Report comparison; Ignore page heading

[TheShadow]

That works.

Thank you for your help

[Aaron]

Hello,

The file input into Beyond Compare is a single .txt that is a single, long scroll of multiple pages, correct? I was hoping to get a larger example in order to see what the other lines in the file are, as that would help determine the type of definition that can match on the PageHeader but not match on other data.

Given the above screenshot, I would use something like, a Lines grammar defined as:
^.*Date: \d\d/\d\d/\d\d.*Page:.*$
and next 3 lines
Regular Expression Enabled

Given the complexity of Regular Expressions and the wide variety of text that might appear in your file, it is difficult to create an exact solution without access to more information. If you can send in a full example file, with line numbers of the Page Headers you need to ignore, that would help. You can attach here to the forum thread, or email [email protected] and please include a link back to this forum thread in the email for our reference.

[TheShadow]

Per the attached screen shot, each page of the report contains a 'heading' that consists of four lines. The first line will be different on each page as it contains the date and time.

RegEx such as "[0-9][0-9]" will work for the for "Date: 01/30/15 16:39." However, the expression is also cause issues within the body of the report.

Again, I am seeking a solution that will work on a page heading and not cause other issues / conflicts with the data provide within the body of the report

[Aaron]

Hello,

A page number or timestamp could be assigned a RegEx mask as part of the Basic grammar definition. Did you get a chance to review the above video link? To ignore a page header, we would need to identify it. If it occurs at the beginning of the file we could use the Lines grammar, but if it occurs multiple times at random points throughout the file, we need to find an expression that finds that text and matches on it. Once defined, it can be marked as Unimportant.

In the above screenshot you provided, is that whole bit of text the header, or is just the first line the header?

[TheShadow]

No. Each page heading has page number and time and date stamp.

I was lead to believe BC was capable of ignoring page headings within a report. This isn't true?

[Aaron]

Is the page header exactly that specific text every time? You could create a Basic grammar for the first line of the header that is a copy/paste of that line exactly, and another Basic grammar that is a copy/paste of line2 of the header, etc. Then these lines, regardless of where they appear, would be unimportant.

[TheShadow]

here's what it the page heading looks like:

[TheShadow]

Aaron, unlike a data file, I have a report that repeats the page heading every sixty or so lines. The issue I have is how to define each page

BTW, I've watch the video you suggested previously. Very helpful when comparing text files

[Aaron]

Hello,

We have a KB article and video that should help, here:
http://www.scootersoftware.com/suppo..._unimportantv3

One of our grammar types is Lines, which can start at line 1, and go for x lines. Or you can define something more specific using Delimited or a RegEx.

The key step is, once the grammar is defined, it also needs to be marked as Unimportant. By default, the defined grammar is still Important until it is changed to Unimportant in the Session Settings. Then, Ignore Unimportant Differences should be enabled in the toolbar.

The default Session Settings (editable on the Home screen) control the importance of a "Rules-based scan" in the Folder Compare, when comparing multiple files.

[TheShadow]

I'm comparing two reports and wish to ignore the first three or four lines of each page.

I'd appreciate any direction. Thanks