WiktionaryDumps to words

WiktionaryDumps to words is a draft programming task. It is not yet considered ready to be promoted as a complete task, for reasons that should be found in its talk page.

NOTE

Please help addressing the issues about this task on the discussion page.

Use the wiktionary dump (input) to create a file equivalent than "/usr/share/dict/french" (output). This dump is a big bz2'ed XML file of about 800MB. The "/usr/share/dict/french" file contains one word of the French language by line in a text file. This file is available in Ubuntu with the package wfrench.

OCaml

Using the library xmlm:

<lang ocaml>let () =

 let i = Xmlm.make_input ~strip:true (`Channel stdin) in
 let title = ref "" in
 let tag_path = ref [] in
 let push_tag tag =
   tag_path := tag :: !tag_path
 in
 let pop_tag () =
   match !tag_path with [] -> ()
   | _ :: tl -> tag_path := tl
 in
 let last_tag_is tag =
   match !tag_path with [] -> false
   | hd :: _ -> hd = tag
 in
 while not (Xmlm.eoi i) do
   match Xmlm.input i with
   | `Dtd dtd -> ()
   | `El_start ((uri, tag_name), attrs) -> push_tag tag_name
   | `El_end -> pop_tag ()
   | `Data s ->
       if last_tag_is "title"
       then title := s;
       if last_tag_is "text"
       then begin
         let reg = Str.regexp_string "==French==" in
         if Str.string_match reg s 0
         then print_endline !title
       end
 done</lang>

Output:

wget --quiet https://dumps.wikimedia.org/enwiktionary/latest/enwiktionary-latest-pages-articles.xml.bz2 -O - | bzcat | \
  ocaml str.cma -I $(ocamlfind query xmlm) xmlm.cma to_words.ml
livrer
observateur
qui a bu boira
quelque chose
grande parure
obiit
pleuvoir
voir
...