notes

format xml using xmllint (from libxml2-utils package):
xmllint --format filename.xml

zasady techniki prawodawczej
http://isap.sejm.gov.pl/DetailsServlet?id=WDU20160000283
http://www.arslege.pl/zasady-techniki-prawodawczej/k459/s4931/

szablon http://rcl.gov.pl/szablon

co to jest tiret:
http://jezyk-polski.pl/wp-content/download/art_witorska_z-4-2014.pdf

https://forum.kodujdlapolski.pl/t/repozytorium-prawne-w-gicie/2836/22

Super system aktow prawnych z rep. poludniowej afryki:
https://openbylaws.org.za/about.html

wywiad z dr Adamem Zadroznym (graf powiazan w prawie polskim)
http://audycje.tokfm.pl/podcast/Konstytucja-jak-meduza-a-ustawa-o-vacie-jak-dmuchawiec/47709

whitepapers:

Grece, parsing from plain text, ~96% success (only general info)
Mario Koniaris et al: Towards Automatic and Semantic Indexing of Legal Documents
http://www.dbnet.ntua.gr/~mkoniari/docs/KoniarisPCI16.pdf

Italy, parsing using hidden markov models from plain text (only general information)
Lorenzo Bacci et al: Automatic mark-up of legislative documents and its application to parallel text generation
http://ceur-ws.org/Vol-465/paper6.pdf

schemas
akoma ntoso (Republic of South Africa and others)
MetaLex metalex.eu (dutch and uk legislation)

other projects
Agora-Lex - belgium


Furthest point parsed to in pit act
page 57 (take 80) punkt '<82a' spans to subsequent page, 46sec

2017-07-04: page 85 (take 233) 3m38s
2017-07-06: page 85 (take 233) 3m12s (marginally better when working on a pruned xml)
2017-07-08: all except (take 233) annexes 3mm9s on pruned xml
            parsing all pages results in an error: no such cell (6,2)
2017-07-12: all parsed in 3m9s on pruned xml, without content inside anexes


grant for dannis legislative project from Google DNI
https://digitalnewsinitiative.com/dni-projects/poliwatchbot/