Re: Generalizing EVR comparison precedence, preliminaries

[Jeff Johnson]

On Jan 2, 2009, at 4:46 AM, Ralf S. Engelschall wrote:

On Thu, Jan 01, 2009, Jeff Johnson wrote:

On Jan 1, 2009, at 1:04 PM, Ralf S. Engelschall wrote:

A little help vetting the RE's please ;-) My eyes are already tired

gouging.

from writing the toy script, RE's are more effective than XML at eye

I propose the following three entries in "macros"...

  # STEP 1: Match the string and capture regex parts
  #                      1          2           3             4
  #                      X     ":"  X        "-"X          ":"X
  %evr_tuple_match  ^(?:([^:-]+):)?([^:-]+)(?:-([^:-]+))?(?::([^:-]
+))?$

A little more help please.

I'm using POSIX extended RE's.

Oh, sorry. I silemtly assumed PCRE! But ok, a non-PCRE version is also
possible: we just have to replace the non-capturing parenthesis (?:..)
with regular ones and then just ignore them in the selection phase:

with the original pattern and tested with
        rpm -Va --nofiles
But perhaps I screwed the assertion (or mire.c needs fixing) somehow.

It may be time to commit to PCRE as mandatory when building rpm,
and use PCRE patterns consistently. There's a few compatibility issues
on older linux distros, not hard to fix.

There's also the pattern for extracting href's from HTML that you kindly
gave me a year ago to. Both the html and now the EVRD parsers are
rather too important to be lost if/when rpm is built without PCRE.

Hmmm, the PCRE posix emulation seems to *ahem* extend POSIX syntax slightly then. I do have an assertion in place on the parsed results using RRPMMIRE_REGEX

   # STEP 1: Match the string and capture regex parts
   #                      2          3           5             7
   #                      X     ":"  X        "-"X          ":"X
   %evr_tuple_match  ^(([^:-]+):)?([^:-]+)(-([^:-]+))?(:([^:-]+))?$

   # STEP 2: Assemble <E,V,R,D> tuple from regex parts

   #                 X:X-X:X  X:X-X    X:X      X:       X-X      X
   %evr_tuple_select 2357     237_     23__     2___     _23_     __5_


   %evr_tuple_order  EVRD

# <E,V,R,D> case 1: case 2: case 3: case 4: case 5: case 6: # STEP 3: Configure the comparison order of the <E,V,R,D> tuple elements

73 de Jeff

Thanks! I will add both to macros.in for now (and to help me read PCRE patterns).

This should do the trick with regular POSIX regular expressions, too.

                                      Ralf S. Engelschall
                                      r...@engelschall.com
                                      www.engelschall.com

______________________________________________________________________
RPM Package Manager                                    http://rpm5.org
Developer Communication List                        rpm-devel@rpm5.org

smime.p7s
Description: S/MIME cryptographic signature