def test_latex_comments():
# "normal" comment
latex = 'a %x\nb\n'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'a \nb\n'
assert nums == [1, 2, 5, 6, 7, 8]
# join lines
latex = 'a%x\n b\n'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'ab\n'
assert nums == [1, 7, 8, 9]
# join lines: protect macro name
latex = 'a\\aa%x\nb\n'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'aƄb\n'
assert nums == [1, -2, 8, 9, 10]
# do not join lines, if next line empty
latex = 'a%x\n\nb\n'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'a\n\nb\n'
assert nums == [1, 4, 5, 6, 7, 8]
# remove pure comment lines
latex = 'a %x\n %x\nb\n'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'a \nb\n'
assert nums == [1, 2, 5, 10, 11, 12]
def test_remove_blank_lines_left_by_macros():
# a normal macro: \label
latex = 'a\n\\label{x}\nb\n'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'a\nb\n'
assert nums == [1, 2, 13, 14, 15]
# macro plus comment
latex = 'a\n\\label{x} %x\nb\n'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'a\nb\n'
assert nums == [1, 2, 16, 17, 18]
# \begin and \end
latex = 'a\n\\begin{x}\nb\n\\end{x}\nc\n'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'a\nb\nc\n'
assert nums == [1, 2, 13, 14, 23, 24, 25]
# (actually no blank lines)
latex = 'a\n\\begin{x}b\n\\end{x}c\n'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'a\nb\nc\n'
assert nums == [1, 2, 12, 13, 21, 22, 23]
def test_nested_table():
latex = '\\begin{table}A\\begin{table}B\\end{table}C\\end{table}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == '[Tabelle].'
latex = '\\begin{table}A\\begin{tablx}B\\end{table}C\\end{table}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == '[Tabelle].C'
def test_cite():
latex = '\\cite{x}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == '[1]'
latex = '\\cite[y]{x}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == '[1, y]'
def test_proof():
latex = '\\begin{proof}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'Proof.'
latex = 'A \\begin{proof}[Test] B'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'A Test. B'
assert nums == [1, 2, 17, 18, 19, 20, 21, 22, 23, 24]
def test_unknown_macro():
latex = 'a\\xxx b'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'ab'
latex = 'a\\xxx \n{b} c'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'ab c'
latex = 'a\\xxx\n\n{b} c'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'a\n\nb c'
def test_renewcommand():
latex = 'a\\newcommand\n{\\x}\n{y}b'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'ab'
latex = 'a\\newcommand{\\x}[1][z]{y}b'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'ab'
latex = 'a\\newcommand*{\\x}[1][z]{y}b'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'ab'
latex = 'a\\renewcommand{\\x}[1][z]{y}b'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'ab'
latex = 'a\\newcommand\n\n{\\x}[1]{y}b'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'a\n\n[1]yb'
latex = 'a\\newcommand{\\x}\n\n[1]{y}b'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'a\n\n[1]yb'
latex = 'a\\newcommand{\\x}[1]\n\n{y}b'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'a[1]\n\nyb'
def test_footnotemark():
latex = 'a\\footnotemark b'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'ab'
latex = 'a\\footnotemark[1] b'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'a b'
latex = 'a\\footnotemark\n[1]b'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'ab'
# do not cross paragraph border
latex = '\\footnotemark\n\n[1]'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == '\n[1]'
def test_macro_arguments():
# normal expansion
latex = '\\textcolor\n{red}\n{blue}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'blue'
assert nums == [19, 20, 21, 22, 24]
# no expansion: argument in next paragraph
latex = '\\textcolor\n{red}\n \n{blue}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'red\n \nblue'
# no expansion: argument in next paragraph
latex = '\\textcolor\n\n{red}\n{blue}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == '\nred\nblue'
# expansion: comment line
latex = '\\textcolor\n %x\n{red}\n{blue}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'blue'
def run_proofreader(file):
sys.stderr.write('=== ' + file + '\n')
sys.stderr.flush()
f = tex2txt.myopen(file, encoding=cmdline.encoding)
tex = f.read()
f.close()
if not tex.endswith('\n'):
tex += '\n'
if cmdline.plain:
(plain, charmap) = (tex, list(range(1, len(tex) + 1)))
else:
(plain, charmap) = tex2txt.tex2txt(tex, options)
if cmdline.list_unknown:
# only look for unknown macros and environemnts
return (tex, plain, charmap, [])
# here, we could dispatch to other tools, see for instance
# - https://textgears.com/api
# - Python package prowritingaid.python
#
if cmdline.textgears:
matches = run_textgears(plain)
else:
matches = run_languagetool(plain)
matches += create_single_letter_matches(plain)
matches += create_equation_punct_messages(plain)
# sort matches according to position in LaTeX text
#
def f(m):
beg = json_get(m, 'offset', int)
if beg < 0 or beg >= len(charmap):
tex2txt.fatal('run_proofreader():' +
' bad message read from proofreader')
return abs(charmap[beg])
matches.sort(key=f)
return (tex, plain, charmap, matches)
def test_system_macros():
latex = 'a\\footnote[2]{x}b'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'ab\n\n\nx\n'
def test_nested_macro():
latex = '\\textcolor{x}{\\textcolor{y}{z}}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'z'
return cmdline.skip and re.search(r'\A' + cmdline.skip + r'\Z', fn)
todo = cmdline.file
done = []
while todo:
f = todo.pop(0)
if f in done or skip_file(f):
continue
done.append(f)
if not cmdline.include:
continue
fp = tex2txt.myopen(f, encoding=cmdline.encoding)
tex = fp.read()
fp.close()
(plain, _) = tex2txt.tex2txt(tex, opts)
for f in plain.split():
if not f.endswith('.tex'):
f += '.tex'
if f not in done + todo and not skip_file(f):
todo.append(f)
cmdline.file = done
if cmdline.include:
sys.stderr.write(', '.join(cmdline.file) + '\n')
sys.stderr.flush()
# prepare options for tex2txt()
#
options = tex2txt.Options(char=True,
repl=cmdline.replace,
def test_verbatim():
# extra blank lines: see LAB:VERBATIM in tex2txt.py
latex = '\\begin{verbatim}\\verb?%\\x?\n\\end{verbatim}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == '\n\n\\verb?%\\x?\n\n'
# encoding=input_encoding),
# defs=tex2txt.read_definitions('Tools/LT/defs.py',
# encoding='utf-8'),
lang='en')
for file in sys.argv[1:]:
sys.stderr.write('=== ' + file + '\n')
sys.stderr.flush()
# read file and call tex2txt()
#
f = tex2txt.myopen(file, encoding=input_encoding)
tex = f.read()
f.close()
(plain, charmap) = tex2txt.tex2txt(tex, options)
starts = tex2txt.get_line_starts(plain)
# call LanguageTool
#
out = subprocess.run(ltcmd,
input=plain.encode(encoding='utf-8'),
stdout=subprocess.PIPE)
if 'Windows' in os.getenv('OS', default=''):
# under Windows, LanguageTool produces Latin-1 output
msg = out.stdout.decode(encoding='latin-1')
else:
msg = out.stdout.decode(encoding='utf-8')
lines = msg.splitlines(keepends=True)
if lines:
\textcolor{red}{redx colour.}}
is lazy.
"""
plain_t = r"""
Only few people
is lazy.
We use
redx colour.
"""
nums_t = [
1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 64, 65, 66, 67, 68,
69, 70, 71, 72, 73, 27, 27, 27, 27, 28, 29, 30, 31, 32, 33, 50, 51, 52, 53,
54, 55, 56, 57, 58, 59, 60, 61, 63, 74
]
options = tex2txt.Options(lang='en', char=True)
plain, nums = tex2txt.tex2txt(latex, options)
def test_text():
assert plain == plain_t
def test_nums():
assert nums == nums_t
def test_verb():
latex = '\\verb?%x\\y?\\label{z}?'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == '%x\\y?'