Legacy What is the purpose of the "-a" trailing on urls with title that ends in a number?

[bzcomputers]

What is the purpose of the "-a" trailing on urls with title that ends in a number?

What is the purpose of the "-a" trailing on urls with titles that end in a number?

If it is not essential can you add a setting so that we can disable this feature?

I have quite a few urls that end in a date and everyone is appended with the "-a" which seems counter intuitive to place the "a" at the end of the url since it is usually the first stopword everyone tries to remove and in this case we are actually adding it back in.

 

[Fillip H.]

It's to avoid confusing the regular expressions that try to map URLs back to their original components.


Assume you have two URL formats:
f[forum_id]/[thread_title]-{thread_id].html
f[forum_id]/[thread_title]-{thread_id]-[page].html

To a regular expression, these look like:
f[number]/[any amount of any characters]-[number].html
f[number]/[any amount of any characters]-[number]-[number].html

Let's assume we post a thread called "Best Of 2013" (thread ID 8) in forum ID 4:
f4/best-of-2013-8.html
f4/best-of-2013-8-2.html

If you notice, the second URL format (reserved for pages) matches the first "actual URL" (not a paginated thread). This would lead to the mod attempting to resolve thread ID 2013, page 8.

We cannot rely on users not creating the "wrong" URL formats. Users could avoid this by prefixing the "[page]" format with "page" (so it reads "page[page]", however users may choose not to do so.

Going back to the thread called "Best Of 2013" (thread ID 8) in forum ID 4:
f4/best-of-2013-a-8.html
f4/best-of-2013-a-8-2.html

You will now see that the regular expressions will only ever match the intended URLs.


Making this "filler" customisable would introduce additional confusion, as we would have to validate that the "filler" did not end in a number.

 

[bzcomputers]

Thanks for taking the time to explain - it makes perfect sense.