**Warning: Don&#39;t do this if you&#39;ve already pushed**

You want to do:

    git reset HEAD~

If you don&#39;t want the changes and blow everything away:

    git reset --hard HEAD~

In Java, we can use `System.currentTimeMillis()` to get the current timestamp in Milliseconds since epoch time which is -

&gt; the difference, measured in milliseconds, between the current time and
&gt; midnight, January 1, 1970 UTC.


In C++ how to get the same thing?

Currently I am using this to get the current timestamp - 

    struct timeval tp;
    gettimeofday(&amp;tp, NULL);
    long int ms = tp.tv_sec * 1000 + tp.tv_usec / 1000; //get current timestamp in milliseconds

    cout &lt;&lt; ms &lt;&lt; endl;

This looks right or not?

How to get current timestamp in milliseconds since 1970 just the way Java gets

Follow up question of
https://serverfault.com/questions/191331/should-servers-have-their-timezone-set-to-gmt-utc

Should the MySQL timezone be set to UTC or should it be set to be the same timezone as the server or PHP is set? (If it is not UTC)

What are the pros and cons?

Should MySQL have its timezone set to UTC?

I did the following comments

    git add /file1/path
    git rm /file/path
    git commit -m &quot;message&quot;

how do I undo my last commit using git?

Like I don&#39;t want to have those files committed.


How to undo last commit

I did the following comments
<pre><code class="hljs language-bash">git add /file1/path
git rm /file/path
git commit -m "message"
</code></pre>
how do I undo my last commit using git?
Like I don't want to have those files committed.

I&#39;m quite new to Git.

I&#39;m creating a branch and then want to push it to `origin`.

I think that simply issuing `git push` (while standing on my branch) should be sufficient.

Is it possible/reasonable to do that (by specifying push.default simple)?

git push vs git push origin &lt;branchname&gt;

My Node app is running fine locally, but has run into an error when deploying to Heroku. The app uses Sequelize in a `/models` folder, which contains `index.js`, `Company.js` and `Users.js`. Locally, I am able to import the models using the following code in `/models/index.js`:

    // load models
    var models = [
      &#39;Company&#39;,
      &#39;User&#39;
    ];
    models.forEach(function(model) {
      module.exports[model] = sequelize.import(__dirname + &#39;/&#39; + model);
    });

This works fine, however, when I deploy to Heroku the app crashes with the following error:

    Error: Cannot find module &#39;/app/models/Company&#39;
       at Function.Module._resolveFilename (module.js:338:15)
       at Function.Module._load (module.js:280:25)
       at Module.require (module.js:364:17)
       at require (module.js:380:17)
       at module.exports.Sequelize.import (/app/node_modules/sequelize/lib/sequelize.js:219:24)
       at module.exports.sequelize (/app/models/index.js:60:43)
       at Array.forEach (native)
       at Object.&lt;anonymous&gt; (/app/models/index.js:59:8)
       at Module._compile (module.js:456:26)
       at Object.Module._extensions..js (module.js:474:10)
    Process exited with status 8


Initially I thought it was due to case sensitivity (local mac vs heroku linux), but I moved the file, made a git commit, and then moved back and committed again to ensure `Company.js` is capitalized in the git repository. This didn&#39;t solve the problem and I&#39;m not sure what the issue could be.

Heroku + Node: Cannot find module error

We use [git to distribute an operating system and keep it upto date](http://webconverger.org/upgrade). We can&#39;t distribute the full repository since it&#39;s too large (&gt;2GB), so we have been using shallow clones (~300M). However [recently when fetching from a shallow clone, it&#39;s now inefficiently fetches the entire &gt;2GB repository](https://github.com/Webconverger/webc/issues/174). This is an untenable waste of bandwidth for deployments.

The git documentation says you cannot fetch from a shallow repository, though that&#39;s strictly not true. Are there any workarounds to make a `git clone --depth 1` able to fetch just what&#39;s changed from it? Or some other strategy to keep the **distribution size as small as possible** whilst having all the bits git needs to do an update?

I have unsuccessfully tried cloning from `--depth 20` to see if it will upgrade more efficiently, that didn&#39;t work. I did also look into http://git-scm.com/docs/git-bundle, but that seems to create huge bundles.

How to git fetch efficiently from a shallow clone

In the book &quot;Git: Version Control for Everyone. Beginner&#39;s Guide&quot; on page 69 there is a suggestion: &quot;As an alternative to `git pull`, we can also use `git fetch` followed by `git merge @{u}`&quot;.

What does `@{u}` mean here?

A search in Google for `git merge @{u}` provides a link to this page https://mislav.net/2013/02/merge-vs-rebase/ where `@{u}` can also be found.

What does the argument @{u} mean in Git?

I work on feature branches that have annoying big names, so I often end up grepping my current branch name when I need to push up to the feature branch from my local feature branch, like so: 

    git branch | grep &#39;*&#39; | sed &#39;s/* //&#39; | xargs git push origin

This works fine. I want to make this an alias, so I did this in `~/.gitconfig`: 

    [alias]
       pushcur = branch | grep &#39;*&#39; | sed &#39;s/* //&#39; | xargs git push origin

Now, when I run `git pushcur`, I get the following error: 

    usage: git branch [options] [-r | -a] [--merged | --no-merged]

Leading me to believe that the alias is not properly parsing the pipes. Is there something else I should do to achieve the desired alias? 

Content Type	Original Author	Original Content on Stackoverflow
Question	Kermit the Frog	View Question on Stackoverflow
Solution 1 - Git	manojlds	View Answer on Stackoverflow

How to undo last commit

Git Problem Overview

Git Solutions

Solution 1 - Git

Should MySQL have its timezone set to UTC?

How to get current timestamp in milliseconds since 1970 just the way Java gets

Attributions