Recent changes to bugs

Help please

2003-08-30T17:18:27Z

http://www.my3q.com/home2/21/cluo/88170.phtml

Please forgive me to borther you with my short on-line
questionnaire, although I've known that you probably
have received the mails like this before.

Currently, I am a student of Durham Business School
(UK), striving on completing the writing of my MA
dissertation. I just need to borrow your
SEVEN minutes (I promise, have timed it)to help me run
to the completion of my
dissertation. Please............

The aim of this questionnaire is to understand how
people reflect on the
importance of trust and collaboration to the success of
projects (e.g.
Open Source Software Development) carried out in a
virtual environment.

And that will be much much appreaciated if you could
forward the above questionnaire link to your project
members, colleagues, and friends.

Thank you.............

Best regards,
Charles Luo

File publish broken under Linux Opera 5

2001-11-16T14:42:43Z

When choosing a file to publish Opera adds quotation marks around the filename, which are not parsed properly. The result is:

ERROR: File does not exist:
""Somefile""

Tested under Opera 5.0 for Linux - 20010510 Build 024 -[5]

This may be due to Opera allowing you to select multiple files simultaneously for publishing.

Python exception on Open/Save

2001-11-08T04:08:17Z

When I attempt to Open/Save something of interest the
attempt fails (no page displayed in the browser). In
the log the following is printed:

2001-11-07_23:37:40 (MojoHTTPD) ERROR: do_GET caught
this exception:
2001-11-07_23:37:40 () Traceback (most recent call
last):
2001-11-07_23:37:40 () File "mojohttpd.pyo", line 337,
in do_GET
2001-11-07_23:37:40 () File "mojohttpd.pyo", line 439,
in _handle_request
2001-11-07_23:37:40 () File "mojohttpd.pyo", line
1002, in handle_mojoid_get
2001-11-07_23:37:40 () File "mojohttpd.pyo", line
1027, in handle_dinode_stream
2001-11-07_23:37:40 () File "mojohttpd.pyo", line
1376, in download_dinode
2001-11-07_23:37:40 () File "unsplit.pyo", line 212,
in __init__
2001-11-07_23:37:40 () File "unsplit.pyo", line 258,
in open
2001-11-07_23:37:40 () File "sharemap.pyo", line 453,
in parse_sharemap
2001-11-07_23:37:40 () File "xml\sax\__init__.pyo",
line 41, in parseString
2001-11-07_23:37:40 () File "xml\sax\sax2exts.pyo",
line 34, in make_parser
2001-11-07_23:37:40 () File "xml\sax\saxexts.pyo",
line 77, in make_parser
2001-11-07_23:37:40 () SAXReaderNotAvailable: No
parsers found

I've read in the newsgroups that this has something to
do with a missing module (ie: expat). Is there
something I'm missing that didn't come with the
Windows 2000 .MSI install?

Publish by proxy

2001-09-10T15:28:13Z

As I understand it, when I publish something, I
duplicate the data I publish so that I end up sending
out twice as much data as I'm actually publishing. If
I'm on a slow link, that stinks.

I'd like to be able to send a block to a broker on a
fast link and ask it to duplicate it. That way, my
slow link isn't used as much. I send it out once, and
it ends up on the net twice.

As a requesting broker, I'd send the mojo needed to
cover the publication plus a little more for the
server's trouble.

As a serving broker, I need to make sure I don't spend
more on publication than was offered for the service.

restarting downloads

2001-09-06T13:03:07Z

Mojonation doesn't seem to save sufficient state to
re-start a download over a broker re-start. Further,
Mojonation doesn't notice the existing file when I
attempt to manually restart it, which leads to a great
deal of frustration as I can never successfully
download extrememly large files as they never complete
before the broker hangs of I run out of VM.

Won't compile: cryptlib.cpp

2001-05-26T16:21:59Z

I've tried compiling MojoNation from CVS, but
compilation stops with these errors:

g++ -fpermissive -O2 -w -pipe -c cryptlib.cpp
cryptlib.cpp: In method `long unsigned int
CryptoPP::BufferedTransformation::MaxRetrievable ()
const':
cryptlib.cpp:169: cannot allocate an object of type
`CryptoPP::BufferedTransformation'
cryptlib.cpp:169: since the following virtual
functions are abstract:
cryptlib.cpp:100: void
CryptoPP::BufferedTransformation::Put (const
byte *, unsigned int)
cryptlib.cpp:94: void
CryptoPP::BufferedTransformation::Put (unsigned
char)
cryptlib.cpp:169: could not convert
`BufferedTransformation((&BitBucket()))' to
`CryptoPP::BufferedTransformation &'
cryptlib.h:318: in passing argument 1 of
`CryptoPP::BufferedTransformation::CopyTo
(CryptoPP::BufferedTransformation &, long unsigned
int) const'
cryptlib.cpp: In method `unsigned int
CryptoPP::BufferedTransformation::Get (byte *,
unsigned int)':
cryptlib.cpp:196: cannot allocate an object of type
`CryptoPP::BufferedTransformation'
cryptlib.cpp:196: since type
`CryptoPP::BufferedTransformation' has
abstract virtual functions
cryptlib.cpp:196: could not convert
`BufferedTransformation((&ArraySink(outString,
getMax)))' to
`CryptoPP::BufferedTransformation &'
cryptlib.h:312: in passing argument 1 of
`CryptoPP::BufferedTransformation::TransferTo
(CryptoPP::BufferedTransformation &, long unsigned
int)'
cryptlib.cpp: In method `unsigned int
CryptoPP::BufferedTransformation::Peek (byte *,
unsigned int) const':
cryptlib.cpp:212: cannot allocate an object of type
`CryptoPP::BufferedTransformation'
cryptlib.cpp:212: since type
`CryptoPP::BufferedTransformation' has
abstract virtual functions
cryptlib.cpp:212: could not convert
`BufferedTransformation((&ArraySink(outString,
peekMax)))' to
`CryptoPP::BufferedTransformation &'
cryptlib.h:318: in passing argument 1 of
`CryptoPP::BufferedTransformation::CopyTo
(CryptoPP::BufferedTransformation &, long unsigned
int) const'
cryptlib.cpp: In method `long unsigned int
CryptoPP::BufferedTransformation::Skip (long unsigned
int)':
cryptlib.cpp:220: cannot allocate an object of type
`CryptoPP::BufferedTransformation'
cryptlib.cpp:220: since type
`CryptoPP::BufferedTransformation' has
abstract virtual functions
cryptlib.cpp:220: could not convert
`BufferedTransformation((&BitBucket()))' to
`CryptoPP::BufferedTransformation &'
cryptlib.h:312: in passing argument 1 of
`CryptoPP::BufferedTransformation::TransferTo
(CryptoPP::BufferedTransformation &, long unsigned
int)'
cryptlib.cpp: In method `unsigned int
CryptoPP::BufferedTransformation::NumberOfMessages ()
const':
cryptlib.cpp:252: cannot allocate an object of type
`CryptoPP::BufferedTransformation'
cryptlib.cpp:252: since type
`CryptoPP::BufferedTransformation' has
abstract virtual functions
cryptlib.cpp:252: could not convert
`BufferedTransformation((&BitBucket()))' to
`CryptoPP::BufferedTransformation &'
cryptlib.h:340: in passing argument 1 of
`CryptoPP::BufferedTransformation::CopyMessagesTo
(CryptoPP::BufferedTransformation &, unsigned int)
const'
cryptlib.cpp: In method `unsigned int
CryptoPP::BufferedTransformation::SkipMessages
(unsigned int)':
cryptlib.cpp:276: cannot allocate an object of type
`CryptoPP::BufferedTransformation'
cryptlib.cpp:276: since type
`CryptoPP::BufferedTransformation' has
abstract virtual functions
cryptlib.cpp:276: could not convert
`BufferedTransformation((&BitBucket()))' to
`CryptoPP::BufferedTransformation &'
cryptlib.h:338: in passing argument 1 of
`CryptoPP::BufferedTransformation::TransferMessagesTo
(CryptoPP::BufferedTransformation &, unsigned int)'
make[2]: *** [cryptlib.o] Fehler 1
make[2]: Verlassen des Verzeichnisses Verzeichnis
ť/home/guido/extsrc/cryptoppŤ
make[1]: *** [inst_cryptopp_40] Fehler 2
make[1]: Verlassen des Verzeichnisses Verzeichnis
ť/home/guido/evilŤ
gmake: ***
[/home/guido/extsrc//cryptopp/libcryptopp.a] Fehler 2

I'm running Linux Mandrake 8.0 with Python 2.0 and
gcc 2.96

Downloads should not be added twice

2001-05-25T10:23:52Z

It should not be possible to add a download twice.
Clicking a link to a mojoID a second time, adds the
download again.
Im not sure if this is just a cosmetic flaw or if the
file (or parts of it) are really downloaded twice.

Wrong Invalid Privacy Code Error

2001-05-21T13:19:17Z

Linux version (beta 980 redhat package running on a
slackware) is unable to get right privacy code on
files.

I'm getting an "invalid privacy code" error, if i try
the win version on the same files it works well.

Excessive memory use

2001-04-17T19:46:37Z

I didn't see this in the FAQ, nor in the list of open
bugs.

I left my broker running over a 4 day weekend while I
was out of town. When I came back, it was sitting at
around 156MB in size, which seems a little excessive,
especially since I hadn't done any searches or
downloads myself.

I've noticed that when I do searches and downloads, my
broker typically grows quite quickly. I'm *guessing* at
circular references, but it could also just be terrible
fragmentation.

ciphertext stealing of small messages

2001-03-24T05:55:09Z

Mojonation uses ciphertext stealing to encode its
messages; well and good, but for plaintexts that are
less than a block long, mojonation tries to change the
IV in place, violating the strings-never-change
assumption of python. (See file
evil/common/crypto/Python/desx_openssl.c, line 188).
Later, when messages are encrypted (see, eg,
evil/common/mesgen.py, line 582) the IV is usually
packed before the encryption is done, making it
impossible to decrypt the output when messages smaller
than 8 bytes are sent.

Suggested solution 1: assert message length > 7

Suggested solution 2: make tripledescbc return a pair
(iv, crypted_string)

Suggested solution 3: abandon ciphertext stealing and
use padding or CFB mode.