><br></div><div>bar</div></div></body></html>',
+                "html": '<html><head></head><body><div class="gmail_extra"><div class="gmail_quote"><div>Hi,<br>\n<br>This is the reply<br>\n<br>\nThanks a lot!<br>\n<span class="HOEnZb"><font color="#888888">Foo</font></span></div></div></div></body></html>',
+                "html_bottom": '<html><head></head><body><div class="gmail_extra">-- <br><div class="gmail_signature"><div dir="ltr"><div><div dir="ltr"><b>John Doe</b></div><div dir="ltr"><b>Senior Director</b><div>Some Company</div></div></div></div></div>\n</div>\n</body></html>',
+            },
+        ),
+    ],
+)
+def test_unwrap_html_file(read_file, file, expected):
+    html = read_file(file)
+    assert unwrap_html(html) == expected
+
+
+def test_outlook_forward(read_file):
+    data = read_file("outlook_forward.html")
+    result = unwrap_html(data)
+    assert result["type"] == "forward"
+    assert result["from"] == "John Doe"
+    assert result["to"] == "Foo Bar (foo@bar.example)"
+    assert result["date"] == "Wednesday, July 09, 2014 10:27 AM"
+    assert result["subject"] == "The subject!"
+    assert result["html"] == read_file("outlook_forward_unwrapped.html")
+    assert result["html_top"] == read_file(
+        "outlook_forward_unwrapped_top.html"
+    )
+    assert "html_bottom" not in result
+
+
+def test_thunderbird_forward(read_file):
+    data = read_file("thunderbird_forward.html")
+    result = unwrap_html(data)
+    assert result["type"] == "forward"
+    assert result["from"] == "John Doe <johndoe@example.com>"
+    assert result["to"] == "Foo Bar <foobar@example.com>"
+    assert result["date"] == "Tue, 3 May 2016 14:54:27 +0200 (CEST)"
+    assert result["subject"] == "Re: Example subject"
+    assert "html_top" not in result
+    assert result["html"] == read_file("thunderbird_forward_unwrapped.html")
+    assert "html_bottom" not in result
+
+
+def test_mailru_forward(read_file):
+    data = read_file("mailru_forward.html")
+    result = unwrap_html(data)
+    assert result["type"] == "forward"
+    assert result["from"] == "Иван Иванов <ivanivanov@example.com>"
+    assert result["to"] == "Петр Петров <petrpetrov@example.com>"
+    assert result["date"] == "Среда, 14 июня 2017, 15:19 +03:00"
+    assert result["subject"] == "Тестовая тема"
+    assert "html_top" not in result
+    assert result["html"] == read_file("mailru_forward_unwrapped.html")
+    assert "html_bottom" not in result