Specifying Revisions

Mercurial supports several ways to specify revisions.

Specifying single revisions

A plain integer is treated as a revision number. Negative integers are treated as sequential offsets from the tip, with -1 denoting the tip, -2 denoting the revision prior to the tip, and so forth.

A 40-digit hexadecimal string is treated as a unique revision identifier. A hexadecimal string less than 40 characters long is treated as a unique revision identifier and is referred to as a short-form identifier. A short-form identifier is only valid if it is the prefix of exactly one full-length identifier.

Any other string is treated as a bookmark, tag, or branch name. A bookmark is a movable pointer to a revision. A tag is a permanent name associated with a revision. A branch name denotes the tipmost open branch head of that branch - or if they are all closed, the tipmost closed head of the branch. Bookmark, tag, and branch names must not contain the ":" character.

The reserved name "tip" always identifies the most recent revision.

The reserved name "null" indicates the null revision. This is the revision of an empty repository, and the parent of revision 0.

The reserved name "." indicates the working directory parent. If no working directory is checked out, it is equivalent to null. If an uncommitted merge is in progress, "." is the revision of the first parent.

Finally, commands that expect a single revision (like "hg update") also accept revsets (see below for details). When given a revset, they use the last revision of the revset. A few commands accept two single revisions (like "hg diff"). When given a revset, they use the first and the last revisions of the revset.

Specifying multiple revisions

Mercurial supports a functional language for selecting a set of revisions. Expressions in this language are called revsets.

The language supports a number of predicates which are joined by infix operators. Parenthesis can be used for grouping.

Identifiers such as branch names may need quoting with single or double quotes if they contain characters like "-" or if they match one of the predefined predicates.

Special characters can be used in quoted identifiers by escaping them, e.g., "\n" is interpreted as a newline. To prevent them from being interpreted, strings can be prefixed with "r", e.g. "r'...'".

Operators

There is a single prefix operator:

"not x"

Changesets not in x. Short form is "! x".

These are the supported infix operators:

"x::y"

A DAG range, meaning all changesets that are descendants of x and ancestors of y, including x and y themselves. If the first endpoint is left out, this is equivalent to "ancestors(y)", if the second is left out it is equivalent to "descendants(x)".

An alternative syntax is "x..y".

"x:y"

All changesets with revision numbers between x and y, both inclusive. Either endpoint can be left out, they default to 0 and tip.

"x and y"

The intersection of changesets in x and y. Short form is "x & y".

"x or y"

The union of changesets in x and y. There are two alternative short forms: "x | y" and "x + y".

"x - y"

Changesets in x but not in y.

"x % y"

Changesets that are ancestors of x but not ancestors of y (i.e. ::x - ::y). This is shorthand notation for "only(x, y)" (see below). The second argument is optional and, if left out, is equivalent to "only(x)".

"x^n"

The nth parent of x, n == 0, 1, or 2. For n == 0, x; for n == 1, the first parent of each changeset in x; for n == 2, the second parent of changeset in x.

"x~n"

The nth first ancestor of x; "x~0" is x; "x~3" is "x^^^". For n < 0, the nth unambiguous descendent of x.

"x ## y"

Concatenate strings and identifiers into one string.

All other prefix, infix and postfix operators have lower priority than "##". For example, "a1 ## a2~2" is equivalent to "(a1 ## a2)~2".

For example:

[revsetalias]
issue(a1) = grep(r'\bissue[ :]?' ## a1 ## r'\b|\bbug\(' ## a1 ## r'\)')

"issue(1234)" is equivalent to "grep(r'\bissue[ :]?1234\b|\bbug\(1234\)')" in this case. This matches against all of "issue 1234", "issue:1234", "issue1234" and "bug(1234)".

There is a single postfix operator:

"x^"

Equivalent to "x^1", the first parent of each changeset in x.

Patterns

Where noted, predicates that perform string matching can accept a pattern string. The pattern may be either a literal, or a regular expression. If the pattern starts with "re:", the remainder of the pattern is treated as a regular expression. Otherwise, it is treated as a literal. To match a pattern that actually starts with "re:", use the prefix "literal:".

Matching is case-sensitive, unless otherwise noted. To perform a case- insensitive match on a case-sensitive predicate, use a regular expression, prefixed with "(?i)".

For example, "tag(r're:(?i)release')" matches "release" or "RELEASE" or "Release", etc.

Predicates

The following predicates are supported:

"adds(pattern)"

Changesets that add a file matching pattern.

The pattern without explicit kind like "glob:" is expected to be relative to the current directory and match against a file or a directory.

"all()"

All changesets, the same as "0:tip".

"ancestor(*changeset)"

A greatest common ancestor of the changesets.

Accepts 0 or more changesets. Will return empty list when passed no args. Greatest common ancestor of a single changeset is that changeset.

"ancestors(set[, depth])"

Changesets that are ancestors of changesets in set, including the given changesets themselves.

If depth is specified, the result only includes changesets up to the specified generation.

"author(string)"

Alias for "user(string)".

"bisect(string)"

Changesets marked in the specified bisect status:

• "good", "bad", "skip": csets explicitly marked as good/bad/skip
• "goods", "bads" : csets topologically good/bad
• "range" : csets taking part in the bisection
• "pruned" : csets that are goods, bads or skipped
• "untested" : csets whose fate is yet unknown
• "ignored" : csets ignored due to DAG topology
• "current" : the cset currently being bisected

"bookmark([name])"

The named bookmark or all bookmarks.

Pattern matching is supported for 'name'. See 'hg help revisions.patterns'.

"branch(string or set)"

All changesets belonging to the given branch or the branches of the given changesets.

Pattern matching is supported for 'string'. See 'hg help revisions.patterns'.

"branchpoint()"

Changesets with more than one child.

"bundle()"

Changesets in the bundle.

Bundle must be specified by the -R option.

"children(set)"

Child changesets of changesets in set.

"closed()"

Changeset is closed.

"commonancestors(set)"

Changesets that are ancestors of every changeset in set.

"conflictlocal()"

The local side of the merge, if currently in an unresolved merge.

"merge" here includes merge conflicts from e.g. 'hg rebase' or 'hg graft'.

"conflictother()"

The other side of the merge, if currently in an unresolved merge.

"merge" here includes merge conflicts from e.g. 'hg rebase' or 'hg graft'.

"contains(pattern)"

The revision's manifest contains a file matching pattern (but might not modify it). See 'hg help patterns' for information about file patterns.

The pattern without explicit kind like "glob:" is expected to be relative to the current directory and match against a file exactly for efficiency.

"contentdivergent()"

Final successors of changesets with an alternative set of final successors. (EXPERIMENTAL)

"converted([id])"

Changesets converted from the given identifier in the old repository if present, or all converted changesets if no identifier is specified.

"date(interval)"

Changesets within the interval, see 'hg help dates'.

"desc(string)"

Search commit message for string. The match is case-insensitive.

Pattern matching is supported for 'string'. See 'hg help revisions.patterns'.

"descendants(set[, depth])"

Changesets which are descendants of changesets in set, including the given changesets themselves.

If depth is specified, the result only includes changesets up to the specified generation.

"destination([set])"

Changesets that were created by a graft, transplant or rebase operation, with the given revisions specified as the source. Omitting the optional set is the same as passing all().

"diffcontains(pattern)"

Search revision differences for when the pattern was added or removed.

The pattern may be a substring literal or a regular expression. See 'hg help revisions.patterns'.

"draft()"

Changeset in draft phase.

"expectsize(set[, size])"

Return the given revset if size matches the revset size. Abort if the revset doesn't expect given size. size can either be an integer range or an integer.

For example, "expectsize(0:1, 3:5)" will abort as revset size is 2 and 2 is not between 3 and 5 inclusive.

"extdata(source)"

Changesets in the specified extdata source. (EXPERIMENTAL)

"extinct()"

Obsolete changesets with obsolete descendants only. (EXPERIMENTAL)

"extra(label, [value])"

Changesets with the given label in the extra metadata, with the given optional value.

Pattern matching is supported for 'value'. See 'hg help revisions.patterns'.

"file(pattern)"

Changesets affecting files matched by pattern.

For a faster but less accurate result, consider using "filelog()" instead.

This predicate uses "glob:" as the default kind of pattern.

"filelog(pattern)"

Changesets connected to the specified filelog.

For performance reasons, visits only revisions mentioned in the file-level filelog, rather than filtering through all changesets (much faster, but doesn't include deletes or duplicate changes). For a slower, more accurate result, use "file()".

The pattern without explicit kind like "glob:" is expected to be relative to the current directory and match against a file exactly for efficiency.

"first(set, [n])"

An alias for limit().

"follow([file[, startrev]])"

An alias for "::." (ancestors of the working directory's first parent). If file pattern is specified, the histories of files matching given pattern in the revision given by startrev are followed, including copies.

"followlines(file, fromline:toline[, startrev=., descend=False])"

Changesets modifying 'file' in line range ('fromline', 'toline').

Line range corresponds to 'file' content at 'startrev' and should hence be consistent with file size. If startrev is not specified, working directory's parent is used.

By default, ancestors of 'startrev' are returned. If 'descend' is True, descendants of 'startrev' are returned though renames are (currently) not followed in this direction.

"grep(regex)"

Like "keyword(string)" but accepts a regex. Use "grep(r'...')" to ensure special escape characters are handled correctly. Unlike "keyword(string)", the match is case-sensitive.

"head()"

Changeset is a named branch head.

"heads(set)"

Members of set with no children in set.

"hidden()"

Hidden changesets.

"id(string)"

Revision non-ambiguously specified by the given hex string prefix.

"keyword(string)"

Search commit message, user name, and names of changed files for string. The match is case-insensitive.

For a regular expression or case sensitive search of these fields, use "grep(regex)".

"last(set, [n])"

Last n members of set, defaulting to 1.

"limit(set[, n[, offset]])"

First n members of set, defaulting to 1, starting from offset.

"matching(revision [, field])"

Changesets in which a given set of fields match the set of fields in the selected revision or set.

To match more than one field pass the list of fields to match separated by spaces (e.g. "author description").

Valid fields are most regular revision fields and some special fields.

Regular revision fields are "description", "author", "branch", "date", "files", "phase", "parents", "substate", "user" and "diff". Note that "author" and "user" are synonyms. "diff" refers to the contents of the revision. Two revisions matching their "diff" will also match their "files".

Special fields are "summary" and "metadata": "summary" matches the first line of the description. "metadata" is equivalent to matching "description user date" (i.e. it matches the main metadata fields).

"metadata" is the default field which is used when no fields are specified. You can match more than one field at a time.

"max(set)"

Changeset with highest revision number in set.

"merge()"

Changeset is a merge changeset.

"min(set)"

Changeset with lowest revision number in set.

"modifies(pattern)"

Changesets modifying files matched by pattern.

The pattern without explicit kind like "glob:" is expected to be relative to the current directory and match against a file or a directory.

"named(namespace)"

The changesets in a given namespace.

Pattern matching is supported for 'namespace'. See 'hg help revisions.patterns'.

"nodefromfile(path)"

An alias for "::." (ancestors of the working directory's first parent). If file pattern is specified, the histories of files matching given pattern in the revision given by startrev are followed, including copies.

"none()"

No changesets.

"obsolete()"

Mutable changeset with a newer version. (EXPERIMENTAL)

"only(set, [set])"

Changesets that are ancestors of the first set that are not ancestors of any other head in the repo. If a second set is specified, the result is ancestors of the first set that are not ancestors of the second set (i.e. ::<set1> - ::<set2>).

"origin([set])"

Changesets that were specified as a source for the grafts, transplants or rebases that created the given revisions. Omitting the optional set is the same as passing all(). If a changeset created by these operations is itself specified as a source for one of these operations, only the source changeset for the first operation is selected.

"orphan()"

Non-obsolete changesets with obsolete ancestors. (EXPERIMENTAL)

"outgoing([path])"

Changesets not found in the specified destination repository, or the default push location.

If the location resolve to multiple repositories, the union of all outgoing changeset will be used.

"p1([set])"

First parent of changesets in set, or the working directory.

"p2([set])"

Second parent of changesets in set, or the working directory.

"parents([set])"

The set of all parents for all changesets in set, or the working directory.

"phasedivergent()"

Mutable changesets marked as successors of public changesets.

Only non-public and non-obsolete changesets can be 'phasedivergent'. (EXPERIMENTAL)

"present(set)"

An empty set, if any revision in set isn't found; otherwise, all revisions in set.

If any of specified revisions is not present in the local repository, the query is normally aborted. But this predicate allows the query to continue even in such cases.

"public()"

Changeset in public phase.

"remote([id [,path]])"

Local revision that corresponds to the given identifier in a remote repository, if present. Here, the '.' identifier is a synonym for the current local branch.

"removes(pattern)"

Changesets which remove files matching pattern.

The pattern without explicit kind like "glob:" is expected to be relative to the current directory and match against a file or a directory.

"rev(number)"

Revision with the given numeric identifier.

"reverse(set)"

Reverse order of set.

"revset(set)"

Strictly interpret the content as a revset.

The content of this special predicate will be strictly interpreted as a revset. For example, "revset(id(0))" will be interpreted as "id(0)" without possible ambiguity with a "id(0)" bookmark or tag.

"roots(set)"

Changesets in set with no parent changeset in set.

"secret()"

Changeset in secret phase.

"sort(set[, [-]key... [, ...]])"

Sort set by keys. The default sort order is ascending, specify a key as "-key" to sort in descending order.

The keys can be:

• "rev" for the revision number,
• "branch" for the branch name,
• "desc" for the commit message (description),
• "user" for user name ("author" can be used as an alias),
• "date" for the commit date
• "topo" for a reverse topographical sort
• "node" the nodeid of the revision

The "topo" sort order cannot be combined with other sort keys. This sort takes one optional argument, "topo.firstbranch", which takes a revset that specifies what topographical branches to prioritize in the sort.

"stack([revs])"

Experimental revset for the stack of changesets or working directory parent. (EXPERIMENTAL)

"subrepo([pattern])"

Changesets that add, modify or remove the given subrepo. If no subrepo pattern is named, any subrepo changes are returned.

"successors(set)"

All successors for set, including the given set themselves. (EXPERIMENTAL)

"tag([name])"

The specified tag by name, or all tagged revisions if no name is given.

Pattern matching is supported for 'name'. See 'hg help revisions.patterns'.

"unstable()"

Changesets with instabilities. (EXPERIMENTAL)

"user(string)"

User name contains string. The match is case-insensitive.

Pattern matching is supported for 'string'. See 'hg help revisions.patterns'.

"wdir()"

Working directory. (EXPERIMENTAL)

Aliases

New predicates (known as "aliases") can be defined, using any combination of existing predicates or other aliases. An alias definition looks like:

<alias> = <definition>

in the "revsetalias" section of a Mercurial configuration file. Arguments of the form 'a1', 'a2', etc. are substituted from the alias into the definition.

For example,

[revsetalias]
h = heads()
d(s) = sort(s, date)
rs(s, k) = reverse(sort(s, k))

defines three aliases, "h", "d", and "rs". "rs(0:tip, author)" is exactly equivalent to "reverse(sort(0:tip, author))".

Equivalents

Command line equivalents for 'hg log':

-f    ->  ::.
-d x  ->  date(x)
-k x  ->  keyword(x)
-m    ->  merge()
-u x  ->  user(x)
-b x  ->  branch(x)
-P x  ->  !::x
-l x  ->  limit(expr, x)

Examples

Some sample queries:

• Changesets on the default branch:

  hg log -r "branch(default)"
  

• Changesets on the default branch since tag 1.5 (excluding merges):

  hg log -r "branch(default) and 1.5:: and not merge()"
  

• Open branch heads:

  hg log -r "head() and not closed()"
  

• Changesets between tags 1.3 and 1.5 mentioning "bug" that affect "hgext/*":

  hg log -r "1.3::1.5 and keyword(bug) and file('hgext/*')"
  

• Changesets committed in May 2008, sorted by user:

  hg log -r "sort(date('May 2008'), user)"
  

• Changesets mentioning "bug" or "issue" that are not in a tagged release:

  hg log -r "(keyword(bug) or keyword(issue)) and not ancestors(tag())"
  

• Update to the commit that bookmark @ is pointing to, without activating the bookmark (this works because the last revision of the revset is used):

  hg update :@
  

• Show diff between tags 1.3 and 1.5 (this works because the first and the last revisions of the revset are used):

  hg diff -r 1.3::1.5