A tool to search a grid given as a text file for words present in a file.
wordsearch.py [-h] [--trie] [--multiprocess] grid words
with the following arguments:
- Positional:
grid: The file containing the grid of wordswords: The list of words to check
- Optional:
--trie: Use trie data structure.--multiprocess: Use multiple processes to search for words/generate Trie.
Implemented in utils/grid.py this splits the grid into a list of rows and
also transposes them to get a list of columns.
Word presence is then checked using Python's in statement on each row &
column until the word is found or all rows & columns have been searched.
- Initialisation is relatively quick. Transposing rows is the most intensive operation.
- Grid has to be searched for every word.
- Multiprocessing implementation is slower. I have been unable to fix this.
In my opinon you should use this without the --multiprocessing flag
for the best results
Implemented in utils/trie.py this iterates through the grid, extracting
the possible words at each point and storing them in the form of a
Trie.
Word presence is then checked by descending the Trie until all letters in the word have been found or the next letter can't be found.
- Search is quite quick as it has essentially 'found' all words already.
- Entire grid is processed even if looking for one word.
- Processing grid is quite resource intesive and can taka a long time.
Don't. This is sadly the worst option by far. Use Grid. Although I do admit I did find this one the most interesting to implement.
If you have to use this use it with the --multiprocessing flag.
There is a test grid with a words file in tests/data.
However, these require setting the ROW_LENGTH to 1000.
These were generated using utils/generation.py.
For usage run utils/generation.py --help.