Get started using mercurial without cloning

Harry Putnam reader at newsguy.com
Wed Apr 7 15:42:27 CDT 2010


TK Soh <teekaysoh at gmail.com> writes:

[...]

>>> >> If you want all of that structure in one repository, simply cd to the
>>> >> root folder, then run:
>>> >>
>>> >> hg init
>>> >> hg add
>>> >> hg commit -m "initial commit"

[...]

>> That'd be under the "Starting A New Project" section of Chapter 2.
>
> Somehow I feel that this section should appear earlier in the chapter.
> Perhaps it should be moved to somewhere closely after the "Making a local
> copy of a repository" section.

I agree.

I don't know if any changes are being taken in that online book, but
you'd think it would be right close to the main `Getting started'
header in what people are calling chapter 2.

Should be right there where cloning is shown as a way to get started.

In fact the init command is actually explained right where I'm fussing
about, by way of showing how to use `hg help <cmd>'

I guess it just wasn't that clear to me that it was being suggested as
a way to `get started'.  But also a hold over from `cvs' background
where initializing a repo only creates `cvsroot' with admin files.
And does nothing to populate it.  That requires adding files somehow.
Often with `cvs import'.

Showing the output of `hg help init' is NOT the same thing as saying.

"One way to start a new repo from a collection of files is with the
`hg init' command . . ."

I'm not sure how others read, but once I saw it was a help command I
blew right past the content of the green box, where it says:

   create a new repository in the given directory

So it may just be very sloppy reading habits on my part, but I have a
hunch lots of people skip along looking for the startup stuff.

Its a small suggestion. 

The book is very helpful as is.



More information about the Mercurial mailing list