Cmake export tests. Add validation of lyxhtml exported files

ATM, the validation is commented out.
To use it, one has first to install html5validator with
  $ pip3 install html5validator>=0.4.2
and uncomment the line 335 in export.cmake

diff --git a/development/autotests/ExportTests.cmake b/development/autotests/ExportTests.cmake
index 60d813b48d7db39d6e2cdf7a8b68316d81c00dfd..e0b31ff9533b4162169807d9b28676e207cf85b4 100644 (file)
--- a/development/autotests/ExportTests.cmake
+++ b/development/autotests/ExportTests.cmake
@@ -362,6 +362,7 @@ foreach(libsubfolderx autotests/export lib/doc lib/examples lib/templates lib/ta
           -Dinverted=${inverted}
           -DTOP_SRC_DIR=${TOP_SRC_DIR}
           -DPERL_EXECUTABLE=${PERL_EXECUTABLE}
+          -DLYX_PYTHON_EXECUTABLE=${LYX_PYTHON_EXECUTABLE}
           -P "${TOP_SRC_DIR}/development/autotests/export.cmake")
         setmarkedtestlabel(${TestName} ${mytestlabel})
       endif()
@@ -460,11 +461,11 @@ foreach(libsubfolderx autotests/export lib/doc lib/examples lib/templates lib/ta
           endif()
           string(REGEX REPLACE "[\\(\\)]" "_" TestName "${TestName1}")
           maketestname(TestName inverted invertedTests ignoredTests unreliableTests mytestlabel)
-         if (format MATCHES "docbook5")
-           set(f_extension "xml")
-         else()
-           set(f_extension ${format})
-         endif()
+          if (format MATCHES "docbook5")
+            set(f_extension "xml")
+          else()
+            set(f_extension ${format})
+          endif()
           if(TestName)
             add_test(NAME ${TestName}
               WORKING_DIRECTORY "${CMAKE_CURRENT_BINARY_DIR}/${LYX_HOME}"
@@ -483,6 +484,7 @@ foreach(libsubfolderx autotests/export lib/doc lib/examples lib/templates lib/ta
               -DXMLLINT_EXECUTABLE=${XMLLINT_EXECUTABLE}
               -DJAVA_EXECUTABLE=${jingjava}
               -DENCODING=${_enc2}
+              -DLYX_PYTHON_EXECUTABLE=${LYX_PYTHON_EXECUTABLE}
               -P "${TOP_SRC_DIR}/development/autotests/export.cmake")
             setmarkedtestlabel(${TestName} ${mytestlabel}) # check for suspended pdf/dvi exports
           endif()

diff --git a/development/autotests/export.cmake b/development/autotests/export.cmake
index 43d6df8166b4eabfd644a3b4e68d86299e50b187..0185d5b39d6e1ab6916b2d0b8d5dac327c4fa295 100755 (executable)
--- a/development/autotests/export.cmake
+++ b/development/autotests/export.cmake
@@ -26,6 +26,7 @@
 #       -DTOP_SRC_DIR=${TOP_SRC_DIR} \
 #       -DIgnoreErrorMessage=(ON/OFF) \
 #       -DPERL_EXECUTABLE=${PERL_EXECUTABLE} \
+#       -DLYX_PYTHON_EXECUTABLE=${LYX_PYTHON_EXECUTABLE} \
 #       -DXMLLINT_EXECUTABLE=${XMLLINT_EXECUTABLE} \
 #       -DJAVA_EXECUTABLE=${JAVA_EXECUTABLE} \
 #       -DENCODING=xxx \
@@ -57,6 +58,7 @@ get_filename_component(updir_ "${LYX_ROOT}" DIRECTORY)
 get_filename_component(updir2_ "${LYX_ROOT}" NAME)
 set(file "${updir2_}/${file}")
 set(LYX_ROOT "${updir_}")
+set(NO_FAILES 0)
 
 if(format MATCHES "dvi|pdf")
   message(STATUS "LYX_TESTS_USERDIR = ${LYX_TESTS_USERDIR}")
@@ -72,6 +74,7 @@ if(format MATCHES "dvi|pdf")
     RESULT_VARIABLE _err)
   string(COMPARE EQUAL  ${_err} 0 _erg)
   if(NOT _erg)
+    set(NO_FAILES 1)
     message(FATAL_ERROR "Export failed while converting")
   endif()
   # We only need "_${ENCODING}" for unicode tests (because multiple encodings
@@ -86,8 +89,9 @@ else()
     RESULT_VARIABLE _err)
   string(COMPARE EQUAL  ${_err} 0 _erg)
   if(NOT _erg)
+    set(NO_FAILES 1)
     message(FATAL_ERROR "Export failed while converting")
-    endif()
+  endif()
   if(extension MATCHES "\\.lyx$")
     # Font-type not relevant for lyx16/lyx2[0123] exports
     set(result_file_base "${TempDir}/${file}")
@@ -114,12 +118,96 @@ endfunction()
 
 macro(Summary _err _msg)
   if (${_err})
+    MATH(EXPR NO_FAILES "${NO_FAILES}+1")
     list(APPEND _TestResultMessage "Error: ${_msg}")
   else()
     list(APPEND _TestResultMessage "OK: ${_msg}")
   endif()
 endmacro()
 
+macro(check_xhtml_validate xhtml_file)
+  message(STATUS "Calling ${LYX_PYTHON_EXECUTABLE} \"${TOP_SRC_DIR}/development/autotests/simplehtml_validity.py\" \"${TempDir}/${xhtml_file}\"")
+  set(_outputfile "${TempDir}/${xhtml_file}.validate_out")
+  execute_process(
+    COMMAND ${LYX_PYTHON_EXECUTABLE} "${TOP_SRC_DIR}/development/autotests/simplehtml_validity.py" "${TempDir}/${xhtml_file}"
+    WORKING_DIRECTORY "${TempDir}"
+    OUTPUT_VARIABLE xmlout
+    ERROR_VARIABLE xmlerr
+    RESULT_VARIABLE _err)
+  file(WRITE "${_outputfile}" ${xmlout})
+  if (_err)
+    Summary(_err "${_err} validating \"${_outputfile}\"")
+    MATH(EXPR NO_FAILES "${NO_FAILES}+1")
+  endif()
+  if (NOT "${xmlout}" STREQUAL "")
+    message(STATUS "${xmlout}")
+    set(_err -1)
+    Summary(_err "Non empty output \"${_outputfile}\" with \"symplehtml_validity.py\"")
+  endif()
+endmacro()
+
+macro(check_xhtml_xmllint xhtml_file)
+  set(xmllint_params --loaddtd --noout)
+  string(REPLACE ";" " " xmllint_params2 " ${xmllint_params}")
+  message(STATUS "Calling: " ${XMLLINT_EXECUTABLE} ${xmllint_params2} " \"${TempDir}/${xhtml_file}\"")
+  set(_outputfile "${TempDir}/${xhtml_file}.sax_out")
+  execute_process(
+    COMMAND ${XMLLINT_EXECUTABLE} ${xmllint_params} "${xhtml_file}"
+    WORKING_DIRECTORY "${TempDir}"
+    OUTPUT_VARIABLE xmlout
+    ERROR_VARIABLE xmlerr
+    RESULT_VARIABLE _err)
+  file(WRITE "${_outputfile}" ${xmlout})
+  Summary(_err "Checking \"${TempDir}/${xhtml_file}\" with ${XMLLINT_EXECUTABLE}")
+  if (NOT _err)
+    # check if parser output contains error messages
+    message(STATUS "Check the output: ${PERL_EXECUTABLE} ${TOP_SRC_DIR}/development/autotests/examineXmllintOutput.pl")
+    execute_process(
+      COMMAND ${PERL_EXECUTABLE} "${TOP_SRC_DIR}/development/autotests/examineXmllintOutput.pl" "${_outputfile}"
+      WORKING_DIRECTORY "${TempDir}"
+      OUTPUT_VARIABLE xmlout
+      RESULT_VARIABLE _err)
+    Summary(_err "Parse messages of ${XMLLINT_EXECUTABLE} for errors")
+  else()
+    message(STATUS "Errors from xmllint: ${xmlerr}")
+  endif()
+  if (NOT _err)
+    if (NOT "${xmlout}" STREQUAL "")
+      message(STATUS "${xmlout}")
+      set(_err -1)
+      Summary(_err "Non empty output \"${_outputfile}\" of \"${XMLLINT_EXECUTABLE}\"")
+    endif()
+  endif()
+endmacro()
+
+macro(check_xhtml_xmlparser xhtml_file)
+  message(STATUS "Calling ${PERL_EXECUTABLE} \"${TOP_SRC_DIR}/development/autotests/xmlParser.pl\" \"${xhtml_file}\"")
+  execute_process(
+    COMMAND ${PERL_EXECUTABLE} "${TOP_SRC_DIR}/development/autotests/xmlParser.pl" "${result_file_name}"
+    WORKING_DIRECTORY "${TempDir}"
+    OUTPUT_VARIABLE parserout
+    ERROR_VARIABLE parsererr
+    RESULT_VARIABLE _err
+  )
+  if (_err)
+    message(STATUS "${parsererr}")
+  endif()
+  Summary(_err "Checking \"${TempDir}/${xhtml_file}\" with xmlParser.pl")
+endmacro()
+
+macro(check_xhtml_jing xhtml_file)
+  message(STATUS "Calling: ${JAVA_EXECUTABLE} -jar \"${TOP_SRC_DIR}/development/tools/jing.jar\" \"https://docbook.org/xml/5.2b09/rng/docbook.rng\" \"${TempDir}/${xhtml_file}\"")
+  set(_outputfile "${TempDir}/${xhtml_file}.jing_out")
+  execute_process(
+    COMMAND ${JAVA_EXECUTABLE} -jar "${TOP_SRC_DIR}/development/tools/jing.jar" "https://docbook.org/xml/5.2b09/rng/docbook.rng" "${xhtml_file}"
+    WORKING_DIRECTORY "${TempDir}"
+    OUTPUT_VARIABLE jingout
+    RESULT_VARIABLE _err)
+  file(WRITE "${_outputfile}" ${jingout})
+  message(STATUS "_err = ${_err}, jingout = ${jingout}")
+  Summary(_err "Checking for empty output \"${_outputfile}\" of ${JAVA_EXECUTABLE} -jar \"${TOP_SRC_DIR}/development/tools/jing.jar\"")
+endmacro()
+
 set(ENV{${LYX_USERDIR_VER}} "${LYX_TESTS_USERDIR}")
 set(ENV{LANG} "en_US.UTF-8") # to get all error-messages in english
 set(ENV{LANGUAGE} "US:en")
@@ -234,69 +322,21 @@ else()
     else()
       message(STATUS "Expected result file \"${TempDir}/${result_file_name}\" exists")
       if (extension MATCHES "^x(ht)?ml$")
-        if (format MATCHES "xhtml")
-          set(xmllint_params --loaddtd --noout)
-          set(executable_ ${XMLLINT_EXECUTABLE})
-        else()
-          set(xmllint_params)
-          set(executable_ ${PERL_EXECUTABLE} "${TOP_SRC_DIR}/development/autotests/filterXml4Sax.pl")
+        if (NOT format MATCHES "xhtml")
           # Check with perl xml-parser
           # needs XML::Parser module
-          message(STATUS "Calling ${PERL_EXECUTABLE} \"${TOP_SRC_DIR}/development/autotests/xmlParser.pl\" \"${result_file_name}\"")
-          execute_process(
-            COMMAND ${PERL_EXECUTABLE} "${TOP_SRC_DIR}/development/autotests/xmlParser.pl" "${result_file_name}"
-           WORKING_DIRECTORY "${TempDir}"
-            OUTPUT_VARIABLE parserout
-            ERROR_VARIABLE parsererr
-            RESULT_VARIABLE _err
-          )
-          if (_err)
-            message(STATUS "${parsererr}")
-          endif()
-          Summary(_err "Checking \"${TempDir}/${result_file_name}\" with xmlParser.pl")
+          check_xhtml_xmlparser(${result_file_name})
         endif()
-        if (XMLLINT_EXECUTABLE)
-          string(REPLACE ";" " " xmllint_params2 " ${xmllint_params}")
-          message(STATUS "Calling: " ${executable_} ${xmllint_params2} " ${WORKDIR}/${result_file_name}")
-          # check the created xhtml file
-          execute_process(
-            COMMAND ${executable_} ${xmllint_params}  "${result_file_name}"
-           WORKING_DIRECTORY "${TempDir}"
-            OUTPUT_VARIABLE xmlout
-            ERROR_VARIABLE xmlerr
-            RESULT_VARIABLE _err)
-          file(WRITE "${TempDir}/${result_file_name}.sax_out" ${xmlout})
-          Summary(_err "Checking \"${TempDir}/${result_file_name}\" with ${XMLLINT_EXECUTABLE}")
-          if (NOT _err)
-            # check if parser output contains error messages
-            message(STATUS "Check the output: ${PERL_EXECUTABLE} ${TOP_SRC_DIR}/development/autotests/examineXmllintOutput.pl")
-            execute_process(
-              COMMAND ${PERL_EXECUTABLE} "${TOP_SRC_DIR}/development/autotests/examineXmllintOutput.pl" "${result_file_name}.sax_out"
-             WORKING_DIRECTORY "${TempDir}"
-              OUTPUT_VARIABLE xmlout
-              RESULT_VARIABLE _err)
-            Summary(_err "Parse messages of ${XMLLINT_EXECUTABLE} for errors")
+        if (JAVA_EXECUTABLE)
+          if (format MATCHES "docbook5")
+            # check with jing
+            check_xhtml_jing(${result_file_name})
           else()
-            message(STATUS "Errors from xmllint: ${xmlerr}")
-          endif()
-          if (NOT _err)
-            if (NOT "${xmlout}" STREQUAL "")
-              message(STATUS "${xmlout}")
-              set(_err -1)
-              Summary(_err "Non empty output of \"${XMLLINT_EXECUTABLE}\"")
-            endif()
+            #check_xhtml_validate(${result_file_name})
           endif()
         endif()
-        if (NOT _err AND format MATCHES "docbook5" AND JAVA_EXECUTABLE)
-          # check with jing
-          message(STATUS "Calling: ${JAVA_EXECUTABLE} -jar \"${TOP_SRC_DIR}/development/tools/jing.jar\" https://docbook.org/xml/5.2b09/rng/docbook.rng \"${WORKDIR}/${result_file_name}\"")
-          execute_process(
-            COMMAND ${JAVA_EXECUTABLE} -jar "${TOP_SRC_DIR}/development/tools/jing.jar" "https://docbook.org/xml/5.2b09/rng/docbook.rng" "${result_file_name}"
-           WORKING_DIRECTORY "${TempDir}"
-            OUTPUT_VARIABLE jingout
-            RESULT_VARIABLE _err)
-          message(STATUS "_err = ${_err}, jingout = ${jingout}")
-          Summary(_err "Checking for empty output of ${JAVA_EXECUTABLE} -jar \"${TOP_SRC_DIR}/development/tools/jing.jar\"")
+        if (XMLLINT_EXECUTABLE)
+          check_xhtml_xmllint(${result_file_name})
         endif()
       endif()
     endif()
@@ -304,9 +344,9 @@ else()
 endif()
 
 if(inverted)
-  string(COMPARE EQUAL  ${_err} 0 _erg)
+  string(COMPARE EQUAL  ${NO_FAILES} 0 _erg)
 else()
-  string(COMPARE NOTEQUAL  ${_err} 0 _erg)
+  string(COMPARE NOTEQUAL  ${NO_FAILES} 0 _erg)
 endif()
 
 if ($ENV{LYX_DEBUG_LATEX})

diff --git a/development/autotests/simplehtml_validity.py b/development/autotests/simplehtml_validity.py
new file mode 100644 (file)
index 0000000..0cc71f3
--- /dev/null
+++ b/development/autotests/simplehtml_validity.py
@@ -0,0 +1,53 @@
+# Stricter version of the export tests: validate the XHTML code produced by
+# LyX' lyxhtml output as HTML5. It also validates the CSS and MathML parts.
+# Validation errors usually are mistakes in the generator.
+#
+# Call:
+#     python simplehtml_validity.py PATH_TO_HTML5_SOURCE
+#
+# Written with Python 3.8.8.
+# Requirements:
+# - Python package: html5validator: at least v0.4.2
+# - Java runtime engine (JRE): at least v8 (depending on html5validator)
+# Run:
+#     pip install html5validator>=0.4.2
+
+import collections
+import glob
+import sys
+import tempfile
+import os
+
+import html5validator
+
+
+if len(sys.argv) != 2:
+    print('Expecting one argument, the path to the LyX-create xhtml file')
+    sys.exit(-1)
+if not os.path.exists(sys.argv[1]):
+    print('The given path does not point to an existing file')
+    sys.exit(-1)
+
+
+xhtml_file_name = sys.argv[1]
+xhtml_list = [xhtml_file_name]
+
+validator = html5validator.Validator(format='text')
+error_count = validator.validate(xhtml_list)
+
+n_invalid = 0
+n_valid = 0
+if error_count == 0:
+    n_valid += 1
+    print(f'> Found no validation error!')
+    sys.exit(0)
+else:
+    n_invalid += 1
+    print(f'> Found {error_count} validation error{"" if error_count == 1 else "s"}!')
+    sys.exit(-2)
+
+if n_invalid == 0:
+    print("That's excellent! Give yourself a pat on the back!")
+    sys.exit(0)
+else:
+    sys.exit(-3)