No announcement yet.

/qc text compare unimportant differences

  • Filter
  • Time
  • Show
Clear All
new posts

  • /qc text compare unimportant differences

    We have v3.2.2. I have built out a website to show results of the compares. I run BC several times, once with filters to ignore unimportant differences and again without filters. This builds ther HTML files for each.

    Next I run BC using /qc with 2 text files trapping the errorlevel, and using the errorlevel, I can create a companion file (if html compare files is ABC.csv.html then the comparison file would be ABC.csv.html.match, or .nomatch, or .diffokay) depending on errorlevel set by /qc compare. Then on website when I'm formatting the results, I use these files to color code the links, so you can quickly see matches, mismatches, and differences. I also support missing file on right or left, but that part works fine.

    My /qc compare returns errorlevel=12 which I treat as similar - unimportant differences. However it only seems to catch its' own unimporttant differences like extra blank lines or upper/lower case differences, not my unimportant string differences even if they are session defaults (a few literal strins, dates, percentages).

    If there a way in v3.3.2 for my session defined unimportant differences to be used on a file by file basis that I can trap and detect in batch (Windows)?

    Is this feature available in more current releases?

  • #2
    Session Settings are bound per Windows User account (AppData). If you setup and then use a Portable Install, does this help propagate your settings so the defaults are used regardless of the user/script scenario running BC3?

    All BC3.x updates are free for 3.x users and contain a wide variety of bug fixes and enhancements. If you can update, I would suggest doing so to be running the latest release.
    Aaron P Scooter Software


    • #3
      Our shop is a tad bit anal regarding software installs. Policy is we do not install anything, software, by version gets tested and approved for use (plays well with the other apps in the pool), then packaged such that the push/install is done remotely. Free or not, any different versions incurs the same process from getting approval thru packaging.

      I do not believe I can create a portable BC install. I will try.

      If my issue was fixed in a later release I would go thru the process of requesting that version upgrade, getting it approved, and getting it packaged.

      Just to comfirm, a /qc test of comparing files, does not allow for pointing to specific session, but if the logged on user (userid job is running under as owner), and doesn't offer explicite ignore dirrerences option, but would use session default values if they existed?


      • #4
        Correct, the session defaults should be used with /qc, as long as the same Windows User is used (so that their configured defaults are used). You shouldn't need to update to get this working. A Portable Install helps just because the .xml setting files are then shared regardless of the user account that runs the application, stored in the same directory as the bcompare.exe.
        Aaron P Scooter Software