Re: [fossil-users] Fossil 2.1

[K. Fossil user]

1/ Git is better than Fossil when it comes to the job to be done. period.
2/ nine days after first commit is not evidence of efficiency.

a) it could be an evidence of issues. Windows compilation issue was reported 
after commits in the past.That kind of issue is totally abnormal.
b) The version numbering is rubbish X.Y when everyone [say serious projects] do 
x.y.z ...Did someone notice Google Chrome numbering ? Huh ?

c) debian and slackware do take very long time before stable releases.IMHO 
debian and slackware teams are (very far) better than will ever be the Fossil 
team whatever you may think.
d) not to mention that when it comes to security, the Fossil Team is not aware 
about it...(who is stupid enough to use inetd ? Fossil claim they do ! wow !)
3/ If Fossil is so the best, why don't people would like to use it ? There are 
many website that host Fossil repository ...
4/ Ok it's Warren comments ... So I don't take it too seriously. I like to read 
his comments when I've got times so I could laugh, sometimes loudly...I urge 
Warren to sometimes talk about security : it's VERY amusing...(The MD5 discuss 
: wow)I say : SOMETIMES, not every day...  
Regards

K.

  De : Warren Young 
 À : Fossil SCM user's discussion  
 Envoyé le : Vendredi 10 mars 2017 17h32
 Objet : Re: [fossil-users] Fossil 2.1
   
On Mar 10, 2017, at 10:26 AM, Richard Hipp  wrote:
> 
> Fossil version 2.1 is now available from the download pages. 

…after 9 days from first commit to final release.

Beat that, Git!

___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


   ___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

Re: [fossil-users] Fossil can not import its own git export

[Martin Gagnon]

On Fri, Mar 10, 2017 at 02:02:23PM -0600, Zakero wrote:
> In an effort to do some experimenting with the latest release of Fossil:
> 
> ./fossil-2.1 version
> This is fossil version 2.1 [83e3445f67] 2017-03-10 17:07:08 UT
> 
> I wanted to export the fossil repository to git.  Then import the "git
> file" back into fossil.  Being the first time to attempt this, there is a
> good chance I am doing something wrong.  Below are the 2 fossil commands:
> 
> ./fossil-2.1 export --git -R ./fossil-scm.fossil > ./fossil-scm.git
> ./fossil-2.1 import --git fossil-sha3.fossil ./fossil-scm.git
> 
> The import fails with an error: "bad fast-import line: [def.txt]"

It seems to be a problem with a weird filename (with a newline in it),
added in this checkin: 
  http://fossil-scm.org/index.html/info/ac1af2306b70c537

  (It's all what this checkin was about: to test how fossil can handle
  some weird filenames.)

One of the file in this checkin is: "abc\ndef.txt" (with a newline)

Looks like the export/import command doesn't support file with newline,
we can see the filename is truncated in the error message.

I even test using the version of fossil from branch: ticket-d17d6e5b17
(from 2013) that was made to fix some issues with filenames and I get
the same error when exporting/importing a test repository with a file
containing a newline in it.

  [snip]

Attached to this email (hopefully it will not get removed) is my test
script that I use to create a new repository with a file containing a
newline in it and trying to do an export/import cycle.

If you look at the resulting "test.git-export" file, you can see that
the newline on the filename produce a line with a part of the filename
in it, which is seen as garbage on the git-fastexport file.

I have no clue how to fix that.

Regards

-- 
Martin G.


test_file_with_newline.sh
Description: Bourne shell script
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

[fossil-users] Fossil can not import its own git export

[Zakero]

In an effort to do some experimenting with the latest release of Fossil:

./fossil-2.1 version
This is fossil version 2.1 [83e3445f67] 2017-03-10 17:07:08 UT

I wanted to export the fossil repository to git.  Then import the "git
file" back into fossil.  Being the first time to attempt this, there is a
good chance I am doing something wrong.  Below are the 2 fossil commands:

./fossil-2.1 export --git -R ./fossil-scm.fossil > ./fossil-scm.git
./fossil-2.1 import --git fossil-sha3.fossil ./fossil-scm.git

The import fails with an error: "bad fast-import line: [def.txt]"

Is there another option that I am supposed to be using?  Has anyone ever
successfully done this in the past?  I have tried this export->import with
fossil 2.0 and "version 1.37 [377fbf2ca7] 2017-01-19 13:46:04 UTC" which I
built from trunk awhile ago.  All failed to import what they exported.

An observation, the fossil init/new command supports the "--sha1" option.
Should the fossil import command support that as well?

The reason for the git import is that I was planning to use that as
baseline to help see the difference between sha1 and sha3 fossil
repositories (with no tickets and wiki artifacts).  What is the size impact
of the larger hash-id?  How much overhead is involved with the extra
calculations?  An import from the same data-set should effectively isolate
the sha1/sha3 changes and give an idea to the answers.  The basis of the
unfinished script is below and comments are welcome.

#!/bin/sh
# Download Fossil 2.0 as fossil-2.0
# Download Fossil 2.1 as fossil-2.1

# Use Fossil 2.1 to clone the fossil repository
./fossil-2.1 clone http://fossil-scm.org ./fossil-scm.fossil

# Export the fossil repository to git
./fossil-2.1 export --git -R ./fossil-scm.fossil > ./fossil-scm.git

# The git repo is now the "base-line"
# Phase 1: Do a full import
# Phase 2: Do a import without a rebuild and vacuum.  Time the rebuild
separately.

# Convert the base-line into a fresh fossil repo using Hardend SHA1
time ./fossil-2.0 import --git fossil-sha1.fossil ./fossil-scm.git

# Convert the base-line into a fresh fossil repo using SHA3
time ./fossil-2.1 import --git fossil-sha3.fossil ./fossil-scm.git
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

Re: [fossil-users] Fossil 2.1

[Warren Young]

On Mar 10, 2017, at 10:26 AM, Richard Hipp  wrote:
> 
> Fossil version 2.1 is now available from the download pages. 

…after 9 days from first commit to final release.

Beat that, Git!

___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

[fossil-users] Fossil 2.1

[Richard Hipp]

Fossil version 2.1 is now available from the download pages.  Please
let me know if you find any problems.

The next check-in to the self-hosting Fossil repository will likely
use a SHA3 hash.
-- 
D. Richard Hipp
d...@sqlite.org
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

Re: [fossil-users] The SHA3 transition as firewall

[Warren Young]

On Mar 10, 2017, at 2:53 AM, Joerg Sonnenberger  wrote:
> 
> On Thu, Mar 09, 2017 at 01:37:35PM -0700, Warren Young wrote:
>> 
>> 
>> That can be gotten around with a MITM attack
> 
> It still only matters if you can *introduce* objects.

You’re telling me it’s of no consequence if, during a sync from one repository 
to another, a MITM is able to replace the artifact representing the tip of 
trunk for one file with a different artifact having the same SHA-1 hash?
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

Re: [fossil-users] Modify username after git import

[Joerg Sonnenberger]

On Fri, Mar 10, 2017 at 12:40:34PM +, Pietro Cerutti wrote:
> On 2017-Mar-10, 12:35, Joerg Sonnenberger wrote:
> > On Fri, Mar 10, 2017 at 11:30:15AM +, Pietro Cerutti wrote:
> > > I would like to change the user associated with commits from
> > > g...@gahr.ch to simply gahr. Do to that, I updated the 'user' column of
> > > the 'event' table.
> > 
> > That's not enough. You have to actually "amend" the commit.
> 
> Thanks Joerg, I understand. Still, I can't seem to figure out how fossil
> recovered the original information.

Because it is stored in the manifests. Most database tables are
effectively a decomposed cache of the manifest contents.

Joerg
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

Re: [fossil-users] Modify username after git import

[Pietro Cerutti]

On 2017-Mar-10, 12:35, Joerg Sonnenberger wrote:
> On Fri, Mar 10, 2017 at 11:30:15AM +, Pietro Cerutti wrote:
> > I would like to change the user associated with commits from
> > g...@gahr.ch to simply gahr. Do to that, I updated the 'user' column of
> > the 'event' table.
> 
> That's not enough. You have to actually "amend" the commit.

Thanks Joerg, I understand. Still, I can't seem to figure out how fossil
recovered the original information.

Anyway, I fixed it by modifying the git export before importing.

-- 
Pietro Cerutti
g...@gahr.ch


signature.asc
Description: PGP signature
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

Re: [fossil-users] Modify username after git import

[Joerg Sonnenberger]

On Fri, Mar 10, 2017 at 11:30:15AM +, Pietro Cerutti wrote:
> I would like to change the user associated with commits from
> g...@gahr.ch to simply gahr. Do to that, I updated the 'user' column of
> the 'event' table.

That's not enough. You have to actually "amend" the commit.

Joerg
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

[fossil-users] Modify username after git import

[Pietro Cerutti]

I have converted a simple git repo to fossil using
git fast-export --all | fossil import --git tcl-scgi.fossil

I would like to change the user associated with commits from
g...@gahr.ch to simply gahr. Do to that, I updated the 'user' column of
the 'event' table.

Both fossil timeline and fossil server look fine:
https://code.ptrcrt.ch/tcl-scgi/timeline

However, when I clone the repo, I still get g...@gahr.ch in the timeline
and in the event table.

I have tried this on different hosts and I still don't see where it's
coming from. Any hintys?

-- 
Pietro Cerutti
g...@gahr.ch


signature.asc
Description: PGP signature
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

Re: [fossil-users] The SHA3 transition as firewall

[Joerg Sonnenberger]

On Thu, Mar 09, 2017 at 01:37:35PM -0700, Warren Young wrote:
> On Mar 9, 2017, at 1:03 PM, Richard Hipp  wrote:
> > 
> > If a new artifact Y' which has the
> > same SHA1 hash as Y comes along, it will be discarded, since an
> > artifact with that same hash is already in the repository.
> 
> That can be gotten around with a MITM attack, as I’ve already brought
> up several times on the list.  Many Fossil instances won’t have TLS
> protection against MITM attacks, and those that do have it may be
> weakened by some well-intentioned TLS-busting middlebox or antimalware
> package.

It still only matters if you can *introduce* objects. MITM for a given
repository with sensible content requires a second preimage attachk.
Those are not possible on any kind of (un)realistic budget.

Joerg
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users