UniTO/tesi/conv.py

61 lines
2.2 KiB
Python
Raw Normal View History

2020-02-21 19:13:13 +01:00
import json
2020-02-24 14:36:26 +01:00
import re
2020-02-21 19:13:13 +01:00
from sys import argv
allsymbols = json.load(open('./unicode-latex.json'))
2020-04-05 20:33:38 +02:00
mysymbols = ['', '', '', '', '', '', '', '', '', '', 'ε','', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', 'ʲ', '', 'π', 'α', 'β', '', 'σ', '', '', '', '', '', '', '', '', '' ]
2020-04-02 14:14:39 +02:00
extrasymbols = {'': '\llbracket', '': '\rrbracket'}
2020-02-21 19:13:13 +01:00
2020-02-24 14:36:26 +01:00
symbols = {s: allsymbols[s] for s in mysymbols}
2020-04-02 14:14:39 +02:00
symbols.update(extrasymbols)
mathsymbols = {s: '$'+v+'$' for s, v in symbols.items()}
2020-02-21 19:13:13 +01:00
def read_by_char(fname):
2020-02-24 14:36:26 +01:00
# Yield character and True/False if inside mathmode block
mathmode = False
mathmode_begin = set(['\\begin{equation*}', '\\begin{equation}'])
mathmode_end = set(['\\end{equation*}', '\\end{equation}'])
cnt = 0
2020-02-21 19:13:13 +01:00
with open(fname, 'r') as fp:
for line in fp.readlines():
2020-02-24 14:36:26 +01:00
cnt += 1
words = [w.strip() for w in line.split(' ')]
2020-02-24 19:46:00 +01:00
2020-02-24 14:36:26 +01:00
if mathmode_begin.intersection(words):
assert mathmode == False
mathmode = True
2020-02-24 19:46:00 +01:00
if mathmode_end.intersection(words):
2020-02-24 14:36:26 +01:00
assert mathmode == True, f'Line: {words}, number: {cnt}'
mathmode = False
2020-02-21 19:13:13 +01:00
for ch in line:
2020-02-24 14:36:26 +01:00
yield ch, mathmode
def convert(ch, mathmode):
if not mathmode:
return mathsymbols[ch] if ch in mathsymbols else ch
else:
return symbols[ch] if ch in symbols else ch
2020-02-21 19:13:13 +01:00
2020-03-02 14:46:37 +01:00
def latex_errors_replacements(charlist):
text = ''.join(charlist).split(' ')
replacements = {'\n\end{comment}\n\end{enumerate}\n\end{enumerate}\n\n\subsection{Symbolic':
'\n\end{comment}\n\n\subsection{Symbolic'}
r_set = set(replacements.keys())
for word in text:
it = r_set.intersection(set([word]))
if it:
yield from replacements[it.pop()]
else:
yield from word
yield ' '
2020-02-24 14:36:26 +01:00
# convert symbols except the one requiring math mode modifiers
2020-02-24 14:37:50 +01:00
firstpass = [convert(*c) for c in read_by_char(argv[1])]
2020-03-02 14:46:37 +01:00
# remove a latex error
secondpass = latex_errors_replacements(firstpass)
2020-02-21 19:13:13 +01:00
2020-03-02 14:46:37 +01:00
newfile = ''.join(secondpass)
2020-02-21 19:13:13 +01:00
with open(argv[2], 'w') as f:
2020-02-24 14:36:26 +01:00
f.write(newfile)