- Directory statement/ contains generated latex for problems in VOJ
- Can we parse tex files into legend, input, output and sample test data? (Done)
- I (t-rekttt) decided to run an analyze on the statements's subsections (by counting how much of
subsubsection
inside each statement) and saw that they have this kind of distribution (in the format of[count, frequency]
).
By examination (subsectionsExtractor.py
) I saw that most of the statement have a typical format of 3 subsections
, beautifully ordered by the format: statement, input, output, notes. So I wrote an extractor for that specific case (inside statementExtractor.py
).
- Done extractor for 3+ subsections. Now we need to write the extractor for the others as well.
- We need to extract sample tests from the extracted statements. Note that many of them lie within
\begin{verbatim}
and\end{verbatim}
(Done, need to verify) - VOJ tests downloader (Working)
- Place your statements into
statement
folder - Run
pip3 install -r requirements.txt
- Run
python3 subsectionsExtractor.py
to extract information about statements based on subsections - Run
python3 statementExtractor.py
to extract statements into parts Extracted statements are saved inextracted
folder. Each statement would be broken into 4 parts:statement.tex
,input.tex
,output.tex
,notes.tex
, the same format as Codeforces Polygon
- Place your VOJ problem setter accounts into
accounts.csv
, one per line - Run
python3 testCrawler.py