[chore] folder/file naming conventions about leetcode HOT 25 CLOSED

@Ahmad-A0 understood, thank you!

do you mind if we add preceding zeros where applicable (e.g. 0001_two_sum.java)? i think it'd be really nice to make the github file sorting add some navigability (this way problems would actually be ordered by number ascending)

from leetcode.

I'd be cool with snake case. Let's wait for @neetcode-gh to drop their two pennies before we start the refactor, though!

from leetcode.

Neet's probably been getting a ton of pings on this repo, it's blown up a bit since the video about the website update. It could be a while before there's a decision, I'd recommend finding other places to contribute in the meantime!

from leetcode.

Hi, I would like to contribute to your work and was wondering if snake_case(words separated by underscores) will be a good fit for the issue, since you don't have a good feeling about spinal-case?

from leetcode.

Hi. Are there any updates? Should I proceed with the work

from leetcode.

Thanks for looking into this! Yeah im fine with having a standard naming convention, but updating the file names will break the links on the site. The links will mostly have to be updated manually, along with the file names so I'm wondering if it's worth it. If there are any PRs that change file names, please cc me on them so I know to update the links, thanks!

from leetcode.

one other thought to add to this. in some languages we're using a directory structure that maps to the categories on neetcode.io (e.g. neetcode_150/backtracking/).

we should likely decide on a single convention? In my opinion, it's easier to search w/ CMD + F if all the solutions are in one top level directory for that language (not everyone is familiar with the t shortcut on github.com). I'm not sure I see a benefit of using the nested dirs structure?

from leetcode.

I'm not sure I see a benefit of using the nested dirs structure?

In my Opinion the nested directory structure is quite compact and easy to use it really gives a brief knowledge about the domains of DSA in a glance.

from leetcode.

I also find it easier to CTRL/CMD + F the repo to find solutions. The website already does a good job of showcasing the problems sorted into their respective domains.

from leetcode.

I think a nested directory structure is more organized and less cluttered. If we need to find which directory it belongs to, we could always add a list in the Readme to indicate which category each question is under.

from leetcode.

Yep! Nested Directory Structure looks well organized and much easier to navigate around

from leetcode.

Hello ! Could I contribute to this?

from leetcode.

thanks for all of the thoughts and input, everyone!

i'm on board with snake_case and the same nested directory structure as the website (this seems like the general consensus/preference).

@Ahmad-A0 would these types of changes break the fancy new "Missing Solutions" feature in the readme (renaming files?)?

i think people are free to start work on this refactor, with these requirements:

1. tag @neetcode-gh as the reviewer on your PR. it can't merge until he's taken the time to update the website accordingly, and for that reason he should be the only person merging these PRs
2. also tag @Ahmad-A0 as a reviewer on the PR, so he can update the "Missing Solutions" script accordingly (this depends on whether or not these changes would actually break that feature, ahmad should be able to answer that for us)
3. please change at most one language + problem set (blind 75 or neetcode 150) per PR. for example, you might decide to make one PR for java/blind_75 (though you're free to make smaller PRs if you'd like). if PRs get too large it's hard to ensure to quality of review can stay high. this will also help neet and ahmad's updates be reasonably scoped
4. ensure your changes are all of the snake case form, including the appropriate subdirectory and preceding zeros, if any. e.g. java/1-Two-Sum.java would become java/neetcode_150/01_arrays_&_hashing/0001_two_sum.java

we should also update the https://github.com/neetcode-gh/leetcode/blob/main/CONTRIBUTING.md#contributing-guidelines to make sure they reflect the decided convention.

from leetcode.

Since the script uses the <problem-number> prefix it should be able to continue working after a quick fix to read nested directories, which I will likely push tomorrow.

However I wanted to suggest using the name leetcode assigns in URLs to avoid any confusion or discrepancies across solutions eg. 0242-valid-anagram.<extension> for https://leetcode.com/problems/valid-anagram/. This change could be incorporated with the nested directory structure, or we could maintain the flat directory structure and add an automated README component to organise the problems into problem type.

@mitchellirvin and others, thoughts?

from leetcode.

@Ahmad-A0 understood, thank you!

do you mind if we add preceding zeros where applicable (e.g. 0001_two_sum.java)? i think it'd be really nice to make the github file sorting add some navigability (this way problems would actually be ordered by number ascending)

Yep this is fine, preceding zeroes won't affect the script.

from leetcode.

does the name in leetcode's URLs ever differ from the name in the title of the problem? I'm not sure I understand the impact of that change. what would the before/after look like?

from leetcode.

The name in the URLs always matches the name of the problem, you can view the full list in the table script on the .github dir.

const URLS = { 
     'NeetCode 150': [ 
         'https://leetcode.com/problems/contains-duplicate/', 
         'https://leetcode.com/problems/valid-anagram/', 
         'https://leetcode.com/problems/two-sum/', 
         'https://leetcode.com/problems/group-anagrams/', 
         'https://leetcode.com/problems/top-k-frequent-elements/', 
         'https://leetcode.com/problems/product-of-array-except-self/', 
         'https://leetcode.com/problems/valid-sudoku/', 
         'https://leetcode.com/problems/encode-and-decode-strings/', 
         'https://leetcode.com/problems/longest-consecutive-sequence/', 
         'https://leetcode.com/problems/valid-palindrome/', 
         'https://leetcode.com/problems/two-sum-ii-input-array-is-sorted/', 
         'https://leetcode.com/problems/3sum/', 
         'https://leetcode.com/problems/container-with-most-water/', 
         'https://leetcode.com/problems/trapping-rain-water/', 
         'https://leetcode.com/problems/best-time-to-buy-and-sell-stock/', 
         'https://leetcode.com/problems/longest-substring-without-repeating-characters/', 
         'https://leetcode.com/problems/longest-repeating-character-replacement/', 
         'https://leetcode.com/problems/permutation-in-string/', 
         'https://leetcode.com/problems/minimum-window-substring/', 
         'https://leetcode.com/problems/sliding-window-maximum/', 
         'https://leetcode.com/problems/valid-parentheses/', 
         'https://leetcode.com/problems/min-stack/', 
         'https://leetcode.com/problems/evaluate-reverse-polish-notation/', 
         'https://leetcode.com/problems/generate-parentheses/', 
         'https://leetcode.com/problems/daily-temperatures/', 
         'https://leetcode.com/problems/car-fleet/', 
         'https://leetcode.com/problems/largest-rectangle-in-histogram/', 
         'https://leetcode.com/problems/binary-search/',

Visually the layout would look pretty similar to what we have now without the casing issues and occasional symbol usage in problem names. Something like this could also be automated since I already have the problem numbers linked to the URLs.

from leetcode.

that's a good idea.

i'd prefer to keep snake case for consistency (as opposed to spinal case), though. e.g. 0242_valid_anagram.<extension> just so we're not mixing and matching naming conventions

from leetcode.

Yep, that sounds great. I can write a script to rename the files although I don't know how practical that is until something like #763 gets done.

from leetcode.

I'm wondering if the nested directory structure is still necessary with the changes made to the README to reflect problem category, having a different folder for the <problem-set> could especially be problematic as some problems would now appear three times in the NeetCode 150, Blind 75 and NeetCode All lists.

from leetcode.

This is long overdue, but I'll have some time over the next week to implement dynamic URLs from #763, so I'm happy to review and merge any PRs related to renaming/organizing. I'll time it to merge the PRs the same time I push my changes.

The main thing I would need is to be able to know if a solution exists for a given language (e.g. LC 1 has a solution in Java). But the readme table that @Ahmad-A0 implemented pretty much solves that, but I would just need to convert it to JSON. Btw is there a script you're using for that? At first glance i couldnt find it.

This will also result in adding many more solutions & lanugages to the site!

from leetcode.

@neetcode-gh

I'm currently using the updateCompletionTable.js script hidden away in ./.github/workflows, it just grabs the problem number for each leetcode question, which I have stored in this JSON file, and sees if the language folder e.g c, cpp, python, etc. has any solutions that start with that number.

Depending on how you want to implement this on your side it might be helpful for me to add another JSON file to this repo that keeps track of which problems are completed alongside the table.

from leetcode.

Thanks, I'm almost done with the site changes required. Do you know if there's already some open PRs for doing the renaming? If not, i can do them.

from leetcode.

There are no open PRs! Unrelated but, can I also ask how files are found on the site's end so I don't break things in the future. @neetcode-gh

from leetcode.

@Ahmad-A0 Just opened #1607 - the PR has a file (.problemSiteData.json) which is the file that contains code links for the site. Lmk if you have any questions, and thank you!

from leetcode.