Meaglyn's perl 2DA diff utility

Meaglyn's 2DA diffing tool.

This little utility is a perl script to manage 2da files.

It can renumber and reformat single 2das, add/remove columns, extract specified lines and/or columns.

It can show the differences between two 2da files. 

I find it extremely useful for dealing with 2das from different haks that have modified the same lines (baseitems for example). I thought others might find it useful.

It requires the Number/Range perl package which can be found
at http://search.cpan.org/~larrysh/Number-Range-0.10/lib/Number/Range.pm

If /usr/bin/perl is not where your perl lives you will want to edit to
first line of the script. It may be possible to make this work on windows. Never tried.

There are currently 2 basic modes of operation.

1) Fix/validate/renumber a single 2da.   The tool will report on the status of
the given file. It will check for line numbers being off and report extra or missing columns.
The headers line is assumed to have the correct entries.

The -v verbose flag will report more information (what is in the missing or extra column etc).
Use -f to fix any errors. Missing columns in a row are filled with ****. Extras are removed from
the end obviously. There is no way to remove columns in the middle.

use -p to cause the new file to be printed out. If -o is not specified then stdout is used. All
the information messages go to stderr. The file specified by -o must not be the same as the input file.

The reprinted file will have all the columns lined up based on the size of the largest entry in each column and
a small pad for legibility.

Report on errors in spells.2da
./diff2da.pl -ar spells.2da

Fixup spells.2da and put the results in t.2da (these are the same)
./diff2da.pl  -pfv spells.2da > t.2da  
./diff2da.pl  -pfv -o t.2da spells.2da  

Printing can also be in semi-colon separated format (use -S). Input file can be in either format (whitespace or semicolon).


2) Show differences between 2 2da files.  (use -u at least)

Because it operates only on the data and not the spaces differences in formatting are irrelevant.

Basic diff of 2 files will report differences in number of lines and columns and for common lines a statement that they differ.
./diff2da.pl -u test2.2da test3.2da
Diffing test2.2da and test3.2da
test3.2da has 1 more row(s) than test2.2da
Row 0: 1 column(s) differ
Row 2: num cols (2|3)  1 column(s) differ
Row 3: Only in test3.2da


verbose diff of 2 files will report differences in number of lines and columns and for common lines a full view of the differences.
./diff2da.pl -uv test2.2da test3.2da
Diffing test2.2da and test3.2da
test3.2da has 1 more row(s) than test2.2da
Row 0: 1 column(s) differ  Value(1|9)
Row 2: num cols (2|3)  1 column(s) differ  Label(third|"the third")
Row 3: Only in test3.2da: "21 fourth ****"


In either mode the lines and columns used can be selected or ignored. So a subset of a 2da can be printed (to extract some lines).
All the lines with only the label can be extracted (useful to create an extension 2da file). Specific lines and columns may be diffed etc.

Renumbering 2da files can be set to start at a given number. This is useful to format and validate an extracted set of 2da lines.

Here is the full usage message:

Usage : ./diff2da.pl [-afhHpqrSuv] [-n start_line] [-o outfile] [-lL <list>] [ -c <list>] file1 [file2]
         a: report all errors when validating
         f: try to fix row and column errors, implies -ar
         h: display this message
         H: suppress printout of the header lines (including column labels)
         p: print the resulting file (defaults to normal tabbed output)
         q: suppress any extra informational messages
         r: report file1's row number and column errors
         S: print in semicolon separated format (useless without -p)
         u: diff file1 and file2. Files should be valid 2da files with no errors.
         v: verbose output. Longer error messages and column differences
         n <start_line>: use given starting line number for validation. Useful when operating on a subset of a 2da
         o <outfile>: Write printed 2da to outfile. Must not be the same as file1 or file2
         l <list>: operate on only the listed rows. E.g. -l 1,3,100..110  
         L <list>: do not operate on the listed rows. E.g. -L 1,3,100..110  
         Note: l and L are mutually exclusive
         c <list>: operate on only the listed columns. E.g. -c Label,Value  
         Note: column headers in <list>  case INsensitive (Label == label == LABEL)
         Note: columns listed effect diffing and printout but not validation.


NOTE:  Any values with spaces MUST be double quoted.  There are a few entries in the CEP appeareances 2da file that are or were single quoted. Or had a double open and a single close. These should be fixed by hand before using.

NOTE2:Any apostrophes are removed. This is due to a limitation in the string parsing library. It get's confused and treats them as quotes. This effects a small number of labels and names. At some point I may have time and energy to address this.

You can find the latest version of this on github (https://github.com/meaglyn/nwn_utils)