def _reorder_following_line(self, reorder_lines):
# Reordering from a line is a bit more complex than it seems. We have to find intersection
# points from all lines in all rects (when there's an interestion, of course). Then, for
# each intersection, we compute the distance of it from the origin of the order arrow.
# We sort our elements by that distance and we have our new order!
neworder = []
for reorder_line in reorder_lines:
neworder += self._get_intersections(reorder_line)
neworder = dedupe(neworder)
if len(neworder) < 2:
return # nothing to reorder
# ok, we have our new order. That was easy huh? Now, what we have to do is to insert that
# new order in the rest of the elements, which might not all be in our new order. That's
# a bit complex too because when drawing arrows, a naive re-ordering is counter intuitive.
# I won't explain why what is done below is done thus because it's hard to describe, but
# here's what we do: We compute the min and max of old orders in our elems. Whatever came
# before min stays before. Whatever came after max stays *right* after, but not at the end.
# The end of the order is constituted of what is between min and max, but isn't in the new
# order. Arrow-wise, these elements have been "orphaned" by the new link and the most
# intuitive thing to do with them is to put them at the end of the order.
minorder = min(e.order for e in neworder)
maxorder = max(e.order for e in neworder)
all_elems = list(self._ordered_elems())
affected_elems = set(neworder)
unaffected_elems = [e for e in all_elems if e not in affected_elems]
before = [e for e in unaffected_elems if e.order < minorder]
after = [e for e in unaffected_elems if e.order > maxorder]
inbetween = [e for e in unaffected_elems if minorder < e.order < maxorder]
# we also add ignore elems to our big concat so that their order value doesn't conflict
ignored = list(self._ignored_elems())
concat = before + neworder + after + inbetween + ignored
for i, elem in enumerate(concat):
elem.order = i
def get_dupe_groups(self, files, j=job.nulljob):
j = j.start_subjob([8, 2])
for f in (f for f in files if not hasattr(f, 'is_ref')):
f.is_ref = False
files = remove_dupe_paths(files)
logging.info("Getting matches. Scan type: %d", self.scan_type)
matches = self._getmatches(files, j)
logging.info('Found %d matches' % len(matches))
j.set_progress(100, tr("Removing false matches"))
# In removing what we call here "false matches", we first want to remove, if we scan by
# folders, we want to remove folder matches for which the parent is also in a match (they're
# "duplicated duplicates if you will). Then, we also don't want mixed file kinds if the
# option isn't enabled, we want matches for which both files exist and, lastly, we don't
# want matches with both files as ref.
if self.scan_type == ScanType.Folders and matches:
allpath = {m.first.path for m in matches}
allpath |= {m.second.path for m in matches}
sortedpaths = sorted(allpath)
toremove = set()
last_parent_path = sortedpaths[0]
for p in sortedpaths[1:]:
if p in last_parent_path:
toremove.add(p)
else:
last_parent_path = p
matches = [m for m in matches if m.first.path not in toremove or m.second.path not in toremove]
if not self.mix_file_kind:
matches = [m for m in matches if get_file_ext(m.first.name) == get_file_ext(m.second.name)]
matches = [m for m in matches if m.first.path.exists() and m.second.path.exists()]
matches = [m for m in matches if not (m.first.is_ref and m.second.is_ref)]
if self.ignore_list:
j = j.start_subjob(2)
iter_matches = j.iter_with_progress(matches, tr("Processed %d/%d matches against the ignore list"))
matches = [
m for m in iter_matches
if not self.ignore_list.AreIgnored(str(m.first.path), str(m.second.path))
]
logging.info('Grouping matches')
groups = engine.get_groups(matches, j)
matched_files = dedupe([m.first for m in matches] + [m.second for m in matches])
if self.scan_type in {ScanType.Filename, ScanType.Fields, ScanType.FieldsNoOrder, ScanType.Tag}:
self.discarded_file_count = len(matched_files) - sum(len(g) for g in groups)
else:
# Ticket #195
# To speed up the scan, we don't bother comparing contents of files that are both ref
# files. However, this messes up "discarded" counting because there's a missing match
# in cases where we end up with a dupe group anyway (with a non-ref file). Because it's
# impossible to have discarded matches in exact dupe scans, we simply set it at 0, thus
# bypassing our tricky problem.
# Also, although ScanType.FuzzyBlock is not always doing exact comparisons, we also
# bypass ref comparison, thus messing up with our "discarded" count. So we're
# effectively disabling the "discarded" feature in PE, but it's better than falsely
# reporting discarded matches.
self.discarded_file_count = 0
groups = [g for g in groups if any(not f.is_ref for f in g)]
logging.info('Created %d groups' % len(groups))
j.set_progress(100, tr("Doing group prioritization"))
for g in groups:
g.prioritize(self._key_func, self._tie_breaker)
return groups
def __Perform(self, copy=False, job=nulljob):
try:
cd_operations = [
t for t in self.name_list if t[0][0].startswith('!')
]
normal_operations = [
t for t in self.name_list if t not in cd_operations
]
cds = dedupe(t[0][0][1:] for t in cd_operations)
job_count = len(cds)
if normal_operations:
job_count += 1
newjob = job.start_subjob(job_count)
if normal_operations:
self.__ProcessNormalList(normal_operations, copy, newjob)
for cd in cds:
cd_location = [
location for location in self.cd_locations
if location.name == cd
][0]
cd_path = self.OnNeedCD(cd_location)
if not cd_path:
return False
name_list = [(t[0][1:], t[1]) for t in cd_operations
if t[0][0][1:] == cd]
if not self.__ProcessCDList(name_list, Path(cd_path),
cd_location, newjob):
return False
return True
except JobCancelled:
return False
def _perform_action(self, import_action, apply=ActionSelectionOptions.ApplyToPane):
if self.selected_pane is None:
return
action_params = self._collect_action_params(import_action, apply)
if not action_params:
return
panes = dedupe(flatten((grp[2] for grp in action_params)))
for pane in panes:
pane.match_flag = False
pane.import_document.cook_flag = False
for action_param in action_params:
import_action.perform_action(*action_param)
for pane in panes:
if not pane.import_document.cook_flag:
pane.import_document.cook()
for pane in panes:
if not pane.match_flag:
pane.match_entries()
self.import_table.refresh()
def _reorder_following_line(self, reorder_lines):
# Reordering from a line is a bit more complex than it seems. We have to find intersection
# points from all lines in all rects (when there's an interestion, of course). Then, for
# each intersection, we compute the distance of it from the origin of the order arrow.
# We sort our elements by that distance and we have our new order!
neworder = []
for reorder_line in reorder_lines:
neworder += self._get_intersections(reorder_line)
neworder = dedupe(neworder)
if len(neworder) < 2:
return # nothing to reorder
# ok, we have our new order. That was easy huh? Now, what we have to do is to insert that
# new order in the rest of the elements, which might not all be in our new order. That's
# a bit complex too because when drawing arrows, a naive re-ordering is counter intuitive.
# I won't explain why what is done below is done thus because it's hard to describe, but
# here's what we do: We compute the min and max of old orders in our elems. Whatever came
# before min stays before. Whatever came after max stays *right* after, but not at the end.
# The end of the order is constituted of what is between min and max, but isn't in the new
# order. Arrow-wise, these elements have been "orphaned" by the new link and the most
# intuitive thing to do with them is to put them at the end of the order.
minorder = min(e.order for e in neworder)
maxorder = max(e.order for e in neworder)
all_elems = list(self._ordered_elems())
affected_elems = set(neworder)
unaffected_elems = [e for e in all_elems if e not in affected_elems]
before = [e for e in unaffected_elems if e.order < minorder]
after = [e for e in unaffected_elems if e.order > maxorder]
inbetween = [e for e in unaffected_elems if minorder < e.order < maxorder]
# we also add ignore elems to our big concat so that their order value doesn't conflict
ignored = list(self._ignored_elems())
concat = before + neworder + after + inbetween + ignored
for i, elem in enumerate(concat):
elem.order = i
def guess_date_format(self, str_dates):
totry = DATE_FORMATS[:]
extra = []
if self.NATIVE_DATE_FORMAT:
extra.append(self.NATIVE_DATE_FORMAT)
if self.EXTRA_DATE_FORMATS:
extra += self.EXTRA_DATE_FORMATS
if self.default_date_format:
extra.append(self.default_date_format)
for format in dedupe(extra + totry):
found_at_least_one = False
for str_date in str_dates:
try:
datetime.datetime.strptime(str_date, format)
found_at_least_one = True
except ValueError:
logging.debug(
"Failed try to read the date {0} with the format {1}".
format(str_date, format))
break
else:
if found_at_least_one:
logging.debug("Correct date format: {0}".format(format))
return format
return None
def mimeData(self, indexes):
nodes = dedupe(index.internalPointer() for index in indexes)
paths = [str(self.name + node.ref.path) for node in nodes]
data = '\n'.join(paths).encode('utf-8')
mimeData = QMimeData()
mimeData.setData(MIME_PATHS, QByteArray(data))
return mimeData
def mimeData(self, indexes):
nodes = dedupe(index.internalPointer() for index in indexes)
paths = [str(self.name + node.ref.path) for node in nodes]
data = '\n'.join(paths).encode('utf-8')
mimeData = QMimeData()
mimeData.setData(MIME_PATHS, QByteArray(data))
return mimeData
def _collect_action_params(self, import_action, apply=ActionSelectionOptions.ApplyToPane):
if apply == ActionSelectionOptions.ApplyToSelection:
if not self.selected_pane:
return []
panes = [self.selected_pane]
selected_rows = self.import_table.selected_rows
if not selected_rows:
return []
transactions = [row.imported.transaction for row in selected_rows if row.imported]
transactions = dedupe(transactions)
can_perform_action = import_action.can_perform_action(panes[0].import_document,
transactions,
panes,
selected_rows)
if not can_perform_action:
return []
return [(panes[0].import_document, transactions, panes, selected_rows)]
if apply == ActionSelectionOptions.ApplyToAll:
panes = self.panes.copy()
else:
panes = [self.selected_pane]
results = []
# Groups of panes which share the same import document
pane_groups = unique_groups(panes, lambda p: p.import_document)
selected_group = None
for ps in pane_groups:
transactions = [e.transaction for p in ps for e in p.account.entries]
transactions = dedupe(transactions)
can_perform_action = import_action.can_perform_action(ps[0].import_document, transactions, ps)
if not can_perform_action:
continue
if self.selected_pane in ps:
selected_group = (ps[0].import_document, transactions, ps)
else:
results.append((ps[0].import_document, transactions, ps))
# We want to ensure that the selected pane group is the last to be called
# allowing the name of the plugin to update.
if selected_group:
results.append(selected_group)
return results
def get_dupe_groups(self, files, ignore_list=None, j=job.nulljob):
for f in (f for f in files if not hasattr(f, "is_ref")):
f.is_ref = False
files = remove_dupe_paths(files)
logging.info("Getting matches. Scan type: %d", self.scan_type)
matches = self._getmatches(files, j)
logging.info("Found %d matches" % len(matches))
j.set_progress(100, tr("Almost done! Fiddling with results..."))
# In removing what we call here "false matches", we first want to remove, if we scan by
# folders, we want to remove folder matches for which the parent is also in a match (they're
# "duplicated duplicates if you will). Then, we also don't want mixed file kinds if the
# option isn't enabled, we want matches for which both files exist and, lastly, we don't
# want matches with both files as ref.
if self.scan_type == ScanType.FOLDERS and matches:
allpath = {m.first.path for m in matches}
allpath |= {m.second.path for m in matches}
sortedpaths = sorted(allpath)
toremove = set()
last_parent_path = sortedpaths[0]
for p in sortedpaths[1:]:
if p in last_parent_path:
toremove.add(p)
else:
last_parent_path = p
matches = [m for m in matches if m.first.path not in toremove or m.second.path not in toremove]
if not self.mix_file_kind:
matches = [m for m in matches if get_file_ext(m.first.name) == get_file_ext(m.second.name)]
matches = [m for m in matches if m.first.path.exists() and m.second.path.exists()]
matches = [m for m in matches if not (m.first.is_ref and m.second.is_ref)]
if ignore_list:
matches = [m for m in matches if not ignore_list.are_ignored(str(m.first.path), str(m.second.path))]
logging.info("Grouping matches")
groups = engine.get_groups(matches)
if self.scan_type in {
ScanType.FILENAME,
ScanType.FIELDS,
ScanType.FIELDSNOORDER,
ScanType.TAG,
}:
matched_files = dedupe([m.first for m in matches] + [m.second for m in matches])
self.discarded_file_count = len(matched_files) - sum(len(g) for g in groups)
else:
# Ticket #195
# To speed up the scan, we don't bother comparing contents of files that are both ref
# files. However, this messes up "discarded" counting because there's a missing match
# in cases where we end up with a dupe group anyway (with a non-ref file). Because it's
# impossible to have discarded matches in exact dupe scans, we simply set it at 0, thus
# bypassing our tricky problem.
# Also, although ScanType.FuzzyBlock is not always doing exact comparisons, we also
# bypass ref comparison, thus messing up with our "discarded" count. So we're
# effectively disabling the "discarded" feature in PE, but it's better than falsely
# reporting discarded matches.
self.discarded_file_count = 0
groups = [g for g in groups if any(not f.is_ref for f in g)]
logging.info("Created %d groups" % len(groups))
for g in groups:
g.prioritize(self._key_func, self._tie_breaker)
return groups
def _getData(self):
folder = self.ref
parent_volumes = dedupe(song.original.parent_volume for song in folder.iterallfiles())
return [
folder.name,
','.join(l.name for l in parent_volumes),
folder.get_stat('filecount'),
format_size(folder.get_stat('size'), 2, 2, False),
format_time(folder.get_stat('duration')),
]
def _getData(self):
folder = self.ref
parent_volumes = dedupe(song.original.parent_volume
for song in folder.iterallfiles())
return [
folder.name,
','.join(l.name for l in parent_volumes),
folder.get_stat('filecount'),
format_size(folder.get_stat('size'), 2, 2, False),
format_time(folder.get_stat('duration')),
]
def _invert_amounts(self, apply_to_all):
if apply_to_all:
panes = self.panes
else:
panes = [self.selected_pane]
entries = flatten(p.account.entries for p in panes)
txns = dedupe(e.transaction for e in entries)
for txn in txns:
for split in txn.splits:
split.amount = -split.amount
# Entries, I don't remember why, hold a copy of their split's amount. It has to be updated.
for entry in entries:
entry.amount = entry.split.amount
self.import_table.refresh()
def show_transfer_account(self, row_index=None):
if row_index is None:
if not self.selected_entries:
return
row_index = self.selected_index
entry = self[row_index].entry
splits = entry.transaction.splits
accounts = dedupe(split.account for split in splits if split.account is not None)
if len(accounts) < 2:
return # no transfer
index = accounts.index(entry.account)
if index < len(accounts) - 1:
account_to_show = accounts[index+1]
else:
account_to_show = accounts[0]
self.mainwindow.open_account(account_to_show)
def show_transfer_account(self, row_index=None):
if row_index is None:
if not self.selected_entries:
return
row_index = self.selected_index
entry = self[row_index].entry
splits = entry.transaction.splits
accounts = dedupe(split.account for split in splits
if split.account is not None)
if len(accounts) < 2:
return # no transfer
index = accounts.index(entry.account)
if index < len(accounts) - 1:
account_to_show = accounts[index + 1]
else:
account_to_show = accounts[0]
self.mainwindow.open_account(account_to_show)
def _refresh_candidates(self):
if self.mainwindow is None or not self.attrname:
return
doc = self.mainwindow.document
attrname = self.attrname
if attrname == 'description':
self._candidates = doc.transactions.descriptions
elif attrname == 'payee':
self._candidates = doc.transactions.payees
elif attrname in {'from', 'to', 'account', 'transfer'}:
result = doc.transactions.account_names
# `result` doesn't contain empty accounts' name, so we'll add them.
result += [a.name for a in doc.accounts if not a.inactive]
if attrname == 'transfer' and self.account is not None:
result = [name for name in result if name != self.account.name]
self._candidates = result
self._candidates = dedupe([name for name in self._candidates if name.strip()])
def __init__(self, partial, candidates):
"""Build a completion list.
'partial' is the partial value to be completed
'candidates' is the list of candidate values to be tried, the most likely candidate first."""
if not partial:
self._completions = None
return
partial = sort_string(partial)
candidates = dedupe(c.strip() for c in candidates)
self._completions = []
for candidate in candidates:
normalized = sort_string(candidate)
if normalized.startswith(partial):
self._completions.append(candidate)
self._completions.reverse()
if self._completions:
self._index = len(self._completions) - 1
def _refresh_candidates(self):
if self.mainwindow is None or not self.attrname:
return
doc = self.mainwindow.document
attrname = self.attrname
if attrname == 'description':
self._candidates = doc.transactions.descriptions
elif attrname == 'payee':
self._candidates = doc.transactions.payees
elif attrname in {'from', 'to', 'account', 'transfer'}:
result = doc.transactions.account_names
# `result` doesn't contain empty accounts' name, so we'll add them.
result += [a.name for a in doc.accounts if not a.inactive]
if attrname == 'transfer' and self.account is not None:
result = [name for name in result if name != self.account.name]
self._candidates = result
self._candidates = dedupe(
[name for name in self._candidates if name.strip()])
def __init__(self, partial, candidates):
"""Build a completion list.
'partial' is the partial value to be completed
'candidates' is the list of candidate values to be tried, the most likely candidate first."""
if not partial:
self._completions = None
return
partial = sort_string(partial)
candidates = dedupe(c.strip() for c in candidates)
self._completions = []
for candidate in candidates:
normalized = sort_string(candidate)
if normalized.startswith(partial):
self._completions.append(candidate)
self._completions.reverse()
if self._completions:
self._index = len(self._completions) - 1
def GetNodeData(self, node):
if node.is_container:
img_name = 'folder_conflict_16' if node.allconflicts else 'folder_16'
parent_volumes = dedupe(song.original.parent_volume for song in node.iterallfiles())
return [
node.name,
','.join(l.name for l in parent_volumes),
node.get_stat('filecount'),
format_size(node.get_stat('size'),2,2,False),
format_time(node.get_stat('duration')),
img_name,
]
else:
img_name = 'song_conflict_16' if is_conflicted(node.name) else 'song_16'
return [
node.name,
node.original.parent_volume.name,
0,
format_size(node.size,2,2,False),
format_time(node.duration, with_hours=False),
img_name,
]
def _get_intersections(self, reorder_line):
# return a list of elements that intersect with line, in order. The order depends on the
# distance of the elem's first intersection with the line's origin
intersections = []
for elem in self._active_elems():
rect = self._elem2drawrect[elem]
for line in rect.lines():
inter = reorder_line.intersection_point(line)
if inter is not None:
dist = inter.distance_to(reorder_line.p1)
intersections.append((dist, elem))
if not intersections:
# so we cross no line, but we might be in the middle of an elem, in which case we should
# return that elem. It's even possible that we're inside multiple rects. This case is
# tricky (there's multiple possibilities here). We just choose the elem for which the
# center is closest to our origin
origin = reorder_line.p1
for elem in self._active_elems():
rect = self._elem2drawrect[elem]
if rect.contains_point(origin):
intersections.append((origin.distance_to(rect.center()), elem))
intersections.sort(key=lambda tup: tup[0])
return dedupe([elem for dist, elem in intersections])
def _get_intersections(self, reorder_line):
# return a list of elements that intersect with line, in order. The order depends on the
# distance of the elem's first intersection with the line's origin
intersections = []
for elem in self._active_elems():
rect = self._elem2drawrect[elem]
for line in rect.lines():
inter = reorder_line.intersection_point(line)
if inter is not None:
dist = inter.distance_to(reorder_line.p1)
intersections.append((dist, elem))
if not intersections:
# so we cross no line, but we might be in the middle of an elem, in which case we should
# return that elem. It's even possible that we're inside multiple rects. This case is
# tricky (there's multiple possibilities here). We just choose the elem for which the
# center is closest to our origin
origin = reorder_line.p1
for elem in self._active_elems():
rect = self._elem2drawrect[elem]
if rect.contains_point(origin):
intersections.append((origin.distance_to(rect.center()), elem))
intersections.sort(key=lambda tup: tup[0])
return dedupe([elem for dist, elem in intersections])
def guess_date_format(self, str_dates):
totry = DATE_FORMATS[:]
extra = []
if self.NATIVE_DATE_FORMAT:
extra.append(self.NATIVE_DATE_FORMAT)
if self.EXTRA_DATE_FORMATS:
extra += self.EXTRA_DATE_FORMATS
if self.default_date_format:
extra.append(self.default_date_format)
for format in dedupe(extra + totry):
found_at_least_one = False
for str_date in str_dates:
try:
datetime.datetime.strptime(str_date, format)
found_at_least_one = True
except ValueError:
logging.debug("Failed try to read the date {0} with the format {1}".format(str_date, format))
break
else:
if found_at_least_one:
logging.debug("Correct date format: {0}".format(format))
return format
return None
def GetNodeData(self, node):
if node.is_container:
img_name = 'folder_conflict_16' if node.allconflicts else 'folder_16'
parent_volumes = dedupe(song.original.parent_volume
for song in node.iterallfiles())
return [
node.name,
','.join(l.name for l in parent_volumes),
node.get_stat('filecount'),
format_size(node.get_stat('size'), 2, 2, False),
format_time(node.get_stat('duration')),
img_name,
]
else:
img_name = 'song_conflict_16' if is_conflicted(
node.name) else 'song_16'
return [
node.name,
node.original.parent_volume.name,
0,
format_size(node.size, 2, 2, False),
format_time(node.duration, with_hours=False),
img_name,
]
def criteria_list(self):
dupes = flatten(g[:] for g in self.results.groups)
values = sorted(dedupe(self.extract_value(d) for d in dupes))
return [Criterion(self, value) for value in values]
def _insertItem(self, item):
self._items = dedupe([item] + self._items)[:self._maxItemCount]
def _swap_fields(self, panes, switch_func):
entries = flatten(p.account.entries for p in panes)
txns = dedupe(e.transaction for e in entries)
for txn in txns:
switch_func(txn)
self.import_table.refresh()
def criteria_list(self):
dupes = flatten(g[:] for g in self.results.groups)
values = sorted(dedupe(self.extract_value(d) for d in dupes))
return [Criterion(self, value) for value in values]
def _insertItem(self, item):
self._items = dedupe([item] + self._items)[:self._maxItemCount]
def delete_entries(self, entries):
from_account = first(entries).account
transactions = dedupe(e.transaction for e in entries)
self.delete_transactions(transactions, from_account=from_account)
