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
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
from yalafi.shell import addpacks
dcls, packs = addpacks.addpacks(cmdline, source_defs)
if dcls:
cmdline.documentclass = dcls
if cmdline.packages:
packs += cmdline.packages.split(',')
cmdline.packages = ','.join(packs)
# on option --include: add included files to work list
# otherwise: remove duplicates
#
if cmdline.include:
sys.stderr.write('=== checking for file inclusions ... ')
sys.stderr.flush()
opts = tex2txt.Options(extr=inclusion_macros, repl=cmdline.replace,
defs=cmdline.define, lang=cmdline.language[:2],
dcls=cmdline.documentclass, pack=cmdline.packages,
nosp=cmdline.no_specials)
def skip_file(fn):
# 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
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)
def test_6():
plain, nums = tex2txt.tex2txt(latex_6, tex2txt.Options())
assert plain == plain_6
def test_4():
plain, nums = tex2txt.tex2txt(latex_4, tex2txt.Options(pack='.tests.defs'))
assert plain == plain_4
def test_3():
plain, nums = tex2txt.tex2txt(latex_3, tex2txt.Options(pack='*'))
assert plain == plain_3
#
# - test of option --unkn
#
from yalafi import tex2txt
options = tex2txt.Options(unkn=True)
latex_1 = r"""
A
\newcommand{\zzz}[1]{#1#1}
\xxx
\begin{\zzz Y}
B
"""
plain_1 = r"""\xxx
YY
"""
def test_1():
plain, nums = tex2txt.tex2txt(latex_1, options)
assert plain_1 == plain