How to get the latest Tcl/Tk sources?
Georgios Petasis
How I can checkout the latest sources of Tcl/Tk?
I have found some instructions of this fossil stuff in the past, but I
cannot find them now.
And searching the wiki, didn't return any help.
George
Gerald W. Lester
1) Download executable of fossil from http://www.fossil-scm.org/ (google
term: fossil scm)
2) Do a fossil help (to dig into help system)
3) Do a fossil clone
Done.
--
+------------------------------------------------------------------------+
| Gerald W. Lester, President, KNG Consulting LLC |
| Email: Gerald...@kng-consulting.net |
+------------------------------------------------------------------------+
Georgios Petasis
> On 5/23/11 11:04 AM, Georgios Petasis wrote:
>> Hi all,
>>
>> How I can checkout the latest sources of Tcl/Tk?
>> I have found some instructions of this fossil stuff in the past, but I
>> cannot find them now.
>> And searching the wiki, didn't return any help.
>
> 1) Download executable of fossil from http://www.fossil-scm.org/ (google
> term: fossil scm)
> 2) Do a fossil help (to dig into help system)
> 3) Do a fossil clone
>
> Done.
>
I tried:
fossil pull
fossil update
and I saw some updates. I don't know if these are all the updates, but
anyway..
George
Gerald W. Lester
Sorry, try: fossil sync
Georgios Petasis
> On 5/23/11 1:43 PM, Georgios Petasis wrote:
>> Στις 23/5/2011 20:25, ο/η Gerald W. Lester έγραψε:
>>> On 5/23/11 11:04 AM, Georgios Petasis wrote:
>>>> Hi all,
>>>>
>>>> How I can checkout the latest sources of Tcl/Tk?
>>>> I have found some instructions of this fossil stuff in the past, but I
>>>> cannot find them now.
>>>> And searching the wiki, didn't return any help.
>>>
>>> 1) Download executable of fossil from http://www.fossil-scm.org/ (google
>>> term: fossil scm)
>>> 2) Do a fossil help (to dig into help system)
>>> 3) Do a fossil clone
>>>
>>> Done.
>>>
>> Thanks, for you help, but I had already a clone, and I wanted to update.
>>
>> I tried:
>>
>> fossil pull
>> fossil update
>>
>> and I saw some updates. I don't know if these are all the updates, but
>> anyway..
>
> Sorry, try: fossil sync
>
Thank you very much.
George
Kevin Walzer
>>
>> Sorry, try: fossil sync
>>
>
> Thank you very much.
>
> George
i find this helpful:
--
Kevin Walzer
Code by Kevin
http://www.codebykevin.com
Andreas Leitgeb
> On 5/23/11 3:01 PM, Georgios Petasis wrote:
>>> Sorry, try: fossil sync
>> Thank you very much.
> http://wiki.tcl.tk/28126
I use fossil in the version installable for Ubuntu 10.10:
This is fossil version [15cb835736] 2010-06-17 18:39:10 UTC
fossil clone (from mirror1) seems to have worked,
fossil pull -R (from core) too,
but
fossil open (with path to tcl.fossil ditto for tk.fossil)
creates some files and directories (maybe even all them)
and then SegFaults. :-(
the ChangeLog's most current entry is from 2008-06-25, "tagging
8.5.3 for release", so it seems I'm on a wrong branch...
Wiki suggests using core-8-5-branch, but I'd like 8.6, and
this I hoped to find on the main-branch... with core-8-6-branch
not existing, either.
fossil branch list | grep 8.6 didn't find anything looking
like a version 8.6 branch name, either.
PS: df -kl indicates plenty of space left on device, just to
rule out the most trivial possible cause.
PPS: I'm not a maintainer, and my intention is, to casually do
my own private experiments and casually merging in changes
from upstream. A rather typical usecase, I'd think.
Alexandre Ferrieux
wrote:
>
> PPS: I'm not a maintainer, and my intention is, to casually do
> my own private experiments and casually merging in changes
> from upstream. A rather typical usecase, I'd think.
Yes, I admit that it's nasty that a not-so-old, bundled version of
fossil misbehaves :/
Two options I guess:
(a) fetch and recompile a more recent fossil from http://fossil-scm.org
(b) fetch a tarball of current Tcl from http://core.tcl.tk/tcl/info/tip
and manually maintain your updates as a patch.
Yes, (b) is awfully tedious, Why not (a), then ? ;-)
-Alex
Andreas Leitgeb
> On 24 mai, 10:17, Andreas Leitgeb <a...@gamma.logic.tuwien.ac.at>
> wrote:
>> PPS: I'm not a maintainer, and my intention is, to casually do
>> my own private experiments and casually merging in changes
>> from upstream. A rather typical usecase, I'd think.
> Yes, I admit that it's nasty that a not-so-old, bundled version of
> fossil misbehaves :/
I'd just like to understand the world "as is" a bit better:
- is the segfault a known-bug - known also as to what it really
matters? (I've seen enough exit-segfaults to not assume
this automatically)
- "should" a plain "fossil open /path/to/tcl.fossil" with a correct
fossil binary have given me some 8.6([ab])/head-version?
- Is there some kind of repository-check, that would tell me
whether my local file tcl.fossil is broken, in the first place?
It would be good to add a specific warning to the wiki-page
as to the minimal version of fossil that doesn't have that bug,
or which precompiled versions of various os-distributions are
specifically known *not* to work. (I'd add that to the wiki, myself
as soon as I'm confident about what to write...)
Typically, it is favourable to use the distribution packages for
stuff that is not the primary target of interest (i.e. just a tool),
unless, of course, there exists some stringent counterindication.
Just for the record: using the same fossil to clone & open fossil's
own repository did *not* result in a segfault.
PS: I'll do an ubuntu-update to 11.04 soon, anyway, and retry with
the version offered there, before building from the fossil-repo...
Alexandre Ferrieux
wrote:
> Alexandre Ferrieux <alexandre.ferri...@gmail.com> wrote:
> > On 24 mai, 10:17, Andreas Leitgeb <a...@gamma.logic.tuwien.ac.at>
> > wrote:
> >> PPS: I'm not a maintainer, and my intention is, to casually do
> >> my own private experiments and casually merging in changes
> >> from upstream. A rather typical usecase, I'd think.
> > Yes, I admit that it's nasty that a not-so-old, bundled version of
> > fossil misbehaves :/
>
> I'd just like to understand the world "as is" a bit better:
> - is the segfault a known-bug - known also as to what it really
> matters? (I've seen enough exit-segfaults to not assume
> this automatically)
My own experience at the beginning of our mass conversion to fossil in
Feb-Mar, is a few exchanges with Richard where I'd report similar
things and he'd reply "Aha. Fixed. Fetch fossil HEAD and retry."...
> - "should" a plain "fossil open /path/to/tcl.fossil" with a correct
> fossil binary have given me some 8.6([ab])/head-version?
Yes, unless the 'fossil clone' itself improperly fetched the repo.
Which is what happened to me several times.
> - Is there some kind of repository-check, that would tell me
> whether my local file tcl.fossil is broken, in the first place?
Dunno.
> It would be good to add a specific warning to the wiki-page
> as to the minimal version of fossil that doesn't have that bug,
> or which precompiled versions of various os-distributions are
> specifically known *not* to work. (I'd add that to the wiki, myself
> as soon as I'm confident about what to write...)
Yes. Maybe you can ask Richard directly, to know the proper threshold
version.
> Typically, it is favourable to use the distribution packages for
> stuff that is not the primary target of interest (i.e. just a tool),
> unless, of course, there exists some stringent counterindication.
Agreed 200%. Nasty as I said.
> Just for the record: using the same fossil to clone & open fossil's
> own repository did *not* result in a segfault.
The repo on core.tcl.tk has undergone thorough brain surgery in the
effort to keep as many diamonds (and as little rust) as possible from
the CVS past. Possibly this is what triggered those new bugs on an
otherwise mature Fossil.
> PS: I'll do an ubuntu-update to 11.04 soon, anyway, and retry with
> the version offered there, before building from the fossil-repo...
Be sure to pass the data (of the precise fossil version bundled in
U1104) on to Richard, that'll help.
Thanks !
-Alex