Bug#762829: file: Pascal modules (ISO 10206) wrongly classified as text/x-ruby
Hi, I realise now I was not subscribed to the bug, so format/linkage of this reply might be a bit of a mess. Sorry! > To improve the situation without breaking detection of other text > file types, I need a fine selection of typical pascal programs. Not > necessarily a big number but a huge variety in style. The compiler test files might be useful here... FPC http://svn.freepascal.org/cgi-bin/viewvc.cgi/trunk/tests/ GPC https://github.com/hebisch/gpc/tree/master/p/test Regards, Peter
Bug#762829: file: Pascal modules (ISO 10206) wrongly classified as text/x-ruby
On 16/07/16 23:28, Christoph Biedl wrote: > tags 762829 confirmed upstream help > thanks > > It's been a while but it hasn't been forgotten ... > > Peter wrote... > >> Some Pascal source files are classified as Ruby. >> Maybe file is confused by the keywords 'module', 'import' and export, >> which are valid Pascal extension keywords in ISO 10206? > > Correct, the "module" keyword in the first line of your example > triggers one of the tests for ruby (magic/Magdir/ruby:25). > > However I'm reluctant to change this as it easily might create > false-positives. If there's is a distinct rule to tell Ruby and > Pascal apart, I'll happily forward it to upstream. > > Christoph > Hi Christoph, Thanks for looking at this. I'm afraid I know very little about Ruby. Pascal files often have semicolons at the end of some lines, and the final line is often "end." The Ruby code I have looked at on http://rosettacode.org never has semicolons at the end of lines and the final "end" does not have a period after it. This suggests two tests that could separate Ruby & Pascal. Regards, Peter
Bug#762829: file: Pascal modules (ISO 10206) wrongly classified as text/x-ruby
Christoph Biedl wrote... > tags 762829 help Refining this: The detection of Pascal programs in file is a mess. To improve the situation without breaking detection of other text file types, I need a fine selection of typical pascal programs. Not necessarily a big number but a huge variety in style. The second is the problem, else I'd just pick some random packages from Debian written in Pascal. Christoph signature.asc Description: Digital signature
Bug#762829: file: Pascal modules (ISO 10206) wrongly classified as text/x-ruby
tags 762829 confirmed upstream help thanks It's been a while but it hasn't been forgotten ... Peter wrote... > Some Pascal source files are classified as Ruby. > Maybe file is confused by the keywords 'module', 'import' and export, > which are valid Pascal extension keywords in ISO 10206? Correct, the "module" keyword in the first line of your example triggers one of the tests for ruby (magic/Magdir/ruby:25). However I'm reluctant to change this as it easily might create false-positives. If there's is a distinct rule to tell Ruby and Pascal apart, I'll happily forward it to upstream. Christoph signature.asc Description: Digital signature
Bug#762829: file: Pascal modules (ISO 10206) wrongly classified as text/x-ruby
Package: file
Version: 1:5.19-2
Severity: normal
Dear Maintainer,
Some Pascal source files are classified as Ruby.
Maybe file is confused by the keywords 'module', 'import' and export,
which are valid Pascal extension keywords in ISO 10206?
A short example is attached.
-- System Information:
Debian Release: jessie/sid
APT prefers testing
APT policy: (500, 'testing')
Architecture: amd64 (x86_64)
Foreign Architectures: i386
Kernel: Linux 3.14-2-amd64 (SMP w/2 CPU cores)
Locale: LANG=aa_DJ.utf8, LC_CTYPE=aa_DJ.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Versions of packages file depends on:
ii libc6 2.19-11
ii libmagic1 1:5.19-2
ii zlib1g 1:1.2.8.dfsg-2
file recommends no packages.
file suggests no packages.
-- no debconf information
module Out4m;
(* FLAG --extended-pascal *)
export emil23m = (OK);
procedure OK;
end;
import StandardOutput;
procedure OK;
begin
WriteLn ('OK')
end;
end.
