import re, sets, string, sys import color, commands, default, lex2, method, mode2, regex, tab_perl from point2 import Point from lex2 import Grammar, ConstantRule, PatternRule, ContextPatternRule, \ RegionRule, DualRegionRule class PodGrammar(Grammar): rules = [ PatternRule(name=r'entry', pattern=r'(?<=^=head[1-4]) +.*$'), PatternRule(name=r'entry', pattern=r'(?<=^=over) +.*$'), PatternRule(name=r'entry', pattern=r'(?<=^=item) +.*$'), PatternRule(name=r'entry', pattern=r'(?:(?<=^=begin)|(?<=^=end)) +.*$'), PatternRule(name=r'entry', pattern=r'(?<=^=encoding) +.*$'), ] class StringGrammar(Grammar): rules = [ PatternRule(name=r'escaped', pattern=r'\\.'), PatternRule(name=r'deref', pattern=r"\$\$*[A-Za-z0-9_](?:[A-Za-z0-9_]|::)*(?:->{\$?(?:[a-zA-Z_][a-zA-Z_0-9]*|'(?:\\.|[^'\\])*'|\"(\\.|[^\\\"])*\")}|->\[\$?[0-9a-zA-Z_]+\])+"), PatternRule(name=r'length', pattern=r"\$#[A-Za-z0-9_](?:[A-Za-z0-9_]|::)*"), ContextPatternRule(name=r'scalar', pattern=r"\$[^A-Za-z0-9 %(delim)s](?![A-Za-z0-9_])", fallback=r"\$[^A-Za-z0-9 ](?![A-Za-z0-9_])"), PatternRule(name=r'scalar', pattern=r"\$\$*[A-Za-z0-9_](?:[A-Za-z0-9_]|::)*"), PatternRule(name=r'cast', pattern=r"[\$\@\%\&]{.*?}"), PatternRule(name=r'array', pattern=r"@\$*[A-Za-z_](?:[A-Za-z0-9_]|::)*"), PatternRule(name=r'hash', pattern=r"%\$*[A-Za-z_](?:[A-Za-z0-9_]|::)*"), ], g = Grammar() pg = PodGrammar() sg = StringGrammar() class PerlGrammar(Grammar): rules = [ RegionRule(name=r'heredoc1', start=r"<<(?P[a-zA-Z0-9_]+) *;", grammar=sg, end=r'^%(heredoc)s$'), RegionRule(name=r'heredoc1', start=r'<< *"(?P[a-zA-Z0-9_]+)" *;', grammar=sg, end=r'^%(heredoc)s$'), RegionRule(name=r'heredoc2', start=r"<< *'(?P[a-zA-Z0-9_]+)' *;", grammar=g, end=r'^%(heredoc)s$'), RegionRule(name=r'eval_heredoc', start=r"<< *`(?P[a-zA-Z0-9_]+)` *;", grammar=sg, end=r'^%(heredoc)s$'), RegionRule(name=r'endblock', start=r"^__END__|__DATA__ *$", grammar=g, end=r''), RegionRule(name=r'pod', start=r'^=[a-zA-Z0-9_]+', grammar=pg, end=r'^=cut'), PatternRule(name=r'comment', pattern=r'#.*$'), RegionRule(name=r'string1', start=r'"', grammar=sg, end=r'"'), RegionRule(name=r'string2', start=r"'", grammar=g, end=r"'"), RegionRule(name=r'evalstring', start=r"`", grammar=sg, end=r"`"), PatternRule(name=r'number', pattern=r'0?\.[0-9]+|[0-9]+(?:\.[0-9]+)?'), PatternRule(name=r'keyword', pattern=r"(?)(?:STDIN|STDERR|STDOUT|and|cmp|continue|do|else|elsif|eq|eval|foreach|for|if|last|my|next|ne|not|or|our|package|require|return|sub|undef|unless|until|use|while)(?![a-zA-Z0-9_])"), PatternRule(name=r'hash_key', pattern=r'(?<={)[A-Za-z0-9_]+(?=})'), PatternRule(name=r'hash_key', pattern=r'[A-Za-z0-9_]+(?= *=>)'), PatternRule(name=r'length', pattern=r"\$#[A-Za-z0-9_](?:[A-Za-z0-9_]|::)*"), PatternRule(name=r'cast', pattern=r'[\$\@\%\^\&](?= *{)'), PatternRule(name=r'scalar', pattern=r"\$[][> *$)"), PatternRule(name=r'scalar', pattern=r"\$\$*[A-Za-z0-9_](?:[A-Za-z0-9_]|::)*"), PatternRule(name=r'array', pattern=r"@\$*[A-Za-z_](?:[A-Za-z0-9_]|::)*"), PatternRule(name=r'hash', pattern=r"%\$*[A-Za-z_](?:[A-Za-z0-9_]|::)*"), PatternRule(name=r'deref', pattern=r"[@%\$&\*](?={)"), RegionRule(name=r'quoted', start=r'q[rqwx]? *\(', grammar=g, end=r'$'), RegionRule(name=r'quoted', start=r'q[rqwx]? *{', grammar=g, end=r'}'), RegionRule(name=r'quoted', start=r'q[rqwx]? *<', grammar=g, end=r'>'), RegionRule(name=r'quoted', start=r'q[rqwx]? *\[', grammar=g, end=r'\]'), RegionRule(name=r'quoted', start=r'q[rqwx]? *(?P[^ #])', grammar=g, end=r'%(delim)s'), RegionRule(name=r'quoted', start=r'q[rqwx]?#', grammar=g, end=r'#'), # match regexes RegionRule(name=r'match', start=r'(?:(?<==~)|(?<=!~)|(?<=\()|(?<=split)) *(?P/)', grammar=sg, end=r'/[a-z]*'), RegionRule(name=r'match', start=r'm *(?P[^ #a-zA-Z0-9_])', grammar=sg, end=r'%(delim)s[a-z]*'), RegionRule(name=r'match', start=r'm(?P#)', grammar=sg, end=r'#[a-z]*'), # replace regexes DualRegionRule(name=r'replace', start=r's *(?P[^ a-zA-Z0-9_])', grammar1=sg, middle=r'%(delim)s', grammar2=sg, end=r'%(delim)s[a-z]*'), DualRegionRule(name=r'replace', start=r's(?P#)', grammar1=sg, middle=r'#', grammar2=sg, end=r'#[a-z]*'), # translate operator DualRegionRule(name=r'translate', start=r'(?:y|tr) *(?P[^ a-zA-Z0-9_])', grammar1=g, middle=r'%(delim)s', grammar2=g, end=r'%(delim)s[a-z]*'), DualRegionRule(name=r'translate', start=r'(?:y|tr)#', grammar1=g, middle=r'#', grammar2=g, end=r'#[a-z]*'), # some more basic stuff PatternRule(name=r'package', pattern=r"(?<=package )(?:[a-zA-Z_][a-zA-Z_0-9]*::)*[a-zA-Z_][a-zA-Z_0-9]*"), PatternRule(name=r'sub', pattern=r"(?<=sub )[a-zA-Z_][a-zA-Z_0-9]*"), PatternRule(name=r'use', pattern=r"(?<=use )(?:[a-zA-Z_][a-zA-Z_0-9]*::)*[a-zA-Z_][a-zA-Z_0-9]*"), PatternRule(name=r'label', pattern=r'[a-zA-Z_][a-zA-Z0-9_]*:(?!:)'), PatternRule(name=r'method', pattern=r"(?<=->)[a-zA-Z_][a-zA-Z_0-9]*"), PatternRule(name=r'function', pattern=r"&\$*(?:[a-zA-Z_][a-zA-Z_0-9]*::)*[a-zA-Z_][a-zA-Z_0-9]*"), PatternRule(name=r'function', pattern=r"(?:[a-zA-Z_][a-zA-Z_0-9]*::)*[a-zA-Z_][a-zA-Z_0-9]*(?= *\()"), PatternRule(name=r'builtin', pattern=r"(?)&?(?:write|warn|wantarray|waitpid|wait|vec|values|utime|use|untie|unshift|unpack|unlink|undef|umask|ucfirst|uc|truncate|times|time|tied|tie|telldir|tell|syswrite|system|sysseek|sysread|sysopen|syscall|symlink|substr|sub|study|stat|srand|sqrt|sprintf|split|splice|sort|socketpair|socket|sleep|sin|shutdown|shmwrite|shmread|shmget|shmctl|shift|setsockopt|setservent|setpwent|setprotoent|setpriority|setpgrp|setnetent|sethostent|setgrent|send|semop|semget|semctl|select|seekdir|seek|scalar|rmdir|rindex|rewinddir|reverse|return|reset|require|rename|ref|redo|recv|readpipe|readlink|readline|readdir|read|rand|quotemeta|push|prototype|printf|print|pos|pop|pipe|package|pack|our|ord|opendir|open|oct|no|next|my|msgsnd|msgrcv|msgget|msgctl|mkdir|map|lstat|log|lock|localtime|local|listen|link|length|lcfirst|lc|last|kill|keys|join|ioctl|int|index|import|hex|grep|goto|gmtime|glob|getsockopt|getsockname|getservent|getservbyport|getservbyname|getpwuid|getpwnam|getpwent|getprotoent|getprotobynumber|getprotobyname|getpriority|getppid|getpgrp|getpeername|getnetent|getnetbyname|getnetbyaddr|getlogin|gethostent|gethostbyname|gethostbyaddr|getgrnam|getgrgid|getgrent|getc|formline|format|fork|flock|fileno|fcntl|exp|exit|exists|exec|eval|eof|endservent|endpwent|endprotoent|endnetent|endhostent|endgrent|each|dump|do|die|delete|defined|dbmopen|dbmclose|crypt|cos|continue|connect|closedir|close|chroot|chr|chown|chop|chomp|chmod|chdir|caller|bless|binmode|bind|atan2|alarm|accept|abs)(?![a-zA-Z0-9_])"), PatternRule(name=r'class', pattern=r"(?:[a-zA-Z_][a-zA-Z_0-9]*::)*[a-zA-Z_][a-zA-Z_0-9]*(?=->)"), # nested regions RegionRule(name=r'paren', start=r'$', grammar=None, end=r'$'), RegionRule(name=r'brace', start=r'{', grammar=None, end=r'}'), RegionRule(name=r'bracket', start=r'\[', grammar=None, end=r'\]'), # some basic stuff #PatternRule(name=r'delimiter', pattern=r",|;|->|=>|=|\?|(?|=>|=|\?|$|$|{|}|\[|\](?|=>|(?>=|<<=|\*\*="), PatternRule(name=r'operator', pattern=r"\+|<=>|<>|<<|<=|<|-|>>|>=|>|\*\*|&|\*|\||/|\^|==|//|~|=~|!~|!=|%|!|\."), PatternRule(name=r'bareword', pattern=r'(?:[a-zA-Z_][a-zA-Z_0-9]*::)*[a-zA-Z_][a-zA-Z_0-9]*'), ] class Perl(mode2.Fundamental): def __init__(self, w): mode2.Fundamental.__init__(self, w) self.tag_matching = True self.grammar = lex2_perl.PerlGrammar() self.lexer = lex2.Lexer(self.name(), self.grammar) self.add_action_and_bindings(PerlCheckSyntax(), ('C-c s',)) self.add_action_and_bindings(PerlHashCleanup(), ('C-c h',)) #self.add_action_and_bindings(PerlHashCleanup2(), ('C-c h',)) self.add_action_and_bindings(PerlViewModulePerldoc(), ('C-c v',)) self.add_action_and_bindings(PerlViewWordPerldoc(), ('C-c p',)) self.add_action_and_bindings(PerlWrapLine(), ('M-q',)) self.add_action_and_bindings(PerlGotoFunction(), ('C-c M-g',)) self.add_action_and_bindings(PerlWhichFunction(), ('C-c w',)) self.add_action_and_bindings(PerlListFunctions(), ('C-c W',)) self.default_color = color.build('default', 'default') self.colors = { # basic stuff 'escaped': color.build('magenta', 'default'), 'null': color.build('default', 'default'), 'delimiter': color.build('default', 'default'), 'sub': color.build('cyan', 'default'), 'number': color.build('default', 'default'), 'operator': color.build('default', 'default'), 'endblock': color.build('red', 'default'), 'keyword': color.build('magenta', 'default'), 'cast': color.build('yellow', 'default'), 'scalar': color.build('yellow', 'default'), 'array': color.build('yellow', 'default'), 'deref': color.build('yellow', 'default'), 'hash': color.build('yellow', 'default'), 'hash_key': color.build('green', 'default'), 'comment': color.build('red', 'default'), 'function': color.build('cyan', 'default'), 'builtin': color.build('magenta', 'default'), 'method': color.build('cyan', 'default'), 'bareword': color.build('default', 'default'), 'label': color.build('cyan', 'default'), 'package': color.build('cyan', 'default'), 'class': color.build('cyan', 'default'), 'use': color.build('cyan', 'default'), 'method': color.build('cyan', 'default'), # heredoc 'heredoc1.start': color.build('green', 'default'), 'heredoc1.null': color.build('green', 'default'), 'heredoc1.end': color.build('green', 'default'), 'heredoc2.start': color.build('green', 'default'), 'heredoc2.null': color.build('green', 'default'), 'heredoc2.end': color.build('green', 'default'), 'eval_heredoc.start': color.build('cyan', 'default'), 'eval_heredoc.null': color.build('cyan', 'default'), 'eval_heredoc.end': color.build('cyan', 'default'), # pod 'pod.start': color.build('red', 'default'), 'pod.null': color.build('red', 'default'), 'pod.entry': color.build('magenta', 'default'), 'pod.end': color.build('red', 'default'), # "" strings 'string1.start': color.build('green', 'default'), 'string1.null': color.build('green', 'default'), 'string1.escaped': color.build('magenta', 'default'), 'string1.deref': color.build('yellow', 'default'), 'string1.end': color.build('green', 'default'), # '' strings 'string2.start': color.build('green', 'default'), 'string2.null': color.build('green', 'default'), 'string2.end': color.build('green', 'default'), # `` strings 'evalstring.start.': color.build('cyan', 'default'), 'evalstring.null': color.build('cyan', 'default'), 'string1.escaped': color.build('magenta', 'default'), 'string1.deref': color.build('yellow', 'default'), 'evalstring.end': color.build('cyan', 'default'), # quoted region 'quoted': color.build('cyan', 'default'), 'quoted.start': color.build('cyan', 'default'), 'quoted.null': color.build('cyan', 'default'), 'quoted.end': color.build('cyan', 'default'), # match regex 'match.start': color.build('cyan', 'default'), 'match.end': color.build('cyan', 'default'), 'match.null': color.build('cyan', 'default'), # replace regex 'replace.start': color.build('cyan', 'default'), 'replace.middle': color.build('cyan', 'default'), 'replace.end': color.build('cyan', 'default'), 'replace.null': color.build('cyan', 'default'), 'replace.escaped': color.build('magenta', 'default'), 'replace.deref': color.build('yellow', 'default'), 'replace.length': color.build('yellow', 'default'), 'replace.scalar': color.build('yellow', 'default'), 'replace.hash': color.build('yellow', 'default'), 'replace.cast': color.build('yellow', 'default'), # translate regex 'translate.start': color.build('magenta', 'default'), 'translate.middle': color.build('magenta', 'default'), 'translate.end': color.build('magenta', 'default'), 'translate.null': color.build('magenta', 'default'), } def name(self): return "Perl" def build_function_map(self): b = self.window.buffer self.functions = {} for i in range(0, len(b.lines)): m = regex.perl_function.match(b.lines[i]) if m: self.functions[m.group(1)] = i def get_functions(self): if self.functions is None: self.build_function_map() return self.functions def get_function_names(self): functions = self.get_functions() pairs = [[functions[key], key] for key in functions] pairs.sort() names = [x[1] for x in pairs] return names class PerlWrapLine(method.Method): '''Wrap lines, comments, POD''' margin = 80 comment_re = re.compile('^( *)(#+)( *)([^ ].*)$') def _execute(self, w, **vargs): pcursor = w.physical_cursor() r = w.get_region(pcursor) if r is None: return t = r[4] if t == 'pod': assert False, 'POD: %s' % repr(r) elif t == 'comment': self._wrap_comment(w) else: return def _wrap_comment(self, w): l = w.logical_cursor() m = self.comment_re.match(w.buffer.lines[l.y]) if not m: assert False, 'no match oh geez' pad = m.group(1) + m.group(2) + m.group(3) data = m.group(4) + ' ' start = l.y end = l.y + 1 while end < len(w.buffer.lines): m = self.comment_re.match(w.buffer.lines[end]) if m: data += m.group(4) + ' ' end += 1 else: break words = [word for word in data.split() if word] lines = [pad] for word in words: if len(lines[-1]) == len(pad): lines[-1] += word elif len(lines[-1]) + 1 + len(word) <= self.margin: lines[-1] += ' ' + word else: lines.append(pad + word) # remove the old text and add the new start_p = Point(0, start) end_p = Point(len(w.buffer.lines[end-1]), end-1) w.kill(start_p, end_p) w.insert(start_p, '\n'.join(lines)) class PerlCheckSyntax(method.Method): '''Check the syntax of a perl file''' def _args(self): return [method.Argument("lib", type=type(""), prompt="Location of lib: ", default=default.build_constant("."))] def _execute(self, window, **vargs): a = vargs['lib'] cmd = "perl -c -I '%s' '%s'" % (a, window.buffer.path) (status, output) = commands.getstatusoutput(cmd) if status == 0: window.application.set_error("Syntax OK") window.application.data_buffer("*Perl-Check-Syntax*", output, switch_to=False) else: window.application.data_buffer("*Perl-Check-Syntax*", output) class PerlViewModulePerldoc(method.Method): '''View documentation about this file using perldoc''' def _execute(self, w, **vargs): cmd = "perldoc -t -T '%s'" % w.buffer.path (status, output) = commands.getstatusoutput(cmd) w.application.data_buffer("*Perldoc*", output, switch_to=True) class PerlViewWordPerldoc(method.Method): '''View documentation about a package or function using perldoc''' def _execute(self, w, **vargs): cursor = w.logical_cursor() line = w.buffer.lines[cursor.y] word_chars = string.letters + string.digits + '_:' if line[cursor.x] not in word_chars: w.application.set_error('error: no word selected') return start = cursor.x while start > 0 and line[start - 1] in word_chars: start -= 1 end = cursor.x + 1 while end < len(line) - 1 and line[end] in word_chars: end += 1 word = line[start:end] w.application.set_error('the current word is: %r' % word) ok = False data = '' perl_word_re = re.compile('^[a-zA-Z_][a-zA-Z_0-9]*(?:::[a-zA-Z_][a-zA-Z0-9]*)*$') if not perl_word_re.match(word): w.application.set_error('invalid word: %r' % word) return if '::' in word: # we are probably dealing with a package parts = word.split('::') while len(parts) > 0: newword = '::'.join(parts) cmd = "perldoc -t -T '%s'" % newword (status, data) = commands.getstatusoutput(cmd) if status == 0: word = newword ok = True break parts.pop(-1) elif ':' in word: w.application.set_error('invalid word2222: %r' % word) return else: cmd = "perldoc -t -T -f '%s'" % word (status, data) = commands.getstatusoutput(cmd) if status == 0: ok = True else: cmd = "perldoc -t -T -f '%s'" % word (status, data) = commands.getstatusoutput(cmd) ok = status == 0 if not ok: w.application.set_error('nothing found for %r' % word) else: w.application.data_buffer("*Perldoc*", data, switch_to=True) w.application.set_error('displaying documentation for %r' % word) class PerlGotoFunction(method.Method): '''Jump to a function defined in this module''' def _args(self): return [method.Argument("name", type=type(""), datatype="perlfunction", prompt="Goto Function: ")] def _execute(self, w, **vargs): name = vargs['name'] functions = w.mode.get_functions() if name in functions: number = functions[name] p = Point(0, number) w.goto(p) else: w.application.set_error("Function %r was not found" % name) class PerlListFunctions(method.Method): '''Show the user all functions defined in this module''' def _execute(self, w, **vargs): names = w.mode.get_function_names() output = "\n".join(names) + "\n" w.application.data_buffer("*Perl-List-Functions*", output, switch_to=True) class PerlWhichFunction(method.Method): '''Show the user what function they are in''' def _execute(self, w, **vargs): cursor = w.logical_cursor() i = cursor.y name = None while i >= 0 and name is None: line = w.buffer.lines[i] m = regex.perl_function.match(line) if m: name = m.group(1) else: i -= 1 if name is None: w.application.set_error("None"); else: w.application.set_error("line %d: %s" % (i, name)) class PerlHashCleanup(method.Method): '''Correctly align assignment blocks and literal hashes''' def _execute(self, window, **vargs): cursor = window.logical_cursor() b = window.buffer # so this is where we will store the groups that we find groups_by_line = {} # the regex we will try regexes = [regex.perl_hash_cleanup, regex.perl_assign_cleanup] # if we aren't in a hash, inform the user and exit line = b.lines[cursor.y] myregex = None for r in regexes: if r.match(line): myregex = r if myregex is None: raise Exception, "Not a perl hash line" groups_by_line[cursor.y] = myregex.match(line).groups() # find the beginning of this hash block start = 0 i = cursor.y - 1 while i >= 0: line = b.lines[i] m = myregex.match(line) if not m: start = i + 1 break else: groups_by_line[i] = m.groups() i -= 1 # find the end of this hash block end = len(b.lines) - 1 i = cursor.y + 1 while i < len(b.lines): line = b.lines[i] m = myregex.match(line) if not m: end = i - 1 break else: groups_by_line[i] = m.groups() i += 1 # assume that the least indented line is correct indent_w = min([len(groups_by_line[k][0]) for k in groups_by_line]) # find the longest hash key to base all the other padding on key_w = max([len(groups_by_line[k][1]) for k in groups_by_line]) # for each line, format it correctly keys = groups_by_line.keys() keys.sort() data = '' for i in keys: indent_pad = ' ' * indent_w key = groups_by_line[i][1] sep = groups_by_line[i][3] value = groups_by_line[i][5] key_pad = ' ' * (key_w - len(key)) data += indent_pad + key + key_pad + ' ' + sep + ' ' + value + '\n' # remove the old text and add the new start_p = Point(0, start) end_p = Point(0, end + 1) window.kill(start_p, end_p) window.insert_string(start_p, data) class PerlHashCleanup2(method.Method): '''Correctly align assignment blocks and literal hashes''' def process_line2(self, line_regions, sep=None, indent=None): (pre_toks, sep_tok, post_toks) = ([], None, []) ok = False before = True for r in line_regions: (start, end, attr, s, name) = r if name == "": continue elif before: if len(pre_toks) == 0: pre_toks.append(r) elif (name == "delimiter" and s == sep or (sep is None and (s == "=" or s == "=>"))): sep_tok = r before = False else: pre_toks.append(r) else: post_toks.append(r) ok = True if ok: return (True, sep_tok[3], (pre_toks, sep_tok, post_toks)) else: return (False, "", ([], None, [])) def _execute(self, w, **vargs): cursor = w.logical_cursor() b = w.buffer # so this is where we will store the groups that we find groups_by_line = {} all_regions = w.mode.get_regions() line_regions = all_regions[cursor.y] (ok, sep, group) = self.process_line2(line_regions) if not ok: raise Exception, "Not a perl hash line" groups_by_line[cursor.y] = group # find the beginning of this hash block start = cursor.y while start >= 0: (ok2, sep2, group2) = self.process_line2(all_regions[start - 1], sep) if not ok2: break start -= 1 groups_by_line[start] = group2 # find the end of this hash block end = cursor.y while end < len(b.lines) - 1: (ok2, sep2, group2) = self.process_line2(all_regions[end + 1], sep) if not ok2: break end += 1 groups_by_line[end] = group2 # find the minimum indented line indent_w = None for k in groups_by_line: x = groups_by_line[k][0][0].start if indent_w is None or x < indent_w: indent_w = x # find the max key length key_w = None for k in groups_by_line: x = groups_by_line[k][0][-1].end - groups_by_line[k][0][0].start if key_w is None or x > key_w: key_w = x # for each line, format it correctly keys = groups_by_line.keys() keys.sort() data = '' for i in keys: line = ' ' * indent_w l = groups_by_line[i][0][0].start for t in groups_by_line[i][0]: line += ' ' * max(0, t.start - l) line += t.value l = t.end line += ' ' * max(0, key_w - l + groups_by_line[i][0][0].start) line += ' ' + groups_by_line[i][1].value + ' ' l = groups_by_line[i][2][0].start for t in groups_by_line[i][2]: line += ' ' * max(0, t.start - l) line += t.value l = t.end data += line + '\n' # remove the old text and add the new start_p = Point(0, start) end_p = Point(0, end + 1) w.kill(start_p, end_p) w.insert(start_p, data)