doc/FAQ.txt

Mercurial Frequently Asked Questions
====================================

Section 1: General Usage
------------------------

.Q. I did an "hg pull" and my working directory is empty!

There are two parts to Mercurial: the repository and the working
directory. "hg pull" pulls all new changes from a remote repository
into the local one but doesn't alter the working directory.

This keeps you from upsetting your work in progress, which may not be
ready to merge with the new changes you've pulled and also allows you
to manage merging more easily (see below about best practices).

To update your working directory, run "hg update". If you're sure you
want to update your working directory on a pull, you can also use "hg
pull -u". This will refuse to merge or overwrite local changes.


.Q. What are revision numbers, changeset IDs, and tags?

Mercurial will generally allow you to refer to a revision in three
ways: by revision number, by changeset ID, and by tag.

A revision number is a simple decimal number that corresponds with the
ordering of commits in the local repository. It is important to
understand that this ordering can change from machine to machine due
to Mercurial's distributed, decentralized architecture.

This is where changeset IDs come in. A changeset ID is a 160-bit
identifier that uniquely describes a changeset and its position in the
change history, regardless of which machine it's on. This is
represented to the user as a 40 digit hexadecimal number. As that
tends to be unwieldy, Mercurial will accept any unambiguous substring
of that number when specifying versions. It will also generally print
these numbers in "short form", which is the first 12 digits.

You should always use some form of changeset ID rather than the local
revision number when discussing revisions with other Mercurial users
as they may have different revision numbering on their system.

Finally, a tag is an arbitrary string that has been assigned a
correspondence to a changeset ID. This lets you refer to revisions
symbolically.


.Q. What are branches, heads, and the tip?

The central concept of Mercurial is branching. A 'branch' is simply
an independent line of development. In most other version control
systems, all users generally commit to the same line of development
called 'the trunk' or 'the main branch'. In Mercurial, every developer
effectively works on a private branch and there is no internal concept
of 'the main branch'.

Thus Mercurial works hard to make repeated merging between branches
easy. Simply run "hg pull" and "hg update -m" and commit the result.

'Heads' are simply the most recent commits on a branch. Technically,
they are changesets which have no children. Merging is the process of
joining points on two branches into one, usually at their current
heads. Use "hg heads" to find the heads in the current repository.

The 'tip' is the most recently changed head, and also the highest
numbered revision. If you have just made a commit, that commit will be
the head. Alternately, if you have just pulled from another
repository, the tip of that repository becomes the current tip.

The 'tip' is the default revision for many commands such as update,
and also functions as a special symbolic tag.


.Q. How does merging work?

The merge process is simple. Usually you will want to merge the tip
into your working directory. Thus you run "hg update -m" and Mercurial
will incorporate the changes from tip into your local changes.

The first step of this process is tracing back through the history of
changesets and finding the 'common ancestor' of the two versions that
are being merged. This is done on a project-wide and a file by file
basis.

For files that have been changed in both projects, a three-way merge
is attempted to add the changes made remotely into the changes made
locally. If there are conflicts between these changes, the user is
prompted to interactively resolve them.

Mercurial uses a helper tool for this, which is usually found by the
hgmerge script. Example tools include tkdiff, kdiff3, and the classic
RCS merge.

After you've completed the merge and you're satisfied that the results
are correct, it's a good idea to commit your changes. Mercurial won't
allow you to perform another merge until you've done this commit as
that would lose important history that will be needed for future
merges.


.Q. How do tags work in Mercurial?

Tags work slightly differently in Mercurial than most revision
systems. The design attempts to meet the following requirements:

- be version controlled and mergeable just like any other file
- allow signing of tags
- allow adding a tag to an already committed changeset
- allow changing tags in the future

Thus Mercurial stores tags as a file in the working dir. This file is
called .hgtags and consists of a list of changeset IDs and their
corresponding tags. To add a tag to the system, simply add a line to
this file and then commit it for it to take effect. The "hg tag"
command will do this for you and "hg tags" will show the currently
effective tags.

Note that because tags refer to changeset IDs and the changeset ID is
effectively the sum of all the contents of the repository for that
change, it is impossible in Mercurial to simultaneously commit and add
a tag. Thus tagging a revision must be done as a second step.


.Q. What if I want to just keep local tags?

You can add a section called "[tags]" to your .hg/hgrc which contains
a list of tag = changeset ID pairs. Unlike traditional tags, these are
only visible in the local repository, but otherwise act just like
normal tags.


.Q. How do tags work with multiple heads?

The tags that are in effect at any given time are the tags specified
in each head, with heads closer to the tip taking precedence. Local
tags override all other tags.


.Q. What are some best practices for distributed development with Mercurial?

First, merge often! This makes merging easier for everyone and you
find out about conflicts (which are often rooted in incompatible
design decisions) earlier.

Second, don't hesitate to use multiple trees locally. Mercurial makes
this fast and light-weight. Typical usage is to have an incoming tree,
an outgoing tree, and a separate tree for each area being worked on.

The incoming tree is best maintained as a pristine copy of the
upstream repository. This works as a cache so that you don't have to
pull multiple copies over the network. No need to check files out here
as you won't be changing them.

The outgoing tree contains all the changes you intend for merger into
upsteam. Publish this tree with 'hg serve" or hgweb.cgi or use 'hg
push" to push it to another publicly availabe repository.

Then, for each feature you work on, create a new tree. Commit early
and commit often, merge with incoming regularly, and once you're
satisfied with your feature, pull the changes into your outgoing tree.


.Q. How do I import from a repository created in a different SCM?

Take a look at contrib/convert-repo. This is an extensible
framework for converting between repository types.


.Q. What about Windows support?

Patches to support Windows are being actively integrated, a fully
working Windows version is probably not far off


Section 2: Technical
--------------------

.Q. What limits does Mercurial have?

Mercurial currently assumes that single files, indices, and manifests
can fit in memory for efficiency.

Offsets in revlogs are currently tracked with 32 bits, so a revlog for
a single file can currently not grow beyond 4G.

There should otherwise be no limits on file name length, file size,
file contents, number of files, or number of revisions.

The network protocol is big-endian.

File names cannot contain the null character. Committer addresses
cannot contain newlines.

Mercurial is primarily developed for UNIX systems, so some UNIXisms
may be present in ports.


.Q. How does Mercurial store its data?

The fundamental storage type in Mercurial is a "revlog". A revlog is
the set of all revisions of a named object. Each revision is either
stored compressed in its entirety or as a compressed binary delta
against the previous version. The decision of when to store a full
version is made based on how much data would be needed to reconstruct
the file. This lets us ensure that we never need to read huge amounts
of data to reconstruct a object, regardless of how many revisions of it
we store.

In fact, we should always be able to do it with a single read,
provided we know when and where to read. This is where the index comes
in. Each revlog has an index containing a special hash (nodeid) of the
text, hashes for its parents, and where and how much of the revlog
data we need to read to reconstruct it. Thus, with one read of the
index and one read of the data, we can reconstruct any version in time
proportional to the object size.

Similarly, revlogs and their indices are append-only. This means that
adding a new version is also O(1) seeks.

Revlogs are used to represent all revisions of files, manifests, and
changesets. Compression for typical objects with lots of revisions can
range from 100 to 1 for things like project makefiles to over 2000 to
1 for objects like the manifest.


.Q. How are manifests and changesets stored?

A manifest is simply a list of all files in a given revision of a
project along with the nodeids of the corresponding file revisions. So
grabbing a given version of the project means simply looking up its
manifest and reconstruction all the file revisions pointed to by it.

A changeset is a list of all files changed in a check-in along with a
change description and some metadata like user and date. It also
contains a nodeid to the relevent revision of the manifest.


.Q. How do Mercurial hashes get calculated?

Mercurial hashes both the contents of an object and the hash of its
parents to create an identifier that uniquely identifies an object's
contents and history. This greatly simplifies merging of histories
because it avoid graph cycles that can occur when a object is reverted
to an earlier state.

All file revisions have an associated hash value. These are listed in
the manifest of a given project revision, and the manifest hash is
listed in the changeset. The changeset hash is again a hash of the
changeset contents and its parents, so it uniquely identifies the
entire history of the project to that point.


.Q. What checks are there on repository integrity?

Every time a revlog object is retrieved, it is checked against its
hash for integrity. It is also incidentally doublechecked by the
Adler32 checksum used by the underlying zlib compression.

Running 'hg verify' decompresses and reconstitutes each revision of
each object in the repository and cross-checks all of the index
metadata with those contents.

But this alone is not enough to ensure that someone hasn't tampered
with a repository. For that, you need cryptographic signing.


.Q. How does signing work with Mercurial?

Take a look at the hgeditor script for an example. The basic idea is
to use GPG to sign the manifest ID inside that changelog entry. The
manifest ID is a recursive hash of all of the files in the system and
their complete history, and thus signing the manifest hash signs the
entire project contents.


.Q. What about hash collisions? What about weaknesses in SHA1?

The SHA1 hashes are large enough that the odds of accidental hash collision
are negligible for projects that could be handled by the human race.
The known weaknesses in SHA1 are currently still not practical to
attack, and Mercurial will switch to SHA256 hashing before that
becomes a realistic concern.

Collisions with the "short hashes" are not a concern as they're always
checked for ambiguity and are still long enough that they're not
likely to happen for reasonably-sized projects (< 1M changes).