converttolatex.pl

比对算法的具体应用DNA序列分析 ——基因序列 ——基因表达调控信息 寻找基因牵涉到两个方面的工作 ： 识别与基因相关的特殊序列信号 预测基因的编码区域 结合两个方面的结果确定基因的位置和结构 基因表

# perl script to convert genExpDocs readme files to a reasonable latex format
# they will probably still need cleaning up afterwards...
use strict;

$#ARGV==1 or die " \n\nperl convertToLatex.pl D:\\pathToGenExpDocsDir D:\\pathToGenExpDocsTex\n\n";

my @docFiles = glob($ARGV[0]."\\*.txt");
my $outDir = $ARGV[1]."\\";

foreach (@docFiles)
{
	# get filename
	my $outf = "tmp";
	if(/([\w]+)\.txt$/)
	{
		$outf = $outDir . $1 . ".tex";
	}

	# read in every line, and if it is special in some way, do something to it!
	open IN, $_;
	my $tex = "";
	$/ = "---";
	my $count = 0;
	my $secCount=0;
	while(<IN>)
	{
		# crop the --- off the end
		chomp; chomp; chomp;
		# replace any non latex characters
		s/\\/\$\\backslash\$/g;  # backslash
		s/_/\\_/g;               # underscore
		s/ '/ `/g;               # opening quote
		if($count==1)
		{
			# This is the section header
			my @words = split(/ +/);
			$tex .= "\n\\section{";
			foreach my $w (@words)
			{
				$w = lc($w);
				$tex .= ucfirst($w)." ";
			}
			$tex .= "}\n";
		}
		elsif($count>1)
		{
			if(!/\n/)
			{
				# this must be a main list item
				$tex .= "\n\\subsection{";
				/(.*)/;
				/^[^\)]+\) +(.*)/;
				$tex .= "$1}\n";
			}
			else
			{
				# this is the bulk of a section
				if($secCount==1 || $secCount==4)
				{
					# split by lines
					my @lines = split(/\n/);
					my $list = 0;
					my $openBracket=0;
					my $listType = "";
					foreach my $line (@lines)
					{
						# strip all leading and trailing spaces
						$line =~ s/^[\s]+//;
						$line =~ s/[\s]+$//;
						# if a ) comes before a ) this is a list
						if($line=~/^([^(]+)\)(.*)/)
						{
							my $num = $1;
							my $rest = $2;
							
							if($openBracket)
							{
								$openBracket=0;
								$tex .= "$line";								
								if(!$list){ $tex .= "\n"; }
							}
							else
							{
								if($list == 0) 
								{
									#$tex .= "BEGIN LIST\n";
									# work out what type of list this is
									$num = ($num =~ /[\s]+(.*)/);
									if (length($num)==1 || $num=~/^\d+$/)
									{
										# just make it a normal enumerate
										$listType = "enumerate";
										$tex .= "\n\\begin{$listType}\n";
										$tex .= "\n\t\\item\n";
										$tex .= "\t\t$rest ";
									}
									else
									{
										# just make it a normal enumerate
										$listType = "enumerate";
										$tex .= "\n\\begin{$listType}\n";
										$tex .= "\n\t\\item\n";
										$tex .= "\t\t$rest ";										
									}
								}
								else 
								{
									$tex .= "\n\t\\item\n";
									$tex .= "\t\t$rest ";
									#$tex .= "CONTINUE LIST\n";
								}	
								$list = 1;
							}
						}
						elsif ($line=~/^[\s]*$/) 
						{
							if($list > 0)
							{
								$list = 2;
							}
							else
							{ 
								if(!($tex=~/noindent $/))
								{
									$tex .= "$line\n\\noindent ";
								}
								
							}
						}
						else
						{
							# really have found a normal line
							if($list == 2)
							{
								#$tex .= "END LIST\n";
								$tex .= "\n\\end{$listType}\n";
								$list = 0;
							}
							$tex .= "$line";
							if(!$list){$tex.="\n";}
						}
						if($line=~/\([^)]+$/)
						{
							$openBracket=1;
						}

					}
					if($list)
					{
						$tex .= "\n\\end{$listType}\n";
						$list = 0;
						#$tex .= "END LIST\n";						
					}
					
					# first section, leave exactly as it is
					#$tex .= "$_\n";
				}
				elsif($secCount==2)
				{
					# build up a descriptive list
					# split by lines
					my @lines = split(/\n/);
					my $list = 0;
					my $listType = "description";
					$tex .= "\n\t\\begin{$listType}";
					my $itemCount=0;
					foreach my $line (@lines)
					{
						# strip all leading and trailing spaces
						my $lineLength = length($line);
						$line =~ s/^[\s]+//;
						my $spaceCount = $lineLength - length($line);
						$line =~ s/[\s]+$//;
						# if a ) comes before a ) this is a list
						if($spaceCount < 5 && $line=~/^(.+\w)\s\s+(.*)/)
						{
							my $setting = $1;
							my $description = $2;
							$tex .= "\n\t\t\\item{\\emph{$setting}} $description ";
							$itemCount++;
						}
						else
						{
							if($itemCount==0 && length($line)>0)
							{
								$tex .= "\n\t\t\\item{\\emph{$line}} ";							
							}
							else
							{
								$tex .= "$line ";
							}
						}
					}
					$tex .= "\n\t\\end{$listType}\n";
				}
				elsif($secCount==3) # user interface
				{
					# build up a descriptive list
					# split by lines
					my @lines = split(/\n/);
					my $list = 0;
					my $listType = "description";
					$tex .= "\n\t\\begin{$listType}";
					my $itemCount=0;
					my @listItems = ();
					my @listDescriptions = ();
					foreach my $line (@lines)
					{
						# strip all leading and trailing spaces
						my $lineLength = length($line);
						$line =~ s/^[\s]+//;
						my $spaceCount = $lineLength - length($line);
						$line =~ s/[\s]+$//;
						if(($spaceCount < 3) && length($line)>0)
						{
							if($line=~/(.*) - (.*)/)
							{
								#print "push 1 ";
								my $setting = $1;
								my $description = $2;
								push(@listItems, $setting);
								push(@listDescriptions, $description."\n");
							}
							elsif($line=~/(.*):(.*)/)
							{
								#print "push 2 ";
								my $setting = $1;
								my $description = $2;
								push(@listItems, $setting);
								push(@listDescriptions, $description."\n");
							}
							else
							{
								#print "push 3 ";
								push(@listItems, $line);
								push(@listDescriptions, "");
							}
							$itemCount++;
						}
						elsif(length($line)>0)
						{
							if($itemCount==0)
							{
								#print "push 4 ";
								push(@listItems, $line);
								push(@listDescriptions, "");
								$itemCount++;
							}
							else
							{
								#print "push 5 ";
								my $desc = pop(@listDescriptions);
								$desc .= "$line\n";
								push(@listDescriptions, $desc);
								
							}
						}
					}
					for(my $i=0; $i<=$#listItems;$i++)
					{
						$tex .= "\n\t\t\\item{\\emph{$listItems[$i]}}";
						if(length($listDescriptions[$i])>0)
						{
							$tex.="\\\\";
							# check for inner lists...
							# split by lines
							my @lines = split(/\n/, $listDescriptions[$i]);
							my $list = 0;
							my $openBracket=0;
							my $listType2 = "";
							foreach my $line (@lines)
							{
								# strip all leading and trailing spaces
								$line =~ s/^[\s]+//;
								$line =~ s/[\s]+$//;
								# if a ) comes before a ) this is a list
								if($line=~/^([^(]+)\)(.*)/)
								{
									my $num = $1;
									my $rest = $2;

									if($openBracket)
									{
										$openBracket=0;
										$tex .= "$line";								
										if(!$list){ $tex .= "\n"; }
									}
									else
									{
										if($list == 0) 
										{
											#$tex .= "BEGIN LIST\n";
											# work out what type of list this is
											$num = ($num =~ /[\s]+(.*)/);
											if (length($num)==1 || $num=~/^\d+$/)
											{
												# just make it a normal enumerate
												$listType2 = "itemize";
												$tex .= "\n\\begin{$listType2}\n";
												$tex .= "\n\t\\item\n";
												$tex .= "\t\t$rest ";
											}
											else
											{
												# just make it a normal enumerate
												$listType2 = "itemize";
												$tex .= "\n\\begin{$listType2}\n";
												$tex .= "\n\t\\item\n";
												$tex .= "\t\t$rest ";										
											}
										}
										else 
										{
											$tex .= "\n\t\\item\n";
											$tex .= "\t\t$rest ";
											#$tex .= "CONTINUE LIST\n";
										}	
										$list = 1;
									}
								}
								elsif ($line=~/^[\s]*$/) 
								{
									if($list == 1)
									{
										$list = 2;
									}
									$tex .= "$line";
								}
								else
								{
									# really have found a normal line
									if($list == 2)
									{
										#$tex .= "END LIST\n";
										$tex .= "\n\\end{$listType2}\n";
										$list = 0;
									}
									$tex .= "$line";
									if(!$list){$tex.="\n";}
								}
								if($line=~/\([^)]+$/)
								{
									$openBracket=1;
								}

							}
							if($list)
							{
								$tex .= "\n\\end{$listType2}\n";
								$list = 0;
								#$tex .= "END LIST\n";						
							}							
							
							
							#$tex .= "\\\\$listDescriptions[$i]";
						}
					}
					$tex .= "\n\t\\end{$listType}\n";
				}									
				$secCount++;
			}	
		}
		$count++;
	}
	open OUT, ">$outf";
	print OUT $tex;
	print $outf . "\n";
	close OUT;

}