def _getNumberSelectionSplittedNumber(self, firstPartNumber,
lastPartNumber):
escapedFirstPartNumber = helper.escapeForRegex(
firstPartNumber) #Not super necessary, but doesn't hurt
escapedLastPartNumber = helper.escapeForRegex(
lastPartNumber) #Here necessary because of dot
#Get all chunks that start with *last* part of number.
allSelectionsLastPartNumber = self.cutter.filter(
auto_regex='^{}'.format(escapedLastPartNumber)
) # Returns all Selections that have Chunks which start with the number
#Sort them from highest to lowest
sortedAllSelecionsLastPartNumber = sorted(
allSelectionsLastPartNumber,
key=lambda x: x.doc_top) #Sort by appearance
firstPartNumberSelection = None
for selection in sortedAllSelecionsLastPartNumber: #Start with highest selection
#All Chunks that are "slightly" (strict) above last part number chunk
aboveSelections = self.cutter.all().filter(
doc_top__gte=selection.doc_top - 50,
doc_bottom__lte=selection.doc_top,
)
#Any Chunk slightly above that starts with *first* part of number?
maybeFirstPartNumberAboveSelection = aboveSelections.filter(
auto_regex='^{}$'.format(escapedFirstPartNumber))
if len(maybeFirstPartNumberAboveSelection
) == 1: #There is exactly one such selection
#Therefore. return it (per Definition) as number chunk and stop
firstPartNumberSelection = maybeFirstPartNumberAboveSelection
break
return firstPartNumberSelection
def _getNumberSelection(self, number):
numberWithoutPoint = number[:-1]
escapedNum = helper.escapeForRegex(numberWithoutPoint)
allSelectionsNumber = self.cutter.filter(
auto_regex='^{}'.format(escapedNum)
) # Returns all Selections that have Chunks which start with the number
return self._getHighestSelection(allSelectionsNumber)
def _getSubpartSelectionNonStrictBelowNumberSelection(
self, subpart, numberSelection):
escapedSubpart = helper.escapeForRegex(subpart)
numberUpperBorder = self.cutter.all().filter(
doc_top__gte=numberSelection.doc_top - 50, )
allSelectionsSubpartNonStrictBelowNumber = numberUpperBorder.filter(
regex="[^G]" + escapedSubpart
) #Disallow G in Selection for TOP 18. b) because of "(LFGB)" getting attention too (although upper case?)
return self._getHighestSelection(
allSelectionsSubpartNonStrictBelowNumber)
def _getNumberSelection(self, number):
escapedNum = helper.escapeForRegex(number)
allSelectionsNumber = self.cutter.filter(
auto_regex='^{}'.format(escapedNum)
).filter( # Returns all Selections that have Chunks which start with the number
left__lte=self.
TOPRight, #Can't do anything if whole line is one chunk (therefore right__lte bad), but it should at least start before TOPRight
top__gte=self.page_heading,
)
highestSelection = self._getHighestSelection(allSelectionsNumber)
#dVis.showCutter(highestSelection)
return highestSelection
def _getNumberSelection(self, number):
formatedNumber = number.split(
"."
)[0] #46. -> 46 , Done with split because also use it here in SA for Subparts as well
escapedNum = helper.escapeForRegex(formatedNumber)
allSelectionsNumber = self.cutter.filter(
auto_regex='^{}'.format(escapedNum)
).filter(
left__lte=
160 # Dont match e.g. "980 Sitzung" in title for TOP 9 (happens in 989 TOP 9)
)
return self._getHighestSelection(allSelectionsNumber)
def _getSubpartSelectionNonStrictBelowNumberSelection(
self, subpart, numberSelection):
escapedSubpart = helper.escapeForRegex(subpart)
numberUpperBorder = self.cutter.all().filter(
doc_top__gte=numberSelection.doc_top -
50, #Return all Chunks below given numer chunk and the number chunk itself. subpart chunk could be same as number chunk
) # INFO a) for 1. a) NS 970 in same chunk, for 34. a) not
# All Chunks non-strict below number chunk that contain given subpart
allSelectionsSubpartNonStrictBelowNumber = numberUpperBorder.filter(
auto_regex=escapedSubpart
).filter( #46. b) -> b\) because of regex brackets
left__lte=self.
TOPRight, #Can't do anything if whole line is one chunk (therefore right__lte bad), but it should at least start before TOPRight
top__gte=self.page_heading,
)
#Return highest of these
#INFO adding number chunk as upperbound can break this when subpart chunk == number chunk
return self._getHighestSelection(
allSelectionsSubpartNonStrictBelowNumber)
def _getPrefixStringSelection(self, s):
escapedS = helper.escapeForRegex(s)
allSelectionsS = self.cutter.filter(
auto_regex='{}'.format(escapedS)
) # Returns all Selections that have Chunks which *contain* s
return self._getHighestSelection(allSelectionsS)