# Split() won't Split? **URL:** \ **Category:** Ask the Community\ **Created:** [January 19, 2020, 11:19pm UTC](https://connect.superhuman.com/t/split-wont-split/13351 "2020-01-19T23:19:45Z")\ **Posts on this page:** 11\ **Page:** 1
**Author:** ![Johg\_Ananda](https://yyz1.discourse-cdn.com/coda1/user_avatar/connect.superhuman.com/johg_ananda/32/13930_2.png) [@Johg\_Ananda](https://connect.superhuman.com/u/Johg_Ananda)\ **Post date:** [January 19, 2020, 11:19pm UTC](https://connect.superhuman.com/t/split-wont-split/13351/1 "2020-01-19T23:19:45Z")
I want to take some URLs and chop off all the tracking. I noticed in the URLs the mess comes after the ? > **[Daemon](https://www.audible.com/pd/Daemon-Audiobook/B002V5B9SO?pf_rd_p=6a5ce8e4-798e-4a64-8bc5-71dcf66d673f&pf_rd_r=1ZDSSVFBR7B6F32RF5KD&ref=a_lib_c4_libItem_B002V5B9SO)** > > Check out this great listen on Audible.com. Daniel Suarez's New York Times best-selling debut high-tech thriller is “so frightening even the government has taken note” (Entertainment Weekly). > Daemons: computer programs that silently run in the... To pull up the book I only need: [https://www.audible.com/pd/Daemon-Audiobook/B002V5B9SO](https://www.audible.com/pd/Daemon-Audiobook/B002V5B9SO) Me thought it clever to `split()` the array and then take `first()`. For some reason it won’t split! After question mark wouldn’t work I thought maybe ‘/’ - see below. Thoughts? ![Screen Shot 2020-01-19 at 3.16.54 PM](https://global.discourse-cdn.com/coda1/original/2X/1/14dce09e53d8ebc4e63386dbf8c9520196317db4.png) ---
**Author:** ![Federico\_Stefanato](https://yyz1.discourse-cdn.com/coda1/user_avatar/connect.superhuman.com/federico_stefanato/32/10092_2.png) [@Federico\_Stefanato](https://connect.superhuman.com/u/Federico_Stefanato)\ **Post date:** [January 19, 2020, 11:35pm UTC](https://connect.superhuman.com/t/split-wont-split/13351/2 "2020-01-19T23:35:01Z")
Hi @Johg_Ananda, I’m not able to reproduce it… What is the actual content of the cell? Have you tried with `Link.toText().Split("/")`? ---
**Author:** ![Johg\_Ananda](https://yyz1.discourse-cdn.com/coda1/user_avatar/connect.superhuman.com/johg_ananda/32/13930_2.png) [@Johg\_Ananda](https://connect.superhuman.com/u/Johg_Ananda)\ **Post date:** [January 19, 2020, 11:39pm UTC](https://connect.superhuman.com/t/split-wont-split/13351/3 "2020-01-19T23:39:57Z")
@Federico_Stefanato !!! yes it needed to be converted to text. Although I pasted it in as text, Coda did an end run and transformed it into not-text. Thank you! 🙏 ---
**Author:** ![Federico\_Stefanato](https://yyz1.discourse-cdn.com/coda1/user_avatar/connect.superhuman.com/federico_stefanato/32/10092_2.png) [@Federico\_Stefanato](https://connect.superhuman.com/u/Federico_Stefanato)\ **Post date:** [January 19, 2020, 11:47pm UTC](https://connect.superhuman.com/t/split-wont-split/13351/4 "2020-01-19T23:47:46Z")
Great! Even though if you need to take out the query parameters, maybe it’s easier go with [Left()](https://coda.io/formulas#Left) and [Find()](https://coda.io/formulas#Find) functions searching for the “`?`”. Unfortunately, [RegexReplace()](https://coda.io/formulas#RegexReplace) cannot work with groups (please, somebody tell me otherwise); it would have been a good candidate. Cheers! ---
**Author:** ![Johg\_Ananda](https://yyz1.discourse-cdn.com/coda1/user_avatar/connect.superhuman.com/johg_ananda/32/13930_2.png) [@Johg\_Ananda](https://connect.superhuman.com/u/Johg_Ananda)\ **Post date:** [January 20, 2020, 12:14am UTC](https://connect.superhuman.com/t/split-wont-split/13351/5 "2020-01-20T00:14:07Z")
Any reason you think `left()` + `find()` are better than `split()` & `first()`? ---
**Author:** ![Federico\_Stefanato](https://yyz1.discourse-cdn.com/coda1/user_avatar/connect.superhuman.com/federico_stefanato/32/10092_2.png) [@Federico\_Stefanato](https://connect.superhuman.com/u/Federico_Stefanato)\ **Post date:** [January 20, 2020, 7:21am UTC](https://connect.superhuman.com/t/split-wont-split/13351/6 "2020-01-20T07:21:40Z")
Well, just common sense. I guess that splitting an URL in slashes and taking the first occurrence would return “`http`:” 😬 You’re not searching for “/” but for “?”, as far as I understood. If you need a string _before_ a certain string/character, it is usually better search for the occurrence and cut it after. Does it make sense? Cheers! ---
**Author:** ![Federico\_Stefanato](https://yyz1.discourse-cdn.com/coda1/user_avatar/connect.superhuman.com/federico_stefanato/32/10092_2.png) [@Federico\_Stefanato](https://connect.superhuman.com/u/Federico_Stefanato)\ **Post date:** [January 20, 2020, 7:54am UTC](https://connect.superhuman.com/t/split-wont-split/13351/7 "2020-01-20T07:54:51Z")
Just realised that perhaps you intended splitting by “`?`”. Which does work 🙃 ---
**Author:** ![Paul\_Danyliuk](https://yyz1.discourse-cdn.com/coda1/user_avatar/connect.superhuman.com/paul_danyliuk/32/27228_2.png) [@Paul\_Danyliuk](https://connect.superhuman.com/u/Paul_Danyliuk)\ **Post date:** [January 20, 2020, 1:45pm UTC](https://connect.superhuman.com/t/split-wont-split/13351/8 "2020-01-20T13:45:09Z")
> [@Federico\_Stefanato](#): > > Unfortunately, [RegexReplace()](https://coda.io/formulas#RegexReplace) cannot work with groups (please, somebody tell me otherwise) Ahem 🙂 ![image](https://global.discourse-cdn.com/coda1/original/2X/f/f75d19ee51e618f8181440cf05c41837da9b7033.png) ---
**Author:** ![Federico\_Stefanato](https://yyz1.discourse-cdn.com/coda1/user_avatar/connect.superhuman.com/federico_stefanato/32/10092_2.png) [@Federico\_Stefanato](https://connect.superhuman.com/u/Federico_Stefanato)\ **Post date:** [January 20, 2020, 1:49pm UTC](https://connect.superhuman.com/t/split-wont-split/13351/9 "2020-01-20T13:49:36Z")
> [@Paul\_Danyliuk](#): > > image775×119 9.32 KB Well, thank **somebody** , then! 😄 Thanks a lot @Paul_Danyliuk: I actually never dug into that. Happy to know it does work ---
**Author:** ![Paul\_Danyliuk](https://yyz1.discourse-cdn.com/coda1/user_avatar/connect.superhuman.com/paul_danyliuk/32/27228_2.png) [@Paul\_Danyliuk](https://connect.superhuman.com/u/Paul_Danyliuk)\ **Post date:** [January 20, 2020, 1:54pm UTC](https://connect.superhuman.com/t/split-wont-split/13351/10 "2020-01-20T13:54:15Z")
I get the confusion though, the docs sample clearly specifies `$` symbol in the replacement as simply part of the literal `$2000`. I guess it was just a hunch for me to try `\1`, `%1`, `$1` — I never remember which that is in this or that language. ---
**Author:** ![Federico\_Stefanato](https://yyz1.discourse-cdn.com/coda1/user_avatar/connect.superhuman.com/federico_stefanato/32/10092_2.png) [@Federico\_Stefanato](https://connect.superhuman.com/u/Federico_Stefanato)\ **Post date:** [January 20, 2020, 1:57pm UTC](https://connect.superhuman.com/t/split-wont-split/13351/11 "2020-01-20T13:57:09Z")
Exactly: I just read the formula specs and simply interpolated my understanding from that. @Codans: Maybe update the example with @Paul_Danyliuk’s one would help to better see the potential.