metalrules
Feb 20th, 2007, 04:49 PM
I'd just like to leave a comment about my experience using the Spring-LDAP APIs. I've recently needed to provide a set of extracts from an LDAP database, in limited time (as always..). I only have basic knowledge of LDAP, and haven't used Spring before.
I have however used the 'traditional' LDAP APIs previously, and there is just no comparison. The Spring API is levels of magnitude better. At least for what I did anyway. I needed a fraction of the code, and the code I did need to write was much cleaner.
I do have a beef with the Reference Documentation however. I suspect there are a many, many people around who are not LDAP experts (or Spring, yet), but need to get a basic LDAP program up and running. The Spring-LDAP reference document could of been written with those people in mind. For instance, give a COMPLETE example of using the Spring LDAP API as the first example (if possible not using inner classes), including any related XML. It's just not true what's been written : "we get exactly the same functionality with the following code:" .. the extra, required XML is belatedly put at the bottom of the page ! Sure, this is a small point, but when you are trying to get something up and running, you don't want to be sifting thru different sections of a document trying to put it all together.
Also, make sure all example code compiles. If I recall, Example 2.2 didn't compile, I think I needed to change it.
And what a pity you don't explicitly discuss the handling of repeating attributes, in effect making the initial example using "AttributesMapper" irrelevant in anything but the simplest LDAP implementation.
Personally, I think it's a bit lazy to give a reference to an internet article in formal documentation, to explain the software. In this case the problem is compounded because the article mentioned has the same faults as here (particularly, assumed knowledge of Spring).
Most programmers using open source software are well aware the documentation is usually patchy. Like Spring, I'm sure much of it is written by very clever people, to whom documentation is a bit of an annoying afterthought. In the case of Spring however, I'm being particularly critical because it's backed up by a professional company in Interface21. Also, the software is so dammed good it would be such a pity to see it sink because of poor doco...
I have however used the 'traditional' LDAP APIs previously, and there is just no comparison. The Spring API is levels of magnitude better. At least for what I did anyway. I needed a fraction of the code, and the code I did need to write was much cleaner.
I do have a beef with the Reference Documentation however. I suspect there are a many, many people around who are not LDAP experts (or Spring, yet), but need to get a basic LDAP program up and running. The Spring-LDAP reference document could of been written with those people in mind. For instance, give a COMPLETE example of using the Spring LDAP API as the first example (if possible not using inner classes), including any related XML. It's just not true what's been written : "we get exactly the same functionality with the following code:" .. the extra, required XML is belatedly put at the bottom of the page ! Sure, this is a small point, but when you are trying to get something up and running, you don't want to be sifting thru different sections of a document trying to put it all together.
Also, make sure all example code compiles. If I recall, Example 2.2 didn't compile, I think I needed to change it.
And what a pity you don't explicitly discuss the handling of repeating attributes, in effect making the initial example using "AttributesMapper" irrelevant in anything but the simplest LDAP implementation.
Personally, I think it's a bit lazy to give a reference to an internet article in formal documentation, to explain the software. In this case the problem is compounded because the article mentioned has the same faults as here (particularly, assumed knowledge of Spring).
Most programmers using open source software are well aware the documentation is usually patchy. Like Spring, I'm sure much of it is written by very clever people, to whom documentation is a bit of an annoying afterthought. In the case of Spring however, I'm being particularly critical because it's backed up by a professional company in Interface21. Also, the software is so dammed good it would be such a pity to see it sink because of poor doco...