# HG changeset patch # User bos@serpentine.internal.keyresearch.com # Date 1125336701 25200 # Node ID d32b91ebad5d1cda40ade0eed01d4d335be515d0 # Parent 9061f79c6c6f12e6b33f8b8bf875aa33dc26d347# Parent 8a39df05d2c18764ee56d4f2ff575e300ddb101b Merge with TAH. diff --git a/doc/hg.1.txt b/doc/hg.1.txt --- a/doc/hg.1.txt +++ b/doc/hg.1.txt @@ -207,6 +207,31 @@ forget [options] [files]:: -I, --include include names matching the given patterns -X, --exclude exclude names matching the given patterns +grep [options] pattern [files]:: + Search revisions of files for a regular expression. + + This command behaves differently than Unix grep. It only accepts + Python/Perl regexps. It searches repository history, not the + working directory. It always prints the revision number in which + a match appears. + + By default, grep only prints output for the first revision of a + file in which it finds a match. To get it to print every revision + that contains a change in match status ("-" for a match that + becomes a non-match, or "+" for a non-match that becomes a match), + use the --every-match flag. + + options: + -0, --print0 end fields with NUL + -I, --include include names matching the given patterns + -X, --exclude exclude names matching the given patterns + -e, --every-match print every revision that matches + -i, --ignore-case ignore case when matching + -l, --files-with-matches print only file names and revs that match + -n, --line-number print matching line numbers + -r , --rev search in given revision range + -u, --user print user who committed change + heads:: Show all repository head changesets. diff --git a/mercurial/commands.py b/mercurial/commands.py --- a/mercurial/commands.py +++ b/mercurial/commands.py @@ -49,12 +49,31 @@ def walk(repo, pats, opts, head=''): yield r def walkchangerevs(ui, repo, cwd, pats, opts): - # This code most commonly needs to iterate backwards over the - # history it is interested in. Doing so has awful - # (quadratic-looking) performance, so we use iterators in a - # "windowed" way. Walk forwards through a window of revisions, - # yielding them in the desired order, and walk the windows - # themselves backwards. + '''Iterate over files and the revs they changed in. + + Callers most commonly need to iterate backwards over the history + it is interested in. Doing so has awful (quadratic-looking) + performance, so we use iterators in a "windowed" way. + + We walk a window of revisions in the desired order. Within the + window, we first walk forwards to gather data, then in the desired + order (usually backwards) to display it. + + This function returns an (iterator, getchange) pair. The + getchange function returns the changelog entry for a numeric + revision. The iterator yields 3-tuples. They will be of one of + the following forms: + + "window", incrementing, lastrev: stepping through a window, + positive if walking forwards through revs, last rev in the + sequence iterated over - use to reset state for the current window + + "add", rev, fns: out-of-order traversal of the given file names + fns, which changed during revision rev - use to gather data for + possible display + + "iter", rev, None: in-order traversal of the revs earlier iterated + over with "add" - use to display data''' cwd = repo.getcwd() if not pats and cwd: opts['include'] = [os.path.join(cwd, i) for i in opts['include']] @@ -66,6 +85,14 @@ def walkchangerevs(ui, repo, cwd, pats, slowpath = anypats window = 300 fncache = {} + + chcache = {} + def getchange(rev): + ch = chcache.get(rev) + if ch is None: + chcache[rev] = ch = repo.changelog.read(repo.lookup(str(rev))) + return ch + if not slowpath and not files: # No files, no patterns. Display all revs. wanted = dict(zip(revs, revs)) @@ -100,7 +127,7 @@ def walkchangerevs(ui, repo, cwd, pats, def changerevgen(): for i in xrange(repo.changelog.count() - 1, -1, -window): for j in xrange(max(0, i - window), i + 1): - yield j, repo.changelog.read(repo.lookup(str(j)))[3] + yield j, getchange(j)[3] for rev, changefiles in changerevgen(): matches = filter(matchfn, changefiles) @@ -108,20 +135,19 @@ def walkchangerevs(ui, repo, cwd, pats, fncache[rev] = matches wanted[rev] = 1 - for i in xrange(0, len(revs), window): - yield 'window', revs[0] < revs[-1], revs[-1] - nrevs = [rev for rev in revs[i:min(i+window, len(revs))] - if rev in wanted] - srevs = list(nrevs) - srevs.sort() - for rev in srevs: - fns = fncache.get(rev) - if not fns: - fns = repo.changelog.read(repo.lookup(str(rev)))[3] - fns = filter(matchfn, fns) - yield 'add', rev, fns - for rev in nrevs: - yield 'iter', rev, None + def iterate(): + for i in xrange(0, len(revs), window): + yield 'window', revs[0] < revs[-1], revs[-1] + nrevs = [rev for rev in revs[i:min(i+window, len(revs))] + if rev in wanted] + srevs = list(nrevs) + srevs.sort() + for rev in srevs: + fns = fncache.get(rev) or filter(matchfn, getchange(rev)[3]) + yield 'add', rev, fns + for rev in nrevs: + yield 'iter', rev, None + return iterate(), getchange revrangesep = ':' @@ -150,6 +176,7 @@ def revrange(ui, repo, revs, revlog=None except KeyError: raise util.Abort('invalid revision identifier %s', val) return num + seen = {} for spec in revs: if spec.find(revrangesep) >= 0: start, end = spec.split(revrangesep, 1) @@ -157,9 +184,14 @@ def revrange(ui, repo, revs, revlog=None end = fix(end, revcount - 1) step = start > end and -1 or 1 for rev in xrange(start, end+step, step): + if rev in seen: continue + seen[rev] = 1 yield str(rev) else: - yield str(fix(spec, None)) + rev = fix(spec, None) + if rev in seen: continue + seen[rev] = 1 + yield str(rev) def make_filename(repo, r, pat, node=None, total=None, seqno=None, revwidth=None): @@ -265,6 +297,13 @@ def dodiff(fp, ui, repo, node1, node2, f tn = None fp.write(mdiff.unidiff(to, date1, tn, date2, f, r, text=text)) +def trimuser(ui, name, rev, revcache): + """trim the name of the user who committed a change""" + user = revcache.get(rev) + if user is None: + user = revcache[rev] = ui.shortuser(name) + return user + def show_changeset(ui, repo, rev=0, changenode=None, brinfo=None): """show a single changeset or file revision""" log = repo.changelog @@ -467,18 +506,14 @@ def annotate(ui, repo, *pats, **opts): def getnode(rev): return short(repo.changelog.node(rev)) + ucache = {} def getname(rev): - try: - return bcache[rev] - except KeyError: - cl = repo.changelog.read(repo.changelog.node(rev)) - bcache[rev] = name = ui.shortuser(cl[1]) - return name + cl = repo.changelog.read(repo.changelog.node(rev)) + return trimuser(ui, cl[1], rev, ucache) if not pats: raise util.Abort('at least one file name or pattern required') - bcache = {} opmap = [['user', getname], ['number', str], ['changeset', getnode]] if not opts['user'] and not opts['changeset']: opts['number'] = 1 @@ -818,9 +853,9 @@ def grep(ui, repo, pattern, *pats, **opt if opts['ignore_case']: reflags |= re.I regexp = re.compile(pattern, reflags) - sep, end = ':', '\n' + sep, eol = ':', '\n' if opts['print0']: - sep = end = '\0' + sep = eol = '\0' fcache = {} def getfile(fn): @@ -862,20 +897,39 @@ def grep(ui, repo, pattern, *pats, **opt m[s] = s prev = {} + ucache = {} def display(fn, rev, states, prevstates): diff = list(sets.Set(states).symmetric_difference(sets.Set(prevstates))) diff.sort(lambda x, y: cmp(x.linenum, y.linenum)) + counts = {'-': 0, '+': 0} + filerevmatches = {} for l in diff: - if incrementing: + if incrementing or not opts['every_match']: change = ((l in prevstates) and '-') or '+' r = rev else: change = ((l in states) and '-') or '+' r = prev[fn] - ui.write('%s:%s:%s:%s%s\n' % (fn, r, l.linenum, change, l.line)) + cols = [fn, str(rev)] + if opts['line_number']: cols.append(str(l.linenum)) + if opts['every_match']: cols.append(change) + if opts['user']: cols.append(trimuser(ui, getchange(rev)[1], rev, + ucache)) + if opts['files_with_matches']: + c = (fn, rev) + if c in filerevmatches: continue + filerevmatches[c] = 1 + else: + cols.append(l.line) + ui.write(sep.join(cols), eol) + counts[change] += 1 + return counts['+'], counts['-'] fstate = {} - for st, rev, fns in walkchangerevs(ui, repo, repo.getcwd(), pats, opts): + skip = {} + changeiter, getchange = walkchangerevs(ui, repo, repo.getcwd(), pats, opts) + count = 0 + for st, rev, fns in changeiter: if st == 'window': incrementing = rev matches.clear() @@ -884,6 +938,7 @@ def grep(ui, repo, pattern, *pats, **opt mf = repo.manifest.read(change[0]) matches[rev] = {} for fn in fns: + if fn in skip: continue fstate.setdefault(fn, {}) try: grepbody(fn, rev, getfile(fn).read(mf[fn])) @@ -893,8 +948,12 @@ def grep(ui, repo, pattern, *pats, **opt states = matches[rev].items() states.sort() for fn, m in states: - if incrementing or fstate[fn]: - display(fn, rev, m, fstate[fn]) + if fn in skip: continue + if incrementing or not opts['every_match'] or fstate[fn]: + pos, neg = display(fn, rev, m, fstate[fn]) + count += pos + neg + if pos and not opts['every_match']: + skip[fn] = True fstate[fn] = m prev[fn] = rev @@ -902,7 +961,9 @@ def grep(ui, repo, pattern, *pats, **opt fstate = fstate.items() fstate.sort() for fn, state in fstate: + if fn in skip: continue display(fn, rev, {}, state) + return (count == 0 and 1) or 0 def heads(ui, repo, **opts): """show current repository heads""" @@ -1056,8 +1117,9 @@ def log(ui, repo, *pats, **opts): if not pats and cwd: opts['include'] = [os.path.join(cwd, i) for i in opts['include']] opts['exclude'] = [os.path.join(cwd, x) for x in opts['exclude']] - for st, rev, fns in walkchangerevs(ui, repo, (pats and cwd) or '', pats, - opts): + changeiter, getchange = walkchangerevs(ui, repo, (pats and cwd) or '', + pats, opts) + for st, rev, fns in changeiter: if st == 'window': du = dui(ui) elif st == 'add': @@ -1554,13 +1616,15 @@ table = { "hg forget [OPTION]... FILE..."), "grep": (grep, - [('0', 'print0', None, 'end filenames with NUL'), + [('0', 'print0', None, 'end fields with NUL'), ('I', 'include', [], 'include path in search'), ('X', 'exclude', [], 'include path in search'), + ('e', 'every-match', None, 'print every rev with matches'), ('i', 'ignore-case', None, 'ignore case when matching'), - ('l', 'files-with-matches', None, 'print names of files with matches'), - ('n', 'line-number', '', 'print line numbers'), - ('r', 'rev', [], 'search in revision rev')], + ('l', 'files-with-matches', None, 'print names of files and revs with matches'), + ('n', 'line-number', None, 'print line numbers'), + ('r', 'rev', [], 'search in revision rev'), + ('u', 'user', None, 'print user who made change')], "hg grep [OPTION]... PATTERN [FILE]..."), "heads": (heads, diff --git a/mercurial/ui.py b/mercurial/ui.py --- a/mercurial/ui.py +++ b/mercurial/ui.py @@ -80,12 +80,13 @@ class ui: def shortuser(self, user): """Return a short representation of a user name or email address.""" - f = user.find('@') - if f >= 0: - user = user[:f] - f = user.find('<') - if f >= 0: - user = user[f+1:] + if not self.verbose: + f = user.find('@') + if f >= 0: + user = user[:f] + f = user.find('<') + if f >= 0: + user = user[f+1:] return user def expandpath(self, loc): diff --git a/tests/test-grep b/tests/test-grep new file mode 100755 --- /dev/null +++ b/tests/test-grep @@ -0,0 +1,22 @@ +#!/bin/sh + +mkdir t +cd t +hg init +echo import > port +hg add port +hg commit -m 0 -u spam -d '0 0' +echo export >> port +hg commit -m 1 -u eggs -d '1 0' +echo export > port +echo vaportight >> port +echo 'import/export' >> port +hg commit -m 2 -u spam -d '2 0' +echo 'import/export' >> port +hg commit -m 3 -u eggs -d '3 0' +head -3 port > port1 +mv port1 port +hg commit -m 4 -u spam -d '4 0' +hg grep port port +hg grep -enu port port +hg grep import port