Re: [Sugar-devel] Q on tracker cleanup / triage
On 07/04/2010 01:11 AM, Tim McNamara wrote: Hi devs, 500+ unclosed bugs causes problems. I feel intimidated overwhelmed by this number. I also find it difficult to find material sift through bug reports, to find areas where I can help (although, at the moment, I'm quite content with sugar-love level problems). About 500+ bugs being hard to handle, I think that maybe what happens is that either those are not correctly categorized, or our tools are not making use of all info and thus unwanted results appear in our queries making difficult to find what we look for. However, 500+ bugs isn't that accurate. Some have been identified as upstream problems, but their resolution status remains open. As it has been said in another reply, sometimes we want to track upstream resolution status, and sometimes even a workaround in Sugar itself. I suggest that we make more use of the wontfix resolution status. In 5 minutes, I've found some likely candidates. wontfix comes with a high risk of alienating people submitting bugs, but a polite note saying that it's upstream or possibly something like Unfortunately, we have higher priority areas to work on right now. However we would welcome a patch if you are still wiling to work on this issue. It's great that you've taken the time to submit the bug, and we have looked into resolving things for you. Well, if we are willing to accept a patch to solve an issue, then I think that the issue is definitely worth an open ticket. Some candidates: Sorry, no net acces right now, will reply later about specific tickets. Thanks a lot for your interest in this very important topic, Tomeu wontfix - upstream http://bugs.sugarlabs.org/ticket/1307 wontfix - other priorities http://bugs.sugarlabs.org/ticket/288 http://bugs.sugarlabs.org/ticket/292 duplicates http://bugs.sugarlabs.org/ticket/172 http://bugs.sugarlabs.org/ticket/394 I think that resolving bugs, even if by using the wontfix tag will make other areas easier. Also, I don't know if I have the permission to change resolution status when something hasn't been touched for 12+ months. If I were to change status on likely candidates, would this anger the other developers? Tim ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Q on tracker cleanup / triage
On 07/04/2010 05:15 PM, Bernie Innocenti wrote: El Sun, 04-07-2010 a las 02:29 +0100, Gary Martin escribió: Also, I don't know if I have the permission to change resolution status when something hasn't been touched for 12+ months. If I were to change status on likely candidates, would this anger the other developers? I know it's sensitive, for us slightly more delicate souls, but if you feel strongly about a bug status... I'd say go for its change of status. It will trigger either action/indifference from genuinely interested parties, possibly patches/discussion, action, and glory ;) Yes, I would encourage users to apply the Be Bold mantra [1] to the bug tracker as well as the wiki. If in doubt, go ahead and change. If you were wrong, someone will revert your change. [1] http://en.wikipedia.org/wiki/Wikipedia:Be_bold Agreed as well, with the warning that if the changes are in mass and exceed the capacity of the rest of interested parties to react, review or revert those changes, it may not be such a good idea. Normally when developers get angry at some triaging activity it's because of automated changes, not because of spontaneous triaging. It seems fair to say there is some level of (completely understandable) analysis paralysis for some level of bugs. Bug triage activity is at a low at the moment. Don't apologise for bringing that task back up in everyones focus/face. Low? Not really: http://bugs.sugarlabs.org/timeline I'm offline right now so I cannot check (though I have seen lots of activity from you and Silbe in the past weeks, kudos!), but I would like to encourage people interested in triaging that read and improve http://wiki.sugarlabs.org/go/BugSquad/Triage_Guide so we follow a consistent set of guidelines and specially so bugs get out from the triaging queue once someone has been able to look at it. Thanks, Tomeu ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Q on tracker cleanup / triage
Excerpts from Tomeu Vizoso's message of Mon Jul 05 10:47:44 + 2010: [...] or possibly something like Unfortunately, we have higher priority areas to work on right now. Never close bugs because nobody has time to fix them. Rather use the priority field to indicate the maintainer isn't interested in working on the issue, but would welcome patches. Maybe s/maintainer/development team? Actually the priority field is something that nags me about our current bug tracker. It's a single field, but bugs have different priorities to different people. For example the issues with the NetworkManager integration are so annoying that they were (*) of high priority to me, but apparently not to other members of the Development Team; maybe because deployments have different network setups and the issues don't appear there. Guess we have enough bugs to have a triaging team (or bug squad, though I don't like the name) instead of just a bug master. I still think the activity (!= number of open bugs) on our bug tracker is low enough for a single person. It shouldn't take more than say an hour a day. Even Gentoo had a single bug master for a very long time. A triage team might help with cleaning up _old_ bugs, though. (*) I'm currently using system connections and cnetworkmanager so I can use IEEE 802.1x authentication on the university network instead of the buggy VPN client. This works significantly better than using Sugar-managed user connections. With user connections getting thrown out [1] of NetworkManager, quite a few of our problems might get solved for us (or be tackled by the rewrite of the NetworkManager integration that might be required). Sascha [1] http://live.gnome.org/NetworkManager/RemovingUserSettings signature.asc Description: PGP signature ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] Lint sugar sources
Hi all, I'm using/planing-to-use lint tools in several projects. And since I'm not using jhbuild which support pylint/pep8.py tools, I created sugar-lint[1] project. There is single command - sugar-lint which will call pylint (should be installed before) with custom configuration file and pep8.py (comes with sugar-lint sources). [1] http://wiki.sugarlabs.org/go/Activity_Team/Sugar_Lint -- Aleksey ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [PATCH] Bump v0.88.0
Sounds good to me. Regards, Tomeu On Mon, Jul 5, 2010 at 06:21, Aleksey Lim alsr...@member.fsf.org wrote: --- NEWS | 3 +++ configure.ac | 2 +- 2 files changed, 4 insertions(+), 1 deletions(-) diff --git a/NEWS b/NEWS index b20eb59..4aa03d1 100644 --- a/NEWS +++ b/NEWS @@ -1,3 +1,6 @@ +v0.88.0 2010-03-29 +* fix migration from 0.82 data store (#1838) + v0.87.4 2010-03-10 * Do not remove internal metadata fields while rewriting object #1813 * Auto detect mime type in copy-to-journal #586 (tomeu) diff --git a/configure.ac b/configure.ac index ddf7ffb..0fcdb49 100644 --- a/configure.ac +++ b/configure.ac @@ -1,4 +1,4 @@ -AC_INIT([sugar-datastore],[0.87.4],[],[sugar-datastore]) +AC_INIT([sugar-datastore],[0.88.0],[],[sugar-datastore]) AC_PREREQ([2.59]) -- 1.7.0.4 ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Activity title update
On Mon, 2010-07-05 at 15:04 +0200, Tomeu Vizoso wrote: Guess I'm a bit late, but I think I have heard this before and I'm not sure what means to has finished changing an edit widget. Do you know of another GUI toolkit with such a signal? http://doc.qt.nokia.com/4.6/qlineedit.html#editingFinished It's a pretty useful and obvious feature, even Microsoft's MFC did let me do it easily in 1992. -- // Bernie Innocenti - http://codewiz.org/ \X/ Sugar Labs - http://sugarlabs.org/ ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] [PATCH] Journal Volumes Backup and Restore
Add a basic backup and restore feature for the Sugar Journal.
It provides:
- Generic Backup and Restore dialog GUI.
- Process manager class as an abstraction layer between the dialog and
backup/restore scripts. (Allowing to work with many backup and restore
technologies, using the same GUI, with no need for script rewrite).
- Basic file system Volume Restore and Backup scripts implemented in Python.
- New backup and restore options for journal volumes palettes.
This patch is based on Esteban Arias (Plan Ceibal) Volume Backup and Restore
patch, with a few changes:
- Refactor original Backup dialog class into a generic dialog class.
- Create specialized VolumeBackupDialog and VolumeRestoreDialog subclasses.
- Rewrite backup and restore scripts in python for an easier sugar interaction.
- Add backup identification helpers to jarabe.journal.misc.
---
bin/Makefile.am |4 +-
bin/journal-backup-volume | 57
bin/journal-restore-volume| 67 +
src/jarabe/journal/Makefile.am|3 +-
src/jarabe/journal/misc.py| 27
src/jarabe/journal/processdialog.py | 248 +
src/jarabe/journal/volumestoolbar.py |5 +-
src/jarabe/model/Makefile.am |3 +-
src/jarabe/model/processmanagement.py | 98 +
src/jarabe/view/palettes.py | 44 ++
10 files changed, 551 insertions(+), 5 deletions(-)
create mode 100644 bin/journal-backup-volume
create mode 100644 bin/journal-restore-volume
create mode 100644 src/jarabe/journal/processdialog.py
create mode 100644 src/jarabe/model/processmanagement.py
diff --git a/bin/Makefile.am b/bin/Makefile.am
index 05a9215..8cc87b5 100644
--- a/bin/Makefile.am
+++ b/bin/Makefile.am
@@ -5,7 +5,9 @@ python_scripts =\
sugar-install-bundle\
sugar-launch\
sugar-session \
- sugar-ui-check
+ sugar-ui-check \
+ journal-backup-volume \
+ journal-restore-volume
bin_SCRIPTS = \
sugar \
diff --git a/bin/journal-backup-volume b/bin/journal-backup-volume
new file mode 100644
index 000..4f3ec8a
--- /dev/null
+++ b/bin/journal-backup-volume
@@ -0,0 +1,57 @@
+#!/usr/bin/env python
+# Copyright (C) 2010, Paraguay Educa tecnolo...@paraguayeduca.org
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see http://www.gnu.org/licenses/.
+#
+
+import os
+import sys
+import subprocess
+import logging
+
+from sugar import env
+#from sugar.datastore import datastore
+
+backup_identifier = sys.argv[2]
+volume_path = sys.argv[1]
+
+if len(sys.argv) != 3:
+print 'Usage: %s volume_path backup_identifier' % sys.argv[0]
+exit(1)
+
+logging.debug('Backup started')
+
+backup_path = os.path.join(volume_path, 'backup', backup_identifier)
+
+if not os.path.exists(backup_path):
+os.makedirs(backup_path)
+
+#datastore.freeze()
+subprocess.call(['pkill', '-9', '-f', 'python.*datastore-service'])
+
+result = 0
+try:
+cmd = ['tar', '-C', env.get_profile_path(), '-czf', \
+ os.path.join(backup_path, 'datastore.tar.gz'), 'datastore']
+
+subprocess.check_call(cmd)
+
+except Exception, e:
+logging.error('Backup failed: %s', str(e))
+result = 1
+
+#datastore.thaw()
+
+logging.debug('Backup finished')
+exit(result)
diff --git a/bin/journal-restore-volume b/bin/journal-restore-volume
new file mode 100644
index 000..aa14ad0
--- /dev/null
+++ b/bin/journal-restore-volume
@@ -0,0 +1,67 @@
+#!/usr/bin/env python
+# Copyright (C) 2010, Paraguay Educa tecnolo...@paraguayeduca.org
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see http://www.gnu.org/licenses/.
+#
+
+import os
+import sys
+import shutil
+import logging
+import subprocess
+
+from sugar import env
+#from
[Sugar-devel] [PATCH] Journal XS backup and restore
Include a new journal volumes toolbar icon. The icon represents
the schoolserver. The icon palette provides a backup and restore
options.
---
src/jarabe/desktop/favoritesview.py |2 +
src/jarabe/journal/processdialog.py | 30 ++
src/jarabe/journal/volumestoolbar.py | 37 +--
src/jarabe/view/palettes.py | 39 +-
4 files changed, 104 insertions(+), 4 deletions(-)
diff --git a/src/jarabe/desktop/favoritesview.py
b/src/jarabe/desktop/favoritesview.py
index aca945a..a311b52 100644
--- a/src/jarabe/desktop/favoritesview.py
+++ b/src/jarabe/desktop/favoritesview.py
@@ -42,6 +42,7 @@ from jarabe.view import launcher
from jarabe.model.buddy import BuddyModel
from jarabe.model import shell
from jarabe.model import bundleregistry
+from jarabe.journal.journalactivity import get_journal
from jarabe.journal import misc
from jarabe.desktop import schoolserver
@@ -328,6 +329,7 @@ class FavoritesView(hippo.Canvas):
alert.props.msg = _('You are now registered ' \
'with your school server.')
self._my_icon.remove_register_menu()
+get_journal()._volumes_toolbar.setup_xs_button()
ok_icon = Icon(icon_name='dialog-ok')
alert.add_button(gtk.RESPONSE_OK, _('Ok'), ok_icon)
diff --git a/src/jarabe/journal/processdialog.py
b/src/jarabe/journal/processdialog.py
index 8217973..4d8704e 100644
--- a/src/jarabe/journal/processdialog.py
+++ b/src/jarabe/journal/processdialog.py
@@ -246,3 +246,33 @@ class VolumeRestoreDialog(ProcessDialog):
self._message.set_markup('%s %s.\n\n' % (_('Journal content will be
restored from'), volume_path) + \
'bigb%s/b %s/big' % (_('Warning:'),
_('Current Journal content will be deleted!')))
+class XSBackupDialog(ProcessDialog):
+
+def __init__(self, xs_hostname):
+ProcessDialog.__init__(self, 'journal-backup')
+self._resetup_information(xs_hostname)
+
+def _resetup_information(self, xs_hostname):
+self._start_message = '%s %s.' % (_('Please wait, saving Journal
content to'), xs_hostname)
+
+self._finished_message = _('The journal content has been saved.')
+
+self._title.set_markup('bigb%s/b/big' % _('Backup'))
+
+self._message.set_text('%s %s.' % (_('Journal content will be saved
to'), xs_hostname))
+
+class XSRestoreDialog(ProcessDialog):
+
+def __init__(self, xs_hostname):
+ProcessDialog.__init__(self, 'journal-restore')
+self._resetup_information(xs_hostname)
+
+def _resetup_information(self, xs_hostname):
+self._start_message = '%s %s.' % (_('Please wait, restoring journal
content from'), xs_hostname)
+
+self._finished_message = _('The journal content has been restored.')
+
+self._title.set_markup('bigb%s/b/big' % _('Restore'))
+
+self._message.set_text('%s %s.' % (_('Journal content will be restored
from'), xs_hostname))
+
diff --git a/src/jarabe/journal/volumestoolbar.py
b/src/jarabe/journal/volumestoolbar.py
index 2e64fe2..e1f9ce0 100644
--- a/src/jarabe/journal/volumestoolbar.py
+++ b/src/jarabe/journal/volumestoolbar.py
@@ -22,12 +22,14 @@ import gio
import gtk
import gconf
+from sugar.graphics.toolbutton import ToolButton
from sugar.graphics.radiotoolbutton import RadioToolButton
+from sugar.graphics.icon import Icon
from sugar.graphics.palette import Palette
from sugar.graphics.xocolor import XoColor
from jarabe.journal import model
-from jarabe.view.palettes import JournalVolumePalette
+from jarabe.view.palettes import JournalVolumePalette, JournalXSPalette
class VolumesToolbar(gtk.Toolbar):
__gtype_name__ = 'VolumesToolbar'
@@ -52,6 +54,7 @@ class VolumesToolbar(gtk.Toolbar):
self.connect('destroy', self.__destroy_cb)
+self.setup_xs_button()
gobject.idle_add(self._set_up_volumes)
def __destroy_cb(self, widget):
@@ -59,6 +62,21 @@ class VolumesToolbar(gtk.Toolbar):
volume_monitor.disconnect(self._mount_added_hid)
volume_monitor.disconnect(self._mount_removed_hid)
+def setup_xs_button(self):
+client = gconf.client_get_default()
+backup_url = client.get_string('/desktop/sugar/backup_url')
+
+if backup_url:
+xs_button = XSButton()
+xs_button.set_palette(JournalXSPalette(backup_url))
+#xs_button.props.group = self._volume_buttons[0]
+xs_button.show()
+
+self._volume_buttons.append(xs_button)
+position = len(self._volume_buttons) - 1
+self.insert(xs_button, position)
+self.show()
+
def _set_up_volumes(self):
volume_monitor = gio.volume_monitor_get()
self._mount_added_hid = \
@@ -104,8 +122,9 @@ class VolumesToolbar(gtk.Toolbar):
def _get_button_for_mount(self, mount):
mount_point =
[Sugar-devel] [PATCH] School Server Icon
--- icons/scalable/device/Makefile.am |1 + icons/scalable/device/school-server.svg | 20 2 files changed, 21 insertions(+), 0 deletions(-) create mode 100644 icons/scalable/device/school-server.svg diff --git a/icons/scalable/device/Makefile.am b/icons/scalable/device/Makefile.am index 28818ab..198488b 100644 --- a/icons/scalable/device/Makefile.am +++ b/icons/scalable/device/Makefile.am @@ -77,6 +77,7 @@ icon_DATA = \ network-wireless-connected-100.svg \ network-wireless-connected-100.icon \ printer.svg \ + school-server.svg \ speaker-000.svg \ speaker-033.svg \ speaker-066.svg \ diff --git a/icons/scalable/device/school-server.svg b/icons/scalable/device/school-server.svg new file mode 100644 index 000..28a854d --- /dev/null +++ b/icons/scalable/device/school-server.svg @@ -0,0 +1,20 @@ +?xml version=1.0 encoding=UTF-8? +!DOCTYPE svg PUBLIC -//W3C//DTD SVG 1.1//EN http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd; [ + !ENTITY fill_color #FF + !ENTITY stroke_color #010101 +] +svg xmlns=http://www.w3.org/2000/svg; width=55 height=55 +path d=M10,50 S8,50 8,48 L8,35 L27.5,24 L47,35 L47,48 S47,50 45,50 z fill=fill_color; stroke=stroke_color; stroke-width=1.75/ +path d=M33,50 L33,39 S33,37 31,37 L24,37 S22,37 22,39 L22,50 z fill=stroke_color; stroke=stroke_color; stroke-width=1.75/ +path d=M20,23.5 L20,10 L35,10 L35,23.5 z fill=stroke_color; stroke=stroke_color; stroke-width=3.5/ +line x1=6 y1=33.5 x2=27.5 y2=21.5 stroke=stroke_color; stroke-width=6.75 stroke-linecap=round/ +line x1=49 y1=33.5 x2=27.5 y2=21.5 stroke=stroke_color; stroke-width=6.75 stroke-linecap=round/ +line x1=6 y1=33.5 x2=27.5 y2=21.5 stroke=fill_color; stroke-width=3.5 stroke-linecap=round/ +line x1=49 y1=33.5 x2=27.5 y2=21.5 stroke=fill_color; stroke-width=3.5 stroke-linecap=round/ +path d=M20,23.7 L20,10 L35,10 L35,23.7 z fill=fill_color; stroke=none stroke-width=2/ +path d=M18,9.25 L27.5,3.25 L37,9.25 z fill=stroke_color; stroke=stroke_color; stroke-width=2.25/ +circle cx=27.5 cy=15.5 r=3.25 fill=stroke_color; stroke=none stroke-width=2/ +path d=M24.25,19.25 L24.25,15.5 L30.75,15.5 L30.75,19.25 z fill=stroke_color; stroke=none stroke-width=2/ +line x1=24.25 y1=19 x2=30.75 y2=19 stroke=stroke_color; stroke-width=2.5 stroke-linecap=round/ +circle cx=27.5 cy=20.5 r=1.25 fill=stroke_color; stroke=none stroke-width=2/ +/svg -- 1.6.0.4 ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] MANIFEST pointlessness
On Mon, 2010-07-05 at 15:13 +0200, Tomeu Vizoso wrote: Would be good to have some way of tracking what needs to be solved before we can do the switch. Two questions off the top of my head are how first-time activity authors are going to package their software Now they do: ./setup.py fix_manifest ./setup.py dist_xo When we remove the manifest thing, they could do: ./setup.py dist_xo None of the activities I have seen so far store a manually generated MANIFEST in git. In case they'd need to exclude some files, hopefully setup.py would provide extensibility hooks. Actually, I've never seen a setup.py containing anything else than the two lines of copy-pasted code (plus the usual dozen lines of legal crap). Activities which need to build binaries, such as Physics, require a manual build step. This confirms my theory that the xo bundle format is so easy to use just because it doesn't do any of the things that a packaging system needs to do. Currently, setup.py is just a glorified zip -r. and also what those people will have to do to modify an activity installed in their system. (But maybe not discuss these in this thread?). Currently, upgrading an activity causes the new files to be merged with the old ones. This is clearly a bug that should be fixed. The MANIFEST doesn't play any role in activity installation. The version of Sugar shipped in F11-0.88 contains a patch that removes all this code and it is functionally identical from the user's point of view, apart from not logging warnings when the MANIFEST contains inconsistencies. -- // Bernie Innocenti - http://codewiz.org/ \X/ Sugar Labs - http://sugarlabs.org/ ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Activity title update
On Sun, Jul 4, 2010 at 17:08, Bernie Innocenti ber...@codewiz.org wrote: El Fri, 02-07-2010 a las 08:36 +, Sascha Silbe escribió: FWIW, in an unrelated project I'm using the changed signal and have not noticed any missing updates so far (except for SpinButtons which should have been fixed in GTK by now IIRC). The changed signal fires on each keystroke. We can't use this event directly, because updating metadata in the datastore causes a ~1sec freeze on slow hardware such as the XO-1. Also, probably due to event ping-pong, the cursor gets reset to the beginning of the edit-field and all interim keystrokes end up in the wrong position. The old approach was to listen for changed with a 1-second timeout to trigger the metadata update. The timeout logic was also buggy, which is what the patch applied in git fixes. I can't believe GTK wouldn't offer a reliable way to tell when a user has finished changing an edit widget. Guess I'm a bit late, but I think I have heard this before and I'm not sure what means to has finished changing an edit widget. Do you know of another GUI toolkit with such a signal? Thanks, Tomeu -- // Bernie Innocenti - http://codewiz.org/ \X/ Sugar Labs - http://sugarlabs.org/ ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] MANIFEST pointlessness
On Mon, Jul 5, 2010 at 16:04, Bernie Innocenti ber...@codewiz.org wrote: On Mon, 2010-07-05 at 15:13 +0200, Tomeu Vizoso wrote: Would be good to have some way of tracking what needs to be solved before we can do the switch. Two questions off the top of my head are how first-time activity authors are going to package their software Now they do: ./setup.py fix_manifest ./setup.py dist_xo When we remove the manifest thing, they could do: ./setup.py dist_xo None of the activities I have seen so far store a manually generated MANIFEST in git. In case they'd need to exclude some files, hopefully setup.py would provide extensibility hooks. Actually, I've never seen a setup.py containing anything else than the two lines of copy-pasted code (plus the usual dozen lines of legal crap). Activities which need to build binaries, such as Physics, require a manual build step. This confirms my theory that the xo bundle format is so easy to use just because it doesn't do any of the things that a packaging system needs to do. Currently, setup.py is just a glorified zip -r. and also what those people will have to do to modify an activity installed in their system. (But maybe not discuss these in this thread?). Currently, upgrading an activity causes the new files to be merged with the old ones. This is clearly a bug that should be fixed. The MANIFEST doesn't play any role in activity installation. The version of Sugar shipped in F11-0.88 contains a patch that removes all this code and it is functionally identical from the user's point of view, apart from not logging warnings when the MANIFEST contains inconsistencies. Sorry about the confusion, these questions were about the move from xo bundles to packages :( Regards, Tomeu -- // Bernie Innocenti - http://codewiz.org/ \X/ Sugar Labs - http://sugarlabs.org/ ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Clocks on XOs
On 3 July 2010 16:52, Bernie Innocenti ber...@codewiz.org wrote: I checked: olpc-update-query only sets the clock if it's off by more than 24hours, so it cannot serve as a replacement for ntpdate. What's the requirement for super-accurate clocks on the XO? Daniel ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [PATCH] School Server Icon
This is the icon provided by Gary Martin. On Mon, 5 Jul 2010 10:00:10 -0400, Martin Abente mabe...@paraguayeduca.org wrote: --- icons/scalable/device/Makefile.am |1 + icons/scalable/device/school-server.svg | 20 2 files changed, 21 insertions(+), 0 deletions(-) create mode 100644 icons/scalable/device/school-server.svg diff --git a/icons/scalable/device/Makefile.am b/icons/scalable/device/Makefile.am index 28818ab..198488b 100644 --- a/icons/scalable/device/Makefile.am +++ b/icons/scalable/device/Makefile.am @@ -77,6 +77,7 @@ icon_DATA = \ network-wireless-connected-100.svg \ network-wireless-connected-100.icon \ printer.svg \ + school-server.svg \ speaker-000.svg \ speaker-033.svg \ speaker-066.svg \ diff --git a/icons/scalable/device/school-server.svg b/icons/scalable/device/school-server.svg new file mode 100644 index 000..28a854d --- /dev/null +++ b/icons/scalable/device/school-server.svg @@ -0,0 +1,20 @@ +?xml version=1.0 encoding=UTF-8? +!DOCTYPE svg PUBLIC -//W3C//DTD SVG 1.1//EN http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd; [ + !ENTITY fill_color #FF + !ENTITY stroke_color #010101 +] +svg xmlns=http://www.w3.org/2000/svg; width=55 height=55 +path d=M10,50 S8,50 8,48 L8,35 L27.5,24 L47,35 L47,48 S47,50 45,50 z fill=fill_color; stroke=stroke_color; stroke-width=1.75/ +path d=M33,50 L33,39 S33,37 31,37 L24,37 S22,37 22,39 L22,50 z fill=stroke_color; stroke=stroke_color; stroke-width=1.75/ +path d=M20,23.5 L20,10 L35,10 L35,23.5 z fill=stroke_color; stroke=stroke_color; stroke-width=3.5/ +line x1=6 y1=33.5 x2=27.5 y2=21.5 stroke=stroke_color; stroke-width=6.75 stroke-linecap=round/ +line x1=49 y1=33.5 x2=27.5 y2=21.5 stroke=stroke_color; stroke-width=6.75 stroke-linecap=round/ +line x1=6 y1=33.5 x2=27.5 y2=21.5 stroke=fill_color; stroke-width=3.5 stroke-linecap=round/ +line x1=49 y1=33.5 x2=27.5 y2=21.5 stroke=fill_color; stroke-width=3.5 stroke-linecap=round/ +path d=M20,23.7 L20,10 L35,10 L35,23.7 z fill=fill_color; stroke=none stroke-width=2/ +path d=M18,9.25 L27.5,3.25 L37,9.25 z fill=stroke_color; stroke=stroke_color; stroke-width=2.25/ +circle cx=27.5 cy=15.5 r=3.25 fill=stroke_color; stroke=none stroke-width=2/ +path d=M24.25,19.25 L24.25,15.5 L30.75,15.5 L30.75,19.25 z fill=stroke_color; stroke=none stroke-width=2/ +line x1=24.25 y1=19 x2=30.75 y2=19 stroke=stroke_color; stroke-width=2.5 stroke-linecap=round/ +circle cx=27.5 cy=20.5 r=1.25 fill=stroke_color; stroke=none stroke-width=2/ +/svg ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Activity title update
On Mon, Jul 5, 2010 at 15:38, Bernie Innocenti ber...@codewiz.org wrote: On Mon, 2010-07-05 at 15:04 +0200, Tomeu Vizoso wrote: Guess I'm a bit late, but I think I have heard this before and I'm not sure what means to has finished changing an edit widget. Do you know of another GUI toolkit with such a signal? http://doc.qt.nokia.com/4.6/qlineedit.html#editingFinished It's a pretty useful and obvious feature, even Microsoft's MFC did let me do it easily in 1992. But you can do the same with Gtk+, cannot you? This signal is emitted when the Return or Enter key is pressed or the line edit loses focus Regards, Tomeu -- // Bernie Innocenti - http://codewiz.org/ \X/ Sugar Labs - http://sugarlabs.org/ ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] DESIGN
Hello, I am using Sugar on Debian and wanted to report a feature not functioning as desired. Issue: In the control panel, even though I do not have a 3g modem, there is an icon to set up a 3g modem. Steps to reproduce: 1) Start Sugar. 2) Select My Settings. 3) Select Modem Configuration. What I expected: Sugar should do hardware detection prior to showing icon. Regards, Kandarp. ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Activity title update
On Mon, 2010-07-05 at 16:37 +0200, Tomeu Vizoso wrote: But you can do the same with Gtk+, cannot you? This signal is emitted when the Return or Enter key is pressed or the line edit loses focus No, GTK does not seem to send a focus-out-event when the user clicks on the close button. Which is what makes my approach fail. -- // Bernie Innocenti - http://codewiz.org/ \X/ Sugar Labs - http://sugarlabs.org/ ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [PATCH] Add cpu and memory resource indicator to frame
Some comments on top of Sascha's excellent review:
On Fri, Jul 2, 2010 at 12:03, Sascha Silbe
sascha-ml-ui-sugar-de...@silbe.org wrote:
Excerpts from anishmangal2002's message of Thu Jul 01 15:03:16 + 2010:
+ gobject.timeout_add(1000, self.__timer_cb)
The use of polling (i.e. a timer) seems to be necessary as /proc/stat
doesn't provide a better interface. (I.e. the way you do it is fine)
We should check timeout_add() doesn't add a reference to self. I used
to at least some time ago and it would mean leaking palette instances
and timers. It may have been made to use weak references in the mean
time, not sure.
If that's the case, we need to remove the timer on popdown and create
a new one on popup.
+ def __getTimeList(self):
Single underscore should suffice. No one is going to subclass this, and
even then it's not something that needs protection from being overridden
in the subclass.
We use lowercase_separated_by_underscores for method names, see
Naming Conventions in http://www.python.org/dev/peps/pep-0008/ .
+ statFile = file(/proc/stat, r)
Tomeu will nag you about vs. '' ;)
(always use '' unless there's an apostrophe in the string itself).
Also, lowercase_separated_by_underscores for all variables.
+ def __popup_cb(self, gobject):
+ self.updating = True
+
+ def __popdown_cb(self, gobject):
+ self.updating = False
Please stop the timer instead. Calculating values while we don't need them
wastes energy and will prevent the CPU from entering low-power states (i.e.
auto-suspend) in the future.
True, but keep in mind the potential issue with leaking Palette instances.
+ name, value, unit = line.split()[:3]
+ if 'MemTotal:' == name: total = value
+ elif 'MemFree:' == name: free = value
+ elif 'Buffers:' == name: buffers = value
+ elif 'Cached:' == name: cached = value
+ elif 'Active:' == name: break
We try to avoid multiple statements in a single line, so lines need to be split.
A few more nitpicks:
+from gettext import gettext as _
+
+import logging
logging should be grouped with gettext because both are in the python
std library.
+from sugar.graphics import style
+from jarabe.frame.frameinvoker import FrameWidgetInvoker
The jarabe import should be in a separate group because is part of the
currently running program.
+icon_name = 'computer'
You may want to make it a constant similar to FRAME_POSITION_RELATIVE.
+self.updating = False
+self.proc_stat_old = []
+self.proc_stat_new = []
This should be private (leading underscore).
+self._cpu_text.set_label('CPU free: %d%%' % cpu_free)
We don't want to translate this string?
+vbox.pack_start(self._cpu_text, padding=10)
The padding shouldn't be hardcoded to a number of pixels, see the
constants in style.py.
+statFile.close()
Maybe use try..finally?
+for i in range(len(timeList)) :
for i in time_list: doesn't work?
+return 100 * (int(free)+int(buffers)+int(cached)) / int(total)
Operators should have spaces around.
Great patch!
Thanks,
Tomeu
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] [ANNOUNCE] 0.90 Schedule
Hi, I have filled out the missing pieces in the 0.90 development cycle [1]. I want to highlight some important dates: * 16th August: Feature Freeze * 29th September: The final, new stable release * 27th October: Bug fix release The Sugar 0.89.x schedule is based on the GNOME 2.31.x development cycle [2]. One reason is that Sugar as based on the GNOME stack. Another reason is that GNOME makes sure that their release will be able to be shipped by the major distributions. The dates of the release are made accordingly to the GNOME schedule. As there is not much time left for 0.90 we will focus on stabilizing and landing the features that were left over from the last release. As well as landing all the patches that are currently pending. New Features are handled by the Feature process, as it has been the case in the past [3]. Make sure to read the policy what needs to be done in order that your feature can be accepted [4]. It is a sanity check, so don't be too worried. It is there to help the submitter and maintainer in order to get input from the community and the release manager to track features and their status. The actual code gets reviewed by the module maintainer. If there are any questions, about the policy or the schedule, please ask. Regards, Simon [1] http://wiki.sugarlabs.org/go/0.90/Roadmap#Schedule [2] http://live.gnome.org/TwoPointThirtyone [3] http://wiki.sugarlabs.org/go/Features/Policy [4] http://wiki.sugarlabs.org/go/Features/Policy#Acceptance_of_a_feature ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] [PATCH] Delete activities profile data when uninstall
When activities are uninstalled, profile data persists. Profile data can consume a lot of disk space. Therefore it is necessary to delete it. More info http://bugs.sugarlabs.org/ticket/2074 --- src/sugar/bundle/activitybundle.py |5 + 1 files changed, 5 insertions(+), 0 deletions(-) diff --git a/src/sugar/bundle/activitybundle.py b/src/sugar/bundle/activitybundle.py index c83257f..51ad312 100644 --- a/src/sugar/bundle/activitybundle.py +++ b/src/sugar/bundle/activitybundle.py @@ -25,6 +25,7 @@ import locale import os import tempfile import logging +import shutil from sugar import env from sugar import util @@ -417,6 +418,10 @@ class ActivityBundle(Bundle): os.readlink(path).startswith(install_path): os.remove(path) +profile_path = env.get_profile_path(self._bundle_id) +if os.path.exists(profile_path): +shutil.rmtree(profile_path) + self._uninstall(install_path) def is_user_activity(self): -- 1.6.0.4 ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [PATCH] Delete activities profile data when uninstall
Excerpts from Martin Abente's message of Mon Jul 05 19:49:10 + 2010: [src/sugar/bundle/activitybundle.py] @@ -417,6 +418,10 @@ class ActivityBundle(Bundle): [...] +profile_path = env.get_profile_path(self._bundle_id) +if os.path.exists(profile_path): +shutil.rmtree(profile_path) + If the directory contains files or directories with mode 000 (unusual, but possible) or owned by somebody else (e.g. root) this will fail and throw an exception. You could use os.walk() + os.chmod() to handle the 000 case, though I'm unsure whether it's worth the effort. Passing ignore_errors=True to rmtree() to avoid breaking for any other reason is certainly a good idea, though. If you want to check whether rmtree() was successful, just check if profile_path still exists. Sascha -- http://sascha.silbe.org/ http://www.infra-silbe.de/ signature.asc Description: PGP signature ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [PATCH] Delete activities profile data when uninstall
On Mon, 2010-07-05 at 20:19 +, Sascha Silbe wrote: Excerpts from Martin Abente's message of Mon Jul 05 19:49:10 + 2010: [src/sugar/bundle/activitybundle.py] @@ -417,6 +418,10 @@ class ActivityBundle(Bundle): [...] +profile_path = env.get_profile_path(self._bundle_id) +if os.path.exists(profile_path): +shutil.rmtree(profile_path) + If the directory contains files or directories with mode 000 (unusual, but possible) or owned by somebody else (e.g. root) this will fail and throw an exception. You could use os.walk() + os.chmod() to handle the 000 case, though I'm unsure whether it's worth the effort. Passing ignore_errors=True to rmtree() to avoid breaking for any other reason is certainly a good idea, though. If you want to check whether rmtree() was successful, just check if profile_path still exists. Thanks for help, I also found a problem with the patch... When the user upgrades an activity, the older version is being deleted, and so will be the profile data. That could be a problem. Any suggestions? I was thinking about a new parameter to the uninstall ActivityBundle method, something like: reason_upgrade=False. So it wont delete the profile directory in those cases. This solution will also require to add the same parameter to bundleregistry uninstall method. Sascha -- http://sascha.silbe.org/ http://www.infra-silbe.de/ ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] [design] hiding icons in control panel for hardware which is not installed.
I just wanted to raise this to the attention of the design team for the July 10th meeting. I would like to clarify that icons for hardware that is not present can be hidden. david On Mon, Jul 5, 2010 at 9:38 AM, Kandarp Kaushik kand...@seeta.in wrote: Hello, I am using Sugar on Debian and wanted to report a feature not functioning as desired. Issue: In the control panel, even though I do not have a 3g modem, there is an icon to set up a 3g modem. Steps to reproduce: 1) Start Sugar. 2) Select My Settings. 3) Select Modem Configuration. What I expected: Sugar should do hardware detection prior to showing icon. Regards, Kandarp. ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Activity list favorites view
On 6 Jul 2010, at 04:09, Bernie Innocenti wrote: On Tue, 2010-07-06 at 12:47 +1000, James Cameron wrote: The only time I've seen this is when the keyboard F2 key was pressed instead of F3. ;-} lol, that could *really* be the correct explanation! Activity ring in list view with a bogus search string also looks remarkably empty. It persists when switching between views. I wonder why we don't allow searching in the favourites view as well. Long outstanding bug I burnt my fingers years back in my youth :-) Feedback was that folks were accidentally hitting keys while in fav. view. This would trigger an unexpected search and user would be lost – usually left looking at an empty results screen. Problem was that the home fav. view search field was focused by default and any key presses landed there. The canvas could not accept focus, so I submitted a (hack) patch to defocus the search field by focusing on the ring toolbar button instead. Cscott didn't like it (fair enough), and went to watch some baseball match instead – I slagged him off for it – release went ahead without any patch ;-b Search in home fav view has been disabled ever since. I'm sure there are better solutions, probably a drop down menu with all results would be best (rather than switching the user to list view), but switching to list view would be a fair start if the search input text field can be defocused in a friendly was so as not to catch stray key strokes. Regards, --Gary -- // Bernie Innocenti - http://codewiz.org/ \X/ Sugar Labs - http://sugarlabs.org/ ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] [ASLO] Release Write-70
Activity Homepage: http://activities.sugarlabs.org/addon/4201 Sugar Platform: 0.86 - 0.88 Download Now: http://activities.sugarlabs.org/downloads/file/26971/write-70.xo Release notes: Fix startup zoom level (sl#1121) Sugar Labs Activities http://activities.sugarlabs.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] [ASLO] Release Distance-19
Activity Homepage: http://activities.sugarlabs.org/addon/4264 Sugar Platform: 0.82 - 0.88 Download Now: http://activities.sugarlabs.org/downloads/file/26970/distance-19.xo Release notes: Fixes a bug on Sugar 0.84 (and possibly later versions too): http://dev.laptop.org/ticket/10199. Sugar Labs Activities http://activities.sugarlabs.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
