Re: [fossil-users] Is fossil export known to be broken?

[org.fossil-scm.fossil-users]

On 2015-12-12T16:06:20 -0800
jungle Boogie  wrote:
> 
> Great news!
> 
> Did you happen to change your git version at all during either test?

Nope, same version all along (2.6.2).

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

[fossil-users] Is fossil export known to be broken?

[org.fossil-scm.fossil-users]

Hello!

I'm attempting to export a moderately large (241mb) repository to git:

  http://fossil.io7m.com/repo.cgi/io7m-r1/index

The export proceeds without an error:

  /tmp/r1-fossil$ fossil export > /tmp/r1.bin

The import proceeds without an error:

  /tmp/r1-git$ git init
  /tmp/r1-git$ git fast-import < /tmp/r1.bin

However, the resulting git checkout is missing files that are present
in the most recent commit to the fossil repository.

  /tmp/r1-fossil$ ls -alF 
io7m-r1-documentation/src/main/resources/com/io7m/r1/documentation/ | wc -l
  43

  /tmp/r1-git$ ls -alF 
io7m-r1-documentation/src/main/resources/com/io7m/r1/documentation/ | wc -l
  30

In some cases, entire directory hierarchies are missing:

  /tmp/r1-fossil$ ls -alF 
io7m-r1-examples/src/main/resources/com/io7m/r1/examples/results/scenes/ | wc -l
  80

  /tmp/r1-git$ ls -alF 
io7m-r1-examples/src/main/resources/com/io7m/r1/examples/results/scenes/ | wc -l
  ls: cannot access 
io7m-r1-examples/src/main/resources/com/io7m/r1/examples/results/scenes/: No 
such file or directory
  0

Is there some way to get more information about what's going wrong here? I don't
know if fossil or git is at fault, and I have no way of knowing how badly the
history has been corrupted by the export or import.

M
___
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] Is fossil export known to be broken?

[org.fossil-scm.fossil-users]

On 2015-12-12T12:26:19 +
 wrote:
>
> Is there some way to get more information about what's going wrong here? I 
> don't
> know if fossil or git is at fault, and I have no way of knowing how badly the
> history has been corrupted by the export or import.

It seems that somebody else ran into this at the start of the year:

  http://www.mail-archive.com/fossil-users@lists.fossil-scm.org/msg19238.html

M
___
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] Is fossil export known to be broken?

[org.fossil-scm.fossil-users]

On 2015-12-12T14:21:19 +
 wrote:
> 
> Ideally, I'd be able to reproduce this with a somewhat smaller
> repository...

Surprisingly, this turned out to be easier than expected!

http://waste.io7m.com/2015/12/12/fossilexport/

1. Create fossil repository.
2. Add README.txt and commit in trunk.
3. Create branch 'b0' and switch to it.
4. Add README-b0.txt and commit in b0.
5. Switch to trunk.
6. Merge and commit 'b0'.

The test.fossil repository is the repository resulting from the above:

$ fossil timeline -R test.fossil 
=== 2015-12-12 ===
14:26:36 [6ee55fff27] *MERGE* Merge b0 (user: someone tags: trunk)
14:26:14 [5c6acf101e] Add README-b0 (user: someone tags: b0)
14:25:38 [76a31a2f66] Create new branch named "b0" (user: someone tags: b0) 
14:25:17 [79f2f13cf7] *BRANCH* Initial (user: someone tags: trunk)
14:24:46 [fa9a67d8e8] initial empty check-in (user: someone tags: trunk)
+++ no more data (5) +++

Exporting the repository results in this:

---
blob
mark :4
data 7
Hello.

blob
mark :10
data 10
Hello b0.

commit refs/heads/trunk
mark :3
committer someone  1449930286 +
data 22
initial empty check-in
deleteall

commit refs/heads/trunk
mark :7
committer someone  1449930317 +
data 7
Initial
from :3
M 100644 :4 README.txt

commit refs/heads/b0
mark :9
committer someone  1449930338 +
data 28
Create new branch named "b0"
from :7

commit refs/heads/b0
mark :13
committer someone  1449930374 +
data 13
Add README-b0
from :9
M 100644 :10 README-b0.txt

commit refs/heads/trunk
mark :15
committer someone  1449930396 +
data 8
Merge b0
from :7
merge :13
---

Importing that results in no README-b0.txt existing in the 'trunk'
branch of the created git repository.

I don't understand what's going wrong yet, but at least we now have
a repro case.

M
___
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] Is fossil export known to be broken?

[org.fossil-scm.fossil-users]

On 2015-12-12T08:42:48 -0500
Richard Hipp  wrote:
> 
> The import/export functionality of Fossil has been greatly improved in
> the past by the work of volunteers tracking down obscure
> incompatibilities.  If you would like to try to get to the bottom of
> the problems you are seeing, and either identify the root cause, or
> better to suggest patches, that would be greatly appreciated.

Not knowing the fast-import format, I suspect I'm going to be
approaching the git people and asking them why git doesn't seem to
import the produced file properly. I've checked the exported data and
all of the files and commits are present, but it does seem like
something isn't carried across merges.

Ideally, I'd be able to reproduce this with a somewhat smaller
repository...

M
___
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] Is fossil export known to be broken?

[org.fossil-scm.fossil-users]

On 2015-12-12T23:24:08 +0300
Konstantin Khomoutov  wrote:
> 
> Did you report a bug?
> 
> It's just a matter of posting a mail to git at vger.kernel.org;
> subscription is not required.

I sent the git list the same info and test case a couple of hours ago,
but haven't had a response yet.

I still don't know if this is a Fossil or a Git issue, so I thought I'd
let the git people look at the fast-import file to see if they can spot
any obvious mistakes in it.

M
___
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] Is fossil export known to be broken?

[org.fossil-scm.fossil-users]

On 2015-12-12T16:07:23 -0500
Martin Gagnon  wrote:
> 
> What version of git and fossil are you using ?
> 
> I cannot reproduce the problem using your repo case. After the merge,
> the README-b0.txt file is present.
> 
> Here a script that reproduce your repo case, does it correspond to your
> test case ?

Hello!

$ fossil version
This is fossil version 1.32 [715f88811a] 2015-05-02 21:11:26 UTC

$ git version
git version 2.6.2

The script you've attached is exactly what I used to create the test
case. Which versions are you using?

M
___
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] Is fossil export known to be broken?

[org.fossil-scm.fossil-users]

On 2015-12-12T16:21:14 -0500
Martin Gagnon  wrote:
> > 
> > The script you've attached is exactly what I used to create the test
> > case. Which versions are you using?
> 
> This is fossil version 1.34 [a4889252f1] 2015-12-07 18:19:28 UTC
> 
> git version 1.9.1

Hrm, fossil is quite a bit newer and git is quite a bit older. I'll try
the current fossil trunk and see if that eliminates the issue. Is also
possible that fossil generates fast-import data that the older git can
work with but newer versions won't.

M
___
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] Is fossil export known to be broken?

[org.fossil-scm.fossil-users]

On 2015-12-12T21:43:49 +
 wrote:

> On 2015-12-12T16:21:14 -0500
> Martin Gagnon  wrote:
> > > 
> > > The script you've attached is exactly what I used to create the test
> > > case. Which versions are you using?
> > 
> > This is fossil version 1.34 [a4889252f1] 2015-12-07 18:19:28 UTC
> > 
> > git version 1.9.1

Fossil trunk produces an identical fast-import file to the older
version. Could you verify that the fast-import file that your repo case
produces is the same as:

  http://waste.io7m.com/2015/12/12/fossilexport/test.export

It probably won't be byte-for-byte identical, but I'd assume that the
"mark" numbers would be the same, in addition to the order and number
of commits.

M
___
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] Is fossil export known to be broken?

[org.fossil-scm.fossil-users]

On 2015-12-12T17:14:56 -0500
Martin Gagnon  wrote:
> 
> For some reason, mine have an extra line at the end that look like this:
> 
> M 100644 :10 README-b0.txt
> 
> For the rest, it's pretty similar.

Ah!

I've just discovered that if I use the trunk version of fossil and run
it on a freshly created repository, then it does export correctly. If I
use the trunk version of fossil on the repository created by the older
version, it fails. If I 'fossil rebuild' the old repository, exporting
works correctly!

I have no idea as to the cause, but it seems that fixing it requires
using a new version of Fossil AND rebuilding the repository before
exporting.

I have now correctly exported my original repository and all files
appear to be intact!

M
___
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] Query to return non-propagating tags applied to commits?

[org.fossil-scm.fossil-users]

On 2015-08-01T09:50:06 +
org.fossil-scm.fossil-us...@io7m.com wrote:
 
 Ah, thanks, that explains quite a lot. It was the tagtype field that
 was the missing part of the puzzle. I was working from the output of
 the .schema command, so didn't have the documentation above.

Hm, the following gives the history of a tag 'io7m-jparasol-0.5.1':

select
  t.tagid,
  substr(t.tagname,5),
  tx.* 
from tagxref as tx
join tag as t on t.tagid = tx.tagid
where t.tagname = 'sym-io7m-jparasol-0.5.1' 
  and (tx.tagtype = 0 || tx.tagtype = 1) 
order by tx.mtime desc;

tagid|substr(t.tagname,5)|tagid|tagtype|srcid|origid|value|mtime|rid
67|io7m-jparasol-0.5.1|67|0|2804|2804||2456819.16056917|2804
67|io7m-jparasol-0.5.1|67|1|2797|2796||2456818.37702453|2796

(Taken from
http://fossil.io7m.com/repo.cgi/io7m-jparasol/timeline?n=1000y=allv=0)

Unfortunately, it doesn't quite tell the whole story, as that tag was
only cancelled because a branch occurred as the next event after the
tag was created. I'm not sure how to distinguish between a tag being
explicitly removed by the user, and a tag being cancelled just because
that's how fossil works internally.

I feel like I should instead be querying events, rather than the
tagxref table.

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

[fossil-users] Query to return non-propagating tags applied to commits?

[org.fossil-scm.fossil-users]

'Lo.

Is there a straightforward SQL query that will, given a tag name,
show me the commits to which that tag is applied (if any), taking
into account any cancel tag events?

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

[fossil-users] better git migration

[org.fossil-scm.fossil-users]

Hello.

I have a repository that I'd like to migrate to github (*ducks rotten
fruit and vegetables*) but it's not clear to me that the existing
export/import process can satisfy my requirements.

Firstly: Every commit I've ever made to fossil has been PGP signed. The
repository in question spans a couple of years and I've been the only
committer. My PGP keys expire yearly, so all commits in the repository
are signed by one of two keys. For reasons of least privilege, I tend
to use different usernames on different machines, so although all the
committers are me, they aren't necessarily all the same name.
An initial run of fossil export | git fast-import seems to
show no pgp signatures in the resulting git repository. This doesn't
seem unreasonable, given that it's obviously not going to be possible
to forge new signatures!

Secondly: all of the committer names are... sort of bad. Most are of
the form m0 m0 (m0 being the username I used there).

I feel like fast-import may not be the way to go about this. I'm
wondering if there's a way to:

1. Transform *all* committer names to a given value: I'd prefer to use
   a name and email address that matches my github name/email.

2. PGP sign all commits with the key with which they were originally
   signed (obviously, some of these keys have expired).

Obviously, I'd like to preserve the original dates and times too.

I could sort of imagine replaying and re-signing the original commits
one-by-one to achieve the above, but I'm not sure how to go about this
or if it's even feasible.

Any help would be appreciated!
M
___
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] Select specific changes within files

[org.fossil-scm.fossil-users]

On 2015-03-20T07:26:53 -0430
Abilio Marques abili...@gmail.com wrote:

 I personally would like a selective stash.

This seems like the best of all worlds: You get the clean separate
commits typical of git, but with each of those commits actually
compiled and tested.

I tried to put this together ages ago by producing a single diff
from the stash and then interactively patching using some 3rd-party GUI
patch tools, but it never really worked out (due to the 3rd-party tools
not really doing it very well).

It's basically the logic from git add --patch but working in reverse:
Take a large diff, interactively split it into chunks and optionally
apply some of the chunks. Compile. Test. Commit. Repeat until there are
no chunks left.

M
___
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] Dealing with an unreliable remote?

[org.fossil-scm.fossil-users]

On 2015-01-09T16:53:44 -0500
Richard Hipp d...@sqlite.org wrote:
 
 Please try this:  Visit the Admin/Access page and reduce the Download
 packet limit.  Maybe that will allow the CGI to finish.
 

Yep, lowering it to 100 from 500 seems to have helped a lot.
Thanks!

 Clones are specially optimized so that they run faster.  And this is
 probably interfering with the restart.  A push or pull is completely
 restartable.  Perhaps something can be done.  Keep pestering people
 until we have time to work on this.  In the meantime, try lowering the
 Download pack limit as described above.  Or:  Get a better ISP.  ;-)

Which people are you referring to here? Pester people on this list?

M
___
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] Dealing with an unreliable remote?

[org.fossil-scm.fossil-users]

On 2015-01-09T16:47:14 -0700
Andy Bradford
amb-sendok-1423439234.cmfoedahgecpfnpnd...@bradfords.org wrote:

 Thus said org.fossil-scm.fossil-us...@io7m.com on Fri, 09 Jan 2015 23:30:55 
 +:
 
  Yep, lowering it to 100 from 500 seems to have helped a lot.
 
 Drat! I wanted to try to clone  your repo later while it was misbehaving
 so I could see what problem you were hitting. :-) Similar to the problem
 reported  earlier, I'm  not sure  how you  could end  up with  a partial
 clone.

Made a copy of the repos here with the default 500 limit:

  http://fossil.io7m.com/repo.cgi/io7m-r1-500

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

[fossil-users] Dealing with an unreliable remote?

[org.fossil-scm.fossil-users]

'Lo.

I've run up against an unpleasant problem. My fossil
repositories are served over CGI, and the CGI program is running on a
hosting provider that treats CPU time as a precious resource. When
cloning a repository that's larger than a few tens of megabytes,
whatever they're using as a cpu time supervisor usually kills the clone
in process before it can be completed.

This wouldn't be such an enormous problem if the clone could be resumed
where it left off, but fossil seems to get upset at this:

$ fossil sync --repository io7m-r1.fossil
Fossil internal error: infinite loop in DELTA table

This essentially means that many of my public repositories can't be
cloned by anyone!

An example repository that tends to show the problem:

  http://fossil.io7m.com/repo.cgi/io7m-r1/index

Cloning means receiving roughly 130mb of data.

Is there something that can be done to make clones resumable? Given the
transactional nature of repository operations, I'd have expected
resuming to just work.

M
___
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] Excessively long path length breaks working copy on Windows 7

[org.fossil-scm.fossil-users]

On 2014-11-15T16:57:23 +
org.fossil-scm.fossil-us...@io7m.com wrote:

 'Lo.
 
 I seem to have run into a problem that I can reproduce with the
 following repository:
 
   http://waste.io7m.com/2014/11/15/fossil-pathbug.fossil
 

I've filed a bug so that this doesn't get lost to time:

https://fossil-scm.org/index.html/tktview/81b89131083cd01af0547e3cc0a17960512d04aa

M
___
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] Excessively long path length breaks working copy on Windows 7

[org.fossil-scm.fossil-users]

On 2014-12-01T21:23:53 +0100
Stephan Beal sgb...@googlemail.com wrote:

 On Mon, Dec 1, 2014 at 9:22 PM, Stephan Beal sgb...@googlemail.com wrote:
 
  Strange: when i follow the link you posted, Chromium tells me (for the
  first time ever, possibly b/c i've never knowingly accessed it over https):
 
  This server could not prove that it is *fossil-scm.org
  http://fossil-scm.org*; its security certificate is
 
 
 btw: apparently doesn't happen if you prefix the URL with www.
 

Ah, yeah. The certificate is likely for specifically
www.fossil-scm.org as opposed to *.fossil-scm.org.

M
___
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] Excessively long path length breaks working copy on Windows 7

[org.fossil-scm.fossil-users]

On 2014-11-16T23:10:31 +0100
Jan Nijtmans jan.nijtm...@gmail.com wrote:

'Lo.

 
 The win32 api has limitations:
 http://msdn.microsoft.com/en-us/library/aa365247(VS.85).aspx#maxpath

That's an understatement.

 For absolute paths it is possible to workaround this (which fossil
 does), but for relative paths that's a more difficult story. A
 possible solution might be to translate relative paths to absolute
 paths (with extended path prefix) if they contain more than 260
 characters, but it is not trivial to get this right. Is it really
 worth it? (if you can convince me that it's worth it, maybe I
 implement it some day .. )

Hm. I should mention that I discovered this issue in the first place
because I have a real world repository that contains the following path:

 
io7m-r1-examples/src/main/resources/com/io7m/r1/examples/results/com.io7m.r1.examples.scenes.SPShadowVarianceSpecularDiffuseOnly0/com.io7m.r1.examples.ExampleRendererDeferredDefault/0.png

However... That path is only 187 characters long. Even with the full
expansion:

  
c:\cygwin64\home\someone\fossil\fossil-pathbug/io7m-r1-examples/src/main/resources/com/io7m/r1/examples/results/com.io7m.r1.examples.scenes.SPShadowVarianceSpecularDiffuseOnly0/com.io7m.r1.examples.ExampleRendererDeferredDefault/0.png

That's still only 234 characters. I'm not entirely sure why I'm
apparently hitting this limit at all. This is easily reproduceable,
just create the above (relative) path on a non-Windows machine and then
check out on a Windows 7 machine.

Secondly, it's not just an issue of the long path: The real killer is
that it appears to damage the working copy in some way. If someone
checks in a file with a long path, and I have uncomitted changes in my
working copy, and I then fossil update and run into the above
error... What am I supposed to do? The working copy seems to be left in
a bad state. I don't think I'd trust it enough to commit as there are
potentially files missing from the working copy and Fossil doesn't seem
to know or care about it.

M

___
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] Excessively long path length breaks working copy on Windows 7

[org.fossil-scm.fossil-users]

On 2014-11-17T16:23:08 +0100
Stephan Beal sgb...@googlemail.com wrote:

 On Mon, Nov 17, 2014 at 3:31 PM, org.fossil-scm.fossil-us...@io7m.com
 wrote:
 
  That's still only 234 characters. I'm not entirely sure why I'm
  apparently hitting this limit at all. This is easily reproduceable,
  just create the above (relative) path on a non-Windows machine and then
  check out on a Windows 7 machine.
 
 
 Are you on an encrypted filesystem? If so, the encryption might be the
 culprit - the encrypted names are longer than the originals, potentially
 longer than the fs can handle. That's happened to me on Linux before.

'Lo.

Nope, it's an almost pristine Windows 7 install using the default NTFS
settings.

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

[fossil-users] Excessively long path length breaks working copy on Windows 7

[org.fossil-scm.fossil-users]

'Lo.

I seem to have run into a problem that I can reproduce with the
following repository:

  http://waste.io7m.com/2014/11/15/fossil-pathbug.fossil

The repository contains a directory b full of files, and a directory
a containing about 256 nested directories, with a single file at the
deepest point.

On Windows 7, if I'm at:

  c:\cygwin64\home\someone\fossil\fossil-pathbug

(Don't let the name fool you, I'm using the latest fossil binary from
fossil-scm.org and am running it under the ordinary cmd.exe, not
cygwin).

... and I attempt to open the fossil-pathbug.fossil repository:

c:\...\fossil-pathbug fossil open ..\fossil-pathbug.fossil
unable to create directory 
c:/cygwin64/home/someone/fossil/fossil-pathbug/a/a/a/a/a/a/a/a/a/a/a/a/a/a/a/a/a/a/a/a/a/a/a/a/a...

(Extra 'a's omitted as cmd.exe doesn't seem to allow copying and pasting).

This would be fine, except that it seems like it leaves the working copy in 
some sort of inconsistent state: A listing of the directory shows no b 
directory, a fossil stat shows no missing files. None of the commands 
such as fossil update will check out the remaining files (the commands 
will silently return success, apparently).

I'm using fossil 1.29 from the site (3e5ebe2b90).

M
___
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] Ordering ticket priority/severity

[org.fossil-scm.fossil-users]

On 2014-10-13T06:40:37 +0100
Jacek Cała jacek.c...@gmail.com wrote:

   Hi,
 
 Below is one of my ticket report pages. I'm not entirely sure it's the best
 approach but works fine for me. What it does is two selects. The inner
 select classifies status, priority, severity and difficulty so then I can
 order them appropriately. The outer select presents the data in human
 readable way.

'Lo.

Thanks, that looks pretty good. Didn't realize sqlite supported nested
SELECT queries in that manner (in my experience, just about every SQL
implementation on the planet does something different!).

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

[fossil-users] Ordering ticket priority/severity

[org.fossil-scm.fossil-users]

'Lo.

Currently, if I do, in a ticket report:

  ORDER BY priority, severity

I definitely get something ordered by priority and severity, but of
course the ordering relation for both columns is lexicographical. That
is, Important  Critical because Critical appears earlier in the
alphabet.

That's pretty awful! Is there a recommended way to get a better
ordering relation without having to mutilate the ticket system too
much?

M 
___
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] Getting configure to find openssl on FreeBSD

[org.fossil-scm.fossil-users]

Bizarrely, I went back to this today and fossil compiled first time
without me needing to specify any --with-openssl options. I've no idea
why!

On 2014-09-11T17:33:33 -0400
Eric Rubin-Smith eas@gmail.com wrote:
 
 A stripped shared library would not be worth much, since the whole point
 of a shared library is to provide symbols.  Did you pass '-D'?

Nope, didn't know -D existed.

$ file /usr/lib/libssl.so.6 
/usr/lib/libssl.so.6: ELF 64-bit LSB shared object, x86-64, version 1
(FreeBSD), dynamically linked, stripped

$ nm /usr/lib/libssl.so.6
nm: /usr/lib/libssl.so.6: no symbols

$ nm -D /usr/lib/libssl.so.6 | grep SSL_new
00034810 T SSL_new

All binaries in the FreeBSD base are stripped (not sure I really see
the utility of it, but...)

On 2014-09-12T12:15:18 +1000
Steve Bennett ste...@workware.net.au wrote:
 
 ./configure --debug ...
 
 And then look at config.log

Thanks, useful for future reference.

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

[fossil-users] Getting configure to find openssl on FreeBSD

[org.fossil-scm.fossil-users]

'Lo.

The fossil build scripts seem to be unable to find openssl on FreeBSD
9.2. It has a choice of the version included with the base system
(in /usr) or the version available from FreeBSD ports (/usr/local),
but it can't seem to find either of them.

Is there any way to get it to give more information about why it's
failing to find them?

$ ./configure --with-openssl=auto
Host System...x86_64-unknown-freebsd9.2
Build System...x86_64-unknown-freebsd9.2
C compiler... cc -g -O2
C++ compiler... c++ -g -O2
Build C compiler...cc
Checking for stdlib.h...ok
Checking for uint32_t...ok
Checking for uint16_t...ok
Checking for int16_t...ok
Checking for uint8_t...ok
Checking for pread...ok
Checking for tclsh...no
Checking for system ssl...no
Checking for ssl in /usr/sfw...no
Checking for ssl in /usr/local/ssl...no
Checking for ssl in /usr/lib/ssl...no
Checking for ssl in /usr/ssl...no
Checking for ssl in /usr/pkg...no
Checking for ssl in /usr/local...no
Checking for ssl in /usr...no
Error: OpenSSL not found. Consider --with-openssl=none to disable HTTPS
support Try: 'configure --help' for options

$ ./configure --with-openssl=/usr 
...
Checking for ssl in /usr...no
Error: OpenSSL not found. Consider --with-openssl=none to disable HTTPS
support Try: 'configure --help' for options

$ ./configure --with-openssl=/usr/lib
...
Checking for ssl in /usr/lib...no
Error: OpenSSL not found. Consider --with-openssl=none to disable HTTPS
support Try: 'configure --help' for options

$ ./configure --with-openssl=/usr/local
...
Checking for ssl in /usr/local...no
Error: OpenSSL not found. Consider --with-openssl=none to disable HTTPS
support Try: 'configure --help' for options

$ ./configure --with-openssl=/usr/local/lib
...
Checking for ssl in /usr/local/lib...no
Error: OpenSSL not found. Consider --with-openssl=none to disable HTTPS
support Try: 'configure --help' for options

$ ls /usr/local/lib/libssl.so*
/usr/local/lib/libssl.so
/usr/local/lib/libssl.so.8

$ ls /usr/lib/libssl.so*
/usr/lib/libssl.so
/usr/lib/libssl.so.6

... and so on.

M
___
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] Getting configure to find openssl on FreeBSD

[org.fossil-scm.fossil-users]

On 2014-09-11T16:09:42 -0400
Eric Rubin-Smith eas@gmail.com wrote:
 
 I skimmed through the configure code.  Looks like you need to have 
 the subdir and file 'openssl/ssl.h' beneath the dir you specify, and 
 the subdir and files 'lib/libssl.so' and 'lib/libcrypto.so' beneath the 
 dir you specify.  Furthermore, one of those libraries has to provide
 the symbol SSL_new.  
 
 My (very quick and possibly wrong) read of the code is that those are 
 exactly the pass criteria for deciding that SSL is supported on the 
 build box using the dir your specify.

Think that may be partly wrong, as that'd mean that if I specified /usr,
then the following would have to exist:

  /usr/openssl/ssl.h

... and that's obviously not true of any of the other platforms upon
which I've built fossil. I do have (where → indicates a symlink):

  /usr/include/openssl/ssl.h
  /usr/lib/libssl.so → /usr/lib/libssl.so.6
  /usr/lib/libssl.so.6
  /usr/lib/libcrypto.so → /lib/libcrypto.so.6
  /lib/libcrypto.so.6

  /usr/local/include/openssl/ssl.h
  /usr/local/lib/libcrypto.so → /usr/local/lib/libcrypto.so.8
  /usr/local/lib/libcrypto.so.8
  /usr/local/lib/libssl.so → /usr/local/lib/libssl.so.8
  /usr/local/lib/libssl.so.8

The /usr/lib/*.so libraries are stripped, so doesn't appear to have
any symbols, but the /usr/local/lib/*.so libraries aren't, and the
ssl library definitely contains SSL_new.

M
___
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] Stability of fossil test-name-to-id

[org.fossil-scm.fossil-users]

'Lo.

On 2014-08-19T21:17:18 +0200
Stephan Beal sgb...@googlemail.com wrote:

 You might be interested in a standalone alternative:
 ...
 
 and has already been created as a standalone binary with documented
 semantics:
 
 http://fossil.wanderinghorse.net/repos/libfossil/doxygen/fossil-core_8h.html#a265d03fe1e040d7068679f9b2a4d1a11

Looks good, certainly. Problem in my case is that I'd either have to
write language bindings (as I'm not working from C), or require the
person using my program to have installed your standalone utilities.

I'm trying to stick to either calling the fossil binary directly, or
piping SQL into fossil sqlite in the worst case.

Am I to assume that test-name-to-id is definitely not to be depended
upon?

M
___
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] Stability of fossil test-name-to-id

[org.fossil-scm.fossil-users]

On 2014-08-20T14:51:30 -0400
Richard Hipp d...@sqlite.org wrote:
 
 Probably test-name-to-id won't just go away.  If you encounter problems, it
 will likely be because we enhance the command to output additional
 information, which then breaks your parsing logic.  Or the command might be
 promoted to a supported command by removing the test- prefix.

On 2014-08-20T20:50:26 +0200
Stephan Beal sgb...@googlemail.com wrote:

 Nothing named test- should be depended upon.


Ok, thanks to you both!

I'll use SQL for now. Needless to say, I'd love to see this become a
supported command, ideally just producing a single artifact ID as
output to avoid having to parse anything. I'm mainly just using this
information to determine whether the tip of the repository has changed
since I looked at it last.

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

[fossil-users] Stability of fossil test-name-to-id

[org.fossil-scm.fossil-users]

Hi.

I'm writing a program that depends on the functionality of
test-name-to-id. Given that the command is completely undocumented and
doesn't appear in any help texts, is that sufficient warning to suggest
that the command may be removed at any moment and shouldn't be used for
anything?

I could open the repository and perform the SQL query manually, but the
temptation to use a command that's already written and known to work is
tempting...

M
___
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] syncing many repositories

[org.fossil-scm.fossil-users]

On 2014-08-09T21:19:55 -0600
Andy Bradford
amb-sendok-1410232795.nbeloplbpekihcmep...@bradfords.org wrote:

 Thus said org.fossil-scm.fossil-us...@io7m.com on Sat, 09 Aug 2014 16:48:04 
 -:
 
  I  don't suppose  there's  any way  to  make fossil  hit  a given  URI
  whenever the current repository receives artifacts?
 
 You might be able to use the commit hooks for this:
 
 http://fossil-scm.org/index.html/info/1311841a3c201a669bc0e433e24c6caf3fd3ab1d
 
 There is a  common script that gets  run for all syncs,  maybe this will
 work.

Ah, thank you! This looks promising. Turns out that I was using a
version of fossil that was built barely days before this commit reached
trunk. I'm now using a current (as of a few minutes ago) version, so
I'll give it a try.

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

[fossil-users] Sync by URI?

[org.fossil-scm.fossil-users]

'Lo.

Assuming I've got a fossil server listening on http://example.com:8080,
is there a URI I can hit on that address that will cause that repository
to sync with whatever the current remote is? Like:

  $ curl http://example.com:8080/sync
  (possibly with authentication)

M
___
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] Sync by URI?

[org.fossil-scm.fossil-users]

On 2014-08-10T16:58:12 +0200
Stephan Beal sgb...@googlemail.com wrote:

 On Sun, Aug 10, 2014 at 4:53 PM, org.fossil-scm.fossil-us...@io7m.com
 wrote:
 
  Assuming I've got a fossil server listening on http://example.com:8080,
  is there a URI I can hit on that address that will cause that repository
  to sync with whatever the current remote is? Like:
 
 
 The short answer is no, and i'm not aware of any current mechanism it could
 be built upon. Note that not all public repos (in my experience _no_ public
 repos) have a remote URL which they could sync with.

OK, that's fine.

Was just checking that I wasn't re-implementing something.

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

[fossil-users] syncing many repositories

[org.fossil-scm.fossil-users]

'Lo.

First of all, because this doesn't get said enough: Thanks for fossil!
Been a happy user for three years now, and I love the approach taken to
UI simplicity and the focus on reliability and integrity. I have no
idea how I coped before having an integrated ticket system.

However, one aspect of fossil is really starting to become a problem:
Syncing large numbers of repositories.

I'm using what I believe is probably a common setup: I have a server on
my private network here containing the canonical repositories for all
my projects. I have clones of these repositories on all machines here,
and the repositories on the server are backed up nightly. I then have
public-facing repositories on io7m.com which are basically read-only
public mirrors of my own private repositores (I have no contributors):

  http://fossil.io7m.com

The io7m.com site is hosted by site5.com, using their standard shared
web hosting account. I have ssh access, but not root access. I cannot
run any long-running processes on it.

As the number of repositories has grown, this has become a real
problem. I first started using cron to sync repositories. After this
became unwieldly (due to the number of repositories and the way cron
would send me email every time fossil said anything), I ended up
writing a simple program to sync all repositories at staggered, random
intervals, sending email only if any sync failed. The servers at site5
appear to implement some sort of rate limiting: If you open too many
ssh connections in too short a time, they'll throttle and close
connections. In order to try to reduce the load on their side, I've
ended up using longer and longer staggered intervals (to the point
where I'm basically syncing any one repository less frequently than
hourly).

Now that the number of (public) repositories has grown to 32, even this
has become a problem: Due to the number of repositories, I'm still
opening too many connections to the server, and the number of
repositories is only going to grow. All I can do is keep trying to
increase the upper bound on the sync interval.

I believe the sync periodically approach is fundamentally wasteful,
and scales increasingly poorly when the number of repositories grows.
I'm basically burning through gigabytes of bandwidth (32 repositories,
24 hours a day soon adds up) even though I'm generally only committing
to one repository at a time, and nobody is making any changes to
repositories on the io7m.com side. Perhaps 95% of syncs are entirely
pointless!

What are my options here? I'd much rather sync on demand: When a
repository on my private server receives any artifacts, it should then
sync with the io7m.com side. This would massively reduce my own
bandwidth usage and would give the io7m.com servers a break. Is there
any way to achieve this, currently?

Regards,
Mark 
___
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] syncing many repositories

[org.fossil-scm.fossil-users]

On 2014-08-09T12:14:46 +0200
Joerg Sonnenberger jo...@britannica.bec.de wrote:

 On Sat, Aug 09, 2014 at 09:46:37AM +,
 org.fossil-scm.fossil-us...@io7m.com wrote:
  If you open too many ssh connections in too short a time, they'll
  throttle and close connections.
 
 Have you tried using Master mode? Try running:
 ssh -v -MN -o ControlPath=/tmp/socket remote
 and in a separate terminal:
 ssh -o ControlPath=/tmp/socket remote
 That should tell you in the first terminal that a new connection was
 establish. If it works, add the ControlPath option to ~/.ssh/config
 and change your sync script to something like:
 
  ssh -MN remote 
  fossil sync ...
  ssh -O stop remote

I've not used master mode before. From what I can make out from the
ssh_config manual page, this causes ssh to open a single long-running
connection to the server which is re-used by anything connecting to the
control socket? If so, this won't actually work in my case, as this is
considered a long-running process on the site5 side and will eventually
be killed.

In case you're wondering: Yes, this does mean that if I'm attempting to
upload a file to site5 and it ends up taking over an hour or so,
they'll eventually kill the connection. No, I don't like it either, but
I kind of see their point... They have a server full of people running
endless giant piles of php, and need to eliminate the inevitable
runaway processes that result.

I could use a process supervision system (as I already do with the
existing program I use to sync repositories) in order to repeatedly
bring the ssh tunnel back up, but to me doing all of this still seems
like trying to fight a symptom rather than fixing the actual underlying
issue. This also doesn't solve the issue of bandwidth use: I'm on a
rather poor quality ADSL connection and bandwidth is at a premium. I'm
currently wasting rather a lot of bandwidth doing repeated unnecessary
syncs, because there doesn't appear to be another option.

M
___
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] syncing many repositories

[org.fossil-scm.fossil-users]

On 2014-08-09T13:25:21 +0200
Stephan Beal sgb...@googlemail.com wrote:

 You can suppress that: simply redirect all output: /dev/null 21
 cron only mails if the program generates output. Better yet, redirect it
 all to a running log file:
 
  /path/to/log 21

The problem there is that I want it to tell me when a sync fails, but I
don't want to hear sync succeeded over and over. Essentially, I only
want output if the process in question returns a non-zero exit code.
The program I ended up writing to perform syncs does this check, and so
I only get an inbox full of errors instead!

M
___
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] syncing many repositories

[org.fossil-scm.fossil-users]

On 2014-08-09T13:43:32 +0200
Stephan Beal sgb...@googlemail.com wrote:

 On Sat, Aug 9, 2014 at 1:39 PM, org.fossil-scm.fossil-us...@io7m.com
 wrote:
 
  The problem there is that I want it to tell me when a sync fails, but I
  don't want to hear sync succeeded over and over. Essentially, I only
  want output if the process in question returns a non-zero exit code.
  The program I ended up writing to perform syncs does this check, and so
  I only get an inbox full of errors instead!
 
 
 How about:
 
 if ! fossil ... /log 21; then cat /log; else rm -f /log; fi

That's more or less what I wrote. I just didn't do it in /bin/sh and
didn't use temporary files.

M
___
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] syncing many repositories

[org.fossil-scm.fossil-users]

On 2014-08-09T10:23:35 -0600
Andy Bradford
amb-sendok-1410193415.ailbbaenibndbdhmo...@bradfords.org wrote:
 
 If none of  your public-facing repositories accept  commits, wiki edits,
 or tickets, or  in otherwords are completely read-only,  you could stage
 the public-facing repositories  on your local system and  then use rsync
 to copy them to your io7m.com  host. This should be fairly efficient. It
 would use  a single  SSH connection  to the server  and should  be short
 lived  given that  rsync will  only transfer  blocks of  data that  have
 changed.

I have considered this before, and I already use rsync for nightly
backups. However, I'm hesitant to use it for this because, even though
the repositories on io7m.com are conceptually read-only, I do forsee
people eventually filing tickets on them should I one day start
actually promoting any of the software I write and gain users.

Having heard everything suggested on here, I'm leaning towards storing
the ID of the tips of each repository and syncing only when they
change. I'll do an unconditional sync once a day in order to pull in
anything that may have been unexpectedly added on the io7m.com side.

I don't suppose there's any way to make fossil hit a given URI whenever
the current repository receives artifacts?

M
___
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] Sort tickets by priority AND severity?

[org.fossil-scm.fossil-users]

On Sat, 5 Apr 2014 11:34:05 -0400
F. Eugene Aumson feaum...@gmail.com wrote:

 You can do this via editing the query for the ticket view.  Here's a
 snippet from my own All tickets query:
 
 SELECT ticket
 ...
 FROM ticket
 ORDER BY status DESC, priority, severity

Ah, right, thanks.

For some reason I assumed that I couldn't specify my own ordering
(think I assumed that the ticket view would insert its own ORDER BY
clause, as none of the other reports I'd seen had specified them).

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

[fossil-users] Sort tickets by priority AND severity?

[org.fossil-scm.fossil-users]

'Lo.

Subject says it all, really. In the ticket view, is there a way to order
the results by ticket severity AND priority? I can click the headers of
both columns and the tickets are ordered in some manner, but I can't
tell exactly if it's doing what I think it's doing (perhaps some arrows
in the column header indicating the sort order for that column would be
a good idea).

M
___
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] Proposed Fossil interface enhancement.

[org.fossil-scm.fossil-users]

On Mon, 30 Dec 2013 21:25:36 +0100
Stephan Beal sgb...@googlemail.com wrote:

 On Mon, Dec 30, 2013 at 9:20 PM, Michai Ramakers m.ramak...@gmail.comwrote:
 
  I'm not a greybeard console-only type, but occasionally use 'links' to
  browse fossil repos; perhaps JS collapse/expand should not get too
  much in the way there. Or if I'm the only links-/lynx-user out there,
  go ahead :)
 
 
 AFAIR, we have no (important/major) functionality which _requires_ JS in
 order to work. There are some bits which benefit from it (e.g. try clicking
 on two of the little boxes in the timeline, and the wysiwyg editor), but
 nothing major which relies on it.

Unless it's something else causing it here, clicking anything in the timeline 
without javascript enabled will take you to the honeypot... I've gotten used
to browsing fossil repositories with javascript enabled as too much breaks
without it.

M
___
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] Proposed Fossil interface enhancement.

[org.fossil-scm.fossil-users]

On Mon, 30 Dec 2013 22:25:58 +0100
Stephan Beal sgb...@googlemail.com wrote:

 It's apparently not detecting your browser as human...

On Mon, 30 Dec 2013 16:28:17 -0500
Richard Hipp d...@sqlite.org wrote:

 This is a defense against spiders trying to index the entire repository,

Oh, I'm aware, don't worry. I was specifically commenting on the statement:

AFAIR, we have no (important/major) functionality which _requires_ JS in
order to work.

... Which isn't strictly true in the default setup. Browsing a fossil
repository with default settings with javascript disabled is not a great 
experience. I see how the original statement can be taken more than one 
way though. I'll shut up now.

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

[fossil-users] Plain Gray (No Logo) and long ticket texts (missing pre.verbatim in CSS)

[org.fossil-scm.fossil-users]

'Lo.

Unless I'm mistaken, it seems that the Plain Gray (No Logo) skin is missing
the following CSS present in some form in all of the other skins:

pre.verbatim {
  padding: 0.5em;
  white-space: pre-wrap;
}

This means that long ticket comments without manual linebreaks tend to result
in pages that have essentially unbounded width!

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

[fossil-users] not authorized to write - how to debug?

[org.fossil-scm.fossil-users]

'Lo.

I'm suddenly seeing not authorized to write on attempting to push
to a newly created and cloned repository. Is there some way to get
either the server or the client to give more information? I've created
and cloned this repository the same way as the other 44 repositories
here, and have never seen this problem before!

M
___
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] not authorized to write - how to debug?

[org.fossil-scm.fossil-users]

On Fri, 15 Nov 2013 21:53:02 +0100
Stephan Beal sgb...@googlemail.com wrote:

 On Fri, Nov 15, 2013 at 9:47 PM, org.fossil-scm.fossil-us...@io7m.comwrote:
 
  I'm suddenly seeing not authorized to write on attempting to push
  to a newly created and cloned repository. Is there some way to get
  either the server or the client to give more information? I've created
  and cloned this repository the same way as the other 44 repositories
  here, and have never seen this problem before!
 
 
 
 When i've seen this before it's been a genuine problem/mishap in my setup
 of the user's permissions. Make sure the user has (or inherits) the 'i'
 permission. In the normal case this is inherited from the 'developer' ('v')
 permission, so make sure that the 'developer' special user has the 'i'
 permission.
 
 There was one particular case where i saw this intermittently on a
 particular repo, but  a second or third attempt to push succeeded.

Seems that I have the correct capabilities:

$ fossil user capabilities developer -R examples.fossil 
dei
$ fossil user capabilities m0  -R examples.fossil
si

I've even nuked the repository, created a new one, cloned it, and had the
same problem again...

M
___
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] Fwd: not authorized to write - how to debug?

[org.fossil-scm.fossil-users]

On Fri, 15 Nov 2013 22:24:27 +0100
Stephan Beal sgb...@googlemail.com wrote:
 Ooops... unintentionally got the thread running off-list...

I'll add my two responses here too:

 There were recently a number of tweaks made to the user name handling and
 it's possible that something broke in the process. i don't expect that's
 the problem, but i'm out of ideas :/.  

Interesting!

$ fossil clone http://m...@fossil.int.arc7.info/com.io7m.cirrus/jogl-examples 
jogl-examples2.fossil
password for m0: 
remember password (Y/n)? Y
Round-trips: 2   Artifacts sent: 0  received: 4
Clone finished with 601 bytes sent, 1198 bytes received
Rebuilding repository meta-data...
  100.0% complete...
project-id: 88519ecd92d8ad16e7ca2b4a16bd61ddbab883fa
admin-user: m0 (password is b8be21)

$ mkdir jogl-examples2
$ cd jogl-examples2
$ fossil open ../jogl-examples2.fossil 
$ ls
$ ls
$ echo  README.txt
$ fossil add README.txt 
ADDED  README.txt
$ fossil commit -m 'README'
Autosync:  http://m...@fossil.int.arc7.info/com.io7m.cirrus/jogl-examples
Round-trips: 1   Artifacts sent: 0  received: 0
Pull finished with 377 bytes sent, 287 bytes received
New_Version: cf09a2a5e14f2b4f04a94f11ba76203512f03778
Autosync:  http://m...@fossil.int.arc7.info/com.io7m.cirrus/jogl-examples
Round-trips: 1   Artifacts sent: 2  received: 0
Sync finished with 1245 bytes sent, 340 bytes received

So this time, it actually asked me for a password on cloning (which I don't
remember it ever doing before). I entered the password that was printed
when the repository was created on the server side and it seemed happy.

--8--

And then:

--8--

Something's definitely broken.

On the server side:

$ fossil init broken.fossil
project-id: 1010ff8e54729a6fc501aa32aa97485b32cdf79a
server-id:  8ca0cf03909980c31afecd0d53584c7a0886347d
admin-user: fossil (initial password is 63cdac)
$ fossil user new m0 -R broken.fossil 
contact-info: 
password:
Retype new password:
$ ls -alF broken.fossil 
-rw-rw-r--  1 fossil  source  58368 Nov 15 21:32 broken.fossil

On the client side:

$ fossil clone http://m...@fossil.int.arc7.info/com.io7m.cirrus/broken 
broken.fossil
password for m0: 
remember password (Y/n)? 
Round-trips: 2   Artifacts sent: 0  received: 3
Clone finished with 588 bytes sent, 1161 bytes received
Rebuilding repository meta-data...
  100.0% complete...
project-id: 1010ff8e54729a6fc501aa32aa97485b32cdf79a
admin-user: m0 (password is 632d09)

$ mkdir broken
$ cd broken
$ fossil open ../broken.fossil 
$ touch README.txt
$ fossil add README.txt 
ADDED  README.txt
$ fossil commit -m 'Initial'
Autosync:  http://m...@fossil.int.arc7.info/com.io7m.cirrus/broken
Round-trips: 1   Artifacts sent: 0  received: 0
Pull finished with 365 bytes sent, 287 bytes received
New_Version: ab4971a89e310d5f881847f8e5a38f5ae28c5996
Autosync:  http://m...@fossil.int.arc7.info/com.io7m.cirrus/broken
Round-trips: 1   Artifacts sent: 2  received: 0
Error: not authorized to write
Round-trips: 1   Artifacts sent: 2  received: 0
Sync finished with 1240 bytes sent, 313 bytes received
Autosync failed

M
___
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] Fwd: not authorized to write - how to debug?

[org.fossil-scm.fossil-users]

On 15 Nov 2013 16:54:31 -0700
Andy Bradford amb-sendok-1387151671.hehojnjibeaklgdje...@bradfords.org 
wrote:

 Thus said org.fossil-scm.fossil-us...@io7m.com on Fri, 15 Nov 2013 23:37:52 
 +:
 
  Oh dear...  I think we've found  the problem. I'd cloned  the original
  repositories over SSH  on the same server! I've no  idea why I thought
  it was over HTTP before.
 
 That explains a lot. SSH URLs  have full permission when cloning (unless
 restricted using an SSH key and REMOTE_USER environment variable).

Right, makes sense.

  I'm not sure this explains the apparently inconsistent behaviour in my
  last email, but it would at least  explain why I've not run into it to
  date.
 
 Perhaps  I  missed  it,  but  will you  explain  what  you  observed  as
 inconsistent?

I'm not sure I did. I'd been trying various things on the original examples
repository to test Stephan's original off-list suggestions which probably left
me with a repository in a somewhat inconsistent state with regards to the things
we'd been discussing (capabilities, etc), and then Ron Wilson mentioned 
something that I mistakenly took as agreement that something unusual was 
happening.

Anyway, the following seems to work over http:

server$ fossil init notbroken.fossil
project-id: b9ee7d556e6dbd445a29e36c8e1c83299555d6a6
server-id:  e2e18e29ee5ebe54efe992096e308066b194e7b8
admin-user: fossil (initial password is 68accc)
server$ fossil user new -R notbroken.fossil
...
server$ fossil user capabilities m0 -R notbroken.fossil
u
server$ fossil user capabilities m0 v -R notbroken.fossil
v

Client:

$ fossil clone http://m...@fossil.int.arc7.info/com.io7m.cirrus/notbroken 
notbroken.fossil
password for m0: 
remember password (Y/n)? 
Round-trips: 2   Artifacts sent: 0  received: 3
Clone finished with 594 bytes sent, 1160 bytes received
Rebuilding repository meta-data...
  100.0% complete...
project-id: b9ee7d556e6dbd445a29e36c8e1c83299555d6a6
admin-user: m0 (password is 65a74d)
$ mkdir notbroken
$ cd notbroken
$ fossil open ../notbroken.fossil 
$ echo Hello  README.txt
$ fossil add README.txt 
ADDED  README.txt
$ fossil commit -m 'Hello'
Autosync:  http://m...@fossil.int.arc7.info/com.io7m.cirrus/notbroken
Round-trips: 1   Artifacts sent: 0  received: 0
Pull finished with 368 bytes sent, 285 bytes received
New_Version: bdf188dd290370f38f2905959718328a8446a111
Autosync:  http://m...@fossil.int.arc7.info/com.io7m.cirrus/notbroken
Round-trips: 1   Artifacts sent: 2  received: 0
Sync finished with 1245 bytes sent, 341 bytes received

... So I think everything's OK!

M
___
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] interactive patching?

[org.fossil-scm.fossil-users]

On Thu, 31 Oct 2013 13:10:25 +0100
Stephan Beal sgb...@googlemail.com wrote:

 On Thu, Oct 31, 2013 at 1:20 AM, org.fossil-scm.fossil-us...@io7m.comwrote:
 
  4. Use some external tool to interactively apply bits of stash.diff
  ...

  The main problem: Where is the tool to achieve step 4? I've looked
  and am not aware of any standalone tool that can essentially break
  up a patch and interactively apply parts of it to a directory. I
  personally use the 'Team - Apply patch' tool built into Eclipse,
  but I'm curious as to what other people to do achieve the above.
 
 
 Kompare (for Linux) can do that. It's what i always use to (partially)
 apply such patch sets.
 
 http://www.kde.org/applications/development/kompare/

Hm, right!

How is that actually done? I installed Kompare a weeks ago and couldn't see
a way to apply a patch in the UI:

http://waste.io7m.com/2013/10/31/kompare0.png
http://waste.io7m.com/2013/10/31/kompare1.png
http://waste.io7m.com/2013/10/31/kompare2.png

... So I sort of assumed that it couldn't do it. Is there something
obvious I'm missing?

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

[fossil-users] interactive patching?

[org.fossil-scm.fossil-users]

'Lo.

It's fairly common, in git, to do this sort of thing:

1. Make a load of unrelated changes.
2. Use git add --patch to stage commits consisting of
   sets of the changes.

The reason this doesn't work to well is probably obvious to most
fossil users: You don't know that each set of separate staged changes
compiles and works properly, because you're not compiling against a
clean working copy with only those individual changes. I don't know
if git has a better way of doing things now, as it's been quite a
while since I used it.

The somewhat better analogous sequence in fossil is:

1. Make a load of unrelated changes.
2. fossil stash
3. fossil stash diff  stash.diff
4. Use some external tool to interactively apply bits of stash.diff
5. Compile, run unit tests, commit if everything works.
6. If there's anything left in stash.diff, repeat step 4.

This is better because it means each set of changes are tested
against a clean copy of the source (so each subset of unrelated
changes can be shown to not be dependent, etc).

The main problem: Where is the tool to achieve step 4? I've looked
and am not aware of any standalone tool that can essentially break
up a patch and interactively apply parts of it to a directory. I
personally use the 'Team - Apply patch' tool built into Eclipse,
but I'm curious as to what other people to do achieve the above.

I realise that often the answer is don't make a lot of unrelated
changes at once, but for the sake of discussion, let's assume that
this is sometimes unavoidable (and undesirable).

M
___
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] CGI mode with multiple repositories

[org.fossil-scm.fossil-users]

On Wed, 5 Jun 2013 18:52:28 +0200
Stephan Beal sgb...@googlemail.com wrote:

 On Wed, Jun 5, 2013 at 11:51 AM, Kevin Martin ke...@khn.org.uk
 wrote:
 
  fossils/
  symlink_to_/var/fossils/repos1.fossil
  symlink_to_/var/fossils/repos2.fossil
  …
 
 
 I like this solution because:

One minor issue, although I'm not sure if it's a problem: How well does
this work on Windows? I'm aware NTFS has something analogous to
symbolic links, but do they behave consistently (with respect to
UNIX-ish symlinks)?

I don't particularly care, personally, but I'm aware that fossil
strives to work consistently on all platforms. I'd hesitate to put
something in that'd only work on UNIX-likes.

M
___
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] CGI mode with multiple repositories

[org.fossil-scm.fossil-users]

On Tue, 4 Jun 2013 08:28:20 -0400
Richard Hipp d...@sqlite.org wrote:
 
  I've figured it out. The errors happens when I access the cgi url
  without a repository name. When I provide a repository name, it
  works fine.
 
 
 Sounds like a bug in Fossil that needs fixing.
 

Apologies for jumping in, but I'd *love* to see a basic generated
listing of the directories fossil is managing in this case. Seems
as though it'd be fairly easy to grab the project name/description
from each fossil database in the directory tree managed by the server.

It's been a while since I've used gitweb, but I think this would make
fossil feature-comparable to it.

M
___
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] CGI mode with multiple repositories

[org.fossil-scm.fossil-users]

On Tue, 4 Jun 2013 11:25:19 -0400
Richard Hipp d...@sqlite.org wrote:

 On Tue, Jun 4, 2013 at 11:20 AM,
 org.fossil-scm.fossil-us...@io7m.comwrote:
  Apologies for jumping in, but I'd *love* to see a basic generated
  listing of the directories fossil is managing in this case. Seems
  as though it'd be fairly easy to grab the project name/description
  from each fossil database in the directory tree managed by the
  server.
 
 
 It would be.  Why don't you send in a patch?

I'm somewhat snowed under at the moment, but I'll see what I can do.

I think it'd be sensible to limit how far the server is allowed to
recurse into directories. Gitweb handles this by having the user
actually edit the gitweb script, but I'm not sure where this
configurable value should go in Fossil.

M
___
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] CGI mode with multiple repositories

[org.fossil-scm.fossil-users]

On Tue, 4 Jun 2013 17:05:29 +0100
Francis Daly fran...@daoine.org wrote:
 
 Create /var/fossil/index.fossil where the front web page is the
 editable-to-some list of the repos to make easily visible.
 
 fossil server --notfound /index /var/fossil
 
 Manual updates needed; anything not listed is still accessible if the
 name is known; but it is a cheap way of making a friendly list
 available.
 
 Too naive?

No, it's certainly an option.

I was picturing something along these lines:

  http://fossil.io7m.com

The output comes from a small program running as a CGI. It doesn't do
any recursion into subdirectories at all. I was going to attempt to
provide something that worked similarly. One option would be to do
unlimited recursion, but use an optional whitelist.fossil (equivalent
to the index.fossil you suggested) containing the names of fossil
repositories and then only consider those files when generating the
index.

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

[fossil-users] Database is locked on sync after connection loss

[org.fossil-scm.fossil-users]

Hello.

I've been running a set of scripts to periodically sync a set of
repositories to a remote server. Today, we had a few problems with
our connection to the net and it seems that a few syncs were in
progress when the connection went down. Now, for one (or possibly
more, I've not tried them all yet) of the repositories, attempting
to sync always results in a Database locked error:

[fossil@fossil ~]$ ./fossil sync -R com.io7m/io7m-changelog.fossil 
ssh://i...@io7m.com/f-db/io7m-changelog.fossil
./fossil: database is locked
COMMIT

If you have recently updated your fossil executable, you might
need to run fossil all rebuild to bring the repository
schemas up to date.

I don't see any obvious stale lock files in the local or remote
directories.

Any idea what's going on here?

[fossil@fossil ~]$ ./fossil version
This is fossil version 1.25 [558a17a686] 2012-12-22 13:48:31 UTC

M
___
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] Database is locked on sync after connection loss

[org.fossil-scm.fossil-users]

On Sun, 2 Jun 2013 17:57:59 +
org.fossil-scm.fossil-us...@io7m.com wrote:
 I don't see any obvious stale lock files in the local or remote
 directories.
 
 Any idea what's going on here?

It turns out that there was a rogue fossil process still hanging
around. No idea why it didn't time out.

Killing the process unlocked the database and everything appears to be
working now.

M

___
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] Ticket 'subsystem' field?

[org.fossil-scm.fossil-users]

On Sun, 24 Mar 2013 19:20:26 -0400
Doug Franklin nutdriverle...@comcast.net wrote:

  I sort of expected entries to appear there after I'd set the
  fields of a few tickets, but they didn't.
 
 Menu Admin, Tickets, Common.  Edit the values in the 
 subsystem_choices group.
 

Ah, thanks!

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

[fossil-users] Ticket 'subsystem' field?

[org.fossil-scm.fossil-users]

Hello.

I can set the subsystem field of a given ticket to XYZ with fossil
ticket set X subsystem XYZ, but I can't seem to work out how to
populate the drop down subsystem menu in the web interface. Is this
documented anywhere?

I sort of expected entries to appear there after I'd set the fields of
a few tickets, but they didn't.

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

[fossil-users] Request: default ticket resolution 'Open'

[org.fossil-scm.fossil-users]

Hello.

When creating a new ticket (on a fresh repository, using fossil version
4f510b66cb), the resolution field is unset. If I then edit the ticket,
but make no changes, and then save, the resolution field is set to Open.

I feel that setting the resolution to Open on ticket creation might be
slightly more useful behaviour (as if you at some point later on write
an Open tickets report - WHERE resolution = 'Open' - you have to
click through all of your open tickets to correctly set the resolution).

M
___
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] Suggested way to sync to multiple servers?

[org.fossil-scm.fossil-users]

On Fri, 11 Jan 2013 11:36:10 +0100
Remigiusz Modrzejewski l...@maxnet.org.pl wrote:

 On Jan 11, 2013, at 00:47 , org.fossil-scm.fossil-us...@io7m.com
 wrote:
 
  My use case here is that I've got repositories hosted on
  http://fossil.io7m.com but I've also got an exact mirror of that
  site on my internal network here that's used for testing quickly on
  multiple machines/VMs (and will now be polled very frequently for
  CI testing, which would be slow and would start to get expensive in
  terms of data usage with the number of repositores over the WAN).
 
 Mirrors can be easily kept up to date with cron. 
 An empty (no commits exchanged) sync is something like 0.03s for my
 laptop's CPU, so it can be even every minute without overloading the
 machine.

Yes, that's possible. It's not a very satisfying solution though as I
now have to have extra infrastructure running and have to make lots of
frequent redundant pulls/pushes as opposed to only doing them when
they're actually needed.

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

[fossil-users] Suggested way to sync to multiple servers?

[org.fossil-scm.fossil-users]

Hello.

It seems as though it's only really possible to sync to one remote
server at a time. Is there a way to specify that a repository should
push to multiple remote servers?

My use case here is that I've got repositories hosted on
http://fossil.io7m.com but I've also got an exact mirror of that site
on my internal network here that's used for testing quickly on multiple
machines/VMs (and will now be polled very frequently for CI testing,
which would be slow and would start to get expensive in terms of data
usage with the number of repositores over the WAN).

I'd quite like to either be able to push to multiple remotes, or
possibly to be able to push to a remote and then have that remote push
to a further remote.

Right now, it seems the only way I can do this is to maintain a list of
remotes with each repository, and manually push to each URL in the list
with push --once. That obviously doesn't play nicely with fossil all
sync!

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

[fossil-users] Most future-proof way to query fossil repositories (and checked out copies!)

[org.fossil-scm.fossil-users]

Hello.

I'm working on a Fossil plugin for Jenkins (http://jenkins-ci.org), and
in order to implement the plugin, I have to query fossil repositories
in various ways.

I'm calling the fossil executables for cloning, pulling, and opening
repositories, but I'm not exactly sure what I should be doing for
queries such as What version is checked out in the current directory?
and Have any commits been made on branch B since this version was
checked out?

Should I be opening the repositories as SQLite databases and querying
the tables? Is the schema subject to change constantly? For that
matter, should I be parsing the output of the commands? Is the format
of the output subject to unannounced changes?

Personally, I'd rather use SQL than parse text...

Regards,
M
___
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] Most future-proof way to query fossil repositories (and checked out copies!)

[org.fossil-scm.fossil-users]

On Wed, 9 Jan 2013 15:53:25 -0500
Richard Hipp d...@sqlite.org wrote:

 The schema is subject to change.  And years ago it would change on a
 regular basis.  But the schema is mostly stable now.  There may be
 tweaks here and there, but it mostly it will likely be the same.
 
 The schema for the repository is divided into two parts, the content
 schema and
 the auxiliary schema.  The content schema contains primary
 information. The
 auxiliary schema contains values computed from the primary schema
 when you run fossil rebuild or make other changes.  The auxiliary
 schema is basically a
 cache for information that is contained in the content schema but
 which is difficult to compute on-the-fly.
 
 Two entries in the CONFIG table contain version numbers for these two
 schemas:
 
 SELECT * FROM config WHERE name GLOB '*-schema';
 
 So you can monitor those two values to see if the schema changes out
 from under you.
 
 The schema for the .fslckout database is separate.  It is
 unversioned, unfortunately.
 

Thanks!

It should be reasonably easy to stay current as I don't exactly need
much out of the databases.

I think I *may* be reasonably safe with the .fslckout database as it
seems I only need a single value: the 'checkout' row from the 'vvar'
table.

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

[fossil-users] 'SQLITE_ERROR: no such table: user' from web interface

[org.fossil-scm.fossil-users]

Hi.

I've just noticed, on a few repositories, that I'm seeing
SQLITE_ERROR: no such table: user in red at the top of all
pages. This seems to happen when I'm logged into another
repository in the same login group. I've tried fossil all rebuild
to no avail.

Any idea what could cause this?
___
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] 'SQLITE_ERROR: no such table: user' from web interface

[org.fossil-scm.fossil-users]

On Mon, 9 Jul 2012 15:56:10 +
org.fossil-scm.fossil-us...@io7m.com wrote:

 Hi.
 
 I've just noticed, on a few repositories, that I'm seeing
 SQLITE_ERROR: no such table: user in red at the top of all
 pages. This seems to happen when I'm logged into another
 repository in the same login group. I've tried fossil all rebuild
 to no avail.
 
 Any idea what could cause this?

I've just discovered that this is (apparently) caused by someone
moving the database files (thus breaking the links in the login group).
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

[fossil-users] Request: Hierarchical wiki navigation

[org.fossil-scm.fossil-users]

Hello.

One of the main criticisms I have of most wiki software is that it
demands users maintain their own navigational links when creating
categories of pages. This usually results a total mess after a year
or so of editing.

I'd like to make a request that I'm currently unable to implement
due to severe lack of time:

User creates 'directory' page.
User creates {'directory/a' 'directory/b' 'directory/c'} pages.

Now, via some hierarchy page off the main 'directory' page (perhaps
a link in the titlebar, where 'edit' normally appears?), it should be
possible to get a list containing {'directory/a' 'directory/b'
'directory/c'}. On each of {'directory/a' 'directory/b'
'directory/c'}, there should be a 'parent' link to return to
'directory'.

It would also be nice if renaming 'directory' could optionally rename
{'directory/a' 'directory/b' 'directory/c'}.

___
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] List of checkins

[org.fossil-scm.fossil-users]

On Sat, 26 May 2012 18:02:41 -0500
Brian Smith br...@linuxfood.net wrote:

 Another way, just to add to the list:
 
 echo -e .mode csv\nselect uuid from event, blob where
 event.objid=blob.rid and type='ci' and mtime 
 julianday('2012-05-27'); | fsl sqlite
 
 -B
 
 
 On Sat, May 26, 2012 at 5:51 PM, Stephan Beal sgb...@googlemail.com
 wrote:
 
  On Sun, May 27, 2012 at 12:49 AM, Stephan Beal
  sgb...@googlemail.comwrote:
 
   [stephan@host:~/cvs/fossil/fossil]$ f json timeline checkin -b
  2012-05-27 -I 1 | awk -F : '/uuid:/{print $2}' | cut -d'' -f2
 
 
  Slightly more succinct, but should work the same:
 
   f json timeline checkin -b 2012-05-27 -I 2 | grep 'uuid' | cut
  -d'' -f4

Thanks to all who replied.

I've settled on the SQL solution above. The version of fossil I'm using
doesn't have JSON enabled. I did originally attempt an SQL solution but
after having looked at the documentation on the site, I didn't think
it'd be as straightforward as a single SELECT - turns out that it is!

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

[fossil-users] List of checkins

[org.fossil-scm.fossil-users]

Hello.

Is it possible to get a more raw list than 'fossil timeline before
2012-05-27 -t ci' out of the command line tool? Would prefer a straight
list of newline-separated hash values.
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

[fossil-users] losing history in private branch merge?

[org.fossil-scm.fossil-users]

Hello.

1) I created a private branch, made several commits, and then merged the
private branch with the current trunk. Rather than seeing the commits
I made on the private branch in the timeline for the trunk, I only see
the one large commit resulting from accumulating all the smaller commits
into one when merging.

Is it possible to avoid squashing all private commits into one?

2) Additionally, as auto-sync was enabled, the large commit was pushed
to a remote. I realize it's not possible to rewrite history (nor would I
want to), but is there some way to push any changes I make whilst trying
to sort out item 1 above (so that the private commits are visible)?
___
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] losing history in private branch merge?

[org.fossil-scm.fossil-users]

On Fri, 25 May 2012 11:53:35 -0400
Richard Hipp d...@sqlite.org wrote:

 On Fri, May 25, 2012 at 11:42 AM,
 org.fossil-scm.fossil-us...@io7m.comwrote:
  Is it possible to avoid squashing all private commits into one?
 
 
 The branch is private.  If all the individual commits where pushed
 out to the world, it wouldn't be private any more and the whole
 purpose of having a private branch would be defeated, no?
 

That's one interpretation of private, yes. I took it to mean that the
branch wouldn't be synced, or visible, on any remotes. I don't think
that necessarily implies coalescing commits like that...

If it's not possible, I can live with it, I'll just switch to only using
public branches.


___
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] losing history in private branch merge?

[org.fossil-scm.fossil-users]

On Fri, 25 May 2012 16:10:44 +
org.fossil-scm.fossil-us...@io7m.com wrote:

 On Fri, 25 May 2012 11:53:35 -0400
 Richard Hipp d...@sqlite.org wrote:
 
  On Fri, May 25, 2012 at 11:42 AM,
  org.fossil-scm.fossil-us...@io7m.comwrote:
   Is it possible to avoid squashing all private commits into one?
  
  
  The branch is private.  If all the individual commits where pushed
  out to the world, it wouldn't be private any more and the whole
  purpose of having a private branch would be defeated, no?
  
 
 That's one interpretation of private, yes. I took it to mean that the
 branch wouldn't be synced, or visible, on any remotes. I don't think
 that necessarily implies coalescing commits like that...
 
 If it's not possible, I can live with it, I'll just switch to only
 using public branches.

Is it possible to get those commits into trunk in any way? Losing the
details isn't acceptable (due to my misunderstanding of exactly what
a private branch entailed).

___
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] losing history in private branch merge?

[org.fossil-scm.fossil-users]

On Fri, 25 May 2012 10:18:34 -0700
Matt Welland estifo...@gmail.com wrote:

 On Fri, May 25, 2012 at 9:10 AM,
 org.fossil-scm.fossil-us...@io7m.comwrote:
 
  On Fri, 25 May 2012 11:53:35 -0400
  Richard Hipp d...@sqlite.org wrote:
 
   On Fri, May 25, 2012 at 11:42 AM,
   org.fossil-scm.fossil-us...@io7m.comwrote:
Is it possible to avoid squashing all private commits into one?
   
  
   The branch is private.  If all the individual commits where pushed
   out to the world, it wouldn't be private any more and the whole
   purpose of having a private branch would be defeated, no?
  
 
  That's one interpretation of private, yes. I took it to mean that
  the branch wouldn't be synced, or visible, on any remotes. I don't
  think that necessarily implies coalescing commits like that...
 
  If it's not possible, I can live with it, I'll just switch to only
  using public branches.
 
 
 It sounds like the request would be able to switch a branch from
 private to public. That seems like it would be a very powerful
 feature (assuming it is not already possible?).

Yeah, I think the direct approach would be (if the feature existed):

1. Undo the merge commit (perhaps by shunning).
2. Mark the private branch as public.
3. Merge again.


___
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] losing history in private branch merge?

[org.fossil-scm.fossil-users]

On Fri, 25 May 2012 13:34:54 -0400
Richard Hipp d...@sqlite.org wrote:

 I don't think so, not other than checking each one out and
 recommitting them one by one.  To do otherwise would be changing the
 history of the project, which Fossil does not allow (by design).

That's fine.

Which commands should I use to do the above? I want to avoid screwing
things up even further, so I'd like to be sure...
___
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] losing history in private branch merge?

[org.fossil-scm.fossil-users]

On Fri, 25 May 2012 17:41:44 +
org.fossil-scm.fossil-us...@io7m.com wrote:

 On Fri, 25 May 2012 13:34:54 -0400
 Richard Hipp d...@sqlite.org wrote:
 
  I don't think so, not other than checking each one out and
  recommitting them one by one.  To do otherwise would be changing the
  history of the project, which Fossil does not allow (by design).
 
 That's fine.
 
 Which commands should I use to do the above? I want to avoid screwing
 things up even further, so I'd like to be sure...

Expecting disaster, I made a copy of the database.

I shunned the merge commit from the web interface. Then, for each
commit in the private branch, I attempted to merge the private commit
with the current trunk (which had not changed since the start of the
private branch). Within two commits, I somehow had merge conflicts (how
is that even possible?) and files from a commit that I'd not yet merged.

Assuming I'd done something wrong, I went back to the original copy of
the database and tried again. Same result. I give up.

___
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] Names in imported git repositories

[org.fossil-scm.fossil-users]

On Sat, 17 Mar 2012 03:11:56 +0100
Dmitry Chestnykh dmi...@codingrobots.com wrote:

 On 03/15/2012 02:38 PM, org.fossil-scm.fossil-us...@io7m.com wrote:
  Is it possible to tell fossil that commits should be assigned to
  specific user names in the resulting fossil repository?
 
 When I converted my repos, I just did a global search/replace in the 
 file exported from git.
 

That's fine. Just wanted to be sure that this wasn't a bug.
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

[fossil-users] Names in imported git repositories

[org.fossil-scm.fossil-users]

Hello.

I'm attempting to import a test repository from git. Unfortunately, the
names used in the resulting fossil commits are the *email addresses* of
the original committers.

Is it possible to tell fossil that commits should be assigned to
specific user names in the resulting fossil repository?
___
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] Set theme from command line

[org.fossil-scm.fossil-users]

On Wed, 14 Mar 2012 21:10:21 +0100
Stephan Beal sgb...@googlemail.com wrote:

 On Wed, Mar 14, 2012 at 9:08 PM, Stephan Beal sgb...@googlemail.com
 wrote:
 
  specifically, you want to export a config from one
 
 
 e.g.
 
 fossil config export skin my.skin
 fossil config import my.skin
 
 should do the trick.
 

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

[fossil-users] HTTP request on commit?

[org.fossil-scm.fossil-users]

Hello.

I'm evaluating fossil as a replacement for existing git infrastructure.
One feature we do need is the ability to notify a server when a commit
occurs.

Is there a way to get fossil to make an HTTP request to a given address
on commit (with information about the commit such as SHA1, message,
timestamp, etc)?
___
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] HTTP request on commit?

[org.fossil-scm.fossil-users]

On Mon, 5 Mar 2012 09:23:47 -0500
Richard Hipp d...@sqlite.org wrote:
 On Mon, Mar 5, 2012 at 8:07 AM,
 org.fossil-scm.fossil-us...@io7m.comwrote:
  Ah, yeah, should have explained that. It's essentially just for a
  notification service similar to that offered by CIA.vc.
 
 I'm not familiar with CIA.vc.  But if you just want notifications of
 changes, did you know that Fossil servers have an RSS feed for this?
 Example:
 
 http://www.fossil-scm.org/fossil/timeline.rss

That's not too bad, but it lacks some metadata we'd like. It also
seems to lump all events (check-ins, tickets, etc) into one stream,
so we'd have to somehow parse it and strip out anything that doesn't
look like a check-in. It's also rather heavy in that if, for example,
we're running a server that posts notification messages to an XMPP or
IRC channel, that server has to monitor a potentially large number of
feeds as opposed to (almost blindly) accepting notification
messages from the network.

I'm not really sure what the most pleasant way to achieve the desired
effect is, but having fossil send a small request to an external server
seemed the easiest to manage in a platform-independent manner (git
commit hooks become pretty unpleasant if developers are working across
different platforms and have to replicate their scripts across all of
their machines).

___
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] HTTP request on commit?

[org.fossil-scm.fossil-users]

On Mon, 5 Mar 2012 10:04:27 -0500
Richard Hipp d...@sqlite.org wrote:

 On Mon, Mar 5, 2012 at 8:49 AM,
 org.fossil-scm.fossil-us...@io7m.comwrote:
  That's not too bad, but it lacks some metadata we'd like. It also
  seems to lump all events (check-ins, tickets, etc) into one stream,
  so we'd have to somehow parse it and strip out anything that doesn't
  look like a check-in.
 
 
 A query parameter can fix this:
 
 http://www.fossil-scm.org/fossil/timeline.rss?y=ci
 

Ok, thanks. I'll try to get something together with this.
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users