Cmbsolver, what are you trying to do?
I know only the second word is Latin. Anyway, the Liber Primus (LP2) still remain unsolved after a decade. Or least a publicly viewable on it was solved. So after some other failed attempts at getting at the puzzle, I am trying something with cribbing the data. This one may prove unproductive, but it is worth a shot. Other traditional methods frequency analysis have proven to be unproductive.
Basically, I will take sentences in the liber primus and break them apart, then I will get the possible word for each attribute (see file layout below). Then it will try different permutations of the word combinations and run it through a language tokenizer. If it tokenizer shows the probability of the words being a sentence, then it will add them to a text file for review.
Building the word list files.
- Building out the word list.
- I built the word list from three separate sources. I downloaded or mirrored the following locations.
- I wrote a program in Golang that would break apart the word lists.
- I spun up three mysql databases on different ports (https://github.com/cmbsolver/libergo/blob/main/create_podman_db.sh)
- Then I ran the program to generate the word lists in CSV format.
- Used dbeaver to upload the CSV files to the tables in each database.
File Layout
dict_word – Regular english word.
dict_runeglish – The runeglish version of the word.
dict_rune – The rune version of the word.
dict_rune_no_doublet – The rune version without doublets.
gem_sum – The gematria sum.
gem_sum_prime – Whether or not the gematria sum is prime.
gem_product – The gematria product.
gem_product_prime – Whether or not the gematria product is prime.
dict_word_length – The regular word length.
dict_runeglish_length – Runeglish word length.
dict_rune_length – The rune length.
dict_rune_no_doublet_length – The rune length without doublets.
rune_pattern – The rune pattern.
rune_pattern_no_doublet – The rune pattern without doublets.
If you want the word lists, it is available here.