lyx_mirror/development/autotests/invertedTests
Günter Milde 40240ef6a0 Improve error message and ctests for input encoding cp858.
Encoding cp858 supported by only some iconv variants.
Most users will want to change their "encoding" setting instead
of installing/recompiling "iconv" to support this legacy encoding.

ctests are likely will fail with either "vanilla" or "enhanced"
iconv and test a situation that is unlikely to change generally,
so we ignore this test now by default.
2019-05-09 09:28:37 +02:00

363 lines
16 KiB
Plaintext

# Regular expressions for tests that are known to fail
#
# Patterns will be automatically enclosed with '^' and '$'.
# Matching tests get the label "inverted"
# or (if also matching a pattern in "suspendedTests") "suspended".
# They get also the test-property 'WILL_FAIL', i.e.
# they are reported as failing if the export works without error.
# https://cmake.org/cmake/help/v3.0/prop_test/WILL_FAIL.html
# ======================================================================
Sublabel: todo
# LyX bugs and problems with the sample documents that require attention
# ======================================================================
# * minor issues to explore and properly sort,
# * easyfix issues,
# * LyX bugs to report at trac (move pattern to section "lyxbugs" once done).
# missing glyphs: There is no ^^A in font [lmroman12-regular]
# and all the line down to ^^Z and beyond...
# XeTeX artifact? works with LuaTeX, explore:
export/doc/(|de/|es/|fr/)EmbeddedObjects_pdf4_systemF
# iconv fails, if a nomenclature inset contains an uncodable character
export/export/latex/iconv-error-nomenclature_pdf2
# `acmart` document class:
# (see also lyxbugs below and ignoredTests)
# issues with listsof, reported to the maintainer
# https://www.mail-archive.com/search?l=mid&q=of84gj%24l6b%241%40blaine.gmane.org
# LyX gives the following terminal message:
#
# TextClass.cpp (1385): The layout does not provide a list command for
# the float `sidebar'. LyX will not be able to produce a float list.
#
# which makes "check_load" and "lyx2lyx" fail (lyx2lyx for < 2.3 is ignored).
check_load/templates/Articles/Association_for_Computing_Machinery_.*
export/templates/Articles/Association_for_Computing_Machinery_.*_lyx23
# Failing LyX2LyX conversions for old versions:
#
# "beamer" lyx2lyx fails sometimes for versions < 2.1
# Back-conversion results in some layouts without name ("\begin_layout \n")
# which trigger a warning message leading to test failure.
export/examples/(|de/|fr/|ja/)Presentations/Beamer_lyx(16|20)
export/examples/Presentations/Beamer_.*Complex.*_lyx(16|20)
export/templates/(|de/|es/|fr/|ja/)Presentations/Beamer_lyx(16|20)
#
# modernCV fails for versions < 2.1
# * reversed order of \end_layout and \end_inset results in corrupt LyX file
# * reversed order of command arguments results in incompilable LaTeX
export/examples/(|es/)Curricula_Vitae/Modern_CV_(lyx(16|20)|dvi|dvi3.*F|pdf|pdf3)
export/examples/(|es/)Curricula_Vitae/Europe_CV_(dvi|dvi3.*F|pdf|pdf3)
#
# back-conversion fails for rotated float inset in a list
export/export/lyx2lyx/lyx_2_3_test_lyx(16|20|21|22)
# Korean legacy encoding fails (missing fonts)
export/export/latex/CJK/ko_default_pdf2
# New problem with TL18: endless loop?
# ! TeX capacity exceeded, sorry [input stack +size=5000].
export/export/latex/luainputenc-utf8_pdf5_texF
# nonstandard tests failing for unknown reason:
export/templates/Obsolete/ACM_SIGGGRAPH.*_pdf5_(texF|systemF)
export/templates/Articles/International_Union_of_Crystallography.*_pdf4_systemF
export/templates/Articles/R_Journal_(dvi3|pdf[45])_systemF
#export/templates/Articles/Kluwer_.*_systemF
# Babel with non-TeX fonts fails for some supported languages
# Lithuanian: L7x font encoding also set with LuaTeX and non-TeX fonts
# (error but output OK)
export/export/latex/languages/supported-languages_babel_non-tex_(dvi3|pdf5)_systemF
# Thai with CJK package requires the babel option "thaicjk" instead of "thai".
# (see thaifont.txt in the CJK documentation for setup details).
export/export/latex/unicodesymbols/031-thai_th_(euc-|gbk|jis|utf8-cjk).*
# Package clashes when Japanese documents are compiled with non-TeX fonts,
#
# The textclass-loaded jsclasses/okumacro.sty needs format `pLaTeX2e'
# (okumacro provides \ruby support).
export/doc/ja/(Additional|LaTeXConfig|Math|UserGuide).*_systemF
export/doc/ja/(Formula-numbering|Customization|EmbeddedObjects|Intro)_pdf4_systemF
export/doc/ja/Tutorial_(dvi3|pdf5)_systemF
# Test for command \pdfsave fails:
# ! Package xypdf Error: pdfTeX version 1.40.0 or higher is needed for the xypdf
export/examples/ja/Graphics_and_Insets/XY-Pic.*_systemF
# Document class js*.cls requires platex <-> Xe/LuaTeX, fixed with "bxjs*.cls"
# ================================================
Sublabel: lyxbugs
# LyX bugs with a Trac number.
# ================================================
#6197: „amssymb” with Polish babel name clash \lll ("<<<" vs. "ł")
# Wrong output with pdflatex (-> unreliableTests),
# missing glyph with XeTeX/LuaTeX and Babel (OK with Polyglossia)
export/export/latex/6197-polish-amssymb_pdf._texF
#8035 work around a Polyglossia bug:
# An error is reported if the document font does not announce support for a
# script (Cyrillic, Greek, ...) which Polyglossia thinks is required to
# typeset the document language. This is not required with LyX's check for
# missing glyphs and results in many false positives:
# a) with fonts including the required characters but
# not the tag tested by Polyglossia,
# b) if the document language is "Serbian (Latin Script)",
# the Cyrillic script is not required yet Polyglossia tests for support.
# The document "sr/Braille.lyx" is an example of case b).
# (For some reason, the false positive error is not thrown with PDF (luatex).)
export/examples/sr/Modules/Braille_(dvi3|pdf4)_systemF
export/export/latex/language-in-tables_.*_systemF
#9633 Nested language handling causes compilation failures with polyglossia
export/export/latex/languagenesting-quote-branch_.*_systemF
#9871 LyX sends invalid Unicode to iconv when converting to ASCII
# most probably due to BabelPreamble code (language specific headings for
# theorems, problems , ... are written in the language's default encoding
# if they contain non-ASCII characters)
# Error 84 returned from iconv when converting from UCS-4LE to ascii:
# Invalid or incomplete multibyte or wide character.
# see also the discussion at:
# https://www.mail-archive.com/search?l=mid&q=20170228190738.cgs3tl3ny2bxiegk%40steph
export/export/latex/bug9871-slovak-ascii-preambleencoding_pdf2
#10355 xmllint detects failures
export/doc/attic/eu_UserGuide_xhtml
export/doc/(es|ja)/UserGuide_xhtml
export/examples/External_Material/Spreadsheet_xhtml
export/export/WrongDfnTagHandling_xhtml
#10371 Compilation error if longest label marker in labeling style
# contains underscores etc.
export/export/latex/LongestLabelWithUnderscore.*
#11455 "Unable to process argument!" with multiple 1.6.x roundtrips
export/templates/Articles/IEEE_Transactions_Journal_lyx16
#11520 "listerrors" fails with pdflatex, lualatex, and xelatex
# Example for lib/scripts/listerrors
# Requires noweb. Seems to be restricted to plain "latex".
# Find out why and whether the script could be improved.
export/examples/Modules/Noweb_Listerrors_(pdf[245]).*
#11521 Font encoding "LAE" required with farsi.
# The Farsi font encoding LAE must be added also to a custom font encoding
# (similar to the Greek and Cyrillic font encodings LGR and T2A).
export/export/latex/languages/fa-OT1_pdf2
#11522 Support specification of complementing non-TeX fonts
#
# More than one non-TeX font required for the character set in many documents
# to prevent "missing glyph" errors with non-TeX fonts
#
export/export/latex/.*inputenc-.*_systemF
export/doc/(|de/|es/|fr/)Math.*systemF
# minimal example for Math.lyx:
export/export/latex/systemfonts-Math-missingchars.*systemF
#
# Farsi: missing glyphs (em space and "English" quotes)
export/examples/fa/Welcome_pdf4_systemF
#
# CJK-quotes examples (missing in most Latin-script fonts)
export/doc/(|de/|es/|fr/)UserGuide_(dvi3|pdf4|pdf5)_systemF
#
# Japanese
export/examples/ja/Modules/Linguistics.*systemF
#
# Arabic (see also "texissues")
# "! Extra \else" and missing glyph "no ␣ in font Scheherazade":
# (the lyx2lyx test also tests compilation after round-trip)
export/doc/ar/UserGuide_pdf4_systemF
export/doc/ar/UserGuide_lyx(22|23)
#
# Juergen Spitzmueller: They do not have problems with xunicode, but without.
# The reason is that the asterisk glyph is not included in the arabic font
# (Scheherazade), but apparently, xunicode defines an extra asterisk
# glyph (for a different purpose) that is being used here. Clearly the
# problem should be fixed somewhere else.
# Günter Milde: works in TeXLive17 (Debian/stretch)
# export/doc/ar/Intro_pdf4_systemF
#11532 inputencoding desynchronisation
export/export/latex/misplaced-inputenc-switch_pdf2
# ==============================================================
Sublabel: ert
# Export failures due to "raw" LaTeX use in ERT or preamble code
# ==============================================================
#
# "Wontfix" if demonstrating correct use and OK in the default output format.
# ================================================
Sublabel: texissues
# Export fails due to LaTeX limitations,
# ================================================
#
# e.g. non-ASCII in verbatim or listings, incompatible packages, ...
# "Wontfix" if documents demonstrate correct use in the default output format.
# Problem with Czech and Slovak Babel support.
# (reported per mail to Petr Tesa\v{r}\'ik, 2016-09-06)
# All exports using babel fail.
# Without the Czech or Slovak text it compiles
# No problem with inserted German (or other foreign language) text
# No problem if the table is Slovak, too
export/export/latex/TableErrorIfSlovakTextInserted_(dvi|pdf[23]?|.*_texF)
# Xe/LuaTeX
###########
# Babel-Russian uses UTF-8 for auto-strings if it detects Xe/LuaTeX.
# This fails unless the inputencoding is set to utf-8, too.
export/export/latex/languages/ru_auto-legacy-luatex.*_texF
# Babel-Ukrainian uses UTF-8 if it detects Xe/LuaTeX (for "\date").
# With 8-bit TeX fonts this results in:
# * garbled output with PDF (LuaTeX)
# * missing glyphs with XeTeX and DVI (LuaTeX)
# New in TL18 -> cf. unreliableTests.varying_versions.
# Workaround: suppressing auto-date, setting inputenc to "utf8", or using
# date inset (new in LyX 2.4) in the source documents.
export/export/latex/languages/uk-babel-date-missing-chars-with-ascii_(pdf4|dvi3)_texF
# the "dvi3" tests for these pass, but have wrong output.
export/doc/uk/Intro_pdf4_texF
export/examples/uk/Welcome_pdf4_texF
# Some packages fail with XeTeX/LuaTeX
# Reported upstream (https://www.mail-archive.com/lyx-devel@lists.lyx.org/msg184435.html)
# Is there a new version fixing the problem?
export/examples/Articles/Astronomy_%26_Astrophysics_(dvi3|pdf[45])_.*
export/templates/Articles/Astronomy_%26_Astrophysics_(dvi3|pdf[45])_.*
# Gives the following error:
# # ! Package hpstatement Error: H or P statement "H251"
# # (hpstatement) is not defined.
# seems to be polyglossia-related (dvi3 worked with babel)
export/examples/Modules/Hazard_and_Precautionary_Statements_.*_systemF
# These files use packages that are not compatible with the newest LuaTeX. We
# could potentially fix these by loading the luatex85 package early in the
# LyX-generated preamble, but it seems it's better to wait until the packages
# fix the root issue themselves.
# xy package:
export/examples/(|fr/)Graphics_and_Insets/XY-(Pic|Figure)_pdf5.*
# Polyglossia + LuaTeX issue
# Reported upstream here:
# https://github.com/reutenauer/polyglossia/issues/163
# These tests were broken at 3374b854
# See discussion here:
# https://www.mail-archive.com/search?l=mid&q=1478004223.2367.5.camel%40lyx.org
export/doc/(sk|nb)/Intro_(dvi3|pdf5)_systemF
# some Polyglossia languages work only with XeTeX and fail with LuaTeX:
# * The font "NotoSansEthiopic" cannot be found.
# * sanskrit: ! Undefined control sequence.
# l.53 \newXeTeXintercharclass
# * RTL-languages: ! Undefined control sequence.
# \setLR ->\@RTLfalse \luatexpardir
export/export/latex/languages/supported-languages_polyglossia_(dvi3|pdf5)_systemF
# fontspec errors with `slides` class:
# LaTeX error: "kernel/command-already-defined"
# Lines 2684 ... 2690 in fontspec-xetex.sty
# Lines 2608 ... 2614 in fontspec-luatex.sty
export/templates/Presentations/Slides_.*_systemF
# Babel-French fails with Xe/LuaTeX and TeX fonts
# LaTeX Error: Command \at already defined. (and a lot more)
export/export/latex/languages/supported-languages_(dvi|pdf).*_texF
export/export/latex/languages/supported-languages_babel_(dvi|pdf).*_texF
# Babel-Spanish fails sometimes with XeTeX and TeX fonts
export/doc/es/(Math|UserGuide|EmbeddedObjects|Customization)_pdf4_texF
# Arabic
#
# Larger Arabic documents currently have non-TeX fonts and XeTeX
# as default output. To fix pdfTeX export:
# - set all non-Arab text parts to a different language (e.g. English)
# - define alternatives for hyperref (e.g. with textorpdfstring)
# for correct handling of hyperref.
# - solve problem with ToC:
# ! This can't happen (copying).
# \AtBegShi@Output ...ipout \box \AtBeginShipoutBox
# \fi \fi
# l.62 \tableofcontents
# {}
# I'm broken. Please show this to someone who can fix can fix
#
# see discussion at https://www.mail-archive.com/search?l=mid&q=20161114035123.s5lt4ib4x4obtptp%40steph
#
# Babel-Arabic loads "inputenc" which fails with LuaTeX
export/.*/ar/.*_(dvi3|pdf5)_texF
# no Arab charaters with luainputenc
export/export/latex/languages/supported-languages_(dvi3|pdf5)_texF
export/export/latex/languages/supported-languages_babel_auto-legacy_(dvi3|pdf5)_texF
# Farsi
# "invalid utf-8 sequence" with Babel-Farsi, LuaTeX, and TeX fonts.
# (XeTeX works, if inputenc is set to "utf8")
export/examples/fa/Welcome_(dvi3|pdf5)_texF
# Hebrew
# babel-hebrew loads "rlbabel.def" which fails with LuaTeX + TeX fonts:
# "! Right-to-Left Support Error: use TeX--XeT or e-TeX engine."
# (XeTeX works, if inputenc is set to "utf8")
export/.*/he/.*(dvi3|pdf5)_texF
export/export/latex/languages/he-.*(dvi3|pdf5)_texF
export/export/latex/languages/supported-languages_(dvi3|pdf5)_texF
export/export/latex/languages/supported-languages_babel_(dvi3|pdf5)_texF
# Polyglossia bug with LuaTeX and LTR languages (Arab, Hebrew and Farsi):
# Undefined control sequences "\luatexpardir" and "\luatextextdir".
# https://github.com/reutenauer/polyglossia/commit/ccb0e9e2c6411170ad779b05ff5076f1193cc323
export/.*/ar/.*(dvi3|pdf5)_systemF
export/.*/fa/Welcome_(dvi3|pdf5)_systemF
# terminal message: "Language hebrew not found in language.dat.lua"
export/.*/he/.*(dvi3|pdf5)_systemF
export/export/latex/languages/supported-languages_(dvi3|pdf5)_systemF
# Ukrainian
# See comment in unreliableTests
export/export/latex/languages/uk-babel-date-missing-chars-with-ascii_pdf5_texF
# ================================================
Sublabel: externalissues
# Export fails due to non-LaTeX external tool,
# ================================================
#
# e.g. a bug or missing feature in ps2pdf
# ps2pdf gives the following and exits with error:
# Error: /invalidfileaccess in --file--
# ps2pdf succeeds if the -dNOSAFER flag is used. For discussion, see:
# https://www.mail-archive.com/search?l=mid&q=1506851183.3130.15.camel%40lyx.org
export/examples/ja/Modules/LilyPond_Book_pdf
# ======================================================================
Sublabel: attic
# Documents in the attic, kept for reference and format conversion test.
# However, many problems here are "wontfix".
# ======================================================================
# Fails because of missing .png graphic files.
# They were replaced by .svgz, but, as this file is in the attic, we do not
# convert to new lyx format. Plain latex and xetex create output (ignoring
# missing files) while (pdf|lua)latex exits with != 0
#export/doc/attic/eu_UserGuide_pdf[25].*
# Files in the attic with non-default output
# (i.e. could be ERT, package incompatiblity, ...)
export/doc/attic/it_(UserGuide_dvi3|UserGuide_pdf4)_systemF
export/doc/attic/sk_UserGuide_pdf4_texF
export/doc/attic/id_UserGuide_.*systemF