Among the files produced for each dump of each project is a list of all titles in namespace 0. This would include all lemmas for a wiktionary. See, for example, http://dumps.wikimedia.org/ruwiktionary/20120107/ruwiktionary-20120107-all-t...

Ariel

Στις 09-01-2012, ημέρα Δευ, και ώρα 11:21 +0100, ο/η Sébastien Druon έγραψε:

Thanks you all for the tips.

Maybe you have a good tip to get the parsed/html version of these entries in an easy way?

Thanks again

Sebastien

On 9 January 2012 10:51, Jérémie Roquet arkanosis@gmail.com wrote: Hi Sebastien,

    2012/1/9 Sébastien Druon <druon.sebastien@gmail.com>:
    > How is it possible to get the list of all the entries
    (words) of a
    > wiktionary?
    > For example, for the russian wiktionary, I want to get the
    list of all the
    > russian entries (no other languages)


    You can download the last dump¹ and pass it through a simple
    awk
    script that prints the titles of the pages that contain the
    {{-ru-}}
    template, ie. something like:

    ----8<----

    BEGIN {
     ru = 0
    }

    END {
     if (ru) {
       print title
     }
    }

    /<title>.*?</ {
     if (ru) {
       print title
     }

     title = substr($0, 12, length($0) - 19)
     ru = 0
    }

    tolower($0) ~ /{{-ru-}}/ {
     ru = 1
    }

    ----8<----

    You'd still have to filter the output to only keep titles in
    the main namespace.

    It should be possible using categories too, but this wouldn't
    be any
    easier nor more reliable and would be much slower.

    Best regards,

    ¹http://dumps.wikimedia.org/ruwiktionary/20120107/ruwiktionary-20120107-pages-articles.xml.bz2

    --
    Jérémie

---

Xmldatadumps-l mailing list Xmldatadumps-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/xmldatadumps-l

Hi Platonides,

2012/1/9 Platonides platonides@gmail.com:

On 09/01/12 10:51, Jérémie Roquet wrote:

...

It should be possible using categories too, but this wouldn't be any easier nor more reliable and would be much slower.

I disagree. You would only need the much smaller ruwiktionary-20120107-category.sql.gz + ruwiktionary-20120107-page.sql.gz

Good point. Still, you'd have to understand the categories' hierarchy :-)

Best regards,