Good algorithm for reading tokens from a string

metalmario

Hi all,

i'm writing a custom code to export a CSV file of products in a database. Here is the idea:
1) A textfield in the web page. It will contain user's choices.
2) A list of allowed tags
3) Unknown separators

Assuming that my database contains only a product(Chair),if user writes :

       tag_stock|tag_productname|tag_category

the CSV file will contain :

2|Chair|Furniture

Obiviously, it should be done for every product in my database.
For example, a full CSV file should contain:

User defined order must be respected.

The algorithm i thought is something like this

$ops = array("product","chair","furniture") //contains the order of the tags read from the string

foreach(product of my catalog) {
for(i = 0 ; i< ops.lenght ; i++)
Read ops
Execute function contained in ops
}

So here are my questions
1) How can i read multiple tokens sequencely from a string?
2) Is there a way to invoke a function by reading the name from a variable?

Thank you very much. Hints or advices are well accepted 😃

Weedpacket

1) [man]strtok[/man]
2) Variable functions
The manual is a wonderful thing.

metalmario

Ok for answer 2), i already knew strtok() but it's not what i want to do. I do not know string delimiters that user will use...
I could call strtok() many times to check if every tag has been inserted(using the tag as delimiter), but it would be inefficient...
Also,It would be hard to understand the order of the tokens.
Is there a function, in a single pass, checks for the presence of multiple tokens?

Bonesnap

You can do this directly in MySQL using the LOAD DATA INFILE syntax. Let the database engine do the work for you.

NogDog

[man]preg_split/man can split a string into an array using a regexp for the field separator, typically using a character class:

$fields = preg_split('#[.,;:/|-]#', $text);

traq

metalmario;11028699 wrote:
For example, a full CSV file should contain:

2|Chair|Furniture|10|Hammer|Tools|10|Sickle|Tools|... and so on

It would be very beneficial to use a different delimiter between products than you use between details.

Is there any reason you don't want to define which delimiters must be used? It would make your task much easier and more reliable.

For example, it is very common to use commas as field delimiters and newlines as row delimiters. There's even a built-in function for it.

2,Chair,Furniture
10,Hammer,Tools
10,Sickle,Tools

(Side question: does the number {tag_stock} correspond with the {tag_category}? For example, you're using "10" as the tag_stock in both fields where "Tools" is the tag_category. Does "10" mean "Tools"? If so, why are you forcing the user to enter both (and possibly mess them up)?)

metalmario;11028699 wrote:
User defined order must be respected.

How is the user defining order? Just the order they're written in?

metalmario;11028699 wrote:

The algorithm i thought is something like this

$ops = array("product","chair","furniture") //contains the order of the tags read from the string
foreach(product of my catalog) {
     for(i = 0 ; i< ops.lenght ; i++) 
               Read ops[i]
               Execute function contained in ops[i]
}

Your $ops array contains strings, not functions. Do you have user functions with these names?

metalmario;11028699 wrote:
1) How can i read multiple tokens sequencely from a string?

metalmario;11028725 wrote:
I could call strtok() many times to check if every tag has been inserted(using the tag as delimiter), but it would be inefficient... Is there a function, in a single pass, checks for the presence of multiple tokens?

You're contradicting yourself. Do you want to do this in sequence or all at once?

metalmario

I finally managed to write a working code.