parser: accept iterator of tokens instead of tokenizer function and program
This can simplify the interface of parse() function. Our tokenizer tends to
have optional arguments other than the message to be parsed.
Before this patch, the "lookup" argument existed only for the revset, and the
templater had to pack [program, start, end] to be passed to its tokenizer.
--- a/mercurial/fileset.py Sat Jun 20 20:11:53 2015 -0700
+++ b/mercurial/fileset.py Sun Jun 21 00:49:26 2015 +0900
@@ -80,8 +80,8 @@
yield ('end', None, pos)
def parse(expr):
- p = parser.parser(tokenize, elements)
- tree, pos = p.parse(expr)
+ p = parser.parser(elements)
+ tree, pos = p.parse(tokenize(expr))
if pos != len(expr):
raise error.ParseError(_("invalid token"), pos)
return tree
--- a/mercurial/parser.py Sat Jun 20 20:11:53 2015 -0700
+++ b/mercurial/parser.py Sun Jun 21 00:49:26 2015 +0900
@@ -19,8 +19,7 @@
from i18n import _
class parser(object):
- def __init__(self, tokenizer, elements, methods=None):
- self._tokenizer = tokenizer
+ def __init__(self, elements, methods=None):
self._elements = elements
self._methods = methods
self.current = None
@@ -72,12 +71,9 @@
if len(infix) == 3:
self._match(infix[2], pos)
return expr
- def parse(self, message, lookup=None):
- 'generate a parse tree from a message'
- if lookup:
- self._iter = self._tokenizer(message, lookup)
- else:
- self._iter = self._tokenizer(message)
+ def parse(self, tokeniter):
+ 'generate a parse tree from tokens'
+ self._iter = tokeniter
self._advance()
res = self._parse()
token, value, pos = self.current
@@ -87,9 +83,9 @@
if not isinstance(tree, tuple):
return tree
return self._methods[tree[0]](*[self.eval(t) for t in tree[1:]])
- def __call__(self, message):
- 'parse a message into a parse tree and evaluate if methods given'
- t = self.parse(message)
+ def __call__(self, tokeniter):
+ 'parse tokens into a parse tree and evaluate if methods given'
+ t = self.parse(tokeniter)
if self._methods:
return self.eval(t)
return t
--- a/mercurial/revset.py Sat Jun 20 20:11:53 2015 -0700
+++ b/mercurial/revset.py Sun Jun 21 00:49:26 2015 +0900
@@ -2387,9 +2387,9 @@
>>> _parsealiasdecl('foo($1, $2, $1)')
('foo', None, None, 'argument names collide with each other')
"""
- p = parser.parser(_tokenizealias, elements)
+ p = parser.parser(elements)
try:
- tree, pos = p.parse(decl)
+ tree, pos = p.parse(_tokenizealias(decl))
if (pos != len(decl)):
raise error.ParseError(_('invalid token'), pos)
@@ -2478,8 +2478,8 @@
pos)
yield (t, value, pos)
- p = parser.parser(tokenizedefn, elements)
- tree, pos = p.parse(defn)
+ p = parser.parser(elements)
+ tree, pos = p.parse(tokenizedefn(defn))
if pos != len(defn):
raise error.ParseError(_('invalid token'), pos)
return parser.simplifyinfixops(tree, ('or',))
@@ -2609,8 +2609,8 @@
return tuple(foldconcat(t) for t in tree)
def parse(spec, lookup=None):
- p = parser.parser(tokenize, elements)
- tree, pos = p.parse(spec, lookup=lookup)
+ p = parser.parser(elements)
+ tree, pos = p.parse(tokenize(spec, lookup=lookup))
if pos != len(spec):
raise error.ParseError(_("invalid token"), pos)
return parser.simplifyinfixops(tree, ('or',))
--- a/mercurial/templater.py Sat Jun 20 20:11:53 2015 -0700
+++ b/mercurial/templater.py Sun Jun 21 00:49:26 2015 +0900
@@ -27,8 +27,7 @@
"end": (0, None, None),
}
-def tokenizer(data):
- program, start, end = data
+def tokenize(program, start, end):
pos = start
while pos < end:
c = program[pos]
@@ -96,7 +95,7 @@
def compiletemplate(tmpl, context):
parsed = []
pos, stop = 0, len(tmpl)
- p = parser.parser(tokenizer, elements)
+ p = parser.parser(elements)
while pos < stop:
n = tmpl.find('{', pos)
if n < 0:
@@ -111,8 +110,7 @@
if n > pos:
parsed.append(('string', tmpl[pos:n]))
- pd = [tmpl, n + 1, stop]
- parseres, pos = p.parse(pd)
+ parseres, pos = p.parse(tokenize(tmpl, n + 1, stop))
parsed.append(parseres)
return [compileexp(e, context, methods) for e in parsed]