yeshuo | 2023-09-01 19:02:01 UTC | #1

import re
match = re.match('[/:](.*)[/:](.*)[/:](.*)','/usr/home/:lumberjack')
match.groups()
('usr/home', '', 'lumberjack')

Why is the result not ('usr','home','lumberjack')?

-------------------------

MRAB | 2023-09-01 19:58:34 UTC | #2

In order to preserve formatting, please select any code or traceback that you post and then click the `</>` button.

```text
[/:] matches '/'
(.*) wants to match as much as it can, but the later 2 occurences of [/:] require the '/:', so it's forced to match no more than 'usr/home'
[/:] matches '/'
(.*) wants to match as much as it can, but the later occurence of [/:] requires the ':', so it's forced to match ''
[/:] matches ':'
(.*) matches 'lumberjack'
```

-------------------------

rob42 | 2023-09-01 20:34:15 UTC | #3

As an aside, from the thread title: if you're trying to construct a file path, which, on the face of it, is what it looks as if you're trying to do, there's a bunch of ready made tools that you can take advantage of, such as, from `pathlib`, `Path` and  `PurePath`, to name but two.

-------------------------

hansgeunsmeyer | 2023-09-02 12:55:19 UTC | #4

[quote="yeshuo chen, post:1, topic:32768, username:yeshuo"]
import re

match = re.match('/:/:/:’, ‘/usr/home/:lumberjack')
match.groups()
[/quote]
Does not result in ```('usr', 'home', 'lumberjack')``` since the regex does not match the string. The `re.match` method only returns matches if the regex matches the whole string. So, I'm guessing you have some copy paste errors there. What are you trying to find out?

-------------------------

MRAB | 2023-09-02 14:55:51 UTC | #5

It looks wrong because the OP didn't go through the steps to preserve the formatting.

You can see what was actually posted by looking at the underlying raw form:

[discuss.python.org/raw/32768](https://discuss.python.org/raw/32768)

When properly formatted, the code looks like this:

```
import re
match = re.match('[/:](.*)[/:](.*)[/:](.*)','/usr/home/:lumberjack')
match.groups()
('usr/home', '', 'lumberjack')
```

-------------------------

hansgeunsmeyer | 2023-09-02 15:13:49 UTC | #6

Thanks - I didn't realize that! :slight_smile:

-------------------------

gwerbin | 2023-09-02 16:20:33 UTC | #7

One of the things I like about https://regex101.com is that it automatically provides this kind of step-by-step explanation of how your pattern works.

-------------------------

pochmann | 2023-09-02 23:31:43 UTC | #8

[quote="Matthew Barnett, post:5, topic:32768, username:MRAB"]
[discuss.python.org/raw/32768](https://discuss.python.org/raw/32768)
[/quote]

How did you get that?

-------------------------

MRAB | 2023-09-02 23:45:27 UTC | #9

Someone in another post said how to get the raw text.

The URL of this thread is:

`https://discuss.python.org/t/why-is-the-output-different-from-what-i-thought/32768`

so the URL of the raw text is:

`https://discuss.python.org/raw/32768`

Useful when a post's format looks messed up.

-------------------------

pochmann | 2023-09-03 00:09:04 UTC | #10

Hopefully we won't need it again until post 65536. (I'm in the app, so didn't immediately see the normal URL, and the special number threw me off)

-------------------------