tex_log_parser 1.0.0.pre.10 → 1.0.0.pre.14

data/lib/tex_log_parser/patterns/highlighted_messages.rb

@@ -1,101 +1,105 @@
 # frozen_string_literal: true
 
-# Matches messages of these forms:
-#
-#     !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
-#     !
-#     ./plain.tex:5: fontspec error: "font-not-found"
-#     !
-#     ! The font "NoSuchFont" cannot be found.
-#     !
-#     ! See the fontspec documentation for further information.
-#     !
-#     ! For immediate help type H <return>.
-#     !...............................................
-#
-#     l.5 \setmainfont{NoSuchFont}
-#
-# and
-#
-#     !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
-#     !
-#     ! fontspec error: "font-not-found"
-#     !
-#     ! The font "NoSuchFont" cannot be found.
-#     !
-#     ! See the fontspec documentation for further information.
-#     !
-#     ! For immediate help type H <return>.
-#     !...............................................
-#
-#     l.5 \setmainfont{NoSuchFont}
-#
-# and
-#
-#     .................................................
-#     . LaTeX info: "xparse/define-command"
-#     .
-#     . Defining command \fontspec with sig. 'O{}mO{}' on line 472.
-#     .................................................
-class HighlightedMessages
-  include RegExpPattern
+  class TexLogParser
+  # Matches messages of these forms:
+  #
+  #     !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
+  #     !
+  #     ./plain.tex:5: fontspec error: "font-not-found"
+  #     !
+  #     ! The font "NoSuchFont" cannot be found.
+  #     !
+  #     ! See the fontspec documentation for further information.
+  #     !
+  #     ! For immediate help type H <return>.
+  #     !...............................................
+  #
+  #     l.5 \setmainfont{NoSuchFont}
+  #
+  # and
+  #
+  #     !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
+  #     !
+  #     ! fontspec error: "font-not-found"
+  #     !
+  #     ! The font "NoSuchFont" cannot be found.
+  #     !
+  #     ! See the fontspec documentation for further information.
+  #     !
+  #     ! For immediate help type H <return>.
+  #     !...............................................
+  #
+  #     l.5 \setmainfont{NoSuchFont}
+  #
+  # and
+  #
+  #     .................................................
+  #     . LaTeX info: "xparse/define-command"
+  #     .
+  #     . Defining command \fontspec with sig. 'O{}mO{}' on line 472.
+  #     .................................................
+  class HighlightedMessages
+    include LogParser::RegExpPattern
 
-  def initialize
-    super(/^(\!{3,}|\.{3,})$/,
-          { pattern: lambda { |m|
-                       if m[1][0] == '!'
-                         /^l\.(\d+)/
-                       else
-                         /^\.{3,}\s*$/
-                       end
-                     },
-            until: :match,
-            inclusive: true })
-  end
+    # Creates a new instance.
+    def initialize
+      super(/^(!{3,}|\.{3,})$/,
+            { pattern: lambda { |m|
+                         if m[1][0] == '!'
+                           /^l\.(\d+)/
+                         else
+                           /^\.{3,}\s*$/
+                         end
+                       },
+              until: :match,
+              inclusive: true })
+    end
 
-  def read(lines)
-    # @type [LogMessage] msg
-    msg, consumed = super(lines)
+    # (see LogParser::RegExpPattern#read)
+    def read(lines)
+      # @type [Message] msg
+      msg, consumed = super(lines)
 
-    is_error = @start_match[1][0] == '!'
+      is_error = @start_match[1][0] == '!'
 
-    if is_error
-      file_line_match = %r{^(/?(?:.*?/)*[^/]+):(\d+):\s*}.match(msg.message)
-      line = nil
-      if !file_line_match.nil? # if file-line active
-        msg.source_file = file_line_match[1]
-        line = file_line_match[2].to_i
-        msg.message.gsub!(file_line_match[0], '')
-      elsif !@end_match[1].nil?
-        # No file-line format, so use line number from end match
-        line = @end_match[1].to_i
-      end
-      msg.source_lines = { from: line, to: line } unless line.nil?
+      if is_error
+        file_line_match = %r{^(/?(?:.*?/)*[^/]+):(\d+):\s*}.match(msg.message)
+        line = nil
+        if !file_line_match.nil? # if file-line active
+          msg.source_file = file_line_match[1]
+          line = file_line_match[2].to_i
+          msg.message.gsub!(file_line_match[0], '')
+        elsif !@end_match[1].nil?
+          # No file-line format, so use line number from end match
+          line = @end_match[1].to_i
+        end
+        msg.source_lines = { from: line, to: line } unless line.nil?
 
-      msg.level = :error
+        msg.level = :error
 
-      msg.message.gsub!(/^.*?For immediate help type.*$/, '')
-      msg.message.gsub!(/^\!\.+\s*$/, '')
-      msg.message.gsub!(/^l\.\d+\s+.*$/, '')
-    else
-      # BROKEN_BY_LINEBREAKS
-      # TODO: may be split across lines --> remove whitespace before extracting
-      line_match = /on line\s+(\d+)\.$/.match(msg.message)
-      unless line_match.nil?
-        line = line_match[1].to_i
-        msg.source_lines = { from: line, to: line }
-      end
+        msg.message.gsub!(/^.*?For immediate help type.*$/, '')
+        msg.message.gsub!(/^!\.+\s*$/, '')
+        msg.message.gsub!(/^l\.\d+\s+.*$/, '')
+      else
+        # BROKEN_BY_LINEBREAKS
+        # TODO: may be split across lines --> remove whitespace before extracting
+        line_match = /on line\s+(\d+)\.$/.match(msg.message)
+        unless line_match.nil?
+          line = line_match[1].to_i
+          msg.source_lines = { from: line, to: line }
+        end
 
-      msg.level = :info
+        msg.level = :info
 
-      msg.message.gsub!(@end_match[0], '')
-    end
+        msg.message.gsub!(@end_match[0], '')
+      end
 
-    msg.preformatted = true
-    msg.message.gsub!(@start, '')
-    msg.message.gsub!(/^#{@start_match[1][0]}\s+/, '')
-    msg.message.strip!
+      msg.preformatted = true
+      msg.message.gsub!(@start, '')
+      msg.message.gsub!(/^#{@start_match[1][0]}\s+/, '')
+      msg.message.strip!
 
-    [msg, consumed]
+      [msg, consumed]
+    end
   end
-end
+end

data/lib/tex_log_parser/patterns/prefixed_multi_line_pattern.rb

@@ -1,51 +1,55 @@
 # frozen_string_literal: true
 
-# Matches messages of this form:
-#
-#     Package tocbasic Info: omitting babel extension for `toc'
-#     (tocbasic)             because of feature `nobabel' available
-#     (tocbasic)             for `toc' on input line 132.
-#
-# Note: currently fails if lines get broken badly, e.g. in 000.log:634.
-class PrefixedMultiLinePattern
-  include RegExpPattern
-
-  def initialize
-    super(/(Package|Class|\w+TeX)\s+(?:(\w+)\s+)?(Warning|Error|Info|Message)/,
-          { pattern: ->(m) { /^\s*\(#{m[2]}\)/ }, # BROKEN_BY_LINEBREAKS
-            until: :mismatch,
-            inclusive: false })
-  end
-
-  def read(lines)
-    # @type [LogMessage] msg
-    # @type [Int] consumed
-    msg, consumed = super(lines)
-
-    case @start_match[3]
-    when 'Error'
-      msg.level = :error
-    when 'Warning'
-      msg.level = :warning
-    when 'Info', 'Message'
-      msg.level = :info
-    else
-      # TODO: abort?
-      Logger.debug 'Unhandled message type!'
+class TexLogParser
+  # Matches messages of this form:
+  #
+  #     Package tocbasic Info: omitting babel extension for `toc'
+  #     (tocbasic)             because of feature `nobabel' available
+  #     (tocbasic)             for `toc' on input line 132.
+  #
+  # Note: currently fails if lines get broken badly, e.g. in 000.log:634.
+  class PrefixedMultiLinePattern
+    include LogParser::RegExpPattern
+
+    # Creates a new instance.
+    def initialize
+      super(/(Package|Class|\w+TeX)\s+(?:(\w+)\s+)?(Warning|Error|Info|Message)/,
+            { pattern: ->(m) { /^\s*\(#{m[2]}\)/ }, # BROKEN_BY_LINEBREAKS
+              until: :mismatch,
+              inclusive: false })
     end
 
-    # source file from scope, parser does it
-
-    # BROKEN_BY_LINEBREAKS
-    # TODO: may be split across lines --> remove whitespace before extracting
-    suffix_match = /on input line\s+(\d+)(?:\.\s*)?\z/.match(msg.message)
-    unless suffix_match.nil?
-      line = suffix_match[1].to_i
-      msg.source_lines = { from: line, to: line }
+    # (see LogParser::RegExpPattern#read)
+    def read(lines)
+      # @type [Message] msg
+      # @type [Int] consumed
+      msg, consumed = super(lines)
+
+      case @start_match[3]
+      when 'Error'
+        msg.level = :error
+      when 'Warning'
+        msg.level = :warning
+      when 'Info', 'Message'
+        msg.level = :info
+      else
+        # TODO: abort?
+        Logger.debug 'Unhandled message type!'
+      end
+
+      # source file from scope, parser does it
+
+      # BROKEN_BY_LINEBREAKS
+      # TODO: may be split across lines --> remove whitespace before extracting
+      suffix_match = /on input line\s+(\d+)(?:\.\s*)?\z/.match(msg.message)
+      unless suffix_match.nil?
+        line = suffix_match[1].to_i
+        msg.source_lines = { from: line, to: line }
+      end
+
+      # TODO: message itself contains useless line prefixes --> remove, preformatted = false
+
+      [msg, consumed]
     end
-
-    # TODO: message itself contains useless line prefixes --> remove, preformatted = false
-
-    [msg, consumed]
   end
 end

data/lib/tex_log_parser/patterns/runaway_parameter_error.rb

@@ -1,24 +1,28 @@
 # frozen_string_literal: true
 
-# Matches messages of this form:
-#
-#   Runaway argument?
-#   {Test. Also, it contains some \ref {warnings} and \ref {errors} for t\ETC.
-class RunawayParameterError
-  include RegExpPattern
+class TexLogParser
+  # Matches messages of this form:
+  #
+  #     Runaway argument?
+  #     {Test. Also, it contains some \ref {warnings} and \ref {errors} for t\ETC.
+  class RunawayParameterError
+    include LogParser::RegExpPattern
 
-  def initialize
-    super(/^Runaway argument\?/,
-          { pattern: ->(_) { /./ }, until: :match, inclusive: true }
-    )
-  end
+    # Creates a new instance.
+    def initialize
+      super(/^Runaway argument\?/,
+            { pattern: ->(_) { /./ }, until: :match, inclusive: true }
+      )
+    end
 
-  def read(lines)
-    # @type [LogMessage] msg
-    msg, consumed = super(lines)
+    # (see LogParser::RegExpPattern#read)
+    def read(lines)
+      # @type [Message] msg
+      msg, consumed = super(lines)
 
-    msg.level = :error
+      msg.level = :error
 
-    [msg, consumed]
+      [msg, consumed]
+    end
   end
-end
+end

data/lib/tex_log_parser/patterns/standard_error.rb

@@ -1,44 +1,48 @@
 # frozen_string_literal: true
 
-# Matches messages of this form:
-#
-#   ! File ended while scanning use of \@footnotetext.
-#   <inserted text>
-#                   \par
-#   <*> plain.tex
-#
-# and
-#
-#   ! Font TU/NoSuchFont(0)/m/n/9=NoSuchFont at 9.0pt not loadable: Metric (TFM) fi
-#   le or installed font not found.
-#   <to be read again>
-#                    relax
-#   l.40 \end{document}
-class StandardError
-  include RegExpPattern
+class TexLogParser
+  # Matches messages of this form:
+  #
+  #     ! File ended while scanning use of \@footnotetext.
+  #     <inserted text>
+  #                     \par
+  #     <*> plain.tex
+  #
+  # and
+  #
+  #     ! Font TU/NoSuchFont(0)/m/n/9=NoSuchFont at 9.0pt not loadable: Metric (TFM) fi
+  #     le or installed font not found.
+  #     <to be read again>
+  #                      relax
+  #     l.40 \end{document}
+  class StandardError
+    include LogParser::RegExpPattern
 
-  def initialize
-    super(/^\! \w+/,
-          { pattern: ->(_) { /^\s*<\*>\s+([^\s]+)|^l\.(\d+)\s+/ }, until: :match, inclusive: true }
-    )
-  end
+    # Creates a new instance.
+    def initialize
+      super(/^! \w+/,
+            { pattern: ->(_) { /^\s*<\*>\s+([^\s]+)|^l\.(\d+)\s+/ }, until: :match, inclusive: true }
+      )
+    end
 
-  def read(lines)
-    # @type [LogMessage] msg
-    msg, consumed = super(lines)
+    # (see LogParser::RegExpPattern#read)
+    def read(lines)
+      # @type [Message] msg
+      msg, consumed = super(lines)
 
-    msg.level = :error
-    # Remove last line
-    msg.message.gsub!(@ending[:pattern][nil], '')
-    msg.message.rstrip!
+      msg.level = :error
+      # Remove last line
+      msg.message.gsub!(@ending[:pattern][nil], '')
+      msg.message.rstrip!
 
-    file = @end_match[1]
-    line = @end_match[2].to_i
+      file = @end_match[1]
+      line = @end_match[2].to_i
 
-    msg.source_file = file unless file.nil?
-    msg.source_lines = { from: line, to: line } unless line.nil? || line.zero?
-    msg.preformatted = true
+      msg.source_file = file unless file.nil?
+      msg.source_lines = { from: line, to: line } unless line.nil? || line.zero?
+      msg.preformatted = true
 
-    [msg, consumed]
+      [msg, consumed]
+    end
   end
-end
+end

data/lib/tex_log_parser.rb

@@ -1,24 +1,30 @@
 # frozen_string_literal: true
 
-require 'logger'
-require 'tex_log_parser/log_buffer'
-require 'tex_log_parser/log_message'
-require 'tex_log_parser/log_parser'
-require 'tex_log_parser/log_pattern'
+require 'log_parser/log_parser'
 Dir["#{File.expand_path(__dir__)}/tex_log_parser/patterns/*.rb"].each { |p| require p }
 
-# TODO: document
+# Parses logs (and output) of LaTeX interpreters, e.g. `pdflatex`, `xelatex` and `lualatex`.
+#
+# *Note:* Due to shortcomings in the native format of those logs, please be
+# aware of these recommendations:
+#  - Use `-file-line-error` if possible; it makes for more robust source file and line reporting.
+#  - Ask for log lines to be broken as rarely as possible; see e.g. [here](https://tex.stackexchange.com/q/52988/3213).
+#     Search the sources for `BROKEN_BY_LINEBREAKS` to find all the nastiness (and potential issues) you can avoid by that.
 class TexLogParser
   include LogParser
 
-  def initialize(log, _options = {})
-    super(log, _options)
+  # (see LogParser#initialize)
+  def initialize(log)
+    super(log)
 
     # BROKEN_BY_LINEBREAKS
     # I'd prefer to have this stateless, but well (see below).
     @pushed_dummy = false
   end
 
+  protected
+
+  # (see LogParser#patterns)
   def patterns
     [HighlightedMessages.new,
      FileLineError.new,
@@ -29,7 +35,22 @@ class TexLogParser
      BadHboxWarning.new]
   end
 
+  # (see LogParser#scope_changes)
+  #
+  # _Implementation note:_ The basic format in LaTeX logs is that
+  #  * `(filename` marks the beginning of messages from that file, and
+  #  * the matching `)` marks the end.
+  # Those nest, of course.
+  #
+  # This implementation is mainly concerned with hacking around problems of how this format is implemented in the logs.
+  #  * Filenames may be broken across lines.
+  #  * Opening "tags" may or may not appear on a dedicated line.
+  #  * Closing "tags" may or may not appear on a dedicated line.
+  #  * Badly line-broken message fragments with parentheses confuse matching heuristics for either.
+  # If inopportune line breaks can be avoided, this method is a lot more reliable.
   def scope_changes(line)
+    pushed_dummy = false
+
     result =
       case line
       when /^\s*\(([^()]*)\)\s*(.*)$/
@@ -55,7 +76,7 @@ class TexLogParser
       when /^[^()]*\)/
         # BROKEN_BY_LINEBREAKS
         # Badly broken lines may push ) prefixed by non-whitespace. Narf.
-        [:pop] if @pushed_dummy ? [:pop] : []
+        @pushed_dummy ? [:pop] : []
       else
         []
       end

data/lib/version.rb

@@ -0,0 +1,7 @@
+# frozen_string_literal: true
+
+# @attr [String] VERSION
+#   The version of TexLogParser.
+class TexLogParser
+  VERSION = '1.0.0.pre.14'
+end

metadata

@@ -1,15 +1,29 @@
 --- !ruby/object:Gem::Specification
 name: tex_log_parser
 version: !ruby/object:Gem::Version
-  version: 1.0.0.pre.10
+  version: 1.0.0.pre.14
 platform: ruby
 authors:
 - Raphael Reitzig
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2018-04-07 00:00:00.000000000 Z
+date: 2018-04-08 00:00:00.000000000 Z
 dependencies:
+- !ruby/object:Gem::Dependency
+  name: github-markup
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.0'
 - !ruby/object:Gem::Dependency
   name: minitest
   requirement: !ruby/object:Gem::Requirement
@@ -38,6 +52,20 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '12.3'
+- !ruby/object:Gem::Dependency
+  name: redcarpet
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.4'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.4'
 - !ruby/object:Gem::Dependency
   name: yard
   requirement: !ruby/object:Gem::Requirement
@@ -76,12 +104,12 @@ files:
 - LICENSE
 - README.md
 - bin/texlogparser
-- lib/logger.rb
+- lib/log_parser/buffer.rb
+- lib/log_parser/log_parser.rb
+- lib/log_parser/logger.rb
+- lib/log_parser/message.rb
+- lib/log_parser/pattern.rb
 - lib/tex_log_parser.rb
-- lib/tex_log_parser/log_buffer.rb
-- lib/tex_log_parser/log_message.rb
-- lib/tex_log_parser/log_parser.rb
-- lib/tex_log_parser/log_pattern.rb
 - lib/tex_log_parser/patterns/bad_hbox_warning.rb
 - lib/tex_log_parser/patterns/fatal_error_occurred.rb
 - lib/tex_log_parser/patterns/file_line_error.rb
@@ -89,7 +117,7 @@ files:
 - lib/tex_log_parser/patterns/prefixed_multi_line_pattern.rb
 - lib/tex_log_parser/patterns/runaway_parameter_error.rb
 - lib/tex_log_parser/patterns/standard_error.rb
-- lib/tex_log_parser/version.rb
+- lib/version.rb
 homepage: http://github.com/reitzig/texlogparser
 licenses:
 - MIT
@@ -111,7 +139,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: 1.3.1
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.5.2.1
+rubygems_version: 2.7.6
 signing_key: 
 specification_version: 4
 summary: Parses log files of (La)TeX engines

data/lib/logger.rb

@@ -1,21 +0,0 @@
-# frozen_string_literal: true
-
-# A simple helper, probably to be replaced by a proper logging library
-# at some point.
-#
-# @attr [true,false] debug
-class Logger
-  class << self
-    attr_accessor :debugging
-
-    def debug?
-      debugging || !ENV['DEBUG'].nil?
-    end
-
-    # Logs the given message to STDOUT if `debug` is true.
-    # @param [String] message
-    def debug(message)
-      puts message if debug?
-    end
-  end
-end

data/lib/tex_log_parser/log_buffer.rb

@@ -1,62 +0,0 @@
-# frozen_string_literal: true
-
-class LogBuffer
-  # @param [Array<String>,IO] log
-  def initialize(log)
-    @buffer = []
-    @stream = nil
-
-    @buffer = log if log.is_a?(Array)
-    @stream = log if log.is_a?(IO) || log.is_a?(StringIO)
-  end
-
-  def empty?
-    @buffer.empty? && stream_is_done?
-  end
-
-  def buffer_size
-    @buffer.size
-  end
-
-  def first
-    self[0]
-  end
-
-  def find_index(starting_from = 0)
-    index = starting_from
-    element = self[index]
-
-    until element.nil?
-      return index if yield(element)
-      index += 1
-      element = self[index]
-    end
-
-    nil
-  end
-
-  def [](offset, length = nil)
-    base_length = length || 1
-    while offset + base_length > @buffer.size
-      return (length.nil? ? nil : @buffer[offset, @buffer.size]) if stream_is_done?
-      @buffer.push(@stream.readline.rstrip)
-    end
-
-    length.nil? ? @buffer[offset] : @buffer[offset, length]
-  end
-
-  def forward(offset = 1)
-    self[offset]
-    @buffer.slice!(0, offset)
-  end
-
-  def close
-    @stream.close unless @stream.nil? || @stream.closed?
-  end
-
-  private
-
-  def stream_is_done?
-    @stream.nil? || @stream.closed? || @stream.eof?
-  end
-end