#!/usr/bin/env python # Copyright (C) 2009-2012 Chris Ball # Gianluca Montecchi # W. Trevor King # # This file is part of Bugs Everywhere. # # Bugs Everywhere is free software: you can redistribute it and/or modify it # under the terms of the GNU General Public License as published by the Free # Software Foundation, either version 2 of the License, or (at your option) any # later version. # # Bugs Everywhere is distributed in the hope that it will be useful, but # WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or # FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for # more details. # # You should have received a copy of the GNU General Public License along with # Bugs Everywhere. If not, see . """ Convert xml output of `be list --xml` into mbox format for browsing with a mail reader. For example $ be list --xml --status=all | be-xml-to-mbox | catmutt mbox is a flat-file format, consisting of a series of messages. Messages begin with a a From_ line, followed by RFC 822 email, followed by a blank line. """ # from mailbox import mbox, Message # the mailbox people really want an on-disk copy import email.utils from libbe.util.encoding import get_output_encoding from libbe.util.utility import str_to_time as rfc2822_to_gmtime_integer from time import asctime, gmtime import types from xml.etree import ElementTree from xml.sax.saxutils import unescape DEFAULT_DOMAIN = "invalid.com" DEFAULT_EMAIL = "dummy@" + DEFAULT_DOMAIN DEFAULT_ENCODING = get_output_encoding() def rfc2822_to_asctime(rfc2822_string): """Convert an RFC 2822-fomatted string into a asctime string. >>> rfc2822_to_asctime("Thu, 01 Jan 1970 00:00:00 +0000") "Thu Jan 01 00:00:00 1970" """ if rfc2822_string == "": return asctime(gmtime(0)) return asctime(gmtime(rfc2822_to_gmtime_integer(rfc2822_string))) class LimitedAttrDict(dict): """ Dict with error checking, to avoid invalid bug/comment fields. """ _attrs = [] # override with list of valid attribute names def __init__(self, **kwargs): dict.__init__(self) for key, value in kwargs.items(): self[key] = value def __setitem__(self, key, item): self._validate_key(key) dict.__setitem__(self, key, item) def _validate_key(self, key): if key in self._attrs: return elif type(key) not in types.StringTypes: raise TypeError("Invalid attribute type %s for '%s'" % (type(key), key)) else: raise ValueError("Invalid attribute name '%s'" % key) class Bug(LimitedAttrDict): _attrs = [u"uuid", u"short-name", u"severity", u"status", u"assigned", u"reporter", u"creator", u"created", u"summary", u"comments", u"extra-strings"] def print_to_mbox(self): if "creator" in self: # otherwise, probably a `be show` uuid-only bug to avoid # root comments. name, addr = email.utils.parseaddr(self["creator"]) print("From %s %s" % (addr, rfc2822_to_asctime(self["created"]))) print("Message-id: <%s@%s>" % (self["uuid"], DEFAULT_DOMAIN)) print("Date: %s" % self["created"]) print("From: %s" % self["creator"]) print("Content-Type: %s; charset=%s" % ("text/plain", DEFAULT_ENCODING)) print("Content-Transfer-Encoding: 8bit") print("Subject: %s: %s" % (self["short-name"], self["summary"])) if "extra-strings" in self: for estr in self["extra-strings"]: print("X-Extra-String: %s" % estr) print() print(self["summary"]) print() if "comments" in self: for comment in self["comments"]: comment.print_to_mbox(self) def init_from_etree(self, element): assert element.tag == "bug", element.tag for field in element.getchildren(): text = unescape(bytes(field.text).decode("unicode_escape").strip()) if field.tag == "comment": comm = Comment() comm.init_from_etree(field) if "comments" in self: self["comments"].append(comm) else: self["comments"] = [comm] elif field.tag == "extra-string": if "extra-strings" in self: self["extra-strings"].append(text) else: self["extra-strings"] = [text] else: self[field.tag] = text def wrap_id(id): if "@" not in id: return "<%s@%s>" % (id, DEFAULT_DOMAIN) return id class Comment(LimitedAttrDict): _attrs = [u"uuid", u"alt-id", u"short-name", u"in-reply-to", u"author", u"date", u"content-type", u"body", u"extra-strings"] def print_to_mbox(self, bug=None): if bug is None: bug = Bug() bug[u"uuid"] = u"no-uuid" name, addr = email.utils.parseaddr(self["author"]) print("From %s %s" % (addr, rfc2822_to_asctime(self["date"]))) if "uuid" in self: id = self["uuid"] elif "alt-id" in self: id = self["alt-id"] else: id = None if id is not None: print("Message-id: %s" % wrap_id(id)) if "alt-id" in self: print("Alt-id: %s" % wrap_id(self["alt-id"])) print("Date: %s" % self["date"]) print("From: %s" % self["author"]) subject = "" if "short-name" in self: subject += self["short-name"] + u": " if "summary" in bug: subject += bug["summary"] else: subject += u"no-subject" print("Subject: %s" % subject) if "in-reply-to" not in self.keys(): self["in-reply-to"] = bug["uuid"] print("In-Reply-To: %s" % wrap_id(self["in-reply-to"])) if "extra-strings" in self: for estr in self["extra-strings"]: print("X-Extra-String: %s" % estr) if self["content-type"].startswith("text/"): print("Content-Transfer-Encoding: 8bit") print("Content-Type: %s; charset=%s" % (self["content-type"], DEFAULT_ENCODING)) else: print("Content-Transfer-Encoding: base64") print("Content-Type: %s;" % (self["content-type"])) print() print(self["body"]) print() def init_from_etree(self, element): assert element.tag == "comment", element.tag for field in element.getchildren(): text = unescape(bytes(field.text).decode("unicode_escape").strip()) if field.tag == "extra-string": if "extra-strings" in self: self["extra-strings"].append(text) else: self["extra-strings"] = [text] else: if field.tag == "body": text += "\n" self[field.tag] = text def print_to_mbox(element): if element.tag == "bug": b = Bug() b.init_from_etree(element) b.print_to_mbox() elif element.tag == "comment": c = Comment() c.init_from_etree(element) c.print_to_mbox() elif element.tag in ["be-xml"]: for elt in element.getchildren(): print_to_mbox(elt) if __name__ == "__main__": import codecs import sys sys.stdin = codecs.getreader(DEFAULT_ENCODING)(sys.stdin) sys.stdout = codecs.getwriter(DEFAULT_ENCODING)(sys.stdout) if len(sys.argv) == 1: # no filename given, use stdin xml_unicode = sys.stdin.read() else: xml_unicode = codecs.open(sys.argv[1], "r", DEFAULT_ENCODING).read() xml_str = xml_unicode.encode("unicode_escape").replace(r"\n", "\n") elist = ElementTree.XML(xml_str) print_to_mbox(elist)