Re: [tryton-dev] Tryton Mac Client
On Nov 8, 2011, at 1:57 PM, Cédric Krier wrote:
> On 08/11/11 14:18 +0100, lists.jc.mic...@symetrie.com wrote:
>> Hi,
>>
>> Le 8 nov. 2011 à 03:13, Phillip Heller a écrit :
>>>> Strange issue here, on 10.6.8: pressing "t" while entering text in a form
>>>> does not produce the letter, but highlights the "Benutzer" ("User"?) menu
>>>> on top of the screen for some seconds without unfolding it. "T" does work
>>>> as expected.
>>>>
>>>> I can get a "t" in there using copy & paste, but it kinda ruins the
>>>> workflow... ,)
>>>
>>> Steffen,
>>>
>>> Yes, this is a bug which I also came across in my testing a day or so ago.
>>>
>>> I've opened bug https://bugs.tryton.org/issue2258 to track.
>>>
>>> It's a simple fix, and I'll get it in along side the changes to make the
>>> build work again.
>>
>>
>> I had the same pb when I built the 2.0 client and Phillip's patch worked
>> well,
>
> I discover why the code did not work.
> It is because now, os.name return posix under MacOSX.
> So we need a new way to detect broken MOD1 or we must change MOD1 for
> something else perhaps Shift+Control.
sys.platform returns Darwin on Mac OS X.
However, I think we should use the same letter as for Form reload and an
additional modifier. So, perhaps CTRL-R for Form Reload, and CTRL-SHIFT-R for
Menu Reload.
Thoughts?
--phil
--
tryton-dev@googlegroups.com mailing list
Re: [tryton-dev] Tryton Mac Client
On Nov 7, 2011, at 9:08 PM, Steffen Beyer wrote:
> Strange issue here, on 10.6.8: pressing "t" while entering text in a form
> does not produce the letter, but highlights the "Benutzer" ("User"?) menu
> on top of the screen for some seconds without unfolding it. "T" does work
> as expected.
>
> I can get a "t" in there using copy & paste, but it kinda ruins the
> workflow... ,)
Steffen,
Yes, this is a bug which I also came across in my testing a day or so ago.
I've opened bug https://bugs.tryton.org/issue2258 to track.
It's a simple fix, and I'll get it in along side the changes to make the build
work again.
Regards,
Phil
--
tryton-dev@googlegroups.com mailing list
[tryton-dev] Tryton Mac Client
Greetings, I've submitted a code review for the patch to fix the gdk pixbuf loader location with newer versions of gtk. I've built Mac application bundles of the 2.0.0 and 2.2.0 client on 10.7.2 against the 10.5 SDK. Unfortunately, the weeks I spent working with the latest build of gtk+-2 (incorporating key binding fixes) was unsuccessful. I will continue to work on this, but wanted to get some newer client builds out there for those who need them. However, I do not have access to Macs running other versions of Mac OS X to test against (it should work on 10.5 and up). If you wouldn't mind giving these binaries a quick test, I would appreciate it. I know for sure of two problems: 1) When you launch the application, it will not present its window until you click the application icon in the Dock. (This is new, probably related to something in gtk) 2) When you quit the application via "Quit" in the context menu from the Dock icon, the application will throw an exception. (This has been around for quite some time) You can find the 2.0.0 and 2.2.0 disk images here: https://public.me.com/pheller Once the code review is completed, some positive test results received, and the accepted code integrated into trunk, I'll submit the binaries to Cedric for inclusion on the website under downloads. Regards, Phil -- tryton-dev@googlegroups.com mailing list
Re: [tryton-dev] Tryton Client 2.0 Mac Os version
On Sep 15, 2011, at 3:29 AM, Cédric Krier wrote: > Oscar and Phillip please read on top-posting: > https://groups.google.com/d/msg/tryton/-IA9TLq5qag/pv5okIVkr38J My apologies. My reply was composed on my phone. In this case, it would not have been difficult to bottom-post. However it usually is, due to the fact that I must first scroll to the bottom of the post to which I am replying - and they are often quite long. Perhaps GUI and mobile email clients will support a bottom post option at some point. If I didn't have to deal with so many calendar invites, I'd move back to mutt without hesitation Regards, Phil smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Tryton Client 2.0 Mac Os version
GTK made a change to the location of the pixbuf loaders in the latest stable, and changes to improve keyboard accelerators should be integrated to stable very soon. Also, some changes are necessary to the build environment for 10.7. Once gtk is ready, I will update the build environment instructions on the wiki and build both the Neso and Tryton clients. My apologies for not getting to this sooner, school and work have taken away the great majority of my free time. Hopefully once this work is done, someone can cron up a daily build of the clients. Regards, --phil On Sep 14, 2011, at 6:08 PM, oscar_andres_col wrote: > Hi Boys, > > To exist any Tryton Client 2.0 version Mac Os in development? > > Oscar Alvarez > > -- > tryton-dev@googlegroups.com mailing list -- tryton-dev@googlegroups.com mailing list
Re: [tryton-dev] About module country_ar
On Sep 9, 2011, at 3:10 PM, Cédric Krier wrote:
> Hi,
>
> I checked the module country_ar [1] and if I understand correctly you try to
> get
> some kind of autocompletion for the Argentina cities.
>
> First, I don't think it is good to extend the Subdivision to support cities,
> especially because there is already a field in address to store the city and
> it is not part of the ISO3166 nor ISO3166-2.
>
> So I (and Nicolas) thought about how we could provide a generic way to get
> autocompletion on city (or other field in address).
> Here are the result:
>
>- We need a place to store the combination of fields: zip, city, country,
> subdivision.
>- We must have autocompletion [1] on the fields zip and city.
>- The system should learn from already encoding tuple.
>
Worth noting is that some locales in the United States (and I'm sure elsewhere)
may have multiple city names for a zipcode.
As an example:
Denver, CO is 80233
Thornton, CO is 80233
The United States does have an extension to zip code ('plus 4') to disambiguate:
Denver, CO is 80233-1248
Thornton, CO is 80233-9822
However, it is rare that one might know the "plus 4" for their location.
Anyways, if you were planning to enforce uniqueness to the tuple, this might be
a consideration.
Otherwise, I think this is a great idea.
--phil
smime.p7s
Description: S/MIME cryptographic signature
[tryton-dev] Fix setup.py per issue1976 to identify proper source path of gdk-pixbuf-loaders
Please review http://codereview.tryton.org/23001/ The patch does need some cleanup regarding line length, which I'll get to tomorrow. --phil smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Python shebang
I believe that distutils will rewrite the shebang to absolute path of the interpreter used to invoke setup.py, unless it is overridden in some manner (i.e., "python ./setup.py build --executable=path/to/python ) I personally feel this is the proper way to set the interpreter. And, for the record, a python2 symlink is not installed by either MacPorts python builds or Apple's distribution. Raising an exception when invoked by Python 3 is certainly reasonable, though. --phil On Jan 13, 2011, at 6:00 PM, Bertrand Chenal wrote: > > Hi, > > > Currently, the shebang on top of bin/trytond and bin/tryton is > > #!/usr/bin/env python > > I propose to use > > #!/usr/bin/env python2 > > instead, because trytond does not work with python3 and so raise an > exception if python3 is the default (like in archlinux). I checked on > ubuntu 10.04 and python2 exist, but I don't know if it's the same on > other distribution. > > Cheers, > > -- > > Bertrand Chenal > > B2CK SPRL > Rue de Rotterdam, 4 > 4000 Liège > Belgium > Tel: +32 472 54 46 59 > Email: bertrand.che...@b2ck.com > Website: http://www.b2ck.com/ > > -- > tryton-dev@googlegroups.com mailing list smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Handle URLs on Mac OS X
On Dec 30, 2010, at 8:58 AM, Cédric Krier wrote:
> It will require some more code to add support for MacOSX bundle and also some
> registry record for the windows setup.
Greetings,
I've done some work on the gtk-osx project to support custom URLs dispatched
from the Mac.
You can find my fork of ige-mac-integration here
https://github.com/pheller/ige-mac-integration (and, a pygtk example in
bindings/python/gtk_osxapplication/test-integration.py, from which the
following example is pulled).
Basically, within a pygtk app, one would register a handler for the OpenURL
events like so (assuming macapp is an OSXApplication object):
macapp.connect("NSApplicationOpenURL", urlhandler)
...
def urlhandler(self, widget, url):
print "Invoked with URL: %s\n" % url
And, one must add some information to the Info.plist to indicate support of the
URL to the Mac OS X LaunchServers. This is done via py2app and setup.py by
adding the following dict to the plist key within the py2app dict:
'CFBundleURLTypes': [{
'CFBundleURLName': 'GTK-OSX Python Example',
'CFBundleURLSchemes': ['gtk-osx'],
}],
There is one issue, however; If the application is not running when a matching
URL is invoked, the application launches, but the signal is not emitted. If
anyone has Cocoa experience, I'd certainly appreciate another set of eyes.
As far as integration within Tryton goes, I have a proof of concept, which
(mostly) works. The limitations being the aforementioned issue, as well as the
inability to launch a new client to handle connections to a different
server/database. I'd like to solve the issues in the upstream bits first, then
I'll clear up the issues within Tryton, and submit a patch.
Regards,
Phil
smime.p7s
Description: S/MIME cryptographic signature
Re: [tryton-dev] Add autoincrement to sqlite primary key for issue1748 (issue2959041)
Inline... On Nov 7, 2010, at 9:51 AM, Mathias Behrle wrote: > Err. Release is the official announcement, not the upload of some packages. > >>> Since 1.8 is not yet officialy released I agree to include it. >> >> That is ... eh, nonsense. If tagging, branching, > > Tagging and branching never ever constitute an *official* release. If you > don't > wait for the official announcement, it is at your risk to use those packages > and be left with uninstallable situations. What do you want to do with module > account 1.8, if trytond 1.8 is missing? > >> to pypi doesn't constitute releasing 'officially' you've spent too much >> time chatting to your marketing department. > > See above. The problem, though, is that pypi advertises availability. One can subscribe to pypi's feed on twitter, for example, to be notified of new packages. And, while the tryton project might make official announcements regarding new releases, not all pypi packagers do that -- so if pypi is an official channel of release, we cannot expect users to wait for a news announcement (on a list to which they might not subscribe) to consider this release "available". That said, I do feel this bug fix is worthy of inclusion, considering that Neso's usefulness is severely impacted without it. Whether it's 1.8.0 or 1.8.1 is really inconsequential. What matters is that the bug was identified and fixed. --phil smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] adding product/category taxes to invoice lines - or how to call the on_change infrastructure
On Oct 17, 2010, at 11:18 AM, Paul J Stevens wrote:
> Hi all,
>
> I'm working on building a simple module ('contract') for periodically
> creating invoices (like the subscription module discussed earlier).
Paul,
This is a very cool development. Do you plan to publish once you're finished?
A couple things I think might be useful:
A "Contracts" tab within the Party view. This tab would list all of the
party's contracts, with a column "Remaining Contract Value", which would be
computed based on the billing amount and remaining contract length.
Also, it would be useful to represent an Early Termination Fee, which is
perhaps sometimes as simple as a flat rate, or a percentage of the remaining
contract value, or the greater of the two, or some combination.
It might be nice to represent Contracts where the enterprise is the customer
(i.e., like "supplier invoices", but for these contracts), though this is
perhaps beyond the scope of what you're currently working on.
All of these things would be useful to eventually figure into a "Guidance
Report", which would help with projected revenue.
--phil
smime.p7s
Description: S/MIME cryptographic signature
Re: [tryton-dev] Application naming within various operating environments
On Oct 13, 2010, at 12:09 PM, Mathias Behrle wrote: > Supporting this. > If there is the possibility to display some sort of application description it > should show "Tryton Client" (like 'Generic Name' in > http://standards.freedesktop.org/desktop-entry-spec/latest/ar01s05.html) I will do this in the Mac application bundle, if possible. >> However, from the unix command line, the server would remain "trytond". > > For sure, and the client remains "tryton". And not only for *unix* command > line. Yes, tryton and trytond from all command lines. I will adjust the GUI presentation for the Mac client from "tryton" to "Tryton" in the patch set to be applied against trunk for 1.8. Please reply if you have further comments or concerns regarding this initiative. --phil smime.p7s Description: S/MIME cryptographic signature
[tryton-dev] Missed a context during the Transaction migration (Purchase)
Oops, Missed a context during the Transaction migration.
Please review this at http://codereview.appspot.com/2497042/
Affected files:
M purchase.py
Index: purchase.py
===
--- a/purchase.py
+++ b/purchase.py
@@ -1363,7 +1363,7 @@
if not uom:
uom = default_uom
if Transaction().context.get('supplier') and
product.product_suppliers:
-supplier_id = context['supplier']
+supplier_id = Transaction().context['supplier']
for product_supplier in product.product_suppliers:
if product_supplier.party.id == supplier_id:
for price in product_supplier.prices:
smime.p7s
Description: S/MIME cryptographic signature
[tryton-dev] Application naming within various operating environments
Greetings, I wanted to bring a side discussion onto the list for comment. I generally think that we should adopt a friendly naming scheme for the various application bundles. I.e., within Windows, Mac OS, or any Linux windowing system definition, the client name would be capitalized. That is, the icon would read "Tryton". In the case the application must present the name within the UI, it would also be capitalized. And, in the case that the UI needs to present the name of the server (assuming there is at some point server bundles), I propose within the UI the server be shown as "Tryton Server". I certainly don't advocate abandoning "trytond". Rather, within things like the Windows Control panel (services), or MacOS Preferences Pane, etc., I would adopt "Tryton Server" -- merely to achieve a level of uniformity with other such applications. However, from the unix command line, the server would remain "trytond". Thoughts? Regards, Phil On Oct 13, 2010, at 9:17 AM, Mathias Behrle wrote: > * Betr.: " Re: Fix Mac OS X application termination and fully integrate > menubar > (issue1975047)" (Wed, 13 Oct 2010 09:51:01 +0200): > >> On 12/10/10 20:13 -0400, Phillip Heller wrote: >>> I was reviewing this some more, and I'm curious why this is? >>> >>> The all the images on the website use a capital "T" Tryton, as does the >>> image in the client login screen, and the client's "About" screen >>> >>> This specific change in the client application bundle simply changes the >>> Application name (as exposed via the Mac User Interface) to "Tryton". It >>> seemed to match the rest of the branding to me... No? >> >> What we decided was that: >> >> Tryton is the name of the project >> tryton is the name of the client >> trytond is the name of the server >> >> But now we can discuss on mailing list about new definition. > > I think the application button/title (even if it is only the client) should be > Tryton, describing the application suite as a whole. > > BTW: > In tryton.desktop we have: > > Type=Application > Name=Tryton > GenericName=Client for the Tryton Application Platform > Comment=Access Tryton server smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Use pysandbox for safe_eval (issue2352042)
On Oct 6, 2010, at 2:16 PM, Cédric Krier wrote:
> On 02/10/10 00:09 +0200, Cédric Krier wrote:
>> Please review this at http://codereview.appspot.com/2352042/
>
>
> I find it strange that nobody commented on this patch.
> It adds a new dependency that doesn't seem yet to be packages in most
> distributions.
>
> Is there anybody who tried to install it on windows? (There is a C module)
> Should we include in the coming release?
Cédric,
I've given it a quick go on Mac OS X 10.6.4. It is no problem to install
pysandbox with easy_install, and trytond/tests/test_tools.py succeeded.
However, when I tried to create a demo database, there are some issues right
from the beginning:
[Wed Oct 06 18:23:45 2010] INFO:server:starting NetRPC protocol, port 8070
[Wed Oct 06 18:23:45 2010] INFO:server:waiting for connections...
[Wed Oct 06 18:23:47 2010] INFO:database:connect to "template1"
[Wed Oct 06 18:23:53 2010] INFO:database:connect to "a"
[Wed Oct 06 18:23:54 2010] INFO:database:connect to "a"
[Wed Oct 06 18:23:54 2010] ERROR:database:CREATE DB: a failed
[Wed Oct 06 18:23:54 2010] ERROR:database:Exception in call:
Traceback (most recent call last):
File
"/Volumes/Data/Users/pheller/Projects/tryton/server/sandbox-test/trytond/protocols/dispatcher.py",
line 184, in create
database.init(transaction.cursor)
File
"/Volumes/Data/Users/pheller/Projects/tryton/server/sandbox-test/trytond/backend/postgresql/database.py",
line 224, in init
info = safe_eval(open(tryton_file).read())
File
"/Volumes/Data/Users/pheller/Projects/tryton/server/sandbox-test/trytond/tools/misc.py",
line 401, in safe_eval
return sandbox.call(lambda: eval(comp, {'__builtins__': {
File
"/Volumes/Data/Users/pheller/Projects/tryton/server/sandbox-test/lib/python2.6/site-packages/pysandbox-1.0.3-py2.6-macosx-10.6-x86_64.egg/sandbox/sandbox_class.py",
line 44, in call
return self._call(func, args, kw)
File
"/Volumes/Data/Users/pheller/Projects/tryton/server/sandbox-test/lib/python2.6/site-packages/pysandbox-1.0.3-py2.6-macosx-10.6-x86_64.egg/sandbox/sandbox_class.py",
line 31, in _call
return limitedTime(timeout, func, *args, **kw)
File
"/Volumes/Data/Users/pheller/Projects/tryton/server/sandbox-test/lib/python2.6/site-packages/pysandbox-1.0.3-py2.6-macosx-10.6-x86_64.egg/sandbox/timeout.py",
line 45, in limitedTime
old_alarm = signal(SIGALRM, signalHandler)
ValueError: signal only works in main thread
Regards,
Phil
smime.p7s
Description: S/MIME cryptographic signature
Re: [tryton-dev] Codereview: Factor out thread setup/teardown, join threads by name, MacOS fixes [was join() threads by name ...]
Please note, the scope of this codereview has been expanded, and now solves the pydevd issue, two MacOS issues (python bug re: errno 57, and file descriptor exception on closing a closed socket), gives meaningful names to threads for easier debugging with pdb, and factors out some common SocketServer code into a new trytond.protocols.common package. http://codereview.appspot.com/2348042/ Regards, Phil On Oct 2, 2010, at 6:31 PM, Phillip Heller wrote: > I usually develop and debug with Eclipse, which uses pydevd to connect to the > debugged process. > > Unfortunately, the call to threading.enumerate() includes the pydevd thread, > and upon calling it's join() method, sends the trytond process into a zombie > state. > > Please review the linked code review, which documents an adjustment to call > the various service thread's join() method by name directly, which solves > this problem. > > http://codereview.appspot.com/2348042 > > Regards, > > Phil smime.p7s Description: S/MIME cryptographic signature
[tryton-dev] Codereview: join() threads by name rather than threading.enumerate()
I usually develop and debug with Eclipse, which uses pydevd to connect to the debugged process. Unfortunately, the call to threading.enumerate() includes the pydevd thread, and upon calling it's join() method, sends the trytond process into a zombie state. Please review the linked code review, which documents an adjustment to call the various service thread's join() method by name directly, which solves this problem. http://codereview.appspot.com/2348042 Regards, Phil smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] MacOS GTK Client Development Environment How-To
Done. Please see: http://code.google.com/p/tryton/wiki/BuildingMacOSXInstall http://code.google.com/p/tryton/wiki/InstallationMacOSX Naturally, comments appreciated. --phil On Oct 2, 2010, at 4:36 PM, Phillip Heller wrote: > I'd be happy to do so, though I don't seem to have access. (Or, perhaps I > don't understand how to edit a google wiki!) > > --phil > > On Oct 2, 2010, at 4:27 PM, Cédric Krier wrote: > >> You should put this on the wiki. smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] MacOS GTK Client Development Environment How-To
I'd be happy to do so, though I don't seem to have access. (Or, perhaps I don't understand how to edit a google wiki!) --phil On Oct 2, 2010, at 4:27 PM, Cédric Krier wrote: > You should put this on the wiki. smime.p7s Description: S/MIME cryptographic signature
[tryton-dev] MacOS GTK Client Development Environment How-To
Greetings,
After a brief discussion with Sharoon a couple weeks ago, I wanted to publish
some instructions for creating a development environment for MacOS, so that
those who need the bleeding edge client need not rely on a daily build (and the
30 some megabyte download that would accompany it).
Some general notes:
This build is for the 10.5 sdk, against the architecture of the build machine.
(I.e., build on i386, get i386 binaries)
Building dual architectures is not painless, though it's getting there.
The 1st step, is to make git available. Though, if you already have git (from
source, installed via macports, etc), feel free to add its location to the path
set in the next step, though be mindful of the step 2 reasoning...
The 2nd step is to create a user just for the build. This is mostly to ensure
that the user does not pick up macports or fink in their path, which has a high
likelyhood of breaking the build. If you added a macports path for git, you'd
do well to remove it before continuing.
The following steps should basically cut and paste, though proceed at your own
risk. The steps were completed on a 10.6.4 with the most recent Xcode update.
###
#
# 0) Validate you have gcc-4.0 and the 10.5 sdk.
#
/usr/bin/gcc-4.0 -v
ls -ld /Developer/SDKs/MacOSX10.5.sdk
#
# 1) Install git available
#
curl -O
http://git-osx-installer.googlecode.com/files/git-1.7.2.3-intel-leopard.dmg
open git-1.7.2.3-intel-leopard.dmg
sudo installer -pkg /Volumes/Git\ 1.7.2.3\ Intel\
Leopard/git-1.7.2.3-intel-leopard.pkg -target /
hdiutil eject "/Volumes/Git 1.7.2.3 Intel Leopard"
#
# 2) create a user just for this build[1], then become this user, and set some
environment variables -- I use "gtkosxbuilder"
#
sudo su - gtkosxbuilder
cat << EOF > ~/.profile
export
PATH=~/.local/bin:~/gtk/inst/bin:/bin:/usr/bin:/sbin:/usr/sbin:/usr/local/git/bin
export DYLD_FALLBACK_LIBRARY_PATH=~/gtk/inst/lib
EOF
. ~/.profile
#
# 3) Download and run the gtk-osx build script
#
curl -L -O
http://downloads.sourceforge.net/sourceforge/gtk-osx/gtk-osx-build-setup.sh
sh gtk-osx-build-setup.sh
#
# 4) modify .jhbuildrc-custom:
#
sed -i -e 's/^setup_sdk/#setup_sdk/g' .jhbuildrc-custom
cat << EOF >> .jhbuildrc-custom
_gtk_osx_use_jhbuild_python = True
branches["python"] =
"http://www.python.org/ftp/python/2.6.4/Python-2.6.4.tar.bz2";
setup_sdk(target="10.5", sdk_version="10.5", architectures=["i386"])
os.environ["CC"] = "/usr/bin/gcc-4.0"
os.environ["DYLD_LIBRARY_PATH"] = ""
build_policy = "updated-deps"
modules = [ "meta-gtk-osx-bootstrap", "meta-gtk-osx-core",
"meta-gtk-osx-python", "librsvg", "meta-gtk-osx-themes", "gtk-quartz-engine" ]
EOF
#
# 5) Modify jhbuild module set to ignore python file size / checksum, and the
lib64 patch which is applicable to Linux
#
sed -i -e 's/size="9807597" md5sum="afb5451049eda91fbde10bd5a4b7fadc"//'
~/Source/jhbuild/modulesets/bootstrap.modules
sed -i -e 's///'
~/Source/jhbuild/modulesets/bootstrap.modules
#
# 6) Build!
#
# Prepare yourself a beverage and wait; my MacPro1,1 builds the whole thing in
~ 30 minutes or so.
# It could possibly be sped up with some args to make (-j8, etc), but some
components break with parallelized build
#
jhbuild bootstrap --ignore-system && jhbuild build
#
# 7) Install mercurial, and satisfy the tryton client dependencies, fixup py2app
#
mkdir -p tmp
cd ~/tmp
curl -O
http://pypi.python.org/packages/source/s/setuptools/setuptools-0.6c11.tar.gz
tar zxvf setuptools-0.6c11.tar.gz
cd setuptools-0.6c11
python ./setup.py install
easy_install mercurial python-dateutil py2app
cd
~gtkosxbuilder/gtk/inst/lib/python2.6/site-packages/py2app-0.5.2-py2.6.egg/py2app/apptemplate/prebuilt
ln -s main-fat main-i386
#
# 8) checkout tryton client source
#
cd ~
hg clone http://hg.tryton.org/tryton
#
# Finally, you should be able to launch the client (though, as yourself, as
only the owner of the "SystemUIServer" process can launch GUI applications):
#
# This also works well within Eclipse (with PyDev) - simply add a new
interpreter ("~gtkosxbuilder/gtk/inst/bin/python") and create run / debug
configurations
# Clearlooks doesn't seem to work unless bundled. Not a deal breaker for
development work.
#
# It certainly should be possible to run and bundle the client source from a
different path (as a different user even), though the latter raises some errors
in
# py2app that I haven't yet worked through.
#
~gtkosxbuilder/gtk/inst/bin/python ~gtkosxbuilder/tryton/bin/tryton
#
# if you want to bundle the application for distribution:
#
sudo su - gtkosxbuilder
cd tryton
python ./setup.py py2app
#
# Any questions, feel free to drop me an email!
#
# Regards,
# --phil
smime.p7s
Description: S/MIME cryptographic signature
Re: [tryton-dev] Better search GUI
While there is discussion on the subject, I thought I might lodge my vote for a "query-by-example" interface. Specifically, I think it would be useful to show a model's form representation, which a user could populate with values. Perhaps this search "form" could have right-click context menu options to alter the field from a default of "like" to "not like", etc. This would be reminiscent of the Informix 4GL UI, which was very convenient. Thoughts? Regards, Phil smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] About codereview
Greetings, I'm one of the guilty parties for having some stale codereviews and roundup issues. Here is a summary of them; I will update each roundup issue with current status as of today, and I will have time work them this weekend. -- server hangs indefinitely upon receipt of SIGINT, narrowed to thread.join on NETRPC https://bugs.tryton.org/roundup/issue1625 http://codereview.appspot.com/2049042/ -- Button activation on o2m etc. https://bugs.tryton.org/roundup/issue1406 http://codereview.appspot.com/1974045/ -- Change id of uom ounce https://bugs.tryton.org/roundup/issue1414 http://codereview.appspot.com/2018042/ -- Fully integrate Tryton client on Mac OS https://bugs.tryton.org/roundup/issue1681 http://codereview.appspot.com/1975047/ Apologies for the stall. --phil On Sep 8, 2010, at 5:53 PM, Cédric Krier wrote: > Hi, > > There is more and more patch submition on codereview which has some comments > and require some more work. But they seem to be dead. > I would like if you stop working on a patch that you notify otherwise others > are waiting for you instead of working on it. > > So thanks to update status on your patches. And close fixed one. > > It is also the same for issue on roundup, if you don't work any more on an > issue that is assigned to you then remove your assignation. > > Thanks, > -- > Cédric Krier > > B2CK SPRL > Rue de Rotterdam, 4 > 4000 Liège > Belgium > Tel: +32 472 54 46 59 > Email/Jabber: cedric.kr...@b2ck.com > Website: http://www.b2ck.com/ smime.p7s Description: S/MIME cryptographic signature
[tryton-dev] Fix for netrpc hang on Mac OS X, issue1625
Greetings, Please review the patch here: http://codereview.appspot.com/2049042 This patch adds a timeout to the socket.accept() within the netrpc running loop. Regards, Phil smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Fix for Mac OS Application Termination, and full integration of Mac OS X Menubar
Sorry guys, my mistake. I must have had an errant gtkrc floating around somewhere. --phil On Aug 26, 2010, at 5:59 AM, Sharoon Thomas wrote: > Hi, > > I an confirm that (like cedk) i also see the same template. > > Thanks, > > Sharoon > > On 26 Aug 2010, at 07:05, Cédric Krier wrote: >> >> What you show is currently (as far as I saw) the default theme we have for >> Tryton bundle. smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Fix for Mac OS Application Termination, and full integration of Mac OS X Menubar
Inline ... On Aug 25, 2010, at 5:36 PM, Cédric Krier wrote: > On 25/08/10 23:27 +0200, Tobias Paepke wrote: >> On 25.08.2010 23:05, Cédric Krier wrote: >>> On 25/08/10 16:49 -0400, Phillip Heller wrote: >>>> I might have found a gtk theme that is more Mac like than the default. >>>> >>>> Try that Binary I provided with the attached gtkrc ( save as ~/.gtkrc-2.0 >>>> ), and let me know what you think. >>> Theme is just customization. We will never have full agrement on this. >>> It was decided to use standard "Clearlooks" for MacOS X bundle. Understood, though I think most Mac users will like this one better >> as there isn't an obvious way to change the gtk theme with a bundled >> client on windows or osx: > > Edit ~/.gtkrc-2.0 And this is a reasonable solution, though I think it is worth reconsidering this for the default. Here is a gallery for some comparisons between this and the current default: http://gallery.me.com/pheller#100196 Regards, Phil smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Fix for Mac OS Application Termination, and full integration of Mac OS X Menubar
I might have found a gtk theme that is more Mac like than the default. Try that Binary I provided with the attached gtkrc ( save as ~/.gtkrc-2.0 ), and let me know what you think. --phil gtkrc Description: Binary data On Aug 25, 2010, at 3:10 PM, Phillip Heller wrote: > Glad it's working well for you folks. > > I've scoured the internet for better looking Leopard styled gtk themes, but > alas I have not found anything that is better than the default theme. I'm > not talented with graphics, so I'll probably just leave it alone, unless > someone else has a gtk theme they'd like me to try. > > A couple things left to do: > > 1) Fix the accelerators (cmd rather than ctrl) > 2) Fix the dock icon crashing issue > > Regards, > > Phil > > On Aug 25, 2010, at 7:18 AM, Tobias Paepke wrote: > >> And it looks way more like a native apple application. >> amazing! >> >> -- >> tryton-dev@googlegroups.com mailing list > smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Custom URLs
Comments inline... On Aug 25, 2010, at 3:28 PM, Cédric Krier wrote: > On 25/08/10 15:10 -0400, Phillip Heller wrote: >> Greetings, >> >> After some discussion today with Tobias in IRC, I decided to share some >> thoughts on implementation of Custom URL handlers for Tryton. >> >> An initial pass at the structure: >> >> tryton:/[/[username[:passwo...@]server[:port][/protocol]][/database]/[/id] > > - username nor password should be define in the url for security purpose. I generally agree. Thought it might be useful for readonly access. > - the server:port should always mandatory otherwise I don't see where the > client should connect. The thought here was that some reasonable defaults would be assumed. If port is not specified, but protocol is, then choose the default port for that protocol. If server isn't specified, then use any connected server. Though this would not work if there were multiple servers open (when the client can do this). I thought the same idea with the database. Though I do now agree that server and database are mandatory. Protocol and port can both default to something, and optionally be overriden. > > - I don't see the usage of protocol as it is the client that should select > which protocol it wants to use. I can imagine one use case where the client specifying the protocol is important; perhaps when an employee is first hired and then created in tryton. The final action of the workflow is to send the user an email that says: "Your tryton account has been created, access it here: tryton://server/netrpc/database". > I think we should think about more powerfull URI scheme (perhaps REST-like) to > be able to get domain/context inside. The best should be to be able to encode > any of the ir.action. Can you give some examples of how you might encode context variables or invoke other actions? I think context is most useful in the query string: tryton://server/database/sale.order/123?lang=es_ES A good thing to work out would be how to run and render a report via URI. I can see the usefulness in pulling an invoice this way - especially if this URL method can eventually be ported to the web client. > In fact I think we should allow the client to accept an url as argument and he > will open a new one if there is no running one or if the runnings are not > connected to the same server and database. I agree that invoking the application with a URL should, if the client is already running, signal to it that a new window should be opened with the URL. And, if the client is not already running, it should be started, and a window opened with the URL. >> As Tobias brought up, there are occasions where a client might need to >> connect to multiple databases. The client is not currently architected to >> support that. A specific example of this use case would be a "Tryton >> Certified" accountant. The accountant might be retained by many clients, >> and need to occasionally work with multiple tryton instances at once. > > I don't think. Tryton migrates easily so there is no reason to have this > issue. Hmmh, I'm not sure I understand your response here. To further explain my point: Perhaps the Accountant is researching an accounting mismatch at tryton://companyA/database, and he has many windows open, with search results. Then, the accountant gets a call from companyB saying "we need emergency help". He should be able to do "Window" -> "New", and connect to companyB's tryton instance. Similarly, if he gets two emails, with URLs that specify different tryton instances, new windows should be created for each. (rather than running multiple instances of the whole application) > >> This brings up yet one more challenge, which is protocol compatibility. In >> the accountant example, there could be an instance where one of the >> accountant's clients is running a 1.X version, where another is running a >> 2.X version. Backwards compatibility in the client would be needed. >> >> I would certainly appreciate feedback and discussions, which I will >> aggregate into a blueprint. > > I'm against this because it slow down the development and it makes testing > become an exponential task. Well, it really only becomes a concern when there are many more instances of Tryton running in the real world. So it really isn't important to discuss now anyways. :-) Regards, Phil smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Fix for Mac OS Application Termination, and full integration of Mac OS X Menubar
Glad it's working well for you folks. I've scoured the internet for better looking Leopard styled gtk themes, but alas I have not found anything that is better than the default theme. I'm not talented with graphics, so I'll probably just leave it alone, unless someone else has a gtk theme they'd like me to try. A couple things left to do: 1) Fix the accelerators (cmd rather than ctrl) 2) Fix the dock icon crashing issue Regards, Phil On Aug 25, 2010, at 7:18 AM, Tobias Paepke wrote: > And it looks way more like a native apple application. > amazing! > > -- > tryton-dev@googlegroups.com mailing list smime.p7s Description: S/MIME cryptographic signature
[tryton-dev] Custom URLs
Greetings, After some discussion today with Tobias in IRC, I decided to share some thoughts on implementation of Custom URL handlers for Tryton. An initial pass at the structure: tryton:/[/[username[:passwo...@]server[:port][/protocol]][/database]/[/id] Examples: tryton://sale.order tryton://party.party/123 tryton://server/party.contact/456 tryton://localhost/netrpc/mycompany/sale.order/12345 tryton://jeff:supersecr...@tryton.some.tld:/xmlrpc/acme/support.case/4789 tryton://reado...@tryton/xmlrpc/bugtracker/bug.history/27891 Some use cases would be: Embed them in email, as part of a workflow. When an employ submits an expense report, the workflow goes to a manager approval state. A request is created for the employee's manager, with content of "An expense report has been submitted for your approval. Please view it here: tryton://trytonhost/expense.report/789" Open them via an external system. When a customer calls into a call center to inquire about an invoice, an IVR (Interactive Voice Reponse) system asks them for their Invoice Number, and then the numeric passcode associated with the billing party record. These values are validated and then the IVR transfers their call to an agent, and transmits "tryton://invoice.invoice/1234" to an helper application running on the agent's machine, which invokes the appropriate operating system action to open this URL. Enabling it on various platforms: Seems it is done like so under Gnome, and then there must be some internal method to listen for events. gconftool-2 -s /desktop/gnome/url-handlers/foo/command '/path/to/app %s' --type String gconftool-2 -s /desktop/gnome/url-handlers/foo/enabled --type Boolean true Under MacOS, some magic is added to the Application Bundle's Info.plist, and then the application must listen for events. Under Windows, find an example here: http://blogs.msdn.com/b/noahc/archive/2006/10/19/register-a-custom-url-protocol-handler.aspx Challenges: As Tobias brought up, there are occasions where a client might need to connect to multiple databases. The client is not currently architected to support that. A specific example of this use case would be a "Tryton Certified" accountant. The accountant might be retained by many clients, and need to occasionally work with multiple tryton instances at once. This brings up yet one more challenge, which is protocol compatibility. In the accountant example, there could be an instance where one of the accountant's clients is running a 1.X version, where another is running a 2.X version. Backwards compatibility in the client would be needed. I would certainly appreciate feedback and discussions, which I will aggregate into a blueprint. Regards, Phil smime.p7s Description: S/MIME cryptographic signature
[tryton-dev] Fix for Mac OS Application Termination, and full integration of Mac OS X Menubar
Greetings, Please review the patch found at the link below, which eliminates the need to force-quit the tryton client upon "Quit", as well as fully integrates the Menu Bar with the Mac Menubar. http://codereview.appspot.com/1975047 Also, please find a binary build of the client at the following URL, for testing. Please note that this client is built with 1.7 code, and requires a 1.7 server (post Transaction patch), and that the gtk-osx build is against the 10.6 sdk. http://files.me.com/pheller/vpu6xg There is a known issue with this build; when "Quit" is invoked from the Dock icon, the application crashes. I will continue to investigate this crash. Assistance is appreciated. Regards, Phil smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Product Attributes EAV model
Comments inline...
On Aug 21, 2010, at 10:23 AM, Sharoon Thomas wrote:
> This would mea that we have a new model class called ModelEAV which inherits
> from ModelStorage and handles attributes.
Excellent! This is far more extensible to other models for things not yet
considered.
> ir.attribute - Where the attributes for a model are defined
>
> model : fields.Many2one('ir.model')
> name : fields.Char
> type : fields.Selection(
>[
>text/selection/binary/numeric
>]
>)
> selections : fields.One2Many ir.attribute.selections (try using property
> fields)
> active: fields.Boolean
> help: fields.Text
>
> Ex:
> model: `product.product`
> name: `color`
> type: `selection`
> active: True
> help: `Choose the color of the product`
How about relation types? A good example of how this might be useful:
A Cisco 6504-E has 4 slots. Perhaps slots 1 and 2 can only contain a
WS-C6504E-S32-GE or WS-C6504E-S32-10GE.
If the ir.attribute type field could include relational types, a many2one in
this instance, this would be pretty easy to support:
name: Cisco 6504-E Slot 1 Compatible
type: one2many (to product.product)
Then, an ir.attribute.value might contain:
entity: (product.product,12) [the Cisco 6504-E product]
attribute: 'Product Relation'
value_one2many: (14,29) [where 14 and 29 are the record ids for
'WS-C6504E-S32-GE', and 'WS-C6504E-S32-10GE']
> and update the view to use a new tag
>
> which will expand into a one2many list view where the attributes are
> shown/edited.
This is also great, though would become more difficult to render if the this
ModelEAV is extended to include relation types.
This aside, great work! Generalization is certainly the way to go.
--phil
smime.p7s
Description: S/MIME cryptographic signature
Re: [tryton-dev] Product Attributes EAV model
Ok, so the Attributes you list below are perfect examples, and I do agree that Serial Numbers and other *instance* related values are a function of inventory. I suppose this concept is something to propose in a stock_instance module or something of the sort. Thanks. --phil On Aug 19, 2010, at 10:46 AM, Sharoon Thomas wrote: > Well i dont think that this EAV model would fit this purpose of a serial > number. However, in other terms this could be generalised. let me stick to > the example of mac book. > > Attributes could be: > > Screen Size > Weight > Processor > Memory > HDD > > etc. which remain constant for a product. having a different serial number, I > am not sure is an attribute issue. I think its more of an inventory tracking > issue ? smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Product Attributes EAV model
Sharoon et al, One thing I had talked with Cédric about recently was a similar EAV concept, for the purposes of tracking individual instances of a product. So, perhaps there would be a ProductInstance model, which "realizes" all of the attributes associated with a product. A good example would be a manufacturer assigned serial number, or a Hardware MAC address. Perhaps you order 10 Apple Macbooks for resale. In Tryton, the "Macbook" product would have attributes including "Serial Number", "MAC", etc. These would have AttributeOptions that would include "not null", so on instantiation of a ProductInstance for each of these 10 Macbooks, these attributes would need to be populated. Another question is how will you accommodate for relational attribute values (o2m, m2o, m2m)? Thoughts of how these things might fit into this EAV pattern? --phil On Aug 19, 2010, at 7:29 AM, Sharoon Thomas wrote: > Hi Cedk, > > On 19 Aug 2010, at 12:18, Cédric Krier wrote: > >> On 18/08/10 18:25 +0200, Mathias Behrle wrote: >>> * Betr.: " [tryton-dev] Product Attributes EAV model" (Wed, 18 Aug 2010 >>> 13:46:55 +0100): >>> Hello All, We have built a POC for the EAV [1] model of storing product attibutes. Quite often we have to add extra fields to product depending on specific customer installations. Like color, manufacturer etc. This is a general approach to the problem by having a relationship between products and attributes just like contact mechanisms of party. I request your comments and code review on the same. [1] http://en.wikipedia.org/wiki/Entity-attribute-value_model >>> >>> Very interesting concept, thanks a lot! >>> >>> For an even more generic generic way, I would even favor to put models >>> >>> ProductAttributes -> Attribute >>> ProductAttributeOptions -> AttributeOption >>> >>> in a separate module attribute (thus they could be (re)used on a number of >>> different objects where the same concept could be required (like party in >>> [1])) >>> >>> and to do only the product specific implementation in module >>> product_attribute >>> containing models >> >> But what will be the advantage to have all atributes in one table? >> Because you will have the same attribute available for any Model, I find it >> strange. >> > Not exactly: > > Attribute > == > model: - The model for which the attribute is valid for > name: - Name of attribute > type: - Data type of the attribute > options: o2m Attribute option > > AttributeOption > > attribute m2o attribute > values - the option value > > AttributeValue > === > > entity: reference field (model, id) > attribute: m2o Attribute > value: value of attribute > > So infact the attribute is domain[ed] on the model. So only relevant > attributes should get displayed in the corresponding object. > > Hope it sounds ok ? > > Thanks, smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
Submitted: -rw-r--r-- 1 pheller staff 16672 Aug 10 12:03 module-company.Transaction.patch -rw-r--r-- 1 pheller staff 2190 Aug 10 12:03 module-company_work_time.Transaction.patch -rw-r--r-- 1 pheller staff 25423 Aug 10 12:03 module-currency.Transaction.patch -rw-r--r-- 1 pheller staff 22725 Aug 10 12:03 module-party.Transaction.patch -rw-r--r-- 1 pheller staff 73142 Aug 10 12:03 module-sale.Transaction.patch -rw-r--r-- 1 pheller staff 4696 Aug 10 12:03 module-sale_price_list.Transaction.patch -rw-r--r-- 1 pheller staff 16637 Aug 10 12:03 module-stock_forecast.Transaction.patch -rw-r--r-- 1 pheller staff 3274 Aug 10 12:03 module-stock_inventory_location.Transaction.patch -rw-r--r-- 1 pheller staff577 Aug 10 12:03 module-stock_location_sequence.Transaction.patch -rw-r--r-- 1 pheller staff 1368 Aug 10 12:03 module-stock_product_location.Transaction.patch -rw-r--r-- 1 pheller staff 37725 Aug 10 12:03 module-stock_supply.Transaction.patch -rw-r--r-- 1 pheller staff 2667 Aug 10 12:03 module-stock_supply_day.Transaction.patch -rw-r--r-- 1 pheller staff 14699 Aug 10 12:03 module-timesheet.Transaction.patch --phil smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
stock_location_sequence: http://codereview.appspot.com/1917049 Brace yourselves, this one is exciting; only the INSTALL needed to be updated for the new python2.5 requirement. --phil smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
stock_supply_day: http://codereview.appspot.com/1885046 smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
stock_supply: http://codereview.appspot.com/1932041 smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
stock_product_location: http://codereview.appspot.com/1875054 smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
stock_inventory_location: http://codereview.appspot.com/1931041 smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
stock_forecast: http://codereview.appspot.com/1918046 smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
Done. http://codereview.appspot.com/1918044 On Aug 5, 2010, at 6:37 PM, Phillip Heller wrote: > Starting sale_price_list > > --phil smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
Starting sale_price_list --phil smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
First pass at a patch here: http://codereview.appspot.com/1892048/show --phil On Aug 4, 2010, at 12:09 PM, Phillip Heller wrote: > Starting on "Sale". > > > On Jul 28, 2010, at 5:04 PM, Cédric Krier wrote: > >> Hi, >> >> I'm asking your help to speed up the migration to the new transaction design. >> Here is my proposition to not have duplicate work. >> People who wants to help can post (with reply to this email) on which module >> he will work before starting. So by posting it he engages to provide a patch >> for review in 3 days (normally it should take 3-4 hours) after that it will >> be >> considered as abandoned. >> Modules that will not require modification must be also posted here by saying >> no need to change. smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
Starting on "Sale". On Jul 28, 2010, at 5:04 PM, Cédric Krier wrote: > Hi, > > I'm asking your help to speed up the migration to the new transaction design. > Here is my proposition to not have duplicate work. > People who wants to help can post (with reply to this email) on which module > he will work before starting. So by posting it he engages to provide a patch > for review in 3 days (normally it should take 3-4 hours) after that it will be > considered as abandoned. > Modules that will not require modification must be also posted here by saying > no need to change. smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
"Party" converted, including unit tests (all of which pass): http://codereview.appspot.com/1850047 On Jul 28, 2010, at 5:04 PM, Cédric Krier wrote: > Hi, > > I'm asking your help to speed up the migration to the new transaction design. > Here is my proposition to not have duplicate work. > People who wants to help can post (with reply to this email) on which module > he will work before starting. So by posting it he engages to provide a patch > for review in 3 days (normally it should take 3-4 hours) after that it will be > considered as abandoned. > Modules that will not require modification must be also posted here by saying > no need to change. smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
Started and finished "currency", including unit tests (all of which pass): http://codereview.appspot.com/1909044 On Jul 28, 2010, at 5:04 PM, Cédric Krier wrote: > Hi, > > I'm asking your help to speed up the migration to the new transaction design. > Here is my proposition to not have duplicate work. > People who wants to help can post (with reply to this email) on which module > he will work before starting. So by posting it he engages to provide a patch > for review in 3 days (normally it should take 3-4 hours) after that it will be > considered as abandoned. > Modules that will not require modification must be also posted here by saying > no need to change. smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
Starting on "Party". On Jul 28, 2010, at 5:04 PM, Cédric Krier wrote: > Hi, > > I'm asking your help to speed up the migration to the new transaction design. > Here is my proposition to not have duplicate work. > People who wants to help can post (with reply to this email) on which module > he will work before starting. So by posting it he engages to provide a patch > for review in 3 days (normally it should take 3-4 hours) after that it will be > considered as abandoned. > Modules that will not require modification must be also posted here by saying > no need to change. smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
company: http://codereview.appspot.com/1853047 company_work_time: http://codereview.appspot.com/1885043 On Jul 29, 2010, at 2:39 PM, Phillip Heller wrote: > I'm starting on: > > company > company_work_time > > On Jul 28, 2010, at 5:04 PM, Cédric Krier wrote: > >> Hi, >> >> I'm asking your help to speed up the migration to the new transaction design. >> Here is my proposition to not have duplicate work. >> People who wants to help can post (with reply to this email) on which module >> he will work before starting. So by posting it he engages to provide a patch >> for review in 3 days (normally it should take 3-4 hours) after that it will >> be >> considered as abandoned. >> Modules that will not require modification must be also posted here by saying >> no need to change. smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
I'm starting on: company company_work_time On Jul 28, 2010, at 5:04 PM, Cédric Krier wrote: > Hi, > > I'm asking your help to speed up the migration to the new transaction design. > Here is my proposition to not have duplicate work. > People who wants to help can post (with reply to this email) on which module > he will work before starting. So by posting it he engages to provide a patch > for review in 3 days (normally it should take 3-4 hours) after that it will be > considered as abandoned. > Modules that will not require modification must be also posted here by saying > no need to change. smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
http://codereview.appspot.com/1869048 On Jul 29, 2010, at 1:27 PM, Phillip Heller wrote: > I'm starting on "timesheet". > > On Jul 28, 2010, at 5:04 PM, Cédric Krier wrote: > >> Hi, >> >> I'm asking your help to speed up the migration to the new transaction design. >> Here is my proposition to not have duplicate work. >> People who wants to help can post (with reply to this email) on which module >> he will work before starting. So by posting it he engages to provide a patch >> for review in 3 days (normally it should take 3-4 hours) after that it will >> be >> considered as abandoned. >> Modules that will not require modification must be also posted here by saying >> no need to change. smime.p7s Description: S/MIME cryptographic signature
Re: [tryton-dev] Help wanted for Transaction migration
I'm starting on "timesheet". On Jul 28, 2010, at 5:04 PM, Cédric Krier wrote: > Hi, > > I'm asking your help to speed up the migration to the new transaction design. > Here is my proposition to not have duplicate work. > People who wants to help can post (with reply to this email) on which module > he will work before starting. So by posting it he engages to provide a patch > for review in 3 days (normally it should take 3-4 hours) after that it will be > considered as abandoned. > Modules that will not require modification must be also posted here by saying > no need to change. smime.p7s Description: S/MIME cryptographic signature
