ls-git | A happy union of ls and git status | Command Line Interface library

A colleague and I have a big Docker puzzle.

When we run the following commands we get different results.

I want to make use of a promising NN I found at towardsdatascience for my case study.

The data shapes I have are:

I want to develop and run a ruby on rails web app on my Bootcamp windows Mac laptop. I am using rubyinstaller and railsinstaller, but the installation breaks always at some PGP key check which I have no idea about.

Facing issue while trying to install ruby installer

I am trying to deploy a code of mine (that uses fbprophet) as an EC2 instance using Gitlab CI.

my requirements.txt file:

I read this Git Internals and a few things about tree-object are not very clear.

For intance, the current branch is master. The final commit object on master is an empty commit and is denoted by cmt_end. Given by git-commit-tree, a commit object has only one tree object. Hence, the tree object of cmt_end is denoted by tree_end.

My confusion is:

• Given by git-commit, git-commit records changes in repository. Does this indicates that tree_end contains the differences between cmt_end and its parent? If so, tree_end should be an empty tree since cmt_end is an empty commit. If not, is it ok to have an empty tree for the empty commit?
• the --allow-empty option in git-commit, says that an empty commit has the exact same tree as its sole parent commit. Does this indicates that the tree object of a commit records the whole working directory instead of changes when doing git-commit?

If I compare the git/refs when creating a new branch from last commit of a branch versus creating from a detached head state, the git/refs shows different behavior (names) and I like to understand why.

Lets assume the head is on master.

I need a new dev branch. After that, I made some commits to the new branch and some fixes on master.

Now I see another way how to solve the dev branch and I made a second new branch and some commits on that and on master again some fixes.

But now the refs look different! As long as I don't merge plus delete or only delete the second branch still the first two refs called after that second branch.

Why is this a good behavior, cause I cannot easily see that the first refs usually belong to master? Only hashes and parent hashes help with identification, but are a slow method. THX.

EDIT: The information shown here is provided from git log command.

I'm building my own implementation of Git, and have reached a point where my output (and that of git itself) has diverged from Pro Git's.

Specifically, right now I'm following through the construction at https://git-scm.com/book/en/v2/Git-Internals-Git-Objects to build a commit. But I can't seem to get the same commit hash out, even after fixing the entire contents of the commit data. I've written a little program to write out and read in commits, and that program is able to round-trip the commit data I've specified, but it gets the "wrong" commit hash. That "wrong" commit hash is consistent with my local git, but not with Pro Git's printed output.

Reproducing this in zsh for convenience (I get exactly the same commit hash as git here does, when I use my own implementation of Git):

A question from a beginner: there are several ways to search in git history. To me, it is not clear what, the difference between the commands really is:

1. git log, e.g. git log -S 'for what you search' --all to search for a string in all branches and git log -G 'search for regexpr' --all to search for regexpr. See for example here
2. git log --grep 'pattern'
3. git rev-list, e.g. git rev-list --grep='string' --all
4. git grep, e.g. git grep "string or regexpr" $(git rev-list --all) from here
5. gitk, e.g. gitk file.txt a gui interface
6. git log --all --full-history -- **/thefile.* to search for a filename in history from here.
7. git diff --word-diff-regex=. > changes.txt combined with grep -oP '\+[^\+]+\+' changes.txt | tr -d '\+' to display changes on a character basis. From here. This even works if changes are not commited yet.

My poor understanding is that

• git log searches in the commit messages? It seems to be based on rev-list? From the docs:

Shows the commit logs. The command takes options applicable to the git rev-list command to control what is shown and how, and options applicable to the git diff-* commands to control how the changes each commit introduces are shown.

• git-rev-list seems to be a more basic command as it operates on the commit object (or tree object?). From the docs:

git-rev-list - Lists commit objects in reverse chronological order

• git grep seems to search everywhere: in files, blobs, trees and commit objects? From the docs:

git-grep - Print lines matching a pattern. Look for specified patterns in the tracked files in the work tree, blobs registered in the index file, or blobs in given tree objects. Patterns are lists of one or more search expressions separated by newline characters. An empty string as search expression matches all lines.

I used the Git book as reference.

So, I have a repository and suspicions that someone accidentally commited credentials into it

I need to find all commits which contains my credentials: tiWEythzNEX8N3sktiWEythzNEX8N3sk

Let's see some commits. There is small repo to reproduce

Use command git log --oneline -n 10