swedish_poetry

Poetic Corpus for Swedish Language

Database

Current corpus is in an SQLite database that can be viewed with, for example, DB Browser for SQLite.

Database has following schema:

Corpus has texts from three swedish authors:

• Karin Boye, volumes:
  • MOLN
  • GÖMDA LAND
  • HÄRDARNA
  • FÖR TRÄDETS SKULL
  • DE SJU DÖDSSYNDERNA
• Nils Ferlin, volumes:
  • GOGGLES
  • BARFOTABARN
  • FRÅN MITT EKORRHJUL
  • EN DÖDDANSARES VISOR
  • KEJSARENS PAPEGOJA
  • MED MÅNGA KULÖRTA LYKTOR
• Esaias Tegnér, volumes:
  • MINDRE DIKTER
  • FRITHIOFS SAGA

Source code

Available source code by file:

• get_texts_boye.py:
  • Crawl http://www.karinboye.se/verk/dikter/dikter/ to save poems as .txt files.
  • List of poems to collect is in dikt.txt
• get_texts_tegner.py:
  • Crawl https://svenskadikter.com/ to collect texts of Frithiofs saga in JSON format to the output file tegner_frithiofs_saga_source_text.json.
  • List of poems to collect is in file tegner_saga.txt
• count_strofika.py:
  • Split poems in stanzas.
  • Split poems into individual lines.
  • Determine whether stanza pattern is regular vs. irregular.
  • Update number of syllables per line.
• count_syllabs.py:
  • Insert poems into DB.
  • Create connection between poem and author.
  • Split poems into stanzas and lines.
• add_words.py:
  • Add individual words from each poem to the database table word, including its gloss, lemma and line id it came from.
  • Stagger was used to obtain words' glosses and lemmas.

Some code (DB creation, determining genre, getting Ferlin's texts) has been lost, most of the other code I wouldn't recommend using as is. It would require significant refactoring for any future use and to adhere to better coding practices.