From 4100339a2ba8b0c0fdf36558222f05c27aa7808a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tim=20R=C3=BChsen?= <tim.ruehsen@gmx.de>
Date: Sat, 27 Jan 2024 19:50:13 +0100
Subject: [PATCH] Parse 'srcset' HTML attr for 'source' HTML tag.
* src/html-url.c (struct known_tag): Use tag_handle_img() for 'source' tag.
* testenv/Test-recursive-include.py: Extend test.
---
src/html-url.c | 6 +--
testenv/Test-recursive-include.py | 67 ++++++++++++++++++++-----------
2 files changed, 47 insertions(+), 26 deletions(-)
diff --git a/src/html-url.c b/src/html-url.c
index 896d6fc4..d94899fb 100644
--- a/src/html-url.c
+++ b/src/html-url.c
@@ -105,7 +105,7 @@ static struct known_tag {
{ TAG_FORM, "form", tag_handle_form },
{ TAG_FRAME, "frame", tag_find_urls },
{ TAG_IFRAME, "iframe", tag_find_urls },
- { TAG_IMG, "img", tag_handle_img },
+ { TAG_IMG, "img", tag_handle_img }, // tag_find_urls() plus handling "srcset"
{ TAG_INPUT, "input", tag_find_urls },
{ TAG_LAYER, "layer", tag_find_urls },
{ TAG_LINK, "link", tag_handle_link },
@@ -118,7 +118,7 @@ static struct known_tag {
{ TAG_TH, "th", tag_find_urls },
{ TAG_VIDEO, "video", tag_find_urls },
{ TAG_AUDIO, "audio", tag_find_urls },
- { TAG_SOURCE, "source", tag_find_urls }
+ { TAG_SOURCE, "source", tag_handle_img } // tag_find_urls() plus handling "srcset"
};
/* tag_url_attributes documents which attributes of which tags contain
@@ -170,7 +170,7 @@ static struct {
{ TAG_VIDEO, "poster", ATTR_INLINE },
{ TAG_AUDIO, "src", ATTR_INLINE },
{ TAG_AUDIO, "poster", ATTR_INLINE },
- { TAG_SOURCE, "src", ATTR_INLINE }
+ { TAG_SOURCE, "src", ATTR_INLINE },
};
/* The lists of interesting tags and attributes are built dynamically,
diff --git a/testenv/Test-recursive-include.py b/testenv/Test-recursive-include.py
index 1fe33cd6..d21b305c 100755
--- a/testenv/Test-recursive-include.py
+++ b/testenv/Test-recursive-include.py
@@ -1,56 +1,77 @@
#!/usr/bin/env python3
from sys import exit
from test.http_test import HTTPTest
-from test.base_test import HTTP, HTTPS
+from test.base_test import HTTP
from misc.wget_file import WgetFile
"""
Basic test of --recursive.
"""
-############# File Definitions ###############################################
+# File Definitions ###############################################
File1 = """<html><body>
<a href=\"/a/File2.html\">text</a>
<a href=\"/b/File3.html\">text</a>
+<picture>
+ Hey, a source <source type=\"image/svg+xml\" lolli=\"xxx\" srcset=\"/a/logo.svg\"/>.
+ Hey, a srcset <img src=\"/a/picture.png\" srcset=\"/a/picture1.png, /a/picture2.png 150w,/a/picture3.png 100x\"/>.
+</picture>
</body></html>"""
File2 = "With lemon or cream?"
File3 = "Surely you're joking Mr. Feynman"
+File4 = "logosvg"
+File5 = "picturepng"
+File6 = "picture1png"
+File7 = "picture2png"
+File8 = "picture3png"
-File1_File = WgetFile ("a/File1.html", File1)
-File2_File = WgetFile ("a/File2.html", File2)
-File3_File = WgetFile ("b/File3.html", File3)
+File1_File = WgetFile("a/File1.html", File1)
+File2_File = WgetFile("a/File2.html", File2)
+File3_File = WgetFile("b/File3.html", File3)
+File4_File = WgetFile("a/logo.svg", File4)
+File5_File = WgetFile("a/picture.png", File5)
+File6_File = WgetFile("a/picture1.png", File6)
+File7_File = WgetFile("a/picture2.png", File7)
+File8_File = WgetFile("a/picture3.png", File8)
WGET_OPTIONS = "--recursive --no-host-directories --include-directories=a"
WGET_URLS = [["a/File1.html"]]
Servers = [HTTP]
-Files = [[File1_File, File2_File, File3_File]]
+Files = [[File1_File, File2_File, File3_File, File4_File, File5_File, File6_File, File7_File, File8_File]]
Existing_Files = []
ExpectedReturnCode = 0
-ExpectedDownloadedFiles = [File1_File, File2_File]
-Request_List = [["GET /a/File1.html",
- "GET /a/File2.html"]]
+ExpectedDownloadedFiles = [File1_File, File2_File, File4_File, File5_File, File6_File, File7_File, File8_File]
+Request_List = [[
+ "GET /a/File1.html",
+ "GET /a/File2.html",
+ "GET /a/logo.svg",
+ "GET /a/picture.png",
+ "GET /a/picture1.png",
+ "GET /a/picture2.png",
+ "GET /a/picture3.png",
+]]
-################ Pre and Post Test Hooks #####################################
+# Pre and Post Test Hooks #####################################
pre_test = {
- "ServerFiles" : Files,
- "LocalFiles" : Existing_Files
+ "ServerFiles": Files,
+ "LocalFiles": Existing_Files
}
test_options = {
- "WgetCommands" : WGET_OPTIONS,
- "Urls" : WGET_URLS
+ "WgetCommands": WGET_OPTIONS,
+ "Urls": WGET_URLS
}
post_test = {
- "ExpectedFiles" : ExpectedDownloadedFiles,
- "ExpectedRetcode" : ExpectedReturnCode
+ "ExpectedFiles": ExpectedDownloadedFiles,
+ "ExpectedRetcode": ExpectedReturnCode
}
-err = HTTPTest (
- pre_hook=pre_test,
- test_params=test_options,
- post_hook=post_test,
- protocols=Servers
-).begin ()
+err = HTTPTest(
+ pre_hook=pre_test,
+ test_params=test_options,
+ post_hook=post_test,
+ protocols=Servers
+).begin()
-exit (err)
+exit(err)