pmacs3/etags.py

"""
Etags parser

:author: Dan Williams
"""

import os
import re
from stat import ST_MTIME
from subprocess import Popen, PIPE, STDOUT

class EtagsRunError(Exception): pass

class TagManager(object):
    lang  = None
    prune = set(['SCCS', 'RCS', 'CVS', '.svn', '.hg', '.git', '.bzr'])
    exts  = set()

    def __init__(self, base='.'):
        self.db = None
        self.base  = base
        self.path  = os.path.join(self.base, 'TAGS')
        self.update()

    def has(self, name):
        return name in self.db.tags

    def get(self, name):
        return self.db.get(name)

    def list(self):
        return self.db.records

    def update(self):
        if self.is_outdated():
            self.run()
        self.db = Etags(self.path)
        self.db.parse()

    def run(self):
        lf      = '--language-force=%s' % self.lang
        args    = ['ctags', '-e', '-f', self.path, lf, '-L-']
        pipe    = Popen(args, stdin=PIPE, stdout=PIPE, stderr=STDOUT)
        indata  = '\n'.join(self.get_paths()) + '\n'
        outdata = pipe.communicate(indata)
        if pipe.returncode != 0:
            raise EtagsRunError(outdata)

    def get_paths(self):
        return list(self._walk(mtime=0))

    def is_outdated(self):
        if not os.path.exists(self.path):
            return True
        mtime = os.stat(self.path)[ST_MTIME]
        itr   = self._walk(mtime)
        try:
            itr.next()
            return True
        except StopIteration:
            return False

    def _walk(self, mtime=0):
        paths = []
        for root, dirs, files in os.walk(self.base):
            for d in dirs:
                if d in self.prune:
                    dirs.remove(d)
            for f in files:
                path = os.path.join(root, f)
                if not self._match(path):
                    continue
                elif os.stat(path)[ST_MTIME] < mtime:
                    continue
                else:
                    yield os.path.join(root, f)
        raise StopIteration

    def _match(self, path):
        _, ext = os.path.splitext(path)
        return ext in self.exts

class Etags(object):
    def __init__(self, fname=None):
        self.fname = fname
        self.rawdata = None
        self.records = []
        self.tags = {}

    def _load(self):
        fd = file(self.fname, 'r')
        self.rawdata = fd.read()
        fd.close()

    def get(self, name):
        return self.tags.get(name, [])

    def parse(self, fname=None):
        """
        Parser is based on the little info found in
        Wikipedia: http://en.wikipedia.org/wiki/Ctags
        """
        if fname:
            self.fname = fname
        self._load()
        i = 0
        data_len = len(self.rawdata)
        data = self.rawdata
        while i < data_len:
            if ord(data[i]) == 0xc:
                i = self._parse_block(data, i+2)

    def _add_record(self, record):
        self.records.append(record)
        name = record.name
        if name is None:
            return
        self.tags.setdefault(name, [])
        self.tags[name].append(record)

    def _parse_block(self, data, i):
        n = data[i:].find('\n') + i
        l = data[i:n]
        try:
            filename, size = l.split(',')
        except ValueError, e:
            raise Exception("parse failed(%s): %r %r %r" % (i, l, e, data))
        size = int(size)
        subblock = data[n+1:n+size+1]

        # ...
        for lineitem in subblock.split('\n'):
            if len(lineitem) == 0: continue
            record = self._parse_record(lineitem, filename)
            self._add_record(record)
        return n + size + 1

    def _parse_record(self, lineitem, filename):
        try:
            defn, rest = lineitem.split(chr(0x7f))
        except ValueError:
            print lineitem
            raise
        name = None
        if chr(0x01) in rest:
            name, rest = rest.split(chr(0x01))
        else:
            txt = defn.strip()
            sp = re.split('[ ,;*()\t&=]', txt)
            sp = [x for x in sp if x != '']
            if len(sp):
                name = sp[-1]

        tokens = rest.split(',')
        line   = int(tokens[0])
        byte   = int(tokens[1])

        record = EtagRecord(path=filename, defn=defn, name=name, line=line, byte=byte)
        return record
        
class EtagRecord(object):
    def __init__(self, **kwargs):
        self.path = None
        self.defn = None
        self.name = None
        self.line = -1
        self.byte = None
        self.__dict__.update(kwargs)

    def __repr__(self):
        return "%s [%s:%d]" % (self.name, self.path, self.line)


if __name__ == '__main__':
    import sys 
    from pprint import pprint

    etags = Etags(sys.argv[1])
    etags.parse()

    if len(sys.argv) > 2:
        print etags.get(sys.argv[2])
    else:
        pprint(etags.records)