[pcre-dev] [Bug 1916] New: Advanced data processing for an alternation with a lot of key words

[admin]

https://bugs.exim.org/show_bug.cgi?id=1916

            Bug ID: 1916
           Summary: Advanced data processing for an alternation with a lot
                    of key words
           Product: PCRE
           Version: 10.22 (PCRE2)
          Hardware: x86
                OS: Linux
            Status: NEW
          Severity: bug
          Priority: medium
         Component: Code
          Assignee: p...@hermes.cam.ac.uk
          Reporter: elfr...@users.sourceforge.net
                CC: pcre-dev@exim.org

I tried a bit of static source code analysis out for an other software project.
One result of this approach was that an alternation was automatically generated
by a script. This “list” contained 2911 function names then.

* I converted these data to a test file.

* I built the current software from the subversion trunk with the configuration
parameter “PCRE2_LINK_SIZE=3” on my openSUSE Tumbleweed system successfully.


elfring@Sonne:~/Projekte/Bau/PCRE> ./pcre2test
/home/elfring/Projekte/Bau/NetBSD/Coccinelle/deletions1/pcretest2-NetBSD7.txt
PCRE2 version 10.23-RC1 2016-08-01
/^(?x)
(?:
   (?:kz?|slob_)free
|
   (?:
       zs_get_chan_addr
    |  zs_find_prom
    |  ZlibFree
…
    |  aarch64_prev_real_insn
    |  aarch64_expand_epilogue
   )
)$/I,memory
Failed: error 186 at offset 68720: regular expression is too complicated
acl_free
ldap_close
Xstrdup


It seems that software improvement possibilities were discussed for a similar
text search for the issue “Faster keyword matching” from 2010-10-01.
https://bugs.exim.org/show_bug.cgi?id=1027

Can further technology help to improve the shown situation a bit more nowadays?

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-- 
## List details at https://lists.exim.org/mailman/listinfo/pcre-dev