Parsing text

I'm just learning php and need to parse some large text files. I was going to use Perl for this, but I read somewhere that "php is the new Perl"... I'm guessing that's an overstatement, but it got me to thinking that I may as well do this (if possible) in php so I can learn from it.

The text file is from a database that I need to replicate. I don't have admin access to it, so I'm using print-to-text-file output of a report. The text file looks something like this...

Code: Select all

 
ID_101027   Roger
ID_101028   Michael
ID_101029   Julie
- - more records - -
- - garbage (report header/footer) lines - -
ID_101027   Smith
ID_101028   Robertson
ID_101029   Williams
- - more records - -
- - garbage (report header/footer) lines - -
ID_101027   1010 Mulberry Drive
ID_101028   772 Waverly Lane
ID_101029   3601 Hamilton Way
- - more records - -
- - garbage (report header/footer) lines - -
(you get the idea)

What I need to do is...
1. Find the (first instance of the first) identifier
2. Write it (and the associated field value) to the first two elements of a multidimensional array
3. Find the next line containing the identifier
4. Split the line into the identifier and the field value
5. Write the field value to an array element
6. When the identifier is not found, restart the loop from the next identifier
7. Use an insert query to add the values from the array to the new database.

Gotchas: Some of the fields (e.g. Notes) will be multiple line fields. These may be interrupted by the aforementioned garbage lines. (Seems like the way to get around this is to get rid of the garbage lines first.) Also, some fields may contain null values.

Questions:
1. Is php well-suited (or well enough) for doing line-by-line text parsing?
2. What functions should I be looking into in order to accomplish this?

Both Perl and PHP and others for that matter are suited to your purpose.

You need to be looking at what are known as Regular Expressions. Make a large pot of coffee because you are going to have to do some serious reading and understanding.

Thanks for the reply, ukgeoff!

I'm familiar with regular expresions and have used them before. The part that I find most frustrating is that they always seem to want to select much more than I intend. How greedy are they in php? Do they stop at line breaks in php? (And/or can you tell them how greedy to be/where to stop?)

Beyond regular expressions, I was wondering about what functions to use in order to split lines. Also, what would I use to (for example) select from line 77, column 12 to the end of line 80?

You can specify if their to be greedy or not.

I recommend a regex tool both as a training tool, it comes with great documentation, and as a build and test tool.

RegexBuddy: Learn, Create, Understand, Test, Use and Save Regular Expression

With regard to the second idea. You would probably have to read the bytes into a variable looking for the EOL markers. But I haven't really thought that one through.

About splitting the lines up, i think you could use php's explode to split up the lines and then foreach to go through the array value (would be each line in this case).

I didnt look into it, but i think im along the right lines here.

Is there any way to tell that the data for an ID is a first name, last name, address, etc? (maybe that's what the garbage does some how?) If not, are the first names always first, then the last names, and then the addresses?

Look at:
preg_match
preg_replace
explode
stristr
str_replace