From: Max Carrara <m.carrara@proxmox.com>
To: pve-devel@lists.proxmox.com
Subject: [pve-devel] [PATCH v2 pve-common 07/12] test: add tests for path_parent of PVE::Path
Date: Fri, 20 Dec 2024 19:52:02 +0100 [thread overview]
Message-ID: <20241220185207.519912-8-m.carrara@proxmox.com> (raw)
In-Reply-To: <20241220185207.519912-1-m.carrara@proxmox.com>
What's important to note is that among the cases added are also some
that specifically account for Perl's File::Spec->canonpath treating
paths consisting of a single component that is neither "/" or "." as
the same. This means that e.g. "foo" and "./foo" are both the same
(canonpath returns "foo").
Because canonpath also returns "" instead of "." for the empty path
"", the behaviour of path_parent should be consistent with that of
Perl's canonpath (and in turn therefore also for path_normalize, which
wraps canonpath).
The path_parent function should otherwise work the same as Rust's
std::path::Path::parent [1]. The discrepancies as mentioned above are
checked for by the tests introduced here.
[1]: https://doc.rust-lang.org/std/path/struct.Path.html#method.parent
Signed-off-by: Max Carrara <m.carrara@proxmox.com>
---
Changes v1 --> v2:
* NEW: Split from patch 02
test/Path/Makefile | 1 +
test/Path/path_parent_tests.pl | 153 +++++++++++++++++++++++++++++++++
2 files changed, 154 insertions(+)
create mode 100755 test/Path/path_parent_tests.pl
diff --git a/test/Path/Makefile b/test/Path/Makefile
index 9dd95f1..a2b5bb1 100644
--- a/test/Path/Makefile
+++ b/test/Path/Makefile
@@ -2,6 +2,7 @@ TESTS = \
path_components_tests.pl \
path_is_absolute_relative_tests.pl \
path_join_tests.pl \
+ path_parent_tests.pl \
path_push_tests.pl \
diff --git a/test/Path/path_parent_tests.pl b/test/Path/path_parent_tests.pl
new file mode 100755
index 0000000..9dfa301
--- /dev/null
+++ b/test/Path/path_parent_tests.pl
@@ -0,0 +1,153 @@
+#!/usr/bin/env perl
+
+use lib '../../src';
+
+use strict;
+use warnings;
+
+use Test::More;
+
+use PVE::Path;
+
+my $path_parent_cases = [
+ {
+ name => "empty path",
+ path => "",
+ parent => undef,
+ },
+ {
+ name => "root",
+ path => "/",
+ parent => undef,
+ },
+ {
+ name => "current path reference",
+ path => ".",
+ parent => "",
+ },
+ # Why? Because File::Spec->canonpath interprets "./foo" and "foo"
+ # as the same thing, and the parent of "./foo" is "."
+ {
+ name => "single component, relative",
+ path => "foo",
+ parent => ".",
+ },
+ # Same as above!
+ {
+ name => "parent directory reference",
+ path => "..",
+ parent => ".",
+ },
+ {
+ name => "single component, absolute",
+ path => "/foo",
+ parent => "/",
+ },
+ {
+ name => "multiple components, relative",
+ path => "foo/bar/baz/quo/qux",
+ parent => "foo/bar/baz/quo",
+ },
+ {
+ name => "multiple components, absolute",
+ path => "/foo/bar/baz/quo/qux",
+ parent => "/foo/bar/baz/quo",
+ },
+ {
+ name => "multiple components, with redundant current path references",
+ path => "foo/bar/./baz/././quo/./././qux/././.",
+ parent => "foo/bar/./baz/././quo",
+ },
+ {
+ name => "multiple components, with parent directory references",
+ path => "foo/bar/../baz/../../quo/../../../qux/../../..",
+ parent => "foo/bar/../baz/../../quo/../../../qux/../..",
+ },
+ {
+ name => "multiple components, with redundant path separators",
+ path => "foo//bar///baz////quo/////qux//////",
+ parent => "foo//bar///baz////quo",
+ },
+ {
+ name => "root path, with redundant path separators",
+ path => "///////////",
+ parent => undef,
+ },
+ {
+ name => "root path, with redundant current path references",
+ path => "/././././././././././.",
+ parent => undef,
+ },
+ {
+ name => "current path reference, with redundant path separators",
+ path => ".///////////",
+ parent => "",
+ },
+ {
+ name => "current path reference, with redundant current path references",
+ path => "././././././././././.",
+ parent => "",
+ },
+ {
+ name => "multiple components,"
+ . " with redundant path separators,"
+ . " with redundant current path references",
+ path => "foo/.///./bar//././////././/baz//././///././././///",
+ parent => "foo/.///./bar",
+ },
+ {
+ name => "path is undef",
+ path => undef,
+ parent => undef,
+ should_throw => 1,
+ },
+];
+
+sub test_path_parent : prototype($) {
+ my ($case) = @_;
+
+ my $name = "path_parent: " . $case->{name};
+
+ my $parent = eval { PVE::Path::path_parent($case->{path}); };
+
+ if ($@) {
+ if ($case->{should_throw}) {
+ pass($name);
+ return;
+ }
+
+ fail($name);
+ diag("Failed to get parent from path:\n$@");
+ return;
+ }
+
+ # Note: `!is()` isn't the same as `isnt()` -- we want extra output here
+ # if the check fails; can't do that with `isnt()`
+ if (!is($parent, $case->{parent}, $name)) {
+ diag("=== Expected ===");
+ diag(explain($case->{parent}));
+ diag("=== Got ===");
+ diag(explain($parent));
+ }
+
+ return;
+}
+
+sub main : prototype() {
+ plan(tests => scalar($path_parent_cases->@*));
+
+ for my $case ($path_parent_cases->@*) {
+ eval {
+ # suppress warnings here to make output less noisy for certain tests if necessary
+ # local $SIG{__WARN__} = sub {};
+ test_path_parent($case);
+ };
+ warn "$@\n" if $@;
+ }
+
+ done_testing();
+
+ return;
+}
+
+main();
--
2.39.5
_______________________________________________
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
next prev parent reply other threads:[~2024-12-20 18:53 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-12-20 18:51 [pve-devel] [PATCH v2 pve-common 00/12] Introduce and Package PVE::Path & PVE::Filesystem Max Carrara
2024-12-20 18:51 ` [pve-devel] [PATCH v2 pve-common 01/12] introduce PVE::Path Max Carrara
2024-12-20 18:51 ` [pve-devel] [PATCH v2 pve-common 02/12] test: add directory for tests of PVE::Path module Max Carrara
2024-12-20 18:51 ` [pve-devel] [PATCH v2 pve-common 03/12] test: add tests for path_is_absolute and path_is_relative of PVE::Path Max Carrara
2024-12-20 18:51 ` [pve-devel] [PATCH v2 pve-common 04/12] test: add tests for path_components " Max Carrara
2024-12-20 18:52 ` [pve-devel] [PATCH v2 pve-common 05/12] test: add tests for path_join " Max Carrara
2024-12-20 18:52 ` [pve-devel] [PATCH v2 pve-common 06/12] test: add tests for path_push " Max Carrara
2024-12-20 18:52 ` Max Carrara [this message]
2024-12-20 18:52 ` [pve-devel] [PATCH v2 pve-common 08/12] test: add tests for path_starts_with, path_ends_with, path_equals Max Carrara
2024-12-20 18:52 ` [pve-devel] [PATCH v2 pve-common 09/12] test: add tests for file path ops functions of PVE::Path Max Carrara
2024-12-20 18:52 ` [pve-devel] [PATCH v2 pve-common 10/12] test: add tests for path_normalize " Max Carrara
2024-12-20 18:52 ` [pve-devel] [PATCH v2 pve-common 11/12] introduce PVE::Filesystem Max Carrara
2024-12-20 18:52 ` [pve-devel] [PATCH v2 pve-common 12/12] debian: introduce package libproxmox-fs-path-utils-perl Max Carrara
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20241220185207.519912-8-m.carrara@proxmox.com \
--to=m.carrara@proxmox.com \
--cc=pve-devel@lists.proxmox.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox