Hello!
I’m building a LibreOffice extension to sort alphabetically Sanskrit wordlists (in IAST transcription) . I’m not a professional programmer (just a Linux enthusiast involved in digital humanities), I have some roughly intermediate knowledge of Python.
I managed to build a word sorting macro (with alphabetical mapping of transliteration signs and some exceptions handling pertaining to some phonetic rules of Sanskrit). After that, somehow I built an extension around it. Everything seems to work quite smoothly. So far so good!
Fact is, if I apply the sorting to some elements of the list, which have formatting attributes (such as italics stems or chunks of words somehow marked). Their formatting disappears right after running the sorting function, and that makes all my building efforts useless.
AI obviously provides faulty solutions. Can I ask for help?
Thanks in advance
my main.py is
import uno
import unohelper
from com.sun.star.task import XJobExecutor
ID_EXTENSION = 'org.examples.sskr_wl_sorter.000'
SERVICE = ('com.sun.star.task.Job',)
# The new sorting function for both Calc and Writer documents
def sort_sanskrit_in_document():
"""Sorts Sanskrit words in LibreOffice Calc or Writer. In Calc, it sorts the whole sheet, ignoring selection."""
try:
# Connect to LibreOffice
localContext = uno.getComponentContext()
smgr = localContext.ServiceManager
desktop = smgr.createInstanceWithContext("com.sun.star.frame.Desktop", localContext)
# Get the active document
model = desktop.getCurrentComponent()
# Check if the document is a Calc sheet or Writer document
if hasattr(model, "Sheets"):
# If it's a Calc spreadsheet, sort the whole sheet, ignoring the selection
sheet = model.CurrentController.ActiveSheet
# Find the last non-empty row
last_row = 0
for i in range(sheet.Rows.getCount()):
cell = sheet.getCellByPosition(0, i) # First column
if cell.getString().strip() == "": # If empty, stop
break
last_row = i
# Find the last non-empty column
last_col = 0
for j in range(sheet.Columns.getCount()):
cell = sheet.getCellByPosition(j, 0)
if cell.getString().strip() == "":
break
last_col = j
# Select the entire range
cell_range = sheet.getCellRangeByPosition(0, 0, last_col, last_row)
# Read all rows
rows = []
for i in range(cell_range.Rows.getCount()):
row = []
for j in range(cell_range.Columns.getCount()):
cell = cell_range.getCellByPosition(j, i)
row.append(cell.getString().strip())
rows.append(row)
# Sort the rows using the first column
sorted_rows = sorted(rows, key=lambda row: custom_sort(row[0], mapping))
# Write the sorted rows back to the spreadsheet
for i, row in enumerate(sorted_rows):
for j, value in enumerate(row):
cell = sheet.getCellByPosition(j, i)
cell.setString(value)
elif hasattr(model, "Text"):
# If it's a Writer document
text = model.Text
cursor = model.CurrentController.getViewCursor()
# Get the selected text or the entire document if nothing is selected
if cursor.isCollapsed():
cursor.gotoStart(False)
cursor.gotoEnd(True)
# Read the selected text as lines
selected_text = cursor.getString().strip().splitlines()
# Sort the existing lines
sorted_lines = sorted(selected_text, key=lambda line: custom_sort(line.split()[0], mapping) if line.strip() else "")
# Clear the selected text
cursor.setString("") # Replace the selected text with an empty string
# Insert the sorted text at the current cursor position
text.insertString(cursor, "\n".join(sorted_lines), False) # Insert the sorted text
else:
raise Exception("This document is neither a spreadsheet nor a Writer document.")
except Exception as e:
print(f"Error during sorting: {e}")
# Custom sorting functions and mapping
def custom_sort(word, mapping):
if not word:
return word # Return the empty word if there's nothing to sort
word_for_sorting = word.replace('√', '') # Remove the root symbol for sorting
preprocessed_word = preprocess_word(word_for_sorting, mapping)
transformed_word = "".join(mapping.get(char, char) for char in preprocessed_word)
# Handle hierarchy between hyphen (-) and equals (=)
hierarchy_weight = ''
if '-' in word and '=' in word:
hierarchy_weight = '1' if word.index('-') < word.index('=') else '2'
elif '-' in word:
hierarchy_weight = '1'
elif '=' in word:
hierarchy_weight = '2'
return transformed_word + hierarchy_weight
def preprocess_word(word, mapping):
"""Pre-processes the word for sorting, removing the root symbol."""
preprocessed_word = ""
i = 0
while i < len(word):
if word[i] == '√':
i += 1
continue
if word[i] == 'ṃ':
if i < len(word) - 1 and word[i + 1] in 'kgcṭtdpb':
nasal_replacements = {'k': 'ṅ', 'g': 'ṅ', 'c': 'ñ', 'j': 'ñ', 'ṭ': 'ṇ', 'ḍ': 'ṇ', 't': 'n', 'd': 'n', 'p': 'm', 'b': 'm'}
preprocessed_word += nasal_replacements.get(word[i + 1], 'ṃ')
else:
preprocessed_word += 'ṃ'
elif word[i] == 'ḥ':
if i < len(word) - 1 and word[i + 1] in 'śṣs':
preprocessed_word += word[i + 1]
else:
preprocessed_word += 'ḥ'
else:
preprocessed_word += word[i]
i += 1
return preprocessed_word
# Mapping for sorting Sanskrit characters
mapping = {
'a': '01', 'ā': '02', 'i': '03', 'ī': '04', 'u': '05', 'ū': '06',
'ṛ': '07', 'ṝ': '08', 'ḷ': '09', 'e': '10', 'ai': '11', 'o': '12',
'au': '13', 'ṃ': '14', 'ḥ': '15', 'k': '16', 'kh': '17', 'g': '18',
'gh': '19', 'ṅ': '20', 'c': '21', 'ch': '22', 'j': '23', 'jh': '24',
'ñ': '25', 'ṭ': '26', 'ṭh': '27', 'ḍ': '28', 'ḍh': '29', 'ṇ': '30',
't': '31', 'th': '32', 'd': '33', 'dh': '34', 'n': '35', 'p': '36',
'ph': '37', 'b': '38', 'bh': '39', 'm': '40', 'y': '41', 'r': '42',
'l': '43', 'v': '44', 'ś': '45', 'ṣ': '46', 's': '47', 'h': '48',
}
# Main Extension class
class MyFirstExtension(unohelper.Base, XJobExecutor):
def __init__(self, ctx):
self.ctx = ctx
print("MyFirstExtension initialized") # Debugging output
# XJobExecutor trigger function
def trigger(self, event):
print(f"Triggered by event: {event}") # Debugging output
sort_sanskrit_in_document() # Call the Sanskrit sorting function for both Calc and Writer
return
# Register the implementation
g_ImplementationHelper = unohelper.ImplementationHelper()
g_ImplementationHelper.addImplementation(MyFirstExtension, ID_EXTENSION, SERVICE)
