view mercurial/util.py @ 1031:503aaf19a040

Rewrite log command. New version is faster and more featureful. The original implementation of log walked backwards through history, which had terrible behaviour. It took several minutes to view complete kernel change history on a fast machine, for example. The rewrite uses a windowed approach to walk hunks of history forwards, while still giving results in reverse order. This reduces run time from five minutes to five seconds on my system. In addition, the rewrite uses our normal name handling mechanisms, so you can run a command like "hg log net/ipv4/**.c" and get a useful answer. It optimises for three different cases (no arguments, only files, and anything goes), so it performs well in all circumstances I've tested.
author Bryan O'Sullivan <bos@serpentine.com>
date Wed, 24 Aug 2005 12:39:10 -0700
parents 22571b8d35d3
children 6d5a62a549fa
line wrap: on
line source

# util.py - utility functions and platform specfic implementations
#
# Copyright 2005 K. Thananchayan <thananck@yahoo.com>
#
# This software may be used and distributed according to the terms
# of the GNU General Public License, incorporated herein by reference.

import os, errno
from demandload import *
demandload(globals(), "re")

def binary(s):
    if s and '\0' in s[:4096]:
        return True
    return False

def unique(g):
    seen = {}
    for f in g:
        if f not in seen:
            seen[f] = 1
            yield f

class Abort(Exception):
    """Raised if a command needs to print an error and exit."""

def always(fn): return True
def never(fn): return False

def globre(pat, head = '^', tail = '$'):
    "convert a glob pattern into a regexp"
    i, n = 0, len(pat)
    res = ''
    group = False
    def peek(): return i < n and pat[i]
    while i < n:
        c = pat[i]
        i = i+1
        if c == '*':
            if peek() == '*':
                i += 1
                res += '.*'
            else:
                res += '[^/]*'
        elif c == '?':
            res += '.'
        elif c == '[':
            j = i
            if j < n and pat[j] in '!]':
                j += 1
            while j < n and pat[j] != ']':
                j += 1
            if j >= n:
                res += '\\['
            else:
                stuff = pat[i:j].replace('\\','\\\\')
                i = j + 1
                if stuff[0] == '!':
                    stuff = '^' + stuff[1:]
                elif stuff[0] == '^':
                    stuff = '\\' + stuff
                res = '%s[%s]' % (res, stuff)
        elif c == '{':
            group = True
            res += '(?:'
        elif c == '}' and group:
            res += ')'
            group = False
        elif c == ',' and group:
            res += '|'
        else:
            res += re.escape(c)
    return head + res + tail

_globchars = {'[': 1, '{': 1, '*': 1, '?': 1}

def pathto(n1, n2):
    '''return the relative path from one place to another.
    this returns a path in the form used by the local filesystem, not hg.'''
    if not n1: return localpath(n2)
    a, b = n1.split('/'), n2.split('/')
    a.reverse(), b.reverse()
    while a and b and a[-1] == b[-1]:
        a.pop(), b.pop()
    b.reverse()
    return os.sep.join((['..'] * len(a)) + b)

def canonpath(repo, cwd, myname):
    rootsep = repo.root + os.sep
    name = myname
    if not name.startswith(os.sep):
        name = os.path.join(repo.root, cwd, name)
    name = os.path.normpath(name)
    if name.startswith(rootsep):
        return pconvert(name[len(rootsep):])
    elif name == repo.root:
        return ''
    else:
        raise Abort('%s not under repository root' % myname)

def matcher(repo, cwd, names, inc, exc, head = ''):
    def patkind(name):
        for prefix in 're:', 'glob:', 'path:', 'relpath:':
            if name.startswith(prefix): return name.split(':', 1)
        for c in name:
            if c in _globchars: return 'glob', name
        return 'relpath', name

    def regex(kind, name, tail):
        '''convert a pattern into a regular expression'''
        if kind == 're':
            return name
        elif kind == 'path':
            return '^' + re.escape(name) + '(?:/|$)'
        elif kind == 'relpath':
            return head + re.escape(name) + tail
        return head + globre(name, '', tail)

    def matchfn(pats, tail):
        """build a matching function from a set of patterns"""
        if pats:
            pat = '(?:%s)' % '|'.join([regex(k, p, tail) for (k, p) in pats])
            return re.compile(pat).match

    def globprefix(pat):
        '''return the non-glob prefix of a path, e.g. foo/* -> foo'''
        root = []
        for p in pat.split(os.sep):
            if patkind(p)[0] == 'glob': break
            root.append(p)
        return '/'.join(root)

    pats = []
    files = []
    roots = []
    for kind, name in map(patkind, names):
        if kind in ('glob', 'relpath'):
            name = canonpath(repo, cwd, name)
            if name == '':
                kind, name = 'glob', '**'
        if kind in ('glob', 'path', 're'):
            pats.append((kind, name))
        if kind == 'glob':
            root = globprefix(name)
            if root: roots.append(root)
        elif kind == 'relpath':
            files.append((kind, name))
            roots.append(name)

    patmatch = matchfn(pats, '$') or always
    filematch = matchfn(files, '(?:/|$)') or always
    incmatch = always
    if inc:
        incmatch = matchfn(map(patkind, inc), '(?:/|$)')
    excmatch = lambda fn: False
    if exc:
        excmatch = matchfn(map(patkind, exc), '(?:/|$)')

    return (roots,
            lambda fn: (incmatch(fn) and not excmatch(fn) and
                        (fn.endswith('/') or
                         (not pats and not files) or
                         (pats and patmatch(fn)) or
                         (files and filematch(fn)))),
            (inc or exc or (pats and pats != [('glob', '**')])) and True)

def system(cmd, errprefix=None):
    """execute a shell command that must succeed"""
    rc = os.system(cmd)
    if rc:
        errmsg = "%s %s" % (os.path.basename(cmd.split(None, 1)[0]),
                            explain_exit(rc)[0])
        if errprefix:
            errmsg = "%s: %s" % (errprefix, errmsg)
        raise Abort(errmsg)

def rename(src, dst):
    try:
        os.rename(src, dst)
    except:
        os.unlink(dst)
        os.rename(src, dst)

def copytree(src, dst, copyfile):
    """Copy a directory tree, files are copied using 'copyfile'."""
    names = os.listdir(src)
    os.mkdir(dst)

    for name in names:
        srcname = os.path.join(src, name)
        dstname = os.path.join(dst, name)
        if os.path.isdir(srcname):
            copytree(srcname, dstname, copyfile)
        elif os.path.isfile(srcname):
            copyfile(srcname, dstname)
        else:
            pass

def _makelock_file(info, pathname):
    ld = os.open(pathname, os.O_CREAT | os.O_WRONLY | os.O_EXCL)
    os.write(ld, info)
    os.close(ld)

def _readlock_file(pathname):
    return file(pathname).read()

# Platfor specific varients
if os.name == 'nt':
    nulldev = 'NUL:'

    def is_exec(f, last):
        return last

    def set_exec(f, mode):
        pass

    def pconvert(path):
        return path.replace("\\", "/")

    def localpath(path):
        return path.replace('/', '\\')

    def normpath(path):
        return pconvert(os.path.normpath(path))

    makelock = _makelock_file
    readlock = _readlock_file

    def explain_exit(code):
        return "exited with status %d" % code, code

else:
    nulldev = '/dev/null'

    def is_exec(f, last):
        return (os.stat(f).st_mode & 0100 != 0)

    def set_exec(f, mode):
        s = os.stat(f).st_mode
        if (s & 0100 != 0) == mode:
            return
        if mode:
            # Turn on +x for every +r bit when making a file executable
            # and obey umask.
            umask = os.umask(0)
            os.umask(umask)
            os.chmod(f, s | (s & 0444) >> 2 & ~umask)
        else:
            os.chmod(f, s & 0666)

    def pconvert(path):
        return path

    def localpath(path):
        return path

    normpath = os.path.normpath

    def makelock(info, pathname):
        try:
            os.symlink(info, pathname)
        except OSError, why:
            if why.errno == errno.EEXIST:
                raise
            else:
                _makelock_file(info, pathname)

    def readlock(pathname):
        try:
            return os.readlink(pathname)
        except OSError, why:
            if why.errno == errno.EINVAL:
                return _readlock_file(pathname)
            else:
                raise

    def explain_exit(code):
        """return a 2-tuple (desc, code) describing a process's status"""
        if os.WIFEXITED(code):
            val = os.WEXITSTATUS(code)
            return "exited with status %d" % val, val
        elif os.WIFSIGNALED(code):
            val = os.WTERMSIG(code)
            return "killed by signal %d" % val, val
        elif os.WIFSTOPPED(code):
            val = os.WSTOPSIG(code)
            return "stopped by signal %d" % val, val
        raise ValueError("invalid exit code")