sourcerer-io / sourcerer-app Goto Github PK

[e] No line at 35; commit: 6a0a139cdabe9ad24f2dcee3044bb62f8738df15; 'Upgrade to Normalize v3-rc1 in prep for v3 final': Index: 35, Size: 35
java.lang.IndexOutOfBoundsException: Index: 35, Size: 35

I do Functional programming

I have [count of lines of comments] per 100 lines of code.

[Underscores|CamelCase] are my most used file naming convention

My average line of code is [average count of characters in a line of code]

My average function has [count of lines of code] lines of code

My TODOs live on average X days

I typically use [inline|multi-line] braces

My preferred build system is [build system user uses].

from readme

The most used library besides STL is Boost.

Persiyanov commit to tensorflow repo (Commit link) has 39 additions and 22 deletions and Python code only.
But APP finds 223427 lines added and 32053 deleted. Also it finds 3036 CPP, 5k tensorflow and so on.
Also this commit leads to 2 'tensorflow' tech entities in commit_stats table.
Anatoly supposes that the reason of such behavior is problem in JGit.

Here's the timing of CommitCrawler:
is file binary check takes ~20%
get file content (old and new versions): ~40%
toFileHeader: ~25%
Old and new versions of a file are computed for each commit, which is an excess work, since old eventually becomes a new one. If a file content was stored, then it will make 20% faster.
Also it's worth to check whether is_binary check is any faster if we feed a file content instead of stream. I tested on a repo that has a low binary files ratio: 30 vs 2415, and it took 20% of time. I suppose binary check may load a whole file.

I use inline assembly in my code.

When I run sourcerer command, can not find repos I have because path is relative from where I run the 'add' command. So, to update all repos I need to run sourcerer command from all paths I used 'add' command.

I write, on average, # functions per file.

My code to empty line ratio is X

Kotlin can be compiled to native binary. Maybe we could do that?
OR
Java can be bundled. Maybe we could do that?

I contribute to open source projects

This task is about parsing Kotlin tech stats

Guys, let's go through the customary github process and make an actual release: https://github.com/sourcerer-io/sourcerer-app/releases

In the console output, whenever we print an error that should not happen, let's include a link to opening a new issue in this repo, e..g

[e] Unhandled exception. Please file an issue at https://github.com/sourcerer-io/sourcerer-app/issues/new and include this console output as well as some account of what happened. Thank you!

Example errors:
Hashing error: Missing unknown b3de5226c057d0588316c8bb0450990605e035d1
https://github.com/persiyanov/AgentNet
Error while hashing: Missing unknown fb714b3606bd663b823f6960a73d052f97283b74
https://github.com/persiyanov/temporal-topic-models-bigartm/
Fixes for CommitCrawler for the same errors: commit
Dirty fixes for LongevityHasher: commit @asurkov , take a look

I make a lot of commits with small changes

I like [CamelCase|SnakeCase]

I like list comprehensions (python)

I use template metaprogramming.

The bash script that gets downloaded to ~/.sourcerer/sourcerer is missing the #!/usr/bin/env bash header.

So when I run sourcerer in my terminal I get the following error message:

Failed to execute process '/usr/local/bin/sourcerer'. Reason:
exec: Exec format error
The file '/usr/local/bin/sourcerer' is marked as an executable but could not be run by the operating system.

I like to write multithreaded programs.

I delete a lot of code

When entering a repository path, users should be able to hit TAB for autocomplete, like in bash.

I do a lot of [CSS2|CSS3|CSS4]

The oldest file in my repository that I edited recently is [number of years, months] years old.

I primarily program in [language most used].

My average file is [count of lines used in a file] lines long

On average I have [count of commits] per file.

I mostly use Spaces overs Tabs for indentation
or
I mostly use Tabs overs Spaces for indentation

I use a lot of [Google|Facebook|Twitter] open source libs

[number of files] files account for 80% of my commits in the past 3 months.

We count number of files in commits, and then say 20 files account for 80% of my commits in last three months.

In one day I have been able to remove # lines of code!

On average, my commits are [count of lines] lines long.

I follow up on TODOs

The longest name of my classes is [count of characters] characters long.

It would be great to have a contribution document, something like this. This would help with upcoming PRs/issues from the community.

I frequently work with a team of [count of team members in most frequently used repo]

There's a tiny error - all longevity is collected for a person who removed a line, author earns nothing. So it's easy to swim if another holds up your head или на чужом горбу в рай ехать.

I typically indent my code blocks with [count of spaces or tabs used to indent code blocks]

I use [tool X] to autogenerate documentation for me