Hi!
I realise that at least for now, PHP is sticking with SVN. No problems.
However...I find it much easier to work in Mercurial to put together
patches, find bugs, etc.. And in fact, I find SVN really awkward--I was
on the verge of switching back to CVS when DVCSes such as Mercurial and
Git became popular. I would put in a vote for Mercurial if PHP decides
to move to a DVCS. Mercurial is more similar to CVS/SVN in its command
set than other DVCSes, is easy and friendly to use, generally does what
you would expect, makes it pretty hard to shoot yourself in the foot,
and includes good Windows support. Operating with the Mozilla and Vim
Mercurial repositories is a lovely experience, and I'd love to add PHP
to the list of projects I follow this way.
However, pushing the team towards a DVCS is not what I'm writing for.
Certainly not to push for a particular one.
I'd just like to be able to use my tool of choice (Mercurial) locally.
Would it be possible to set up a Mercurial mirror of the SVN repo on,
say, bitbucket.org, so I could pull from there? Alternatively, I can
access a git repo somewhere like github.com (I found
https://github.com/php/php-src but the last activity seems to be about 9
months ago--is it dead? are there any plans to revive it?). Of course,
as a last resort, I could just access SVN using Mercurial. However, it
may result in a lower server load if PHP can provide a mirror somewhere,
updated by a commit hook or daily cron job or something, rather than me
(and potentially many others, as DVCSes become even more popular)
accessing SVN directly.
Any possibility? (I'm happy to help, if I can be given appropriate
access to resources to do so.)
Ben.
However, it
may result in a lower server load if PHP can provide a mirror somewhere,
updated by a commit hook or daily cron job or something, rather than me
(and potentially many others, as DVCSes become even more popular)
accessing SVN directly.
The server load was actually what caused us to turn off the git
mirroring. Every time the mirror job ran it slammed the svn server and
there weren't enough people using it to warrant that load.
-Rasmus
I realise that at least for now, PHP is sticking with SVN. No problems.
I realise this is not the topic of discussion but I have to say, that
overall, a switch to DVCS would make a huge difference to PHP development
life cycles. Git for one, makes contributing and integration of patches a
completely different experience. It encourages more community participation
without impinging on quality since you don't need to grant commit
permissions. The whole workflow of creating and integrating patches is much
faster and more simple because you can switch context from branch to branch
almost instantly allowing those with commit permissions to verify if a
contribution is worth merging or not. It's much less work than SVN and the
ease naturally attracts contributors. Merging is not a pita like with SVN.
However, given the recent security breach there is another side: Tampering
with a git repository is virtually impossible because every commit hash is
generated from the previous ones, so if your servers were compromised again,
a change in the past history would require alteration every single commit
hash since that change and the resulting HEAD hash would be different.
Since hashes are based on the commit contents it's just not feasible even
if SHA1 was one day compromised that you could successfully tamper with a
previous commit and engineer the calculations so the current HEAD hash
remained unchanged. If a commit blob (on the file-system) was altered
manually, your git repo would simply fail to validate the next time you use
it. In every scenario you'd know immediately something was wrong and not
have to go looking for it commit by commit.
Something to consider again in the future at least.
Regards,
Drak
I realise that at least for now, PHP is sticking with SVN. No problems.
I realise this is not the topic of discussion but I have to say, that
overall, a switch to DVCS would make a huge difference to PHP development
life cycles. Git for one, makes contributing and integration of patches a
completely different experience. It encourages more community participation
without impinging on quality since you don't need to grant commit
permissions. The whole workflow of creating and integrating patches is much
faster and more simple because you can switch context from branch to branch
almost instantly allowing those with commit permissions to verify if a
contribution is worth merging or not. It's much less work than SVN and the
ease naturally attracts contributors. Merging is not a pita like with SVN.However, given the recent security breach there is another side: Tampering
with a git repository is virtually impossible because every commit hash is
generated from the previous ones, so if your servers were compromised again,
a change in the past history would require alteration every single commit
hash since that change and the resulting HEAD hash would be different.
Since hashes are based on the commit contents it's just not feasible even
if SHA1 was one day compromised that you could successfully tamper with a
previous commit and engineer the calculations so the current HEAD hash
remained unchanged. If a commit blob (on the file-system) was altered
manually, your git repo would simply fail to validate the next time you use
it. In every scenario you'd know immediately something was wrong and not
have to go looking for it commit by commit.Something to consider again in the future at least.
Verifying the svn repository was the least painful part of that experience.
-Rasmus
Hi,
I'm not a frequent poster in the list but I thought I'd really should give
my 1 cent here when I saw "popular" being an argument for using DVCSs, its
not, and its neither fashion nor cargo cult, it is just a plain eye opener
experience of how neither SVN or CVS are the base of all versioning (two of
its creators —Brian Fitzpatrick and Ben Collins-Sussman— have acknowledged
this by saying "sorry about that" with regards to Subversion) and that
better and more natural ways to collaborate and integrate code.
I could provide an epically long argument here, but instead I'll link to the
one I've already made, diagrams and graphics included =):
So, I don't want to make debate here of wether centralized is better than
distributed (because the point is moot), but I think its not a good
situation for the community to have a previously open door to DVCSs now
closed.
Perhaps a solution can be found to even open the door to Mercurial (that is
an excellent place to start with DVCSs because its simplicity and
straight-forwardness) in addition to git in such a way that doesn't stress
the server?.
Regards,
David
I realise that at least for now, PHP is sticking with SVN. No problems.
I realise this is not the topic of discussion but I have to say, that
overall, a switch to DVCS would make a huge difference to PHP development
life cycles. Git for one, makes contributing and integration of patches a
completely different experience. It encourages more community
participation
without impinging on quality since you don't need to grant commit
permissions. The whole workflow of creating and integrating patches is
much
faster and more simple because you can switch context from branch to branch
almost instantly allowing those with commit permissions to verify if a
contribution is worth merging or not. It's much less work than SVN and the
ease naturally attracts contributors. Merging is not a pita like with SVN.However, given the recent security breach there is another side: Tampering
with a git repository is virtually impossible because every commit hash is
generated from the previous ones, so if your servers were compromised
again,
a change in the past history would require alteration every single commit
hash since that change and the resulting HEAD hash would be different.
Since hashes are based on the commit contents it's just not feasible even
if SHA1 was one day compromised that you could successfully tamper with a
previous commit and engineer the calculations so the current HEAD hash
remained unchanged. If a commit blob (on the file-system) was altered
manually, your git repo would simply fail to validate the next time you use
it. In every scenario you'd know immediately something was wrong and not
have to go looking for it commit by commit.Something to consider again in the future at least.
Regards,
Drak
Hi,
I'm not a frequent poster in the list but I thought I'd really should give
my 1 cent here when I saw "popular" being an argument for using DVCSs, its
not, and its neither fashion nor cargo cult, it is just a plain eye opener
experience of how neither SVN or CVS are the base of all versioning (two of
its creators —Brian Fitzpatrick and Ben Collins-Sussman— have acknowledged
this by saying "sorry about that" with regards to Subversion) and that
better and more natural ways to collaborate and integrate code.I could provide an epically long argument here, but instead I'll link to
the one I've already made, diagrams and graphics included =):So, I don't want to make debate here of wether centralized is better than
distributed (because the point is moot), but I think its not a good
situation for the community to have a previously open door to DVCSs now
closed.Perhaps a solution can be found to even open the door to Mercurial (that is
an excellent place to start with DVCSs because its simplicity and
straight-forwardness) in addition to git in such a way that doesn't stress
the server?.
For us, there was huge reluctance in our community to move away from SVN and
we solved all our differenced with a GIT/SVN bridge. I personally wanted to
use GIT so I just checked out our SVN tree using GIT-SVN. Meant I could use
all the features of GIT locally and when I needed to svn update/svn commit,
GIT does that part invisibly. I am sure there are things like this for
Mercurial etc.
For windows: TortoiseGIT (has SVN support built in).
Debian/Ubuntu: apt-get install git-core git-svn
As much as purist for DVCS balk at the SVN bridges, I think they are an
important stepping stone for people who are used to SVN. It allows one to
"try it out" on the project the know and love without forcing the project to
switch. That's the better way. For those who would prefer GIT or whatever
but their project prefers SVN, the bridge allows them to use it without
upsetting the project status quo. They really work well too: locally you
can branch and merge and do all the gymnastics you cant achieve with SVN.
Overall, I think it's only recently that DVCS tools are becoming matured. I
know even a year ago GIT's toolset was horrific, but there have been recent
leaps an bounds in the toolsets and services like github.com have made the
entire experience very easy and social.
My main point is, if you want to use DVCS, there is nothing stopping you
doing right now, without asking the PHP team to change the VCS (although I
would hope one day they do).
Regards,
Drak
dukeofgaming wrote:
So, I don't want to make debate here of wether centralized is better than
distributed (because the point is moot), but I think its not a good
situation for the community to have a previously open door to DVCSs now
closed.
Having been battling with various projects that have decided that 'git is good'
or 'hg is best' ... I've been trying to live in the new world and failing. That
link is looking like a nice CURRENT overview of the situation although things
are still changing mainly for the good.
My problem until recently was that 'subrepo' management in DVCS has always been
playing catchup. Yesterday for the first time I checked out a 'super-project'
matching one of the composite builds provided on an original CVS repo! Still not
fully integrated into Eclipse, but hopefully even that will come with time. But
now I can 'clone' the half dozen different builds and manage them in parallel
with both github and bitbucket 'masters'.
I'm using Mercurial and TortoiseHg which almost work transparently between Linux
and Windows, and with hg-git access to the 'git is good' projects is also
transparent even on windows. Next step this weekend IS to tidy up the SVN setup
so that I have the PHP code locally mirrored as well ....
--
Lester Caine - G8HFL
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php
Hi,
I thought it might be interesting for all to read some feedback I got from
some questions I made to the Joomla community, specifically, the guys and
gals that have been implementing the decoupling of the Joomla framework: the
Joomla Platform.
They currently still use svn officially but they have been experimenting
with both mercurial and git. These are their responses so far:
http://groups.google.com/group/joomla-dev-framework/browse_thread/thread/fd3b0151cc215c0e
It may bring some insight to what a DVCS could do for the PHP community.
Best regards,
David
dukeofgaming wrote:
So, I don't want to make debate here of wether centralized is better than
distributed (because the point is moot), but I think its not a good
situation for the community to have a previously open door to DVCSs now
closed.Having been battling with various projects that have decided that 'git is
good' or 'hg is best' ... I've been trying to live in the new world and
failing. That link is looking like a nice CURRENT overview of the situation
although things are still changing mainly for the good.My problem until recently was that 'subrepo' management in DVCS has always
been playing catchup. Yesterday for the first time I checked out a
'super-project' matching one of the composite builds provided on an original
CVS repo! Still not fully integrated into Eclipse, but hopefully even that
will come with time. But now I can 'clone' the half dozen different builds
and manage them in parallel with both github and bitbucket 'masters'.I'm using Mercurial and TortoiseHg which almost work transparently between
Linux and Windows, and with hg-git access to the 'git is good' projects is
also transparent even on windows. Next step this weekend IS to tidy up the
SVN setup so that I have the PHP code locally mirrored as well ....--
Lester Caine - G8HFLContact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php
dukeofgaming wrote:
I thought it might be interesting for all to read some feedback I got
from some questions I made to the Joomla community, specifically, the
guys and gals that have been implementing the decoupling of the Joomla
framework: the Joomla Platform.They currently still use svn officially but they have been experimenting
with both mercurial and git. These are their responses so far:http://groups.google.com/group/joomla-dev-framework/browse_thread/thread/fd3b0151cc215c0e
It may bring some insight to what a DVCS could do for the PHP community.
I think that there are perhaps still a couple of separate problems here which
the DVCS community need to get behind for any of them to become a 'front runner'.
The main problem here IS 'subrepo' which is a concept that neither HG or GIT
handle particularly well, but to my mind is essential for a very much modular
project like PHP! While I have sort of managed to put together a 'checkout' of a
set of packages which together make a single build of the target project, I'm
still having to pull and commit changes manually to each subrepo - I don't get
the simple overview of outstanding changes that was SO nice to work with in CVS!
To try and put this in a 'PHP' context ... I DID have a list of extensions I was
working with, and a 'sync' in ECLIPSE gave me a nice list of which ones had had
changes pushed ... I could then go and either simply update to those changes on
an extension by extension basis - or review the changes in each file so allow me
to integrate other peoples changes with my own local 'tweaks'. This is something
that DVCS SHOULD be perfect for, but the management tools for bundling a
selection of parts of a project are not there yet, mainly because the core
developers on both HG and GIT "don't see then need because they don't need that
type of working"!
I'm limping along with HG at the moment simply because it works on both Linux
AND Windows, while GIT's windows support is still rather hit and miss when one
is already using other unix tools on windows, but I do not have the same level
of integration in Eclipse that CVS and SVN provide beautify straight out of the
box. At least I can drop out of Eclipse into TortoiseHg and do the code
management, but it's not an ideal way of working.
If I was a 'console jockey' then much of this would probably be a lot less of a
problem. One seems to need to write scripts to do what YOU want to do in both HG
and GIT - and then need to learn Python to do that in HG :( But I never used CVS
from the command line so a LOT of the reasoning for even the switch to SVN was
irrelevant since Eclipse had already solved all of those 'problems'.
At the current time I think that PHP would need to restructure how it is
packaged up to provide a single repo in both HG or GIT. Keeping SVN ( I'd still
prefer CVS here it works BETTER as a master for DVCS! ) as the master from which
we CAN currently sync using HG or GIT is the best of a bad job currently. Unless
some knows better?
--
Lester Caine - G8HFL
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php
At the current time I think that PHP would need to restructure how it is
packaged up to provide a single repo in both HG or GIT. Keeping SVN ( I'd
still prefer CVS here it works BETTER as a master for DVCS! ) as the master
from which we CAN currently sync using HG or GIT is the best of a bad job
currently. Unless some knows better
I know what you mean. There are solutions like submodules in GIT but it
does require some small restructuring usually. By the way, TortoiseGIT is
now quite excellent for Windows GUI. A new version 1.7.0 is just about to
be released too. I'm a bit biased against GIT but overall, GIT is getting
way more traction than Hg and github.com excels in features compared to
anything else.
From what I have seen also, IDE integration with GIT is starting to catchup
both in NetBeans and Eclipse - and already way ahead of the pack is phpStorm
for VCS integration, but it's not free.
Regards,
Drak
Drak wrote:
At the current time I think that PHP would need to restructure how it is packaged up to provide a single repo in both HG or GIT. Keeping SVN ( I'd still prefer CVS here it works BETTER as a master for DVCS! ) as the master from which we CAN currently sync using HG or GIT is the best of a bad job currently. Unless some knows betterI know what you mean. There are solutions like submodules in GIT but it
does require some small restructuring usually. By the way, TortoiseGIT
is now quite excellent for Windows GUI. A new version 1.7.0 is just
about to be released too. I'm a bit biased against GIT but overall, GIT
is getting way more traction than Hg and github.com http://github.com
excels in features compared to anything else.
When TortoiseGIT runs in Linux as well - they will have caught up!
TortoiseHg works transparently in both and on Mac I believe. My customer base is
mainly windows biased while I've been running on Linux here for some years now.
I need both platforms to do the same thing - which Eclipse has provided for
several years, but the DVCS camps still need to address. Windows GIT is
something different from Linux GIT at the moment - which is probably all that is
putting me off it. Actually when I was forced to used GIT the windows side
simply did not work at all, so I HAD to use hggit to get anything!
From what I have seen also, IDE integration with GIT is starting to
catchup both in NetBeans and Eclipse - and already way ahead of the pack
is phpStorm for VCS integration, but it's not free.
hggit + Mercurial Eclipse + Eclipse means I could not care less what the target
is EXCEPT that one can't write a project that is JUST the set of sub-modules
that you want to pull together. IDEALLY - both GIT and HG would just pull stuff
from which ever is providing a particular library? So say Smarty could be on one
while ADOdb is on the other ... The core PHP code does not need to be so
distributed, but being able to pull PECL modules from a number of sources to add
to a core clone of the code could be useful?
--
Lester Caine - G8HFL
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php
Drak wrote:
At the current time I think that PHP would need to restructure how
it is packaged up to provide a single repo in both HG or GIT.
Keeping SVN ( I'd still prefer CVS here it works BETTER as a master
for DVCS! ) as the master from which we CAN currently sync using HG
or GIT is the best of a bad job currently. Unless some knows betterI know what you mean. There are solutions like submodules in GIT but it
does require some small restructuring usually. By the way, TortoiseGIT
is now quite excellent for Windows GUI. A new version 1.7.0 is just
about to be released too. I'm a bit biased against GIT but overall, GIT
is getting way more traction than Hg and github.com http://github.comexcels in features compared to anything else.
When TortoiseGIT runs in Linux as well - they will have caught up!
TortoiseHg works transparently in both and on Mac I believe. My customer
base is mainly windows biased while I've been running on Linux here for some
years now. I need both platforms to do the same thing - which Eclipse has
provided for several years, but the DVCS camps still need to address.
Windows GIT is something different from Linux GIT at the moment - which is
probably all that is putting me off it. Actually when I was forced to used
GIT the windows side simply did not work at all, so I HAD to use hggit to
get anything!From what I have seen also, IDE integration with GIT is starting to
catchup both in NetBeans and Eclipse - and already way ahead of the pack
is phpStorm for VCS integration, but it's not free.hggit + Mercurial Eclipse + Eclipse means I could not care less what the
target is EXCEPT that one can't write a project that is JUST the set of
sub-modules that you want to pull together. IDEALLY - both GIT and HG would
just pull stuff from which ever is providing a particular library? So say
Smarty could be on one while ADOdb is on the other ... The core PHP code
does not need to be so distributed, but being able to pull PECL modules from
a number of sources to add to a core clone of the code could be useful?
Perhaps, it would depend how teams are structured, but even it was a single
repository for everything I think stuff would work fine.
My opinion is that the reason behind sub-repositories is to include code
from vendors, not that much to modularize development of a single big
project, why?, because of the distributed nature of a DVCS: imagine that
there is a PECL team, and that team makes a fork of the main PHP repository,
and they focus on that "fork" (every separate repository can be regarded as
a fork) and they handle all their work integration, and then someone (after
an ammount of work is done) pulls back from the main repo (the one where
last integrations are agreed to be added, "central" if you must call it
that) to ensure everything still runs ok, and then ask to the final
integrator to pull and merge their work: done.
Oh, and HG supports svn and git as subrepos now.
Regards,
David
dukeofgaming wrote:
hggit + Mercurial Eclipse + Eclipse means I could not care less what the target is EXCEPT that one can't write a project that is JUST the set of sub-modules that you want to pull together. IDEALLY - both GIT and HG would just pull stuff from which ever is providing a particular library? So say Smarty could be on one while ADOdb is on the other ... The core PHP code does not need to be so distributed, but being able to pull PECL modules from a number of sources to add to a core clone of the code could be useful?Perhaps, it would depend how teams are structured, but even it was a
single repository for everything I think stuff would work fine.My opinion is that the reason behind sub-repositories is to include code
from vendors, not that much to modularize development of a single big
project, why?, because of the distributed nature of a DVCS: imagine that
there is a PECL team, and that team makes a fork of the main PHP
repository, and they focus on that "fork" (every separate repository can
be regarded as a fork) and they handle all their work integration, and
then someone (after an ammount of work is done) pulls back from the main
repo (the one where last integrations are agreed to be added, "central"
if you must call it that) to ensure everything still runs ok, and then
ask to the final integrator to pull and merge their work: done.
I think this is the MAIN problem with DVCS currently ...
Most of the 'big guns' that are pushing git are ONLY working on a single
repository with a single set of code ( although GNOME and KDE should probably be
'subrepos' of the linux core project? ). The core PHP code can probably be
handled as a single repo, but then you have to clone everything, including all
extensions, while just being able to pick and choose would be nice. Moving to
userland ... being able to follow the latest Smarty, ADOdb and other libraries
in parallel with one's own code base is somewhat useful ... and something I've
actually been doing in Eclipse already for several years. DVCS has currently
distorted the picture because it simply does not integrate well yet :( Adopting
one path for the core code but needing a different one for working with it just
seems wrong to me, so sitting on the fence for the moment seems right. I'm
actively using HG - but the masters I am working into are on github so sitting
on the fence is not a problem :)
Oh, and HG supports svn and git as subrepos now.
It is coming along ... I'm currently following a discussion on getting this
working properly on the mercurial list ... just which directory path should you
be using to link things together ...
--
Lester Caine - G8HFL
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php
dukeofgaming wrote:
hggit + Mercurial Eclipse + Eclipse means I could not care less what
the target is EXCEPT that one can't write a project that is JUST the
set of sub-modules that you want to pull together. IDEALLY - both
GIT and HG would just pull stuff from which ever is providing a
particular library? So say Smarty could be on one while ADOdb is on
the other ... The core PHP code does not need to be so distributed,
but being able to pull PECL modules from a number of sources to add
to a core clone of the code could be useful?Perhaps, it would depend how teams are structured, but even it was a
single repository for everything I think stuff would work fine.My opinion is that the reason behind sub-repositories is to include code
from vendors, not that much to modularize development of a single big
project, why?, because of the distributed nature of a DVCS: imagine that
there is a PECL team, and that team makes a fork of the main PHP
repository, and they focus on that "fork" (every separate repository can
be regarded as a fork) and they handle all their work integration, and
then someone (after an ammount of work is done) pulls back from the main
repo (the one where last integrations are agreed to be added, "central"
if you must call it that) to ensure everything still runs ok, and then
ask to the final integrator to pull and merge their work: done.I think this is the MAIN problem with DVCS currently ...
Most of the 'big guns' that are pushing git are ONLY working on a single
repository with a single set of code ( although GNOME and KDE should
probably be 'subrepos' of the linux core project? ). The core PHP code can
probably be handled as a single repo, but then you have to clone everything,
including all extensions, while just being able to pick and choose would be
nice.
True, that would be a very big clone.
Oh, and HG supports svn and git as subrepos now.
It is coming along ... I'm currently following a discussion on getting this
working properly on the mercurial list ... just which directory path should
you be using to link things together ...
I know it will be a while until any choice of DVCS is adopted by PHP, but
I'll be creating an RFC as soon as I can edit the wiki (I can edit the rfc
namespace but cannot edit the rfc page itself to add the entry) so that we
can start making note of the current challenges and choices.
Regards,
David
dukeofgaming wrote:
I know it will be a while until any choice of DVCS is adopted by PHP,
but I'll be creating an RFC as soon as I can edit the wiki (I can edit
the rfc namespace but cannot edit the rfc page itself to add the entry)
so that we can start making note of the current challenges and choices.
http://lsces.co.uk/wiki/index.php?page=DVCS+adoption+notes is getting rather
long in the tooth and I need to rework my own implementation notes as a number
of things have changed.
http://lsces.co.uk/hg/ will have all of my PHP base added soon, but recursion
into projects is still a 'todo' here, so http://lsces.co.uk/hg/bitweaver2/ is
not linking properly and at some point hopefully it WILL drill down to smarty,
adodb, ckeditor, and the javascript libraries ...
The principles are fine, the mechanics still need to be completed ... and all of
THAT is detracting from writing the PHP code which pays the bills :(
--
Lester Caine - G8HFL
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php
Hi Lester,
AFAIK both mercurial and git support sub-repositories nicely now:
http://mercurial.selenic.com/wiki/Subrepository
http://www.kernel.org/pub/software/scm/git/docs/git-submodule.html
Regarding the graphical tools, I don't use them that much, but I see your
point. Personally, I just occasionally use the Netbeans integration because
it is easier to commit files in separate directories, but that is pretty
much it since I think the rest is easier and quicker to do with the console,
but I guess it will always be a matter of personal preference.
I've been working with hg subrepos and they are quite nice. I'm still not
that familiar with git submodules.
BTW, you don't need to use python to customize/extend mercurial's behavior,
you can use hooks: http://mercurial.selenic.com/wiki/Hook
Best regards,
David
dukeofgaming wrote:
I thought it might be interesting for all to read some feedback I got
from some questions I made to the Joomla community, specifically, the
guys and gals that have been implementing the decoupling of the Joomla
framework: the Joomla Platform.They currently still use svn officially but they have been experimenting
with both mercurial and git. These are their responses so far:http://groups.google.com/group/joomla-dev-framework/browse_thread/thread/fd3b0151cc215c0e
It may bring some insight to what a DVCS could do for the PHP community.
I think that there are perhaps still a couple of separate problems here
which the DVCS community need to get behind for any of them to become a
'front runner'.The main problem here IS 'subrepo' which is a concept that neither HG or
GIT handle particularly well, but to my mind is essential for a very much
modular project like PHP! While I have sort of managed to put together a
'checkout' of a set of packages which together make a single build of the
target project, I'm still having to pull and commit changes manually to each
subrepo - I don't get the simple overview of outstanding changes that was SO
nice to work with in CVS! To try and put this in a 'PHP' context ... I DID
have a list of extensions I was working with, and a 'sync' in ECLIPSE gave
me a nice list of which ones had had changes pushed ... I could then go and
either simply update to those changes on an extension by extension basis -
or review the changes in each file so allow me to integrate other peoples
changes with my own local 'tweaks'. This is something that DVCS SHOULD be
perfect for, but the management tools for bundling a selection of parts of a
project are not there yet, mainly because the core developers on both HG and
GIT "don't see then need because they don't need that type of working"!
I'm limping along with HG at the moment simply because it works on both
Linux AND Windows, while GIT's windows support is still rather hit and miss
when one is already using other unix tools on windows, but I do not have the
same level of integration in Eclipse that CVS and SVN provide beautify
straight out of the box. At least I can drop out of Eclipse into TortoiseHg
and do the code management, but it's not an ideal way of working.If I was a 'console jockey' then much of this would probably be a lot less
of a problem. One seems to need to write scripts to do what YOU want to do
in both HG and GIT - and then need to learn Python to do that in HG :( But I
never used CVS from the command line so a LOT of the reasoning for even the
switch to SVN was irrelevant since Eclipse had already solved all of those
'problems'.At the current time I think that PHP would need to restructure how it is
packaged up to provide a single repo in both HG or GIT. Keeping SVN ( I'd
still prefer CVS here it works BETTER as a master for DVCS! ) as the master
from which we CAN currently sync using HG or GIT is the best of a bad job
currently. Unless some knows better?--
Lester Caine - G8HFLContact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php
On Thu, Apr 28, 2011 at 4:10 AM, Ben Schmidt
mail_ben_schmidt@yahoo.com.auwrote:
Hi!
I realise that at least for now, PHP is sticking with SVN. No problems.
However...I find it much easier to work in Mercurial to put together
patches, find bugs, etc.. And in fact, I find SVN really awkward--I was
on the verge of switching back to CVS when DVCSes such as Mercurial and
Git became popular. I would put in a vote for Mercurial if PHP decides
to move to a DVCS. Mercurial is more similar to CVS/SVN in its command
set than other DVCSes, is easy and friendly to use, generally does what
you would expect, makes it pretty hard to shoot yourself in the foot,
and includes good Windows support. Operating with the Mozilla and Vim
Mercurial repositories is a lovely experience, and I'd love to add PHP
to the list of projects I follow this way.However, pushing the team towards a DVCS is not what I'm writing for.
Certainly not to push for a particular one.I'd just like to be able to use my tool of choice (Mercurial) locally.
Would it be possible to set up a Mercurial mirror of the SVN repo on,
say, bitbucket.org, so I could pull from there? Alternatively, I can
access a git repo somewhere like github.com (I found
https://github.com/php/php-src but the last activity seems to be about 9
months ago--is it dead? are there any plans to revive it?). Of course,
as a last resort, I could just access SVN using Mercurial. However, it
may result in a lower server load if PHP can provide a mirror somewhere,
updated by a commit hook or daily cron job or something, rather than me
(and potentially many others, as DVCSes become even more popular)
accessing SVN directly.Any possibility? (I'm happy to help, if I can be given appropriate
access to resources to do so.)Ben.
anybody out there, who didn't tried github yet, should really do it.
and you can use your svn client to interact with your github repo, if you
didn't feel the need to learn git.
https://github.com/blog/644-subversion-write-support
Tyrael