anaconda-patches September 2013

anaconda-patches@lists.fedorahosted.org

17 participants
168 discussions

[PATCH rhel7-branch] network, ports of patches to r7
by Radek Vykydal 25 Sep '13

25 Sep '13

All of the patches except for [3/9] are ports from master so it is the [3/9] that asks for review the most. https://bugzilla.redhat.com/show_bug.cgi?id=1011826 [1/9], [2/9], [3/9] https://bugzilla.redhat.com/show_bug.cgi?id=1011841 [4/9] https://bugzilla.redhat.com/show_bug.cgi?id=1011855 [5/9] https://bugzilla.redhat.com/show_bug.cgi?id=1011860 [6/9] https://bugzilla.redhat.com/show_bug.cgi?id=1011866 [7/9] https://bugzilla.redhat.com/show_bug.cgi?id=1011928 [8/9], [9/9]

2 10

[master][PATCH] Escape the status before setting it as markup
by Vratislav Podzimek 24 Sep '13

24 Sep '13

The spoke's status may contain characters that need to be escaped before used as markup. Signed-off-by: Vratislav Podzimek <vpodzime(a)redhat.com> --- pyanaconda/ui/gui/hubs/__init__.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/pyanaconda/ui/gui/hubs/__init__.py b/pyanaconda/ui/gui/hubs/__init__.py index 3042b51..f142a18 100644 --- a/pyanaconda/ui/gui/hubs/__init__.py +++ b/pyanaconda/ui/gui/hubs/__init__.py @@ -257,7 +257,7 @@ class Hub(GUIObject, common.Hub): def _updateCompleteness(self, spoke): spoke.selector.set_sensitive(spoke.ready) spoke.selector.set_property("status", spoke.status) - spoke.selector.set_tooltip_markup(spoke.status) + spoke.selector.set_tooltip_markup(GLib.markup_escape_text(spoke.status)) spoke.selector.set_incomplete(not spoke.completed and spoke.mandatory) self._handleCompleteness(spoke) -- 1.7.11.7

2 1

[master][PATCH] Make the keyboard layout preview dialog bigger (#1011140)
by Vratislav Podzimek 24 Sep '13

24 Sep '13

Otherwise it may be hard to read some characters. Signed-off-by: Vratislav Podzimek <vpodzime(a)redhat.com> --- pyanaconda/ui/gui/spokes/keyboard.py | 1 + 1 file changed, 1 insertion(+) diff --git a/pyanaconda/ui/gui/spokes/keyboard.py b/pyanaconda/ui/gui/spokes/keyboard.py index ffab6da..d35eb2f 100644 --- a/pyanaconda/ui/gui/spokes/keyboard.py +++ b/pyanaconda/ui/gui/spokes/keyboard.py @@ -501,6 +501,7 @@ class KeyboardSpoke(NormalSpoke): dialog = Gkbd.KeyboardDrawing.dialog_new() Gkbd.KeyboardDrawing.dialog_set_layout(dialog, self._xkl_wrapper.configreg, layout_row[0]) + dialog.set_size_request(750, 350) with enlightbox(self.window, dialog): dialog.show_all() dialog.run() -- 1.7.11.7

2 3

[master][PATCH] Return switching options with the same order as shown (#1011130)
by Vratislav Podzimek 24 Sep '13

24 Sep '13

The switching options dialog lists all available switching options ordered alphabetically. When returning checked options it should keep the same ordering not to confuse users. Signed-off-by: Vratislav Podzimek <vpodzime(a)redhat.com> --- pyanaconda/ui/gui/spokes/keyboard.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/pyanaconda/ui/gui/spokes/keyboard.py b/pyanaconda/ui/gui/spokes/keyboard.py index d35eb2f..a943d1f 100644 --- a/pyanaconda/ui/gui/spokes/keyboard.py +++ b/pyanaconda/ui/gui/spokes/keyboard.py @@ -238,7 +238,7 @@ class ConfigureSwitchingDialog(GUIObject): def checked_options(self): """Property returning all checked options from the list""" - ret = [row[0] for row in self._switchingOptsStore if row[1] and row[0]] + ret = [row[0] for row in self._switchingOptsSort if row[1] and row[0]] return ret def on_use_option_toggled(self, renderer, path, *args): -- 1.7.11.7

2 1

[initial-setup] EULA agreement
by Vratislav Podzimek 24 Sep '13

24 Sep '13

These patches add EULA agreement support to our tools. Patches for the Initial Setup fix issues that prevent adding spokes to it and introduce the EULA Agreement spoke. The patches for pykickstart and anaconda add and export the eula kickstart command. -- 1.7.11.7

2 13

[PATCH] Add translation support to check_accelerators
by David Shea 24 Sep '13

24 Sep '13

Added a --translate option to check_accelerators.py. When set, it will check all of the translated keyboard accelerators in addition to the English ones. Changed the exception check so that language information can be included as . Added support for the new parameter to the run_check_accelerators.sh wrapper script. `make check' will not check translations. Requires python-polib to parse the .po files. --- tests/accelerators/check_accelerators.py | 127 +++++++++++++++++++++++---- tests/accelerators/run_check_accelerators.sh | 18 +++- 2 files changed, 126 insertions(+), 19 deletions(-) diff --git a/tests/accelerators/check_accelerators.py b/tests/accelerators/check_accelerators.py index 892ff2b..461220a 100755 --- a/tests/accelerators/check_accelerators.py +++ b/tests/accelerators/check_accelerators.py @@ -22,6 +22,8 @@ import argparse import re import os.path import copy +import collections +import locale try: from lxml import etree @@ -30,33 +32,80 @@ except ImportError: sys.exit(1) accel_re = re.compile(r'_(?P<accel>.)') - success = True -def add_check_accel(glade_filename, accels, label): + +# Only used when --translate is requested. +class PODict(collections.Mapping): + def __init__(self, filename): + try: + import polib + except ImportError: + print("You need to install the python-polib package to check translations") + sys.exit(1) + + self._dict = {} + + pofile = polib.pofile(filename) + self.metadata = pofile.metadata + for entry in pofile.translated_entries(): + self._dict[entry.msgid] = entry.msgstr + + def __getitem__(self, key): + return self._dict[key] + + def __iter__(self): + return self._dict.__iter__() + + def __len__(self): + return len(self._dict) + +def is_exception(node, language=None): + if language: + comment_str = "check_accelerators(%s)" % language + else: + comment_str = "check_accelerators" + + return bool(node.xpath("./parent::*/comment()[contains(., '%s')]" % comment_str)) + +def add_check_accel(glade_filename, accels, label, po_map): """Check whether an accelerator conflicts with existing accelerators. and add it to the current accelerator context. """ global success + language = None + if po_map: + if label.text not in po_map: + return + label.text = po_map[label.text] + language = po_map.metadata['Language'] + match = accel_re.search(label.text) if match: accel = match.group('accel').lower() if accel in accels: # Check for an exception comment - prev = label.getprevious() - if (prev is not None) and (prev.tag == etree.Comment) and \ - prev.text.strip().startswith('check_accelerators:'): + if is_exception(label, language): return - print("Accelerator collision for key %s in %s\n line %d: %s\n line %d: %s" %\ - (accel, os.path.normpath(glade_filename), - accels[accel].sourceline, accels[accel].text, - label.sourceline, label.text)) + if language: + lang_str = " for language %s" % language + else: + lang_str = "" + + try: + print(("Accelerator collision for key %s in %s%s\n line %d: %s\n line %d: %s" %\ + (accel, os.path.normpath(glade_filename), lang_str, + accels[accel].sourceline, accels[accel].text, + label.sourceline, label.text)).encode("utf-8")) + except UnicodeEncodeError: + import pdb; pdb.set_trace() + raise success = False else: accels[accel] = label -def combine_accels(glade_filename, list_a, list_b): +def combine_accels(glade_filename, list_a, list_b, po_map): if not list_a: return list_b if not list_b: @@ -67,7 +116,7 @@ def combine_accels(glade_filename, list_a, list_b): for accels_b in list_b: new_accels = copy.copy(accels_a) for accel in accels_b.keys(): - add_check_accel(glade_filename, new_accels, accels_b[accel]) + add_check_accel(glade_filename, new_accels, accels_b[accel], po_map) newlist.append(new_accels) return newlist @@ -80,7 +129,7 @@ def combine_accels(glade_filename, list_a, list_b): # these is compared against the list of accelerators returned for the object's # other GtkNotebook children. -def process_object(glade_filename, interface_object): +def process_object(glade_filename, interface_object, po_map): """Process keyboard shortcuts for a given glade object. The return value from this function is a list of accelerator @@ -94,12 +143,12 @@ def process_object(glade_filename, interface_object): # Add everything that isn't a child of a GtkNotebook for label in interface_object.xpath(".//property[@name='label' and ../property[@name='use_underline']/text() = 'True' and not(ancestor::object[@class='GtkNotebook'])]"): - add_check_accel(glade_filename, accels[0], label) + add_check_accel(glade_filename, accels[0], label, po_map) # For each GtkNotebook tab that is not a child of another notebook, # add the tab to the top-level context for notebook_label in interface_object.xpath(".//object[@class='GtkNotebook' and not(ancestor::object[@class='GtkNotebook'])]/child[@type='tab']//property[@name='label' and ../property[@name='use_underline']/text() = 'True']"): - add_check_accel(glade_filename, accels[0], notebook_label) + add_check_accel(glade_filename, accels[0], notebook_label, po_map) # Now process each non-tab object in each Gtknotebook that is not a child # of another notebook. For each Gtk notebook, each non-tab child represents @@ -129,34 +178,76 @@ def process_object(glade_filename, interface_object): # Create the list of dictionaries for the notebook notebook_list = [] for child in notebook.xpath("./child[not(@type='tab')]"): - notebook_list.extend(process_object(glade_filename, copy.deepcopy(child))) + notebook_list.extend(process_object(glade_filename, copy.deepcopy(child), po_map)) # Now combine this with our list of accelerators - accels = combine_accels(glade_filename, accels, notebook_list) + accels = combine_accels(glade_filename, accels, notebook_list, po_map) return accels -def check_glade(glade_filename): +def check_glade(glade_filename, po_map=None): with open(glade_filename) as glade_file: # Parse the XML glade_tree = etree.parse(glade_file) # Treat each top-level object as a separate context for interface_object in glade_tree.xpath("/interface/object"): - process_object(glade_filename, interface_object) + process_object(glade_filename, interface_object, po_map) def main(argv=None): if argv is None: argv = sys.argv parser = argparse.ArgumentParser("Check for duplicated accelerators") + parser.add_argument("-t", "--translate", action='store_true', + help="Check translated strings") + parser.add_argument("-p", "--podir", action='store', type=str, + metavar='PODIR', help='Directory containing po files', default='./po') parser.add_argument("glade_files", nargs="+", metavar="GLADE-FILE", help='The glade file to check') args = parser.parse_args(args=argv) + # First check the untranslated strings in each file for glade_file in args.glade_files: check_glade(glade_file) + # Now loop over all of the translations + if args.translate: + import langtable + + with open(os.path.join(args.podir, 'LINGUAS')) as linguas: + for line in linguas.readlines(): + if re.match(r'^#', line): + continue + + for lang in line.strip().split(" "): + # Reset the locale to C before parsing the po file because + # polib has erroneous uses of lower(). + # See https://bitbucket.org/izi/polib/issue/54/pofile-parsing-crashes-in-turkish-… + locale.setlocale(locale.LC_ALL, 'C') + po_map = PODict(os.path.join(args.podir, lang + ".po")) + + # Set the locale so that we can use lower() on accelerator keys. + # If the language is of the form xx_XX, use that as the + # locale name. Otherwise use the first locale that + # langtable returns for the language. If that doesn't work, + # just use C and hope for the best. + if '_' in lang: + locale.setlocale(locale.LC_ALL, lang) + else: + locale_list = langtable.list_locales(languageId=lang) + if locale_list: + try: + locale.setlocale(locale.LC_ALL, locale_list[0]) + except locale.Error: + print("No such locale %s, using C" % locale_list[0]) + locale.setlocale(locale.LC_ALL, 'C') + else: + locale.setlocale(locale.LC_ALL, 'C') + + for glade_file in args.glade_files: + check_glade(glade_file, po_map) + if __name__ == "__main__": main(sys.argv[1:]) diff --git a/tests/accelerators/run_check_accelerators.sh b/tests/accelerators/run_check_accelerators.sh index a49cda1..401f081 100755 --- a/tests/accelerators/run_check_accelerators.sh +++ b/tests/accelerators/run_check_accelerators.sh @@ -3,4 +3,20 @@ : "${top_srcdir:=$(dirname "$0")/../..}" srcdir="${top_srcdir}/tests/accelerators" -find "${top_srcdir}" -name '*.glade' -exec "${srcdir}/check_accelerators.py" {} + +# If --translate was specified but not --podir, add --podir +i=0 +translate_set=0 +podir_set=0 +for arg in "$@" ; do + if [ "$arg" = "--translate" -o "$arg" = "-t" ]; then + translate_set=1 + elif [ "$arg" = "--podir" -o "$arg" = "-p" ]; then + podir_set=1 + fi +done + +if [ "$translate_set" -eq 1 -a "$podir_set" -eq 0 ]; then + set -- "$@" --podir "${top_srcdir}/po" +fi + +find "${top_srcdir}" -name '*.glade' -exec "${srcdir}/check_accelerators.py" "$@" {} + -- 1.8.3.1

2 2

[PATCH] Don't rely on X server adding empty variant for its defaults (#1011658)
by Vratislav Podzimek 24 Sep '13

24 Sep '13

The X server keeps "matching" lists of layouts and variants that, when combined together, give a list of what is actually set for input handling. However, for the default layout, the X server doesn't add the empty "" variant to make the lists actually matching. Signed-off-by: Vratislav Podzimek <vpodzime(a)redhat.com> --- pyanaconda/keyboard.py | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/pyanaconda/keyboard.py b/pyanaconda/keyboard.py index e86771a..48a2789 100644 --- a/pyanaconda/keyboard.py +++ b/pyanaconda/keyboard.py @@ -514,7 +514,11 @@ class XklWrapper(object): cur_group = num_groups - 1 layout = self._rec.layouts[cur_group] - variant = self._rec.variants[cur_group] + try: + variant = self._rec.variants[cur_group] + except IndexError: + # X server may have forgotten to add the "" variant for its default layout + variant = "" self._engine.stop_listen(Xkl.EngineListenModes.TRACK_KEYBOARD_STATE) -- 1.7.11.7

2 1

[master][PATCH] Make sure autopart type is handled deterministicaly in text mode (#1010453)
by Vratislav Podzimek 24 Sep '13

24 Sep '13

We should use OrderedDict instead of plain dict if we want to rely on the ordering of its items. Also don't subtract 1 when getting the autopart type, we subtract 1 in the input method and also the default value is set to reflect that. Signed-off-by: Vratislav Podzimek <vpodzime(a)redhat.com> --- pyanaconda/ui/tui/spokes/storage.py | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/pyanaconda/ui/tui/spokes/storage.py b/pyanaconda/ui/tui/spokes/storage.py index 66b58a8..714c590 100644 --- a/pyanaconda/ui/tui/spokes/storage.py +++ b/pyanaconda/ui/tui/spokes/storage.py @@ -38,6 +38,8 @@ from pyanaconda.i18n import _, P_ from pykickstart.constants import CLEARPART_TYPE_ALL, CLEARPART_TYPE_LINUX, CLEARPART_TYPE_NONE +from collections import OrderedDict + import logging log = logging.getLogger("anaconda") @@ -377,8 +379,8 @@ class PartitionSchemeSpoke(NormalTUISpoke): def __init__(self, app, data, storage, payload, instclass): NormalTUISpoke.__init__(self, app, data, storage, payload, instclass) - self.partschemes = {"Standard Partition": AUTOPART_TYPE_PLAIN, - "BTRFS": AUTOPART_TYPE_BTRFS, "LVM": AUTOPART_TYPE_LVM} + self.partschemes = OrderedDict([("Standard Partition", AUTOPART_TYPE_PLAIN), + ("LVM", AUTOPART_TYPE_LVM), ("BTRFS", AUTOPART_TYPE_BTRFS)]) @property def indirect(self): @@ -420,7 +422,7 @@ class PartitionSchemeSpoke(NormalTUISpoke): schemelist = self.partschemes.values() try: - self.data.autopart.type = schemelist[self._selection - 1] + self.data.autopart.type = schemelist[self._selection] except IndexError: # we shouldn't ever see this, but just in case, don't crash. # when autopart.type is detected as None in AutoPartSpoke.apply(), -- 1.7.11.7

2 1

[master][PATCH] Make Keyboard spoke's status consistent with other statuses (#1011166)
by Vratislav Podzimek 24 Sep '13

24 Sep '13

When multiple keyboard layouts are configured, we should list them in the status as we do with the other spokes. Signed-off-by: Vratislav Podzimek <vpodzime(a)redhat.com> --- pyanaconda/ui/gui/spokes/keyboard.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/pyanaconda/ui/gui/spokes/keyboard.py b/pyanaconda/ui/gui/spokes/keyboard.py index 7c575d8..ffab6da 100644 --- a/pyanaconda/ui/gui/spokes/keyboard.py +++ b/pyanaconda/ui/gui/spokes/keyboard.py @@ -298,7 +298,9 @@ class KeyboardSpoke(NormalSpoke): @property def status(self): # We don't need to check that self._store is empty, because that isn't allowed. - return self._xkl_wrapper.get_layout_variant_description(self._store[0][0]) + descriptions = (self._xkl_wrapper.get_layout_variant_description(row[0]) + for row in self._store) + return ", ".join(descriptions) def initialize(self): NormalSpoke.initialize(self) -- 1.7.11.7

2 1

Re: [PATCH] support ppc64le architecture
by Mark Hamzy 24 Sep '13

24 Sep '13

vpodzime(a)redhat.com wrote on 09/24/2013 01:12:35 PM: > This looks good to me. However, is it really all we need to support > ppc64le? As far as I know at this time, yes. The ppc64le architecture is still being developed. -- Mark You must be the change you wish to see in the world. -- Mahatma Gandhi Never let the future disturb you. You will meet it, if you have to, with the same weapons of reason which today arm you against the present. -- Marcus Aurelius

1 0

← Newer
1
2
3
4
5
6
7
...
17
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

anaconda-patches September 2013