Reimplement strip operations # ## % %% with fnmatch().

This is slower in theory, but it's the algorithm that bash and mksh use.

It allows us to process character classes correctly, which is necessary
to run 'abuild'.

core/libstr.py

@@ -15,8 +15,14 @@

import re

from core import glob_
import libc

from core.id_kind import Id
from core import glob_
from core import util

log = util.log
e_die = util.e_die

# Implementation without Python regex:
#
@@ -49,11 +55,8 @@
def DoUnarySuffixOp(s, op, arg):
  """Helper for ${x#prefix} and family."""

  pat_re, err = glob_.GlobToPythonRegex(arg)
  if err:
    e_die("Can't convert glob to regex: %r", arg)

  if pat_re is None:  # simple/fast path for fixed strings
  # Fast path for constant strings.
  if not glob_.LooksLikeGlob(arg):
    if op.op_id in (Id.VOp1_Pound, Id.VOp1_DPound):  # const prefix
      if s.startswith(arg):
        return s[len(arg):]
@@ -70,46 +73,48 @@ def DoUnarySuffixOp(s, op, arg):
    else:  # e.g. ^ ^^ , ,,
      raise AssertionError(op.op_id)

  else:  # glob pattern
    # Extract the group from the regex and return it
    if op.op_id == Id.VOp1_Pound:  # shortest prefix
      # Need non-greedy match
      pat_re2, err = glob_.GlobToPythonRegex(arg, greedy=False)
      r = re.compile(pat_re2)
      m = r.match(s)
      if m:
        return s[m.end(0):]
      else:
        return s

    elif op.op_id == Id.VOp1_DPound:  # longest prefix
      r = re.compile(pat_re)
      m = r.match(s)
      if m:
        return s[m.end(0):]
      else:
        return s
  # For patterns, do fnmatch() in a loop.
  #
  # TODO: Check another fast path first?
  #
  # v=aabbccdd
  # echo ${v#*b}  # strip shortest prefix
  #
  # If the whole thing doesn't match '*b*', then no test can succeed.  So we
  # can fail early.  Conversely echo ${v%%c*} and '*c*'.

  n = len(s)
  if op.op_id == Id.VOp1_Pound:  # shortest prefix
    for i in xrange(1, n):
      #log('Matching pattern %r with %r', arg, s[:i])
      if libc.fnmatch(arg, s[:i]):
        return s[i:]
    else:
      return s

    elif op.op_id == Id.VOp1_Percent:  # shortest suffix
      # NOTE: This is different than re.search, which will find the longest
      # suffix.
      r = re.compile('^(.*)' + pat_re + '$')
      m = r.match(s)
      if m:
        return m.group(1)
      else:
        return s
      
    elif op.op_id == Id.VOp1_DPercent:  # longest suffix
      r = re.compile('^(.*?)' + pat_re + '$')  # non-greedy
      m = r.match(s)
      if m:
        return m.group(1)
      else:
        return s
  elif op.op_id == Id.VOp1_DPound:  # longest prefix
    for i in xrange(-1, -n, -1):
      #log('Matching pattern %r with %r', arg, s[:i])
      if libc.fnmatch(arg, s[:i]):
        return s[i:]
    else:
      return s

  elif op.op_id == Id.VOp1_Percent:  # shortest suffix
    for i in xrange(-1, -n, -1):
      #log('Matching pattern %r with %r', arg, s[:i])
      if libc.fnmatch(arg, s[i:]):
        return s[:i]
    else:
      raise AssertionError(op.op_id)
      return s
    
  elif op.op_id == Id.VOp1_DPercent:  # longest suffix
    for i in xrange(1, n):
      #log('Matching pattern %r with %r', arg, s[:i])
      if libc.fnmatch(arg, s[i:]):
        return s[:i]
    else:
      return s


def PatSub(s, op, pat, replace_str):

core/word_eval_test.py

@@ -16,5 +16,7 @@

class WordEvalTest(unittest.TestCase):
  pass


if __name__ == '__main__':
  unittest.main()

test/gold.sh

@@ -117,6 +117,7 @@ readonly -a PASSING=(
  glob
  no-op
  complex-here-docs
  strip-op-char-class

  count
  one-spec-test

test/spec.sh

@@ -200,7 +200,8 @@ blog-other1() {
}

alias() {
  sh-spec spec/alias.test.sh ${REF_SHELLS[@]} $ZSH $OSH "$@"
  sh-spec spec/alias.test.sh --osh-failures-allowed 10 \
    ${REF_SHELLS[@]} $ZSH $OSH "$@"
}

comments() {