fileset: handle error of string unescaping

mercurial/fileset.py

@@ -1,537 +1,537 b''
 # fileset.py - file set queries for mercurial
 #
 # Copyright 2010 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import re
 
 from .i18n import _
 from . import (
     error,
     merge,
     parser,
     util,
 )
 
 elements = {
     # token-type: binding-strength, primary, prefix, infix, suffix
     "(": (20, None, ("group", 1, ")"), ("func", 1, ")"), None),
     "-": (5, None, ("negate", 19), ("minus", 5), None),
     "not": (10, None, ("not", 10), None, None),
     "!": (10, None, ("not", 10), None, None),
     "and": (5, None, None, ("and", 5), None),
     "&": (5, None, None, ("and", 5), None),
     "or": (4, None, None, ("or", 4), None),
     "|": (4, None, None, ("or", 4), None),
     "+": (4, None, None, ("or", 4), None),
     ",": (2, None, None, ("list", 2), None),
     ")": (0, None, None, None, None),
     "symbol": (0, "symbol", None, None, None),
     "string": (0, "string", None, None, None),
     "end": (0, None, None, None, None),
 }
 
 keywords = set(['and', 'or', 'not'])
 
 globchars = ".*{}[]?/\\_"
 
 def tokenize(program):
     pos, l = 0, len(program)
     while pos < l:
         c = program[pos]
         if c.isspace(): # skip inter-token whitespace
             pass
         elif c in "(),-|&+!": # handle simple operators
             yield (c, None, pos)
         elif (c in '"\'' or c == 'r' and
               program[pos:pos + 2] in ("r'", 'r"')): # handle quoted strings
             if c == 'r':
                 pos += 1
                 c = program[pos]
                 decode = lambda x: x
             else:
-                decode = lambda x: x.decode('string-escape')
+                decode = parser.unescapestr
             pos += 1
             s = pos
             while pos < l: # find closing quote
                 d = program[pos]
                 if d == '\\': # skip over escaped characters
                     pos += 2
                     continue
                 if d == c:
                     yield ('string', decode(program[s:pos]), s)
                     break
                 pos += 1
             else:
                 raise error.ParseError(_("unterminated string"), s)
         elif c.isalnum() or c in globchars or ord(c) > 127:
             # gather up a symbol/keyword
             s = pos
             pos += 1
             while pos < l: # find end of symbol
                 d = program[pos]
                 if not (d.isalnum() or d in globchars or ord(d) > 127):
                     break
                 pos += 1
             sym = program[s:pos]
             if sym in keywords: # operator keywords
                 yield (sym, None, s)
             else:
                 yield ('symbol', sym, s)
             pos -= 1
         else:
             raise error.ParseError(_("syntax error"), pos)
         pos += 1
     yield ('end', None, pos)
 
 def parse(expr):
     p = parser.parser(elements)
     tree, pos = p.parse(tokenize(expr))
     if pos != len(expr):
         raise error.ParseError(_("invalid token"), pos)
     return tree
 
 def getstring(x, err):
     if x and (x[0] == 'string' or x[0] == 'symbol'):
         return x[1]
     raise error.ParseError(err)
 
 def getset(mctx, x):
     if not x:
         raise error.ParseError(_("missing argument"))
     return methods[x[0]](mctx, *x[1:])
 
 def stringset(mctx, x):
     m = mctx.matcher([x])
     return [f for f in mctx.subset if m(f)]
 
 def andset(mctx, x, y):
     return getset(mctx.narrow(getset(mctx, x)), y)
 
 def orset(mctx, x, y):
     # needs optimizing
     xl = getset(mctx, x)
     yl = getset(mctx, y)
     return xl + [f for f in yl if f not in xl]
 
 def notset(mctx, x):
     s = set(getset(mctx, x))
     return [r for r in mctx.subset if r not in s]
 
 def minusset(mctx, x, y):
     xl = getset(mctx, x)
     yl = set(getset(mctx, y))
     return [f for f in xl if f not in yl]
 
 def listset(mctx, a, b):
     raise error.ParseError(_("can't use a list in this context"))
 
 def modified(mctx, x):
     """``modified()``
     File that is modified according to :hg:`status`.
     """
     # i18n: "modified" is a keyword
     getargs(x, 0, 0, _("modified takes no arguments"))
     s = mctx.status().modified
     return [f for f in mctx.subset if f in s]
 
 def added(mctx, x):
     """``added()``
     File that is added according to :hg:`status`.
     """
     # i18n: "added" is a keyword
     getargs(x, 0, 0, _("added takes no arguments"))
     s = mctx.status().added
     return [f for f in mctx.subset if f in s]
 
 def removed(mctx, x):
     """``removed()``
     File that is removed according to :hg:`status`.
     """
     # i18n: "removed" is a keyword
     getargs(x, 0, 0, _("removed takes no arguments"))
     s = mctx.status().removed
     return [f for f in mctx.subset if f in s]
 
 def deleted(mctx, x):
     """``deleted()``
     File that is deleted according to :hg:`status`.
     """
     # i18n: "deleted" is a keyword
     getargs(x, 0, 0, _("deleted takes no arguments"))
     s = mctx.status().deleted
     return [f for f in mctx.subset if f in s]
 
 def unknown(mctx, x):
     """``unknown()``
     File that is unknown according to :hg:`status`. These files will only be
     considered if this predicate is used.
     """
     # i18n: "unknown" is a keyword
     getargs(x, 0, 0, _("unknown takes no arguments"))
     s = mctx.status().unknown
     return [f for f in mctx.subset if f in s]
 
 def ignored(mctx, x):
     """``ignored()``
     File that is ignored according to :hg:`status`. These files will only be
     considered if this predicate is used.
     """
     # i18n: "ignored" is a keyword
     getargs(x, 0, 0, _("ignored takes no arguments"))
     s = mctx.status().ignored
     return [f for f in mctx.subset if f in s]
 
 def clean(mctx, x):
     """``clean()``
     File that is clean according to :hg:`status`.
     """
     # i18n: "clean" is a keyword
     getargs(x, 0, 0, _("clean takes no arguments"))
     s = mctx.status().clean
     return [f for f in mctx.subset if f in s]
 
 def func(mctx, a, b):
     if a[0] == 'symbol' and a[1] in symbols:
         return symbols[a[1]](mctx, b)
 
     keep = lambda fn: getattr(fn, '__doc__', None) is not None
 
     syms = [s for (s, fn) in symbols.items() if keep(fn)]
     raise error.UnknownIdentifier(a[1], syms)
 
 def getlist(x):
     if not x:
         return []
     if x[0] == 'list':
         return getlist(x[1]) + [x[2]]
     return [x]
 
 def getargs(x, min, max, err):
     l = getlist(x)
     if len(l) < min or len(l) > max:
         raise error.ParseError(err)
     return l
 
 def binary(mctx, x):
     """``binary()``
     File that appears to be binary (contains NUL bytes).
     """
     # i18n: "binary" is a keyword
     getargs(x, 0, 0, _("binary takes no arguments"))
     return [f for f in mctx.existing() if util.binary(mctx.ctx[f].data())]
 
 def exec_(mctx, x):
     """``exec()``
     File that is marked as executable.
     """
     # i18n: "exec" is a keyword
     getargs(x, 0, 0, _("exec takes no arguments"))
     return [f for f in mctx.existing() if mctx.ctx.flags(f) == 'x']
 
 def symlink(mctx, x):
     """``symlink()``
     File that is marked as a symlink.
     """
     # i18n: "symlink" is a keyword
     getargs(x, 0, 0, _("symlink takes no arguments"))
     return [f for f in mctx.existing() if mctx.ctx.flags(f) == 'l']
 
 def resolved(mctx, x):
     """``resolved()``
     File that is marked resolved according to :hg:`resolve -l`.
     """
     # i18n: "resolved" is a keyword
     getargs(x, 0, 0, _("resolved takes no arguments"))
     if mctx.ctx.rev() is not None:
         return []
     ms = merge.mergestate(mctx.ctx.repo())
     return [f for f in mctx.subset if f in ms and ms[f] == 'r']
 
 def unresolved(mctx, x):
     """``unresolved()``
     File that is marked unresolved according to :hg:`resolve -l`.
     """
     # i18n: "unresolved" is a keyword
     getargs(x, 0, 0, _("unresolved takes no arguments"))
     if mctx.ctx.rev() is not None:
         return []
     ms = merge.mergestate(mctx.ctx.repo())
     return [f for f in mctx.subset if f in ms and ms[f] == 'u']
 
 def hgignore(mctx, x):
     """``hgignore()``
     File that matches the active .hgignore pattern.
     """
     # i18n: "hgignore" is a keyword
     getargs(x, 0, 0, _("hgignore takes no arguments"))
     ignore = mctx.ctx.repo().dirstate._ignore
     return [f for f in mctx.subset if ignore(f)]
 
 def portable(mctx, x):
     """``portable()``
     File that has a portable name. (This doesn't include filenames with case
     collisions.)
     """
     # i18n: "portable" is a keyword
     getargs(x, 0, 0, _("portable takes no arguments"))
     checkwinfilename = util.checkwinfilename
     return [f for f in mctx.subset if checkwinfilename(f) is None]
 
 def grep(mctx, x):
     """``grep(regex)``
     File contains the given regular expression.
     """
     try:
         # i18n: "grep" is a keyword
         r = re.compile(getstring(x, _("grep requires a pattern")))
     except re.error as e:
         raise error.ParseError(_('invalid match pattern: %s') % e)
     return [f for f in mctx.existing() if r.search(mctx.ctx[f].data())]
 
 def _sizetomax(s):
     try:
         s = s.strip().lower()
         for k, v in util._sizeunits:
             if s.endswith(k):
                 # max(4k) = 5k - 1, max(4.5k) = 4.6k - 1
                 n = s[:-len(k)]
                 inc = 1.0
                 if "." in n:
                     inc /= 10 ** len(n.split(".")[1])
                 return int((float(n) + inc) * v) - 1
         # no extension, this is a precise value
         return int(s)
     except ValueError:
         raise error.ParseError(_("couldn't parse size: %s") % s)
 
 def size(mctx, x):
     """``size(expression)``
     File size matches the given expression. Examples:
 
     - 1k (files from 1024 to 2047 bytes)
     - < 20k (files less than 20480 bytes)
     - >= .5MB (files at least 524288 bytes)
     - 4k - 1MB (files from 4096 bytes to 1048576 bytes)
     """
 
     # i18n: "size" is a keyword
     expr = getstring(x, _("size requires an expression")).strip()
     if '-' in expr: # do we have a range?
         a, b = expr.split('-', 1)
         a = util.sizetoint(a)
         b = util.sizetoint(b)
         m = lambda x: x >= a and x <= b
     elif expr.startswith("<="):
         a = util.sizetoint(expr[2:])
         m = lambda x: x <= a
     elif expr.startswith("<"):
         a = util.sizetoint(expr[1:])
         m = lambda x: x < a
     elif expr.startswith(">="):
         a = util.sizetoint(expr[2:])
         m = lambda x: x >= a
     elif expr.startswith(">"):
         a = util.sizetoint(expr[1:])
         m = lambda x: x > a
     elif expr[0].isdigit or expr[0] == '.':
         a = util.sizetoint(expr)
         b = _sizetomax(expr)
         m = lambda x: x >= a and x <= b
     else:
         raise error.ParseError(_("couldn't parse size: %s") % expr)
 
     return [f for f in mctx.existing() if m(mctx.ctx[f].size())]
 
 def encoding(mctx, x):
     """``encoding(name)``
     File can be successfully decoded with the given character
     encoding. May not be useful for encodings other than ASCII and
     UTF-8.
     """
 
     # i18n: "encoding" is a keyword
     enc = getstring(x, _("encoding requires an encoding name"))
 
     s = []
     for f in mctx.existing():
         d = mctx.ctx[f].data()
         try:
             d.decode(enc)
         except LookupError:
             raise util.Abort(_("unknown encoding '%s'") % enc)
         except UnicodeDecodeError:
             continue
         s.append(f)
 
     return s
 
 def eol(mctx, x):
     """``eol(style)``
     File contains newlines of the given style (dos, unix, mac). Binary
     files are excluded, files with mixed line endings match multiple
     styles.
     """
 
     # i18n: "encoding" is a keyword
     enc = getstring(x, _("encoding requires an encoding name"))
 
     s = []
     for f in mctx.existing():
         d = mctx.ctx[f].data()
         if util.binary(d):
             continue
         if (enc == 'dos' or enc == 'win') and '\r\n' in d:
             s.append(f)
         elif enc == 'unix' and re.search('(?<!\r)\n', d):
             s.append(f)
         elif enc == 'mac' and re.search('\r(?!\n)', d):
             s.append(f)
     return s
 
 def copied(mctx, x):
     """``copied()``
     File that is recorded as being copied.
     """
     # i18n: "copied" is a keyword
     getargs(x, 0, 0, _("copied takes no arguments"))
     s = []
     for f in mctx.subset:
         p = mctx.ctx[f].parents()
         if p and p[0].path() != f:
             s.append(f)
     return s
 
 def subrepo(mctx, x):
     """``subrepo([pattern])``
     Subrepositories whose paths match the given pattern.
     """
     # i18n: "subrepo" is a keyword
     getargs(x, 0, 1, _("subrepo takes at most one argument"))
     ctx = mctx.ctx
     sstate = sorted(ctx.substate)
     if x:
         # i18n: "subrepo" is a keyword
         pat = getstring(x, _("subrepo requires a pattern or no arguments"))
 
         from . import match as matchmod # avoid circular import issues
         fast = not matchmod.patkind(pat)
         if fast:
             def m(s):
                 return (s == pat)
         else:
             m = matchmod.match(ctx.repo().root, '', [pat], ctx=ctx)
         return [sub for sub in sstate if m(sub)]
     else:
         return [sub for sub in sstate]
 
 symbols = {
     'added': added,
     'binary': binary,
     'clean': clean,
     'copied': copied,
     'deleted': deleted,
     'encoding': encoding,
     'eol': eol,
     'exec': exec_,
     'grep': grep,
     'ignored': ignored,
     'hgignore': hgignore,
     'modified': modified,
     'portable': portable,
     'removed': removed,
     'resolved': resolved,
     'size': size,
     'symlink': symlink,
     'unknown': unknown,
     'unresolved': unresolved,
     'subrepo': subrepo,
 }
 
 methods = {
     'string': stringset,
     'symbol': stringset,
     'and': andset,
     'or': orset,
     'minus': minusset,
     'list': listset,
     'group': getset,
     'not': notset,
     'func': func,
 }
 
 class matchctx(object):
     def __init__(self, ctx, subset=None, status=None):
         self.ctx = ctx
         self.subset = subset
         self._status = status
     def status(self):
         return self._status
     def matcher(self, patterns):
         return self.ctx.match(patterns)
     def filter(self, files):
         return [f for f in files if f in self.subset]
     def existing(self):
         if self._status is not None:
             removed = set(self._status[3])
             unknown = set(self._status[4] + self._status[5])
         else:
             removed = set()
             unknown = set()
         return (f for f in self.subset
                 if (f in self.ctx and f not in removed) or f in unknown)
     def narrow(self, files):
         return matchctx(self.ctx, self.filter(files), self._status)
 
 def _intree(funcs, tree):
     if isinstance(tree, tuple):
         if tree[0] == 'func' and tree[1][0] == 'symbol':
             if tree[1][1] in funcs:
                 return True
         for s in tree[1:]:
             if _intree(funcs, s):
                 return True
     return False
 
 # filesets using matchctx.existing()
 _existingcallers = [
     'binary',
     'exec',
     'grep',
     'size',
     'symlink',
 ]
 
 def getfileset(ctx, expr):
     tree = parse(expr)
 
     # do we need status info?
     if (_intree(['modified', 'added', 'removed', 'deleted',
                  'unknown', 'ignored', 'clean'], tree) or
         # Using matchctx.existing() on a workingctx requires us to check
         # for deleted files.
         (ctx.rev() is None and _intree(_existingcallers, tree))):
         unknown = _intree(['unknown'], tree)
         ignored = _intree(['ignored'], tree)
 
         r = ctx.repo()
         status = r.status(ctx.p1(), ctx,
                           unknown=unknown, ignored=ignored, clean=True)
         subset = []
         for c in status:
             subset.extend(c)
     else:
         status = None
         subset = list(ctx.walk(ctx.match([])))
 
     return getset(matchctx(ctx, subset, status), tree)
 
 def prettyformat(tree):
     return parser.prettyformat(tree, ('string', 'symbol'))
 
 # tell hggettext to extract docstrings from these functions:
 i18nfunctions = symbols.values()

tests/test-fileset.t

@@ -1,308 +1,311 b''
   $ fileset() {
   >   hg debugfileset "$@"
   > }
 
   $ hg init repo
   $ cd repo
   $ echo a > a1
   $ echo a > a2
   $ echo b > b1
   $ echo b > b2
   $ hg ci -Am addfiles
   adding a1
   adding a2
   adding b1
   adding b2
 
 Test operators and basic patterns
 
   $ fileset -v a1
   ('symbol', 'a1')
   a1
   $ fileset -v 'a*'
   ('symbol', 'a*')
   a1
   a2
   $ fileset -v '"re:a\d"'
   ('string', 're:a\\d')
   a1
   a2
   $ fileset -v 'a1 or a2'
   (or
     ('symbol', 'a1')
     ('symbol', 'a2'))
   a1
   a2
   $ fileset 'a1 | a2'
   a1
   a2
   $ fileset 'a* and "*1"'
   a1
   $ fileset 'a* & "*1"'
   a1
   $ fileset 'not (r"a*")'
   b1
   b2
   $ fileset '! ("a*")'
   b1
   b2
   $ fileset 'a* - a1'
   a2
   $ fileset 'a_b'
+  $ fileset '"\xy"'
+  hg: parse error: invalid \x escape
+  [255]
 
 Test files status
 
   $ rm a1
   $ hg rm a2
   $ echo b >> b2
   $ hg cp b1 c1
   $ echo c > c2
   $ echo c > c3
   $ cat > .hgignore <<EOF
   > \.hgignore
   > 2$
   > EOF
   $ fileset 'modified()'
   b2
   $ fileset 'added()'
   c1
   $ fileset 'removed()'
   a2
   $ fileset 'deleted()'
   a1
   $ fileset 'unknown()'
   c3
   $ fileset 'ignored()'
   .hgignore
   c2
   $ fileset 'hgignore()'
   a2
   b2
   $ fileset 'clean()'
   b1
   $ fileset 'copied()'
   c1
 
 Test files properties
 
   >>> file('bin', 'wb').write('\0a')
   $ fileset 'binary()'
   $ fileset 'binary() and unknown()'
   bin
   $ echo '^bin$' >> .hgignore
   $ fileset 'binary() and ignored()'
   bin
   $ hg add bin
   $ fileset 'binary()'
   bin
 
   $ fileset 'grep("b{1}")'
   b2
   c1
   b1
   $ fileset 'grep("missingparens(")'
   hg: parse error: invalid match pattern: unbalanced parenthesis
   [255]
 
 #if execbit
   $ chmod +x b2
   $ fileset 'exec()'
   b2
 #endif
 
 #if symlink
   $ ln -s b2 b2link
   $ fileset 'symlink() and unknown()'
   b2link
   $ hg add b2link
 #endif
 
 #if no-windows
   $ echo foo > con.xml
   $ fileset 'not portable()'
   con.xml
   $ hg --config ui.portablefilenames=ignore add con.xml
 #endif
 
   >>> file('1k', 'wb').write(' '*1024)
   >>> file('2k', 'wb').write(' '*2048)
   $ hg add 1k 2k
   $ fileset 'size("bar")'
   hg: parse error: couldn't parse size: bar
   [255]
   $ fileset 'size(1k)'
   1k
   $ fileset '(1k or 2k) and size("< 2k")'
   1k
   $ fileset '(1k or 2k) and size("<=2k")'
   1k
   2k
   $ fileset '(1k or 2k) and size("> 1k")'
   2k
   $ fileset '(1k or 2k) and size(">=1K")'
   1k
   2k
   $ fileset '(1k or 2k) and size(".5KB - 1.5kB")'
   1k
   $ fileset 'size("1M")'
   $ fileset 'size("1 GB")'
 
 Test merge states
 
   $ hg ci -m manychanges
   $ hg up -C 0
   * files updated, 0 files merged, * files removed, 0 files unresolved (glob)
   $ echo c >> b2
   $ hg ci -m diverging b2
   created new head
   $ fileset 'resolved()'
   $ fileset 'unresolved()'
   $ hg merge
   merging b2
   warning: conflicts during merge.
   merging b2 incomplete! (edit conflicts, then use 'hg resolve --mark')
   * files updated, 0 files merged, * files removed, 1 files unresolved (glob)
   use 'hg resolve' to retry unresolved file merges or 'hg update -C .' to abandon
   [1]
   $ fileset 'resolved()'
   $ fileset 'unresolved()'
   b2
   $ echo e > b2
   $ hg resolve -m b2
   (no more unresolved files)
   $ fileset 'resolved()'
   b2
   $ fileset 'unresolved()'
   $ hg ci -m merge
 
 Test subrepo predicate
 
   $ hg init sub
   $ echo a > sub/suba
   $ hg -R sub add sub/suba
   $ hg -R sub ci -m sub
   $ echo 'sub = sub' > .hgsub
   $ hg init sub2
   $ echo b > sub2/b
   $ hg -R sub2 ci -Am sub2
   adding b
   $ echo 'sub2 = sub2' >> .hgsub
   $ fileset 'subrepo()'
   $ hg add .hgsub
   $ fileset 'subrepo()'
   sub
   sub2
   $ fileset 'subrepo("sub")'
   sub
   $ fileset 'subrepo("glob:*")'
   sub
   sub2
   $ hg ci -m subrepo
 
 Test that .hgsubstate is updated as appropriate during a conversion.  The
 saverev property is enough to alter the hashes of the subrepo.
 
   $ hg init ../converted
   $ hg --config extensions.convert= convert --config convert.hg.saverev=True  \
   >      sub ../converted/sub
   initializing destination ../converted/sub repository
   scanning source...
   sorting...
   converting...
   0 sub
   $ hg clone -U sub2 ../converted/sub2
   $ hg --config extensions.convert= convert --config convert.hg.saverev=True  \
   >      . ../converted
   scanning source...
   sorting...
   converting...
   4 addfiles
   3 manychanges
   2 diverging
   1 merge
   0 subrepo
   no ".hgsubstate" updates will be made for "sub2"
   $ hg up -q -R ../converted -r tip
   $ hg --cwd ../converted cat sub/suba sub2/b -r tip
   a
   b
   $ oldnode=`hg log -r tip -T "{node}\n"`
   $ newnode=`hg log -R ../converted -r tip -T "{node}\n"`
   $ [ "$oldnode" != "$newnode" ] || echo "nothing changed"
 
 Test with a revision
 
   $ hg log -G --template '{rev} {desc}\n'
   @  4 subrepo
   |
   o    3 merge
   |\
   | o  2 diverging
   | |
   o |  1 manychanges
   |/
   o  0 addfiles
   
   $ echo unknown > unknown
   $ fileset -r1 'modified()'
   b2
   $ fileset -r1 'added() and c1'
   c1
   $ fileset -r1 'removed()'
   a2
   $ fileset -r1 'deleted()'
   $ fileset -r1 'unknown()'
   $ fileset -r1 'ignored()'
   $ fileset -r1 'hgignore()'
   b2
   bin
   $ fileset -r1 'binary()'
   bin
   $ fileset -r1 'size(1k)'
   1k
   $ fileset -r3 'resolved()'
   $ fileset -r3 'unresolved()'
 
 #if execbit
   $ fileset -r1 'exec()'
   b2
 #endif
 
 #if symlink
   $ fileset -r1 'symlink()'
   b2link
 #endif
 
 #if no-windows
   $ fileset -r1 'not portable()'
   con.xml
   $ hg forget 'con.xml'
 #endif
 
   $ fileset -r4 'subrepo("re:su.*")'
   sub
   sub2
   $ fileset -r4 'subrepo("sub")'
   sub
   $ fileset -r4 'b2 or c1'
   b2
   c1
 
   >>> open('dos', 'wb').write("dos\r\n")
   >>> open('mixed', 'wb').write("dos\r\nunix\n")
   >>> open('mac', 'wb').write("mac\r")
   $ hg add dos mixed mac
 
   $ fileset 'eol(dos)'
   dos
   mixed
   $ fileset 'eol(unix)'
   .hgsub
   .hgsubstate
   a1
   b1
   b2
   c1
   mixed
   $ fileset 'eol(mac)'
   mac