Unpacking sets and ranges from a single string

I was reading the cURL man page on Thursday. Although I didn’t find what I was looking for, I did come across some rather neat syntax for specifying multiple URLs in a single string. Here’s an example:

You can specify multiple URLs or parts of URLs by writing part sets within braces as in:

http://site.**{one,two,three}**.com

or you can get sequences of alphanumeric series by using [] as in:

ftp://ftp.numericals.com/file**[1-100]**.txt
ftp://ftp.numericals.com/file**[001-100]**.txt (with leading zeros)
ftp://ftp.letters.com/file**[a-z]**.txt

The man page has more examples.

I’ve been thinking about something like this for a while. It’s not hard to create these lists using loops in Python, but it’s a bit tedious. It takes time to set up the correct loops, and for anything complicated you end up with an unhealthy level of nesting. I’d rather solve this problem once, and then not have to worry again.

I had a slightly different syntax in mind (simpler, and with more double brackets), but since I had a finished spec in the form of the examples, I decided to use this instead. I’ve written a Python script which takes a single string, and returns all the strings it specifies as an iterator.

It supports the same syntax as the cURL counterpart. Here’s a simple example:

>>> urls = curlparser.parse_string("http://site.{one,two,three}.com")
>>> urls.next()
"http://site.one.com"
>>> urls.next()
"http://site.two.com"
>>> urls.next()
"http://site.three.com"

It can also be invoked from the shell:

$python curlparser.py "ftp://ftp.numericals.com/file[1-5].txt"
ftp://ftp.numericals.com/file1.txt
ftp://ftp.numericals.com/file2.txt
ftp://ftp.numericals.com/file3.txt
ftp://ftp.numericals.com/file4.txt
ftp://ftp.numericals.com/file5.txt

I have this bound to a TextExpander snippet, with a fill-in field for the string.

The script itself is in a Gist, or you can download it directly. I’ve also got a list of examples from the cURL man page.

When I originally thought about this problem, I only had one use-case in mind. It’s another project, which has a database with lots of similar strings. I can save on disk space by encoding strings in the database with this compact notation, and only getting the full list when I actually need the strings.

(At the point of adding an entry to the database, I have a compact string, not the full list. Turning the string into the list is easy. The other way looks much harder. Taking an arbitrary list of strings and finding a set of compact strings which specify exactly that list could be an interesting problem, but it’s not one I have time to investigate.)

Now that I have the script, I’m seeing lots of new use cases. I’ll still be falling back to hand-coding anything particularly complicated, but this seems like a good alternative for anything simple.