def test_latex_comments():
# "normal" comment
latex = 'a %x\nb\n'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'a b\n'
assert nums == [1, 2, 6, 7]
# join lines
latex = 'a%x\n b\n'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'ab\n'
assert nums == [1, 7, 8]
# join lines: protect macro name
latex = 'a\\aa%x\nb\n'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'aåb\n'
assert nums == [1, 2, 8, 9]
# 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]
# remove pure comment lines
latex = 'a %x\n %x\nb\n'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'a b\n'
assert nums == [1, 2, 10, 11]
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]
# macro plus comment
latex = 'a\n\\label{x} %x\nb\n'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'a\n b\n'
assert nums == [1, 2, 12, 16, 17]
# \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]
# (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]
def test_cite():
latex = '\\cite{x}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == '[0]'
latex = '\\cite[y]{x}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == '[0, y]'
def test_proof():
latex = '\\usepackage{amsthm}\n\\begin{proof}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == '\n\nProof.\n'
latex = '\\usepackage{amsthm}\nA \\begin{proof}[Test] B'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'A \n\nTest.\n B'
assert nums == [21, 22, 23, 23, 37, 38, 39, 40, 40, 40, 42, 43]
def test_phantom():
latex = r'A\phantom{XYZ}B\phantom{\label{lab}}C'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'A BC'
latex = r'A\hphantom{XYZ}B\hphantom{\label{lab}}C'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'A BC'
latex = r'A\vphantom{XYZ}B\vphantom{\label{lab}}C'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'ABC'
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 addpacks(cmdline):
packs = '.yalafi.shell.addpacks'
if cmdline.packages.strip(','):
packs = cmdline.packages.strip(',') + ',' + packs
opts = tex2txt.Options(defs=cmdline.define,
lang=cmdline.language[:2],
dcls=cmdline.documentclass,
pack=packs)
f = tex2txt.myopen(cmdline.add_modules, encoding=cmdline.encoding)
latex = f.read()
f.close()
tex2txt.tex2txt(latex, opts)
return documentclass[0], packages
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\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 = '\\usepackage{xcolor}\n\\textcolor\n{red}\n{blue}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'blue'
assert nums == [39, 40, 41, 42]
# no expansion: argument in next paragraph
latex = '\\usepackage{xcolor}\\textcolor\n{red}\n \n{blue}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == ' \nblue'
# no expansion: argument in next paragraph
latex = '\\usepackage{xcolor}\\textcolor\n\n{red}\n{blue}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == '\nred\nblue'
# expansion: comment line
latex = '\\usepackage{xcolor}\\textcolor\n %x\n{red}\n{blue}'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'blue'
def test_3():
plain, nums = tex2txt.tex2txt(latex_3, tex2txt.Options(pack='*'))
assert plain == plain_3
def test_verb():
latex = '\\verb?%x\\y?\\label{z}?'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == '%x\\y?'
def test_verbatim():
# extra blank lines: see LAB:VERBATIM in tex2txt.py
latex = 'A\\begin{verbatim}\\verb?%\\x?\n\\end{verbatim}B'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'A\n\n\\verb?%\\x?\n\n\nB'
def test_newthm():
plain, nums = tex2txt.tex2txt(latex_newthm, options)
assert plain == plain_newthm
def test_delimiters(latex, plain_expected):
prefix = '\\usepackage{.tests.defs}\n'
plain, nums = tex2txt.tex2txt(prefix + latex, options)
assert plain == plain_expected
def run_proofreader_options(tex, source, source_defs, language, disable,
enable, disablecategories, enablecategories,
lt_options):
t2t_options = tex2txt.Options(char=True,
repl=cmdline.replace,
defs=cmdline.define,
lang=language,
extr=cmdline.extract,
unkn=cmdline.list_unknown,
seqs=cmdline.simple_equations,
dcls=cmdline.documentclass,
pack=cmdline.packages,
nosp=cmdline.no_specials)
if cmdline.plain_input:
plain_map = {language: [(tex, list(range(1, len(tex) + 1)))]}
else:
if cmdline.list_unknown:
# only look for unknown macros and environemnts
plain, charmap = tex2txt.tex2txt(tex,
t2t_options,
source=source,
source_defs=source_defs)
return (tex, plain, charmap, [])
if cmdline.multi_language:
def mod_parms(parms):
parms.ml_continue_thresh = cmdline.ml_continue_threshold
plain_map = tex2txt.tex2txt(tex,
t2t_options,
multi_language=True,
modify_parms=mod_parms,
source=source,
source_defs=source_defs)
else:
plain, charmap = tex2txt.tex2txt(tex,
t2t_options,
source=source,
source_defs=source_defs)
plain_map = {language: [(plain, charmap)]}
disa_thresh = disable
if cmdline.ml_disable:
if disa_thresh:
disa_thresh += ','
disa_thresh += cmdline.ml_disable
disacat_thresh = disablecategories
if cmdline.ml_disablecategories:
if disacat_thresh:
disacat_thresh += ','
disacat_thresh += cmdline.ml_disablecategories
delim = '\n\n' # NB: issue #6
matches_tot = []
plain_tot = ''
charmap_tot = []
for lang in plain_map:
for plain, charmap in plain_map[lang]:
if not plain.strip():
continue
# 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:
flag = (cmdline.multi_language
and len(plain.split()) <= cmdline.ml_rule_threshold)
matches = run_languagetool(
plain, lang, disa_thresh if flag else disable, enable,
disacat_thresh if flag else disablecategories,
enablecategories, lt_options)
matches += checks.create_single_letter_matches(plain, cmdline)
matches += checks.create_equation_punct_messages(
plain, cmdline, equation_replacements_display,
equation_replacements_inline, equation_replacements)
for m in matches:
m['offset'] = json_get(m, 'offset', int) + len(plain_tot)
matches_tot += matches
plain_tot += plain
charmap_tot += charmap
plain_tot += delim
charmap_tot += [charmap_tot[-1]] * len(delim)
# sort matches according to position in LaTeX text
#
def f(m):
beg = json_get(m, 'offset', int)
if beg < 0 or beg >= len(charmap_tot):
tex2txt.fatal('run_proofreader():' +
' bad message read from proofreader')
return abs(charmap_tot[beg])
matches_tot.sort(key=f)
return (tex, plain_tot, charmap_tot, matches_tot)
# does file name match regex from option --skip?
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, source=f, source_defs=source_defs)
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()
# helpers for robust JSON evaluation
#
json_decoder = json.JSONDecoder()
def test_6():
plain, nums = tex2txt.tex2txt(latex_6, tex2txt.Options())
assert plain == plain_6
def test_1():
plain, nums = tex2txt.tex2txt(latex_1, options)
assert plain_1 == plain
def test_system_macros():
latex = 'a\\footnote[2]{x}b'
plain, nums = tex2txt.tex2txt(latex, options)
assert plain == 'ab\n\n\nx\n'
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()
# helpers for robust JSON evaluation
#
json_decoder = json.JSONDecoder()
def test_2():
plain, nums = tex2txt.tex2txt(latex_2, options)
assert plain_2 == plain
U_\epsilon(x)
&\subset M \quad\text{for all } x \in \Omega, \notag \\
f(x) % LINE 11
&> 0 \quad\text{for all}\ x \in \Omega \label{l1} \\
f(x)
&= 0 \quad\text{for all} x \in M \setminus \Omega. \label{l2}
\end{align}
"""
plain_t = r"""
We consider
a set C-C-C, a domain D-D-D, andx
a function E-E-E.
With a constant F-F-F, we require
V-V-V equal W-W-W for all X-X-X,
Y-Y-Y equal Z-Z-Z for all U-U-U
U-U-U equal V-V-V for allW-W-W.
Thix is a footnote.
"""
options = tex2txt.Options(lang='en', char=True)
plain, nums = tex2txt.tex2txt(latex, options)
def test_text():
assert plain == plain_t
def test_4():
plain, nums = tex2txt.tex2txt(latex_4, tex2txt.Options(pack='.tests.defs'))
assert plain == plain_4
