70 lines
2.7 KiB
Elixir
70 lines
2.7 KiB
Elixir
defmodule FeedReader.FormatDetectionTest do
|
|
use ExUnit.Case, async: true
|
|
|
|
alias FeedReader.TestSupport.ExternalFormat
|
|
|
|
test "detects RSS, Atom, and RDF from a bounded XML prefix" do
|
|
assert FeedReader.detect_format("<?xml version='1.0'?><rss version='2.0'>") == :rss
|
|
assert FeedReader.detect_format("<feed xmlns='http://www.w3.org/2005/Atom'>") == :atom
|
|
assert FeedReader.detect_format("<rdf:RDF xmlns:rdf='urn:rdf'>") == :rdf
|
|
assert FeedReader.detect_format("<html><body>not a feed</body></html>") == :unknown
|
|
end
|
|
|
|
test "custom format detection replays the bounded prefix and unread chunks" do
|
|
chunks = ["<external-feed>tail", "continued"]
|
|
|
|
{format, _prefix, replay} =
|
|
FeedReader.FormatDetector.detect_and_replay(chunks,
|
|
formats: [ExternalFormat],
|
|
max_detection_bytes: 15
|
|
)
|
|
|
|
assert format == :external_test
|
|
assert replay |> Enum.to_list() |> IO.iodata_to_binary() == IO.iodata_to_binary(chunks)
|
|
assert FeedReader.detect_format("<feed><rss version='2.0'>") == :rss
|
|
end
|
|
|
|
test "detection does not inspect content beyond the configured prefix" do
|
|
content = String.duplicate("x", 12) <> "<rss version='2.0'>"
|
|
|
|
assert FeedReader.detect_format(content, max_detection_bytes: 12) == :unknown
|
|
assert FeedReader.detect_format(content, max_detection_bytes: 40) == :rss
|
|
end
|
|
|
|
test "detection fails closed for unsupported inputs and invalid limits" do
|
|
assert FeedReader.FormatDetector.detect(:not_binary, []) == :unknown
|
|
refute FeedReader.FormatDetector.validate(nil, :rss, [])
|
|
refute FeedReader.validate("<rss version='2.0'>", :rss, max_detection_bytes: 0)
|
|
|
|
{format, _prefix, _replay} = FeedReader.FormatDetector.detect_and_replay([:not_binary], [])
|
|
assert format == :unknown
|
|
refute FeedReader.FormatDetector.validate("<rss version='2.0'>", "rss", [])
|
|
|
|
source = ["<rss version='2.0'>"]
|
|
|
|
assert FeedReader.FormatDetector.detect_and_replay(source, max_detection_bytes: 0) ==
|
|
{:unknown, "", source}
|
|
end
|
|
|
|
test "enumerable detection finalizes when the source ends before its prefix limit" do
|
|
source = ["<rss version='2.0'>"]
|
|
|
|
{format, prefix, replay} =
|
|
FeedReader.FormatDetector.detect_and_replay(source, max_detection_bytes: 64)
|
|
|
|
assert format == :rss
|
|
assert prefix == hd(source)
|
|
assert Enum.to_list(replay) == [prefix]
|
|
end
|
|
|
|
test "validation checks only the explicitly selected format" do
|
|
rss = "<rss version='2.0'><channel></channel></rss>"
|
|
|
|
assert FeedReader.validate(rss, :rss)
|
|
assert FeedReader.validate(rss, FeedReader.Formats.RSS)
|
|
refute FeedReader.validate(rss, :atom)
|
|
refute FeedReader.validate(rss, :unknown)
|
|
assert FeedReader.validate("<external-feed>", ExternalFormat)
|
|
end
|
|
end
|