Test::File(3)

📦

version 1.993 · 8 versions

1.993 · opensuse-leap 16.0 (latest)
1.99.3 · epel 10
1.995 · arch rolling
1.99.5 · fedora 43
1.992 · ubuntu jammy
1.99.5 · fedora 44
1.993 · debian bookworm, ubuntu noble
1.995 · debian trixie, ubuntu resolute, ubuntu stonking

okay so bestie, welcome to Test::File(3) — the manual page for the module that is literally all about testing file attributes, and no cap it is giving useful. 🎀

✨ NAME ✨

Test::File — test file attributes. That is the whole vibe. 💖

🌸 SYNOPSIS 🌸

Like, the tiniest setup ever:

use Test::File;

💖 DESCRIPTION 💖

This module is a whole collection of test utilities for file attributes — like, a little toolkit for checking what files are up to. 🧚

Some file attributes depend on the owner of the process testing the file in the same way the file test operators do. For instance, root (or super-user or Administrator) may always be able to read files no matter the permissions. So yeah, do not forget that whole vibe — permissions are not the whole story. 💅

Some attributes do not make sense outside of Unix, either, so some tests automatically skip if they think they will not work on the platform. If you have a way to make these functions work on Windows, for instance, please send me a patch. :) If you want to pretend to be Windows on a non-Windows machine (for instance, to test skip()), you can set the PRETEND_TO_BE_WINDOWS environment variable. Iconic little hack, honestly. 🦋

The optional NAME parameter for every function allows you to specify a name for the test. If not supplied, a reasonable default will be generated. So you can name your tests, or just let the module do its thing. ✨

🌟 Functions 🌟

okay, function time — here is every single one, no skips: 💅

`has_symlinks`

Returns true if this module thinks that the current system supports symlinks. 🫶

This is not a test function. It is something that tests can use to determine what it should expect or skip. So like, use it for vibe checks, not as an assertion.

`file_exists_ok( FILENAME [, NAME ] )`

Ok if the file exists, and not ok otherwise. Literally that simple. 💖

`file_not_exists_ok( FILENAME [, NAME ] )`

Ok if the file does not exist, and not okay if it does exist. The absence is the moment. 🙈

`file_empty_ok( FILENAME [, NAME ] )`

Ok if the file exists and has empty size, not ok if the file does not exist or exists with non-zero size. 🍓

⚠️ Previously this tried to test any sort of file. Sometime in the future this will fail if the argument is not a plain file or is a directory.

`file_not_empty_ok( FILENAME [, NAME ] )`

Ok if the file exists and has non-zero size, not ok if the file does not exist or exists with zero size. 💖

⚠️ Previously this tried to test any sort of file. Sometime in the future this will fail if the argument is not a plain file or is a directory.

`file_size_ok( FILENAME, SIZE [, NAME ] )`

Ok if the file exists and has SIZE size in bytes (exactly), not ok if the file does not exist or exists with size other than SIZE. Exact exact. 💅

⚠️ Previously this tried to test any sort of file. Sometime in the future this will fail if the argument is not a plain file or is a directory.

`file_max_size_ok( FILENAME, MAX [, NAME ] )`

Ok if the file exists and has size less than or equal to MAX bytes, not ok if the file does not exist or exists with size greater than MAX bytes. 🎀

⚠️ Previously this tried to test any sort of file. Sometime in the future this will fail if the argument is not a plain file or is a directory.

`file_min_size_ok( FILENAME, MIN [, NAME ] )`

Ok if the file exists and has size greater than or equal to MIN bytes, not ok if the file does not exist or exists with size less than MIN bytes. 🌈

⚠️ Previously this tried to test any sort of file. Sometime in the future this will fail if the argument is not a plain file or is a directory.

`file_line_count_is( FILENAME, COUNT [, NAME ] )`

Ok if the file exists and has COUNT lines (exactly), not ok if the file does not exist or exists with a line count other than COUNT. 🔥

This function uses the current value of $/ as the line ending and counts the lines by reading them and counting how many it read. Yes, $/ — the input record separator — is the main character here. 💅

⚠️ Previously this tried to test any sort of file. Sometime in the future this will fail if the argument is not a plain file or is a directory.

`file_line_count_isnt( FILENAME, COUNT [, NAME ] )`

Ok if the file exists and does not have exactly COUNT lines, not ok if the file does not exist or exists with a line count of COUNT. Read that carefully: the file must exist for this test to pass! 😭

This function uses the current value of $/ as the line ending and counts the lines by reading them and counting how many it read. 🦋

⚠️ Previously this tried to test any sort of file. Sometime in the future this will fail if the argument is not a plain file or is a directory.

`file_line_count_between( FILENAME, MIN, MAX, [, NAME ] )`

Ok if the file exists and has a line count between MIN and MAX, inclusively. Inclusive, bestie — MIN and MAX both count. 🫶

This function uses the current value of $/ as the line ending and counts the lines by reading them and counting how many it read. ✨

⚠️ Previously this tried to test any sort of file. Sometime in the future this will fail if the argument is not a plain file or is a directory.

`file_contains_like ( FILENAME, PATTERN [, NAME ] )`

Ok if the file exists and its contents (as one big string) match PATTERN, not ok if the file does not exist, is not readable, or exists but does not match PATTERN. 💖

Since the file contents are read into memory, you should not use this for large files. Besides memory consumption, test diagnostics for failing tests might be difficult to decipher. However, for short files this works very well. So like, keep it cute and short. 🎀

Because the entire contents are treated as one large string, you can make a pattern that tests multiple lines. Do not forget that you may need to use the /s modifier for such patterns:

# make sure file has one or more paragraphs with CSS class X
file_contains_like($html_file, qr{<p class="X">.*?</p>}s);

Contrariwise, if you need to match at the beginning or end of a line inside the file, use the /m modifier:

# make sure file has a setting for foo
file_contains_like($config_file, qr/^ foo \s* = \s* \w+ $/mx);

If you want to test your file contents against multiple patterns, but do not want to have the file read in repeatedly, you can pass an arrayref of patterns instead of a single pattern, like so:

# make sure our template has rendered correctly
file_contains_like($template_out,
        [
        qr/^ $title_line $/mx,
        map { qr/^ $_ $/mx } @chapter_headings,
        qr/^ $footer_line $/mx,
        ]);

Please note that if you do this, and your file does not exist or is not readable, you will only get one test failure instead of a failure for each pattern. This could cause your test plan to be off, although you may not care at that point because your test failed anyway. If you do care, either skip the test plan altogether by employing Test::More's done_testing() function, or use "file_readable_ok" in conjunction with a "SKIP" block. 💀

Contributed by Buddy Burden "<barefoot@cpan.org>".

`file_contains_unlike ( FILENAME, PATTERN [, NAME ] )`

Ok if the file exists and its contents (as one big string) do not match PATTERN, not ok if the file does not exist, is not readable, or exists but matches PATTERN. 💅

All notes and caveats for "file_contains_like" apply to this function as well. So like, read the section above, bestie — it is all relevant. 🦋

Contributed by Buddy Burden "<barefoot@cpan.org>".

`file_contains_utf8_like ( FILENAME, PATTERN [, NAME ] )`

The same as "file_contains_like", except the file is opened as UTF-8. Unicode girlies rise. 🌸

`file_contains_utf8_unlike ( FILENAME, PATTERN [, NAME ] )`

The same as "file_contains_unlike", except the file is opened as UTF-8. 💜

`file_contains_encoded_like ( FILENAME, ENCODING, PATTERN [, NAME ] )`

The same as "file_contains_like", except the file is opened with ENCODING. Pick your encoding, main character. ✨

`file_contains_encoded_unlike ( FILENAME, ENCODING, PATTERN [, NAME ] )`

The same as "file_contains_unlike", except the file is opened with ENCODING. 🎀

`file_readable_ok( FILENAME [, NAME ] )`

Ok if the file exists and is readable, not ok if the file does not exist or is not readable. 👑

`file_not_readable_ok( FILENAME [, NAME ] )`

Ok if the file exists and is not readable, not ok if the file does not exist or is readable. 🙈

`file_writable_ok( FILENAME [, NAME ] )` and `file_writeable_ok( FILENAME [, NAME ] )`

Ok if the file exists and is writable, not ok if the file does not exist or is not writable. ✍️

The original name is "file_writeable_ok" with that extra e. That still works but there is a function with the correct spelling too. So both spellings are slaying, do not worry. 💅

`file_not_writeable_ok( FILENAME [, NAME ] )` and `file_not_writable_ok( FILENAME [, NAME ] )`

Ok if the file exists and is not writable, not ok if the file does not exist or is writable. 🚫

The original name is "file_not_writeable_ok" with that extra e. That still works but there is a function with the correct spelling too. Both exist, no cap. 🫶

`file_executable_ok( FILENAME [, NAME ] )`

Ok if the file exists and is executable, not ok if the file does not exist or is not executable. 🔥

This test automatically skips if it thinks it is on a Windows platform. Windows said no. 😭